texlive[73695] Build/source/libs/luajit: Updated LuaJIT to LuaJIT
commits+lscarso at tug.org
commits+lscarso at tug.org
Sun Feb 2 18:39:28 CET 2025
Revision: 73695
https://tug.org/svn/texlive?view=revision&revision=73695
Author: lscarso
Date: 2025-02-02 18:39:28 +0100 (Sun, 02 Feb 2025)
Log Message:
-----------
Updated LuaJIT to LuaJIT 2.1.1736781742 .
Modified Paths:
--------------
trunk/Build/source/libs/luajit/LuaJIT-src/COPYRIGHT
trunk/Build/source/libs/luajit/LuaJIT-src/Makefile
trunk/Build/source/libs/luajit/LuaJIT-src/README
trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad-print.css
trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad.css
trunk/Build/source/libs/luajit/LuaJIT-src/doc/contact.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_c_api.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_api.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_semantics.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_tutorial.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_jit.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_profiler.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/extensions.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/install.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/luajit.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/running.html
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_proto.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.h
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.lua
trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dynasm.lua
trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.1
trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.pc
trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile
trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.dep
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_asm.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_fold.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_lib.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_libbc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_peobj.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genlibbc.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genminilua.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/minilua.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bc.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bcsave.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64be.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64el.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mipsel.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_ppc.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x64.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x86.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dump.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/p.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/v.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/zone.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/lauxlib.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_aux.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_base.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_bit.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_debug.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_ffi.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_init.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_io.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_jit.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_math.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_os.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_package.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_string.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_table.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_api.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_arch.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm64.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_mips.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_ppc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_x86.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcdump.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcread.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcwrite.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_def.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm64.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_mips.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_ppc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_x86.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_errmsg.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ff.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_frame.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ircall.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_iropt.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_jit.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_load.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_dce.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_fold.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_loop.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_mem.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_narrow.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_sink.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_split.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_profile.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_profile.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_record.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_record.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_snap.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_snap.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_state.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_state.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_str.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_str.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_strfmt.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_strfmt.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_strfmt_num.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_strscan.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_strscan.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_tab.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_tab.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target_arm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target_arm64.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target_mips.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target_ppc.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_target_x86.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_trace.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_trace.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_traceerr.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_udata.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_udata.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_vm.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_vmevent.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_vmevent.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_vmmath.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/ljamalg.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lua.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/luaconf.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/luajit.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lualib.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/msvcbuild.bat
trunk/Build/source/libs/luajit/LuaJIT-src/src/ps4build.bat
trunk/Build/source/libs/luajit/LuaJIT-src/src/psvitabuild.bat
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_arm.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_arm64.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_mips.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_mips64.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_ppc.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_x64.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/vm_x86.dasc
trunk/Build/source/libs/luajit/LuaJIT-src/src/xb1build.bat
trunk/Build/source/libs/luajit/LuaJIT-src/src/xedkbuild.bat
trunk/Build/source/libs/luajit/Makefile.am
trunk/Build/source/libs/luajit/Makefile.in
trunk/Build/source/libs/luajit/TLpatches/ChangeLog
trunk/Build/source/libs/luajit/configure
trunk/Build/source/libs/luajit/configure.ac
trunk/Build/source/libs/luajit/include/Makefile.am
trunk/Build/source/libs/luajit/include/Makefile.in
trunk/Build/source/libs/luajit/lbitlib.c
trunk/Build/source/libs/luajit/m4/lj-system.m4
trunk/Build/source/libs/luajit/native/Makefile.am
trunk/Build/source/libs/luajit/native/Makefile.in
trunk/Build/source/libs/luajit/native/configure
trunk/Build/source/libs/luajit/native/configure.ac
trunk/Build/source/libs/luajit/version.ac
Added Paths:
-----------
trunk/Build/source/libs/luajit/LuaJIT-src/.relver
trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_buffer.html
trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.std
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/.gitignore
trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genversion.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/.gitignore
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6el.lua
trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_buffer.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_assert.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_prng.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_prng.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_serialize.c
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_serialize.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/luajit_rolling.h
trunk/Build/source/libs/luajit/LuaJIT-src/src/nxbuild.bat
trunk/Build/source/libs/luajit/LuaJIT-src/src/ps5build.bat
trunk/Build/source/libs/luajit/TLpatches/patch-lauxlib
trunk/Build/source/libs/luajit/TLpatches/patch-lib_init
trunk/Build/source/libs/luajit/TLpatches/patch-lib_package
trunk/Build/source/libs/luajit/TLpatches/patch-lj_arch
trunk/Build/source/libs/luajit/TLpatches/patch-lj_def
trunk/Build/source/libs/luajit/TLpatches/patch-lj_load
trunk/Build/source/libs/luajit/TLpatches/patch-lj_pnrg
trunk/Build/source/libs/luajit/TLpatches/patch-lj_str
trunk/Build/source/libs/luajit/TLpatches/patch-lua
trunk/Build/source/libs/luajit/TLpatches/patch-lualib
Removed Paths:
-------------
trunk/Build/source/libs/luajit/LuaJIT-src/doc/changes.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/faq.html
trunk/Build/source/libs/luajit/LuaJIT-src/doc/status.html
trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_bcdef.h_
trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_ffdef.h_
trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_folddef.h_
trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_libdef.h_
trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_recdef.h_
trunk/Build/source/libs/luajit/LuaJIT-src/src/lj.supp
trunk/Build/source/libs/luajit/LuaJIT-src/src/luajit.h
trunk/Build/source/libs/luajit/TLpatches/patch-01
trunk/Build/source/libs/luajit/TLpatches/patch-02
trunk/Build/source/libs/luajit/TLpatches/patch-03
trunk/Build/source/libs/luajit/TLpatches/patch-04
trunk/Build/source/libs/luajit/TLpatches/patch-05
trunk/Build/source/libs/luajit/TLpatches/patch-06
trunk/Build/source/libs/luajit/TLpatches/patch-07
trunk/Build/source/libs/luajit/TLpatches/patch-08
trunk/Build/source/libs/luajit/TLpatches/patch-09
trunk/Build/source/libs/luajit/TLpatches/patch-10
trunk/Build/source/libs/luajit/TLpatches/patch-11
trunk/Build/source/libs/luajit/TLpatches/patch-12
trunk/Build/source/libs/luajit/TLpatches/patch-13
Added: trunk/Build/source/libs/luajit/LuaJIT-src/.relver
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/.relver (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/.relver 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1 @@
+1736781742
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/COPYRIGHT
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/COPYRIGHT 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/COPYRIGHT 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
===============================================================================
-LuaJIT -- a Just-In-Time Compiler for Lua. http://luajit.org/
+LuaJIT -- a Just-In-Time Compiler for Lua. https://luajit.org/
-Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+Copyright (C) 2005-2025 Mike Pall. All rights reserved.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
@@ -21,7 +21,7 @@
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
-[ MIT license: http://www.opensource.org/licenses/mit-license.php ]
+[ MIT license: https://www.opensource.org/licenses/mit-license.php ]
===============================================================================
[ LuaJIT includes code from Lua 5.1/5.2, which has this license statement: ]
@@ -51,6 +51,6 @@
This is a version (aka dlmalloc) of malloc/free/realloc written by
Doug Lea and released to the public domain, as explained at
-http://creativecommons.org/licenses/publicdomain
+https://creativecommons.org/licenses/publicdomain
===============================================================================
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/Makefile
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/Makefile 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/Makefile 2025-02-02 17:39:28 UTC (rev 73695)
@@ -10,16 +10,21 @@
# For MSVC, please follow the instructions given in src/msvcbuild.bat.
# For MinGW and Cygwin, cd to src and run make with the Makefile there.
#
-# Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+# Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
##############################################################################
MAJVER= 2
MINVER= 1
-RELVER= 0
-PREREL= -beta3
-VERSION= $(MAJVER).$(MINVER).$(RELVER)$(PREREL)
ABIVER= 5.1
+# LuaJIT uses rolling releases. The release version is based on the time of
+# the latest git commit. The 'git' command must be available during the build.
+RELVER= $(shell cat src/luajit_relver.txt 2>/dev/null || : )
+# Note: setting it with := doesn't work, since it will change during the build.
+
+MMVERSION= $(MAJVER).$(MINVER)
+VERSION= $(MMVERSION).$(RELVER)
+
##############################################################################
#
# Change the installation path as needed. This automatically adjusts
@@ -32,11 +37,13 @@
DPREFIX= $(DESTDIR)$(PREFIX)
INSTALL_BIN= $(DPREFIX)/bin
INSTALL_LIB= $(DPREFIX)/$(MULTILIB)
-INSTALL_SHARE= $(DPREFIX)/share
-INSTALL_INC= $(DPREFIX)/include/luajit-$(MAJVER).$(MINVER)
+INSTALL_SHARE_= $(PREFIX)/share
+INSTALL_SHARE= $(DESTDIR)$(INSTALL_SHARE_)
+INSTALL_DEFINC= $(DPREFIX)/include/luajit-$(MMVERSION)
+INSTALL_INC= $(INSTALL_DEFINC)
-INSTALL_LJLIBD= $(INSTALL_SHARE)/luajit-$(VERSION)
-INSTALL_JITLIB= $(INSTALL_LJLIBD)/jit
+export INSTALL_LJLIBD= $(INSTALL_SHARE_)/luajit-$(MMVERSION)
+INSTALL_JITLIB= $(DESTDIR)$(INSTALL_LJLIBD)/jit
INSTALL_LMODD= $(INSTALL_SHARE)/lua
INSTALL_LMOD= $(INSTALL_LMODD)/$(ABIVER)
INSTALL_CMODD= $(INSTALL_LIB)/lua
@@ -49,10 +56,10 @@
INSTALL_ANAME= libluajit-$(ABIVER).a
INSTALL_SOSHORT1= libluajit-$(ABIVER).so
INSTALL_SOSHORT2= libluajit-$(ABIVER).so.$(MAJVER)
-INSTALL_SONAME= $(INSTALL_SOSHORT2).$(MINVER).$(RELVER)
+INSTALL_SONAME= libluajit-$(ABIVER).so.$(VERSION)
INSTALL_DYLIBSHORT1= libluajit-$(ABIVER).dylib
INSTALL_DYLIBSHORT2= libluajit-$(ABIVER).$(MAJVER).dylib
-INSTALL_DYLIBNAME= libluajit-$(ABIVER).$(MAJVER).$(MINVER).$(RELVER).dylib
+INSTALL_DYLIBNAME= libluajit-$(ABIVER).$(VERSION).dylib
INSTALL_PCNAME= luajit.pc
INSTALL_STATIC= $(INSTALL_LIB)/$(INSTALL_ANAME)
@@ -65,7 +72,7 @@
INSTALL_DIRS= $(INSTALL_BIN) $(INSTALL_LIB) $(INSTALL_INC) $(INSTALL_MAN) \
$(INSTALL_PKGCONFIG) $(INSTALL_JITLIB) $(INSTALL_LMOD) $(INSTALL_CMOD)
-UNINSTALL_DIRS= $(INSTALL_JITLIB) $(INSTALL_LJLIBD) $(INSTALL_INC) \
+UNINSTALL_DIRS= $(INSTALL_JITLIB) $(DESTDIR)$(INSTALL_LJLIBD) $(INSTALL_INC) \
$(INSTALL_LMOD) $(INSTALL_LMODD) $(INSTALL_CMOD) $(INSTALL_CMODD)
RM= rm -f
@@ -75,9 +82,13 @@
INSTALL_X= install -m 0755
INSTALL_F= install -m 0644
UNINSTALL= $(RM)
-LDCONFIG= ldconfig -n
+LDCONFIG= ldconfig -n 2>/dev/null
SED_PC= sed -e "s|^prefix=.*|prefix=$(PREFIX)|" \
- -e "s|^multilib=.*|multilib=$(MULTILIB)|"
+ -e "s|^multilib=.*|multilib=$(MULTILIB)|" \
+ -e "s|^relver=.*|relver=$(RELVER)|"
+ifneq ($(INSTALL_DEFINC),$(INSTALL_INC))
+ SED_PC+= -e "s|^includedir=.*|includedir=$(INSTALL_INC)|"
+endif
FILE_T= luajit
FILE_A= libluajit.a
@@ -88,7 +99,9 @@
FILES_JITLIB= bc.lua bcsave.lua dump.lua p.lua v.lua zone.lua \
dis_x86.lua dis_x64.lua dis_arm.lua dis_arm64.lua \
dis_arm64be.lua dis_ppc.lua dis_mips.lua dis_mipsel.lua \
- dis_mips64.lua dis_mips64el.lua vmdef.lua
+ dis_mips64.lua dis_mips64el.lua \
+ dis_mips64r6.lua dis_mips64r6el.lua \
+ vmdef.lua
ifeq (,$(findstring Windows,$(OS)))
HOST_SYS:= $(shell uname -s)
@@ -102,6 +115,7 @@
INSTALL_SOSHORT1= $(INSTALL_DYLIBSHORT1)
INSTALL_SOSHORT2= $(INSTALL_DYLIBSHORT2)
LDCONFIG= :
+ SED_PC+= -e "s| -Wl,-E||"
endif
##############################################################################
@@ -109,9 +123,9 @@
INSTALL_DEP= src/luajit
default all $(INSTALL_DEP):
- @echo "==== Building LuaJIT $(VERSION) ===="
+ @echo "==== Building LuaJIT $(MMVERSION) ===="
$(MAKE) -C src
- @echo "==== Successfully built LuaJIT $(VERSION) ===="
+ @echo "==== Successfully built LuaJIT $(MMVERSION) ===="
install: $(INSTALL_DEP)
@echo "==== Installing LuaJIT $(VERSION) to $(PREFIX) ===="
@@ -121,7 +135,7 @@
$(RM) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2)
cd src && test -f $(FILE_SO) && \
$(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) && \
- $(LDCONFIG) $(INSTALL_LIB) && \
+ ( $(LDCONFIG) $(INSTALL_LIB) || : ) && \
$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT1) && \
$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT2) || :
cd etc && $(INSTALL_F) $(FILE_MAN) $(INSTALL_MAN)
@@ -130,18 +144,12 @@
$(RM) $(FILE_PC).tmp
cd src && $(INSTALL_F) $(FILES_INC) $(INSTALL_INC)
cd src/jit && $(INSTALL_F) $(FILES_JITLIB) $(INSTALL_JITLIB)
+ $(SYMLINK) $(INSTALL_TNAME) $(INSTALL_TSYM)
@echo "==== Successfully installed LuaJIT $(VERSION) to $(PREFIX) ===="
- @echo ""
- @echo "Note: the development releases deliberately do NOT install a symlink for luajit"
- @echo "You can do this now by running this command (with sudo):"
- @echo ""
- @echo " $(SYMLINK) $(INSTALL_TNAME) $(INSTALL_TSYM)"
- @echo ""
-
uninstall:
@echo "==== Uninstalling LuaJIT $(VERSION) from $(PREFIX) ===="
- $(UNINSTALL) $(INSTALL_T) $(INSTALL_STATIC) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2) $(INSTALL_MAN)/$(FILE_MAN) $(INSTALL_PC)
+ $(UNINSTALL) $(INSTALL_TSYM) $(INSTALL_T) $(INSTALL_STATIC) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2) $(INSTALL_MAN)/$(FILE_MAN) $(INSTALL_PC)
for file in $(FILES_JITLIB); do \
$(UNINSTALL) $(INSTALL_JITLIB)/$$file; \
done
@@ -155,8 +163,9 @@
##############################################################################
amalg:
- @echo "Building LuaJIT $(VERSION)"
+ @echo "==== Building LuaJIT $(MMVERSION) (amalgamation) ===="
$(MAKE) -C src amalg
+ @echo "==== Successfully built LuaJIT $(MMVERSION) (amalgamation) ===="
clean:
$(MAKE) -C src clean
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/README
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/README 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/README 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,11 +1,11 @@
-README for LuaJIT 2.1.0-beta3
------------------------------
+README for LuaJIT 2.1
+---------------------
LuaJIT is a Just-In-Time (JIT) compiler for the Lua programming language.
-Project Homepage: http://luajit.org/
+Project Homepage: https://luajit.org/
-LuaJIT is Copyright (C) 2005-2017 Mike Pall.
+LuaJIT is Copyright (C) 2005-2025 Mike Pall.
LuaJIT is free software, released under the MIT license.
See full Copyright Notice in the COPYRIGHT file or in luajit.h.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad-print.css
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad-print.css 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad-print.css 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,4 +1,4 @@
-/* Copyright (C) 2004-2017 Mike Pall.
+/* Copyright (C) 2004-2025 Mike Pall.
*
* You are welcome to use the general ideas of this design for your own sites.
* But please do not steal the stylesheet, the layout or the color scheme.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad.css
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad.css 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/bluequad.css 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,4 +1,4 @@
-/* Copyright (C) 2004-2017 Mike Pall.
+/* Copyright (C) 2004-2025 Mike Pall.
*
* You are welcome to use the general ideas of this design for your own sites.
* But please do not steal the stylesheet, the layout or the color scheme.
@@ -206,11 +206,9 @@
.ext {
color: #ff8000;
}
-.new {
- font-size: 6pt;
- vertical-align: middle;
- background: #ff8000;
- color: #ffffff;
+.note {
+ padding: 0.5em 1em;
+ border-left: 3px solid #bfcfff;
}
#site {
clear: both;
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/doc/changes.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/changes.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/changes.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,883 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
-<html>
-<head>
-<title>LuaJIT Change History</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
-<meta name="Language" content="en">
-<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
-<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
-<style type="text/css">
-div.major { max-width: 600px; padding: 1em; margin: 1em 0 1em 0; }
-</style>
-</head>
-<body>
-<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
-</div>
-<div id="head">
-<h1>LuaJIT Change History</h1>
-</div>
-<div id="nav">
-<ul><li>
-<a href="luajit.html">LuaJIT</a>
-<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
-</li><li>
-<a href="install.html">Installation</a>
-</li><li>
-<a href="running.html">Running</a>
-</li></ul>
-</li><li>
-<a href="extensions.html">Extensions</a>
-<ul><li>
-<a href="ext_ffi.html">FFI Library</a>
-<ul><li>
-<a href="ext_ffi_tutorial.html">FFI Tutorial</a>
-</li><li>
-<a href="ext_ffi_api.html">ffi.* API</a>
-</li><li>
-<a href="ext_ffi_semantics.html">FFI Semantics</a>
-</li></ul>
-</li><li>
-<a href="ext_jit.html">jit.* Library</a>
-</li><li>
-<a href="ext_c_api.html">Lua/C API</a>
-</li><li>
-<a href="ext_profiler.html">Profiler</a>
-</li></ul>
-</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a class="current" href="changes.html">Changes</a>
-</li></ul>
-</li><li>
-<a href="faq.html">FAQ</a>
-</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
-</li></ul>
-</div>
-<div id="main">
-<p>
-This is a list of changes between the released versions of LuaJIT.<br>
-The current <span style="color: #0000c0;">stable version</span> is <strong>LuaJIT 2.0.5</strong>.<br>
-</p>
-<p>
-Please check the
-<a href="http://luajit.org/changes.html"><span class="ext">»</span> Online Change History</a>
-to see whether newer versions are available.
-</p>
-
-<div class="major" style="background: #d0d0ff;">
-<h2 id="LuaJIT-2.1.0-beta3">LuaJIT 2.1.0-beta3 — 2017-05-01</h2>
-<ul>
-<li>Rewrite memory block allocator.</li>
-<li>Add various extension from Lua 5.2/5.3.</li>
-<li>Remove old Lua 5.0 compatibility defines.</li>
-<li>Set arg table before evaluating <tt>LUA_INIT</tt> and <tt>-e</tt> chunks.</li>
-<li>Fix FOLD rules for <tt>math.abs()</tt> and FP negation.</li>
-<li>Fix soft-float <tt>math.abs()</tt> and negation.</li>
-<li>Fix formatting of some small denormals at low precision.</li>
-<li>LJ_GC64: Add JIT compiler support.</li>
-<li>x64/LJ_GC64: Add JIT compiler backend.</li>
-<li>x86/x64: Generate BMI2 shifts and rotates, if available.</li>
-<li>Windows/x86: Add full exception interoperability.</li>
-<li>ARM64: Add big-endian support.</li>
-<li>ARM64: Add JIT compiler backend.</li>
-<li>MIPS: Fix <tt>TSETR</tt> barrier.</li>
-<li>MIPS: Support MIPS16 interlinking.</li>
-<li>MIPS soft-float: Fix code generation for <tt>HREF</tt>.</li>
-<li>MIPS64: Add MIPS64 hard-float JIT compiler backend.</li>
-<li>MIPS64: Add MIPS64 hard-float/soft-float support to interpreter.</li>
-<li>FFI: Compile bitfield loads/stores.</li>
-<li>Various fixes common with the 2.0 branch.</li>
-</ul>
-
-<h2 id="LuaJIT-2.1.0-beta2">LuaJIT 2.1.0-beta2 — 2016-03-03</h2>
-<ul>
-<li>Enable trace stitching.</li>
-<li>Use internal implementation for converting FP numbers to strings.</li>
-<li>Parse Unicode escape <tt>'\u{XX...}'</tt> in string literals.</li>
-<li>Add MIPS soft-float support.</li>
-<li>Switch MIPS port to dual-number mode.</li>
-<li>x86/x64: Add support for AES-NI, AVX and AVX2 to DynASM.</li>
-<li>FFI: Add <tt>ssize_t</tt> declaration.</li>
-<li>FFI: Parse <tt>#line NN</tt> and <tt>#NN</tt>.</li>
-<li>Various minor fixes.</li>
-</ul>
-
-<h2 id="LuaJIT-2.1.0-beta1">LuaJIT 2.1.0-beta1 — 2015-08-25</h2>
-<p>
-This is a brief summary of the major changes in LuaJIT 2.1 compared to 2.0.
-Please take a look at the commit history for more details.
-</p>
-<ul>
-<li>Changes to the VM core:
-<ul>
-<li>Add low-overhead profiler (<tt>-jp</tt>).</li>
-<li>Add <tt>LJ_GC64</tt> mode: 64 bit GC object references (really: 47 bit). Interpreter-only for now.</li>
-<li>Add <tt>LJ_FR2</tt> mode: Two-slot frame info. Required by <tt>LJ_GC64</tt> mode.</li>
-<li>Add <tt>table.new()</tt> and <tt>table.clear()</tt>.</li>
-<li>Parse binary number literals (<tt>0bxxx</tt>).</li>
-</ul></li>
-<li>Improvements to the JIT compiler:
-<ul>
-<li>Add trace stitching (disabled for now).</li>
-<li>Compile various builtins: <tt>string.char()</tt>, <tt>string.reverse()</tt>, <tt>string.lower()</tt>, <tt>string.upper()</tt>, <tt>string.rep()</tt>, <tt>string.format()</tt>, <tt>table.concat()</tt>, <tt>bit.tohex()</tt>, <tt>getfenv(0)</tt>, <tt>debug.getmetatable()</tt>.</li>
-<li>Compile <tt>string.find()</tt> for fixed string searches (no patterns).</li>
-<li>Compile <tt>BC_TSETM</tt>, e.g. <tt>{1,2,3,f()}</tt>.</li>
-<li>Compile string concatenations (<tt>BC_CAT</tt>).</li>
-<li>Compile <tt>__concat</tt> metamethod.</li>
-<li>Various minor optimizations.</li>
-</ul></li>
-<li>Internal Changes:
-<ul>
-<li>Add support for embedding LuaJIT bytecode for builtins.</li>
-<li>Replace various builtins with embedded bytecode.</li>
-<li>Refactor string buffers and string formatting.</li>
-<li>Remove obsolete non-truncating number to integer conversions.</li>
-</ul></li>
-<li>Ports:
-<ul>
-<li>Add Xbox One port (<tt>LJ_GC64</tt> mode).</li>
-<li>ARM64: Add port of the interpreter (<tt>LJ_GC64</tt> mode).</li>
-<li>x64: Add separate port of the interpreter to <tt>LJ_GC64</tt> mode.</li>
-<li>x86/x64: Drop internal x87 math functions. Use libm functions.</li>
-<li>x86: Remove x87 support from interpreter. SSE2 is mandatory now.</li>
-<li>PPC/e500: Drop support for this architecture.</li>
-</ul></li>
-<li>FFI library:
-<ul>
-<li>FFI: Add 64 bit bitwise operations.</li>
-<li>FFI: Compile VLA/VLS and large cdata allocations with default initialization.</li>
-<li>FFI: Compile conversions from functions to function pointers.</li>
-<li>FFI: Compile lightuserdata to <tt>void *</tt> conversion.</li>
-<li>FFI: Compile <tt>ffi.gc(cdata, nil)</tt>, too.</li>
-<li>FFI: Add <tt>ffi.typeinfo()</tt>.</li>
-</ul></li>
-</ul>
-</div>
-
-<div class="major" style="background: #ffffd0;">
-<h2 id="LuaJIT-2.0.5">LuaJIT 2.0.5 — 2017-05-01</h2>
-<ul>
-<li>Add workaround for MSVC 2015 stdio changes.</li>
-<li>Limit mcode alloc probing, depending on the available pool size.</li>
-<li>Fix overly restrictive range calculation in mcode allocation.</li>
-<li>Fix out-of-scope goto handling in parser.</li>
-<li>Remove internal <tt>__mode = "K"</tt> and replace with safe check.</li>
-<li>Add "proto" field to <tt>jit.util.funcinfo()</tt>.</li>
-<li>Fix GC step size calculation.</li>
-<li>Initialize <tt>uv->immutable</tt> for upvalues of loaded chunks.</li>
-<li>Fix for cdata vs. non-cdata arithmetics/comparisons.</li>
-<li>Drop leftover regs in 'for' iterator assignment, too.</li>
-<li>Fix PHI remarking in SINK pass.</li>
-<li>Don't try to record outermost <tt>pcall()</tt> return to lower frame.</li>
-<li>Add guard for obscure aliasing between open upvalues and SSA slots.</li>
-<li>Remove assumption that <tt>lj_math_random_step()</tt> doesn't clobber FPRs.</li>
-<li>Fix handling of non-numeric strings in arithmetic coercions.</li>
-<li>Fix recording of <tt>select(n, ...)</tt> with off-trace varargs</li>
-<li>Fix install for cross-builds.</li>
-<li>Don't allocate unused 2nd result register in JIT compiler backend.</li>
-<li>Drop marks from replayed instructions when sinking.</li>
-<li>Fix unsinking check.</li>
-<li>Properly handle OOM in <tt>trace_save()</tt>.</li>
-<li>Limit number of arguments given to <tt>io.lines()</tt> and <tt>fp:lines()</tt>.</li>
-<li>Fix narrowing of <tt>TOBIT</tt>.</li>
-<li>OSX: Fix build with recent XCode.</li>
-<li>x86/x64: Don't spill an explicit <tt>REF_BASE</tt> in the IR.</li>
-<li>x86/x64: Fix instruction length decoder.</li>
-<li>x86/x64: Search for exit jumps with instruction length decoder.</li>
-<li>ARM: Fix <tt>BLX</tt> encoding for Thumb interworking calls.</li>
-<li>MIPS: Don't use <tt>RID_GP</tt> as a scratch register.</li>
-<li>MIPS: Fix emitted code for U32 to float conversion.</li>
-<li>MIPS: Backport workaround for compact unwind tables.</li>
-<li>MIPS: Fix cross-endian jit.bcsave.</li>
-<li>MIPS: Fix <tt>BC_ISNEXT</tt> fallback path.</li>
-<li>MIPS: Fix use of ffgccheck delay slots in interpreter.</li>
-<li>FFI: Fix FOLD rules for <tt>int64_t</tt> comparisons.</li>
-<li>FFI: Fix SPLIT pass for <tt>CONV i64.u64</tt>.</li>
-<li>FFI: Fix <tt>ipairs()</tt> recording.</li>
-<li>FFI: Don't propagate qualifiers into subtypes of complex.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.4">LuaJIT 2.0.4 — 2015-05-14</h2>
-<ul>
-<li>Fix stack check in narrowing optimization.</li>
-<li>Fix Lua/C API typecheck error for special indexes.</li>
-<li>Fix string to number conversion.</li>
-<li>Fix lexer error for chunks without tokens.</li>
-<li>Don't compile <tt>IR_RETF</tt> after <tt>CALLT</tt> to ff with-side effects.</li>
-<li>Fix <tt>BC_UCLO</tt>/<tt>BC_JMP</tt> join optimization in Lua parser.</li>
-<li>Fix corner case in string to number conversion.</li>
-<li>Gracefully handle <tt>lua_error()</tt> for a suspended coroutine.</li>
-<li>Avoid error messages when building with Clang.</li>
-<li>Fix snapshot #0 handling for traces with a stack check on entry.</li>
-<li>Fix fused constant loads under high register pressure.</li>
-<li>Invalidate backpropagation cache after DCE.</li>
-<li>Fix ABC elimination.</li>
-<li>Fix debug info for main chunk of stripped bytecode.</li>
-<li>Fix FOLD rule for <tt>string.sub(s, ...) == k</tt>.</li>
-<li>Fix FOLD rule for <tt>STRREF</tt> of <tt>SNEW</tt>.</li>
-<li>Fix frame traversal while searching for error function.</li>
-<li>Prevent GC estimate miscalculation due to buffer growth.</li>
-<li>Prevent adding side traces for stack checks.</li>
-<li>Fix top slot calculation for snapshots with continuations.</li>
-<li>Fix check for reuse of SCEV results in <tt>FORL</tt>.</li>
-<li>Add PS Vita port.</li>
-<li>Fix compatibility issues with Illumos.</li>
-<li>Fix DragonFly build (unsupported).</li>
-<li>OpenBSD/x86: Better executable memory allocation for W^X mode.</li>
-<li>x86: Fix argument checks for <tt>ipairs()</tt> iterator.</li>
-<li>x86: <tt>lj_math_random_step()</tt> clobbers XMM regs on OSX Clang.</li>
-<li>x86: Fix code generation for unused result of <tt>math.random()</tt>.</li>
-<li>x64: Allow building with <tt>LUAJIT_USE_SYSMALLOC</tt> and <tt>LUAJIT_USE_VALGRIND</tt>.</li>
-<li>x86/x64: Fix argument check for bit shifts.</li>
-<li>x86/x64: Fix code generation for fused test/arith ops.</li>
-<li>ARM: Fix write barrier check in <tt>BC_USETS</tt>.</li>
-<li>PPC: Fix red zone overflow in machine code generation.</li>
-<li>PPC: Don't use <tt>mcrxr</tt> on PPE.</li>
-<li>Various archs: Fix excess stack growth in interpreter.</li>
-<li>FFI: Fix FOLD rule for <tt>TOBIT</tt> + <tt>CONV num.u32</tt>.</li>
-<li>FFI: Prevent DSE across <tt>ffi.string()</tt>.</li>
-<li>FFI: No meta fallback when indexing pointer to incomplete struct.</li>
-<li>FFI: Fix initialization of unions of subtypes.</li>
-<li>FFI: Fix cdata vs. non-cdata arithmetic and comparisons.</li>
-<li>FFI: Fix <tt>__index</tt>/<tt>__newindex</tt> metamethod resolution for ctypes.</li>
-<li>FFI: Fix compilation of reference field access.</li>
-<li>FFI: Fix frame traversal for backtraces with FFI callbacks.</li>
-<li>FFI: Fix recording of indexing a struct pointer ctype object itself.</li>
-<li>FFI: Allow non-scalar cdata to be compared for equality by address.</li>
-<li>FFI: Fix pseudo type conversions for type punning.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.3">LuaJIT 2.0.3 — 2014-03-12</h2>
-<ul>
-<li>Add PS4 port.</li>
-<li>Add support for multilib distro builds.</li>
-<li>Fix OSX build.</li>
-<li>Fix MinGW build.</li>
-<li>Fix Xbox 360 build.</li>
-<li>Improve ULOAD forwarding for open upvalues.</li>
-<li>Fix GC steps threshold handling when called by JIT-compiled code.</li>
-<li>Fix argument checks for <tt>math.deg()</tt> and <tt>math.rad()</tt>.</li>
-<li>Fix <tt>jit.flush(func|true)</tt>.</li>
-<li>Respect <tt>jit.off(func)</tt> when returning to a function, too.</li>
-<li>Fix compilation of <tt>string.byte(s, nil, n)</tt>.</li>
-<li>Fix line number for relocated bytecode after closure fixup</li>
-<li>Fix frame traversal for backtraces.</li>
-<li>Fix ABC elimination.</li>
-<li>Fix handling of redundant PHIs.</li>
-<li>Fix snapshot restore for exit to function header.</li>
-<li>Fix type punning alias analysis for constified pointers</li>
-<li>Fix call unroll checks in the presence of metamethod frames.</li>
-<li>Fix initial maxslot for down-recursive traces.</li>
-<li>Prevent BASE register coalescing if parent uses <tt>IR_RETF</tt>.</li>
-<li>Don't purge modified function from stack slots in <tt>BC_RET</tt>.</li>
-<li>Fix recording of <tt>BC_VARG</tt>.</li>
-<li>Don't access dangling reference to reallocated IR.</li>
-<li>Fix frame depth display for bytecode dump in <tt>-jdump</tt>.</li>
-<li>ARM: Fix register allocation when rematerializing FPRs.</li>
-<li>x64: Fix store to upvalue for lightuserdata values.</li>
-<li>FFI: Add missing GC steps for callback argument conversions.</li>
-<li>FFI: Properly unload loaded DLLs.</li>
-<li>FFI: Fix argument checks for <tt>ffi.string()</tt>.</li>
-<li>FFI/x64: Fix passing of vector arguments to calls.</li>
-<li>FFI: Rehash finalizer table after GC cycle, if needed.</li>
-<li>FFI: Fix <tt>cts->L</tt> for cdata unsinking in snapshot restore.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.2">LuaJIT 2.0.2 — 2013-06-03</h2>
-<ul>
-<li>Fix memory access check for fast string interning.</li>
-<li>Fix MSVC intrinsics for older versions.</li>
-<li>Add missing GC steps for <tt>io.*</tt> functions.</li>
-<li>Fix spurious red zone overflows in machine code generation.</li>
-<li>Fix jump-range constrained mcode allocation.</li>
-<li>Inhibit DSE for implicit loads via calls.</li>
-<li>Fix builtin string to number conversion for overflow digits.</li>
-<li>Fix optional argument handling while recording builtins.</li>
-<li>Fix optional argument handling in <tt>table.concat()</tt>.</li>
-<li>Add partial support for building with MingW64 GCC 4.8-SEH.</li>
-<li>Add missing PHI barrier to <tt>string.sub(str, a, b) == kstr</tt> FOLD rule.</li>
-<li>Fix compatibility issues with Illumos.</li>
-<li>ARM: Fix cache flush/sync for exit stubs of JIT-compiled code.</li>
-<li>MIPS: Fix cache flush/sync for JIT-compiled code jump area.</li>
-<li>PPC: Add <tt>plt</tt> suffix for external calls from assembler code.</li>
-<li>FFI: Fix snapshot substitution in SPLIT pass.</li>
-<li>FFI/x86: Fix register allocation for 64 bit comparisons.</li>
-<li>FFI: Fix tailcall in lowest frame to C function with bool result.</li>
-<li>FFI: Ignore <tt>long</tt> type specifier in <tt>ffi.istype()</tt>.</li>
-<li>FFI: Fix calling conventions for 32 bit OSX and iOS simulator (struct returns).</li>
-<li>FFI: Fix calling conventions for ARM hard-float EABI (nested structs).</li>
-<li>FFI: Improve error messages for arithmetic and comparison operators.</li>
-<li>FFI: Insert no-op type conversion for pointer to integer cast.</li>
-<li>FFI: Fix unroll limit for <tt>ffi.fill()</tt>.</li>
-<li>FFI: Must sink <tt>XBAR</tt> together with <tt>XSTORE</tt>s.</li>
-<li>FFI: Preserve intermediate string for <tt>const char *</tt> conversion.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.1">LuaJIT 2.0.1 — 2013-02-19</h2>
-<ul>
-<li>Don't clear frame for out-of-memory error.</li>
-<li>Leave hook when resume catches error thrown from hook.</li>
-<li>Add missing GC steps for template table creation.</li>
-<li>Fix discharge order of comparisons in Lua parser.</li>
-<li>Improve buffer handling for <tt>io.read()</tt>.</li>
-<li>OSX: Add support for Mach-O object files to <tt>-b</tt> option.</li>
-<li>Fix PS3 port.</li>
-<li>Fix/enable Xbox 360 port.</li>
-<li>x86/x64: Always mark ref for shift count as non-weak.</li>
-<li>x64: Don't fuse implicitly 32-to-64 extended operands.</li>
-<li>ARM: Fix armhf call argument handling.</li>
-<li>ARM: Fix code generation for integer math.min/math.max.</li>
-<li>PPC/e500: Fix <tt>lj_vm_floor()</tt> for Inf/NaN.</li>
-<li>FFI: Change priority of table initializer variants for structs.</li>
-<li>FFI: Fix code generation for bool call result check on x86/x64.</li>
-<li>FFI: Load FFI library on-demand for bytecode with cdata literals.</li>
-<li>FFI: Fix handling of qualified transparent structs/unions.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0">LuaJIT 2.0.0 — 2012-11-08</h2>
-<ul>
-<li>Correctness and completeness:
-<ul>
- <li>Fix Android/x86 build.</li>
- <li>Fix recording of equality comparisons with <tt>__eq</tt> metamethods.</li>
- <li>Fix detection of immutable upvalues.</li>
- <li>Replace error with PANIC for callbacks from JIT-compiled code.</li>
- <li>Fix builtin string to number conversion for <tt>INT_MIN</tt>.</li>
- <li>Don't create unneeded array part for template tables.</li>
- <li>Fix <tt>CONV.num.int</tt> sinking.</li>
- <li>Don't propagate implicitly widened number to index metamethods.</li>
- <li>ARM: Fix ordered comparisons of number vs. non-number.</li>
- <li>FFI: Fix code generation for replay of sunk float fields.</li>
- <li>FFI: Fix signedness of bool.</li>
- <li>FFI: Fix recording of bool call result check on x86/x64.</li>
- <li>FFI: Fix stack-adjustment for <tt>__thiscall</tt> callbacks.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta11">LuaJIT 2.0.0-beta11 — 2012-10-16</h2>
-<ul>
-<li>New features:
-<ul>
- <li>Use ARM VFP instructions, if available (build-time detection).</li>
- <li>Add support for ARM hard-float EABI (<tt>armhf</tt>).</li>
- <li>Add PS3 port.</li>
- <li>Add many features from Lua 5.2, e.g. <tt>goto</tt>/labels.
- Refer to <a href="extensions.html#lua52">this list</a>.</li>
- <li>FFI: Add parameterized C types.</li>
- <li>FFI: Add support for copy constructors.</li>
- <li>FFI: Equality comparisons never raise an error (treat as unequal instead).</li>
- <li>FFI: Box all accessed or returned enums.</li>
- <li>FFI: Check for <tt>__new</tt> metamethod when calling a constructor.</li>
- <li>FFI: Handle <tt>__pairs</tt>/<tt>__ipairs</tt> metamethods for cdata objects.</li>
- <li>FFI: Convert <tt>io.*</tt> file handle to <tt>FILE *</tt> pointer (but as a <tt>void *</tt>).</li>
- <li>FFI: Detect and support type punning through unions.</li>
- <li>FFI: Improve various error messages.</li>
-</ul></li>
-<li>Build-system reorganization:
-<ul>
- <li>Reorganize directory layout:<br>
- <tt>lib/*</tt> → <tt>src/jit/*</tt><br>
- <tt>src/buildvm_*.dasc</tt> → <tt>src/vm_*.dasc</tt><br>
- <tt>src/buildvm_*.h</tt> → removed<br>
- <tt>src/buildvm*</tt> → <tt>src/host/*</tt></li>
- <li>Add minified Lua interpreter plus Lua BitOp (<tt>minilua</tt>) to run DynASM.</li>
- <li>Change DynASM bit operations to use Lua BitOp</li>
- <li>Translate only <tt>vm_*.dasc</tt> for detected target architecture.</li>
- <li>Improve target detection for <tt>msvcbuild.bat</tt>.</li>
- <li>Fix build issues on Cygwin and MinGW with optional MSys.</li>
- <li>Handle cross-compiles with FPU/no-FPU or hard-fp/soft-fp ABI mismatch.</li>
- <li>Remove some library functions for no-JIT/no-FFI builds.</li>
- <li>Add uninstall target to top-level Makefile.</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
- <li>Preserve snapshot #0 PC for all traces.</li>
- <li>Fix argument checks for <tt>coroutine.create()</tt>.</li>
- <li>Command line prints version and JIT status to <tt>stdout</tt>, not <tt>stderr</tt>.</li>
- <li>Fix userdata <tt>__gc</tt> separations at Lua state close.</li>
- <li>Fix <tt>TDUP</tt> to <tt>HLOAD</tt> forwarding for <tt>LJ_DUALNUM</tt> builds.</li>
- <li>Fix buffer check in bytecode writer.</li>
- <li>Make <tt>os.date()</tt> thread-safe.</li>
- <li>Add missing declarations for MSVC intrinsics.</li>
- <li>Fix dispatch table modifications for return hooks.</li>
- <li>Workaround for MSVC conversion bug (<tt>double</tt> → <tt>uint32_t</tt> → <tt>int32_t</tt>).</li>
- <li>Fix FOLD rule <tt>(i-j)-i => 0-j</tt>.</li>
- <li>Never use DWARF unwinder on Windows.</li>
- <li>Fix shrinking of direct mapped blocks in builtin allocator.</li>
- <li>Limit recursion depth in <tt>string.match()</tt> et al.</li>
- <li>Fix late despecialization of <tt>ITERN</tt> after loop has been entered.</li>
- <li>Fix <tt>'f'</tt> and <tt>'L'</tt> options for <tt>debug.getinfo()</tt> and <tt>lua_getinfo()</tt>.</li>
- <li>Fix <tt>package.searchpath()</tt>.</li>
- <li>OSX: Change dylib names to be consistent with other platforms.</li>
- <li>Android: Workaround for broken <tt>sprintf("%g", -0.0)</tt>.</li>
- <li>x86: Remove support for ancient CPUs without <tt>CMOV</tt> (before Pentium Pro).</li>
- <li>x86: Fix register allocation for calls returning register pair.</li>
- <li>x86/x64: Fix fusion of unsigned byte comparisons with swapped operands.</li>
- <li>ARM: Fix <tt>tonumber()</tt> argument check.</li>
- <li>ARM: Fix modulo operator and <tt>math.floor()</tt>/<tt>math.ceil()</tt> for <tt>inf</tt>/<tt>nan</tt>.</li>
- <li>ARM: Invoke SPLIT pass for leftover <tt>IR_TOBIT</tt>.</li>
- <li>ARM: Fix BASE register coalescing.</li>
- <li>PPC: Fix interpreter state setup in callbacks.</li>
- <li>PPC: Fix <tt>string.sub()</tt> range check.</li>
- <li>MIPS: Support generation of MIPS/MIPSEL bytecode object files.</li>
- <li>MIPS: Fix calls to <tt>floor()</tt>/<tt>ceil()</tt><tt>/trunc()</tt>.</li>
- <li>ARM/PPC: Detect more target architecture variants.</li>
- <li>ARM/PPC/e500/MIPS: Fix tailcalls from fast functions, esp. <tt>tostring()</tt>.</li>
- <li>ARM/PPC/MIPS: Fix rematerialization of FP constants.</li>
- <li>FFI: Don't call <tt>FreeLibrary()</tt> on our own EXE/DLL.</li>
- <li>FFI: Resolve metamethods for constructors, too.</li>
- <li>FFI: Properly disable callbacks on iOS (would require executable memory).</li>
- <li>FFI: Fix cdecl string parsing during recording.</li>
- <li>FFI: Show address pointed to for <tt>tostring(ref)</tt>, too.</li>
- <li>FFI: Fix alignment of C call argument/return structure.</li>
- <li>FFI: Initialize all fields of standard types.</li>
- <li>FFI: Fix callback handling when new C types are declared in callback.</li>
- <li>FFI: Fix recording of constructors for pointers.</li>
- <li>FFI: Always resolve metamethods for pointers to structs.</li>
- <li>FFI: Correctly propagate alignment when interning nested types.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
- <li>Add allocation sinking and store sinking optimization.</li>
- <li>Constify immutable upvalues.</li>
- <li>Add builtin string to integer or FP number conversion. Improves cross-platform consistency and correctness.</li>
- <li>Create string hash slots in template tables for non-const values, too. Avoids later table resizes.</li>
- <li>Eliminate <tt>HREFK</tt> guard for template table references.</li>
- <li>Add various new FOLD rules.</li>
- <li>Don't use stack unwinding for <tt>lua_yield()</tt> (slow on x64).</li>
- <li>ARM, PPC, MIPS: Improve <tt>XLOAD</tt> operand fusion and register hinting.</li>
- <li>PPC, MIPS: Compile <tt>math.sqrt()</tt> to sqrt instruction, if available.</li>
- <li>FFI: Fold <tt>KPTR</tt> + constant offset in SPLIT pass.</li>
- <li>FFI: Optimize/inline <tt>ffi.copy()</tt> and <tt>ffi.fill()</tt>.</li>
- <li>FFI: Compile and optimize array/struct copies.</li>
- <li>FFI: Compile <tt>ffi.typeof(cdata|ctype)</tt>, <tt>ffi.sizeof()</tt>, <tt>ffi.alignof()</tt>, <tt>ffi.offsetof()</tt> and <tt>ffi.gc()</tt>.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta10">LuaJIT 2.0.0-beta10 — 2012-05-09</h2>
-<ul>
-<li>New features:
-<ul>
-<li>The MIPS of LuaJIT is complete. It requires a CPU conforming to the
-MIPS32 R1 architecture with hardware FPU. O32 hard-fp ABI,
-little-endian or big-endian.</li>
-<li>Auto-detect target arch via cross-compiler. No need for
-<tt>TARGET=arch</tt> anymore.</li>
-<li>Make DynASM compatible with Lua 5.2.</li>
-<li>From Lua 5.2: Try <tt>__tostring</tt> metamethod on non-string error
-messages..</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>Fix parsing of hex literals with exponents.</li>
-<li>Fix bytecode dump for certain number constants.</li>
-<li>Fix argument type in error message for relative arguments.</li>
-<li>Fix argument error handling on Lua stacks without a frame.</li>
-<li>Add missing mcode limit check in assembler backend.</li>
-<li>Fix compilation on OpenBSD.</li>
-<li>Avoid recursive GC steps after GC-triggered trace exit.</li>
-<li>Replace <tt><unwind.h></tt> definitions with our own.</li>
-<li>Fix OSX build issues. Bump minimum required OSX version to 10.4.</li>
-<li>Fix discharge order of comparisons in Lua parser.</li>
-<li>Ensure running <tt>__gc</tt> of userdata created in <tt>__gc</tt>
-at state close.</li>
-<li>Limit number of userdata <tt>__gc</tt> separations at state close.</li>
-<li>Fix bytecode <tt>JMP</tt> slot range when optimizing
-<tt>and</tt>/<tt>or</tt> with constant LHS.</li>
-<li>Fix DSE of <tt>USTORE</tt>.</li>
-<li>Make <tt>lua_concat()</tt> work from C hook with partial frame.</li>
-<li>Add required PHIs for implicit conversions, e.g. via <tt>XREF</tt>
-forwarding.</li>
-<li>Add more comparison variants to Valgrind suppressions file.</li>
-<li>Disable loading bytecode with an extra header (BOM or <tt>#!</tt>).</li>
-<li>Fix PHI stack slot syncing.</li>
-<li>ARM: Reorder type/value tests to silence Valgrind.</li>
-<li>ARM: Fix register allocation for <tt>ldrd</tt>-optimized
-<tt>HREFK</tt>.</li>
-<li>ARM: Fix conditional branch fixup for <tt>OBAR</tt>.</li>
-<li>ARM: Invoke SPLIT pass for <tt>double</tt> args in FFI call.</li>
-<li>ARM: Handle all <tt>CALL*</tt> ops with <tt>double</tt> results in
-SPLIT pass.</li>
-<li>ARM: Fix rejoin of <tt>POW</tt> in SPLIT pass.</li>
-<li>ARM: Fix compilation of <tt>math.sinh</tt>, <tt>math.cosh</tt>,
-<tt>math.tanh</tt>.</li>
-<li>ARM, PPC: Avoid pointless arg clearing in <tt>BC_IFUNCF</tt>.</li>
-<li>PPC: Fix resume after yield from hook.</li>
-<li>PPC: Fix argument checking for <tt>rawget()</tt>.</li>
-<li>PPC: Fix fusion of floating-point <tt>XLOAD</tt>/<tt>XSTORE</tt>.</li>
-<li>PPC: Fix <tt>HREFK</tt> code generation for huge tables.</li>
-<li>PPC: Use builtin D-Cache/I-Cache sync code.</li>
-</ul></li>
-<li>FFI library:
-<ul>
-<li>Ignore empty statements in <tt>ffi.cdef()</tt>.</li>
-<li>Ignore number parsing errors while skipping definitions.</li>
-<li>Don't touch frame in callbacks with tailcalls to fast functions.</li>
-<li>Fix library unloading on POSIX systems.</li>
-<li>Finalize cdata before userdata when closing the state.</li>
-<li>Change <tt>ffi.load()</tt> library name resolution for Cygwin.</li>
-<li>Fix resolving of function name redirects on Windows/x86.</li>
-<li>Fix symbol resolving error messages on Windows.</li>
-<li>Fix blacklisting of C functions calling callbacks.</li>
-<li>Fix result type of pointer difference.</li>
-<li>Use correct PC in FFI metamethod error message.</li>
-<li>Allow <tt>'typedef _Bool int BOOL;'</tt> for the Windows API.</li>
-<li>Don't record test for bool result of call, if ignored.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta9">LuaJIT 2.0.0-beta9 — 2011-12-14</h2>
-<ul>
-<li>New features:
-<ul>
-<li>PPC port of LuaJIT is complete. Default is the dual-number port
-(usually faster). Single-number port selectable via <tt>src/Makefile</tt>
-at build time.</li>
-<li>Add FFI callback support.</li>
-<li>Extend <tt>-b</tt> to generate <tt>.c</tt>, <tt>.h</tt> or <tt>.obj/.o</tt>
-files with embedded bytecode.</li>
-<li>Allow loading embedded bytecode with <tt>require()</tt>.</li>
-<li>From Lua 5.2: Change to <tt>'\z'</tt> escape. Reject undefined escape
-sequences.</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>Fix OSX 10.7 build. Fix <tt>install_name</tt> and versioning on OSX.</li>
-<li>Fix iOS build.</li>
-<li>Install <tt>dis_arm.lua</tt>, too.</li>
-<li>Mark installed shared library as executable.</li>
-<li>Add debug option to <tt>msvcbuild.bat</tt> and improve error handling.</li>
-<li>Fix data-flow analysis for iterators.</li>
-<li>Fix forced unwinding triggered by external unwinder.</li>
-<li>Record missing <tt>for</tt> loop slot loads (return to lower frame).</li>
-<li>Always use ANSI variants of Windows system functions.</li>
-<li>Fix GC barrier for multi-result table constructor (<tt>TSETM</tt>).</li>
-<li>Fix/add various FOLD rules.</li>
-<li>Add potential PHI for number conversions due to type instability.</li>
-<li>Do not eliminate PHIs only referenced from other PHIs.</li>
-<li>Correctly anchor implicit number to string conversions in Lua/C API.</li>
-<li>Fix various stack limit checks.</li>
-<li>x64: Use thread-safe exceptions for external unwinding (GCC platforms).</li>
-<li>x64: Fix result type of cdata index conversions.</li>
-<li>x64: Fix <tt>math.random()</tt> and <tt>bit.bswap()</tt> code generation.</li>
-<li>x64: Fix <tt>lightuserdata</tt> comparisons.</li>
-<li>x64: Always extend stack-passed arguments to pointer size.</li>
-<li>ARM: Many fixes to code generation backend.</li>
-<li>PPC/e500: Fix dispatch for binop metamethods.</li>
-<li>PPC/e500: Save/restore condition registers when entering/leaving the VM.</li>
-<li>PPC/e500: Fix write barrier in stores of strings to upvalues.</li>
-</ul></li>
-<li>FFI library:
-<ul>
-<li>Fix C comment parsing.</li>
-<li>Fix snapshot optimization for cdata comparisons.</li>
-<li>Fix recording of const/enum lookups in namespaces.</li>
-<li>Fix call argument and return handling for <tt>I8/U8/I16/U16</tt> types.</li>
-<li>Fix unfused loads of float fields.</li>
-<li>Fix <tt>ffi.string()</tt> recording.</li>
-<li>Save <tt>GetLastError()</tt> around <tt>ffi.load()</tt> and symbol
-resolving, too.</li>
-<li>Improve ld script detection in <tt>ffi.load()</tt>.</li>
-<li>Record loads/stores to external variables in namespaces.</li>
-<li>Compile calls to stdcall, fastcall and vararg functions.</li>
-<li>Treat function ctypes like pointers in comparisons.</li>
-<li>Resolve <tt>__call</tt> metamethod for pointers, too.</li>
-<li>Record C function calls with bool return values.</li>
-<li>Record <tt>ffi.errno()</tt>.</li>
-<li>x86: Fix number to <tt>uint32_t</tt> conversion rounding.</li>
-<li>x86: Fix 64 bit arithmetic in assembler backend.</li>
-<li>x64: Fix struct-by-value calling conventions.</li>
-<li>ARM: Ensure invocation of SPLIT pass for float conversions.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Display trace types with <tt>-jv</tt> and <tt>-jdump</tt>.</li>
-<li>Record isolated calls. But prefer recording loops over calls.</li>
-<li>Specialize to prototype for non-monomorphic functions. Solves the
-trace-explosion problem for closure-heavy programming styles.</li>
-<li>Always generate a portable <tt>vmdef.lua</tt>. Easier for distros.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta8">LuaJIT 2.0.0-beta8 — 2011-06-23</h2>
-<ul>
-<li>New features:
-<ul>
-<li>Soft-float ARM port of LuaJIT is complete.</li>
-<li>Add support for bytecode loading/saving and <tt>-b</tt> command line
-option.</li>
-<li>From Lua 5.2: <tt>__len</tt> metamethod for tables
-(disabled by default).</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>ARM: Misc. fixes for interpreter.</li>
-<li>x86/x64: Fix <tt>bit.*</tt> argument checking in interpreter.</li>
-<li>Catch early out-of-memory in memory allocator initialization.</li>
-<li>Fix data-flow analysis for paths leading to an upvalue close.</li>
-<li>Fix check for missing arguments in <tt>string.format()</tt>.</li>
-<li>Fix Solaris/x86 build (note: not a supported target).</li>
-<li>Fix recording of loops with instable directions in side traces.</li>
-<li>x86/x64: Fix fusion of comparisons with <tt>u8</tt>/<tt>u16</tt>
-<tt>XLOAD</tt>.</li>
-<li>x86/x64: Fix register allocation for variable shifts.</li>
-</ul></li>
-<li>FFI library:
-<ul>
-<li>Add <tt>ffi.errno()</tt>. Save <tt>errno</tt>/<tt>GetLastError()</tt>
-around allocations etc.</li>
-<li>Fix <tt>__gc</tt> for VLA/VLS cdata objects.</li>
-<li>Fix recording of casts from 32 bit cdata pointers to integers.</li>
-<li><tt>tonumber(cdata)</tt> returns <tt>nil</tt> for non-numbers.</li>
-<li>Show address pointed to for <tt>tostring(pointer)</tt>.</li>
-<li>Print <tt>NULL</tt> pointers as <tt>"cdata<... *>: NULL"</tt>.</li>
-<li>Support <tt>__tostring</tt> metamethod for pointers to structs, too.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>More tuning for loop unrolling heuristics.</li>
-<li>Flatten and compress in-memory debug info (saves ~70%).</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta7">LuaJIT 2.0.0-beta7 — 2011-05-05</h2>
-<ul>
-<li>New features:
-<ul>
-<li>ARM port of the LuaJIT interpreter is complete.</li>
-<li>FFI library: Add <tt>ffi.gc()</tt>, <tt>ffi.metatype()</tt>,
-<tt>ffi.istype()</tt>.</li>
-<li>FFI library: Resolve ld script redirection in <tt>ffi.load()</tt>.</li>
-<li>From Lua 5.2: <tt>package.searchpath()</tt>, <tt>fp:read("*L")</tt>,
-<tt>load(string)</tt>.</li>
-<li>From Lua 5.2, disabled by default: empty statement,
-<tt>table.unpack()</tt>, modified <tt>coroutine.running()</tt>.</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>FFI library: numerous fixes.</li>
-<li>Fix type mismatches in store-to-load forwarding.</li>
-<li>Fix error handling within metamethods.</li>
-<li>Fix <tt>table.maxn()</tt>.</li>
-<li>Improve accuracy of <tt>x^-k</tt> on x64.</li>
-<li>Fix code generation for Intel Atom in x64 mode.</li>
-<li>Fix narrowing of POW.</li>
-<li>Fix recording of retried fast functions.</li>
-<li>Fix code generation for <tt>bit.bnot()</tt> and multiplies.</li>
-<li>Fix error location within cpcall frames.</li>
-<li>Add workaround for old libgcc unwind bug.</li>
-<li>Fix <tt>lua_yield()</tt> and <tt>getmetatable(lightuserdata)</tt> on x64.</li>
-<li>Misc. fixes for PPC/e500 interpreter.</li>
-<li>Fix stack slot updates for down-recursion.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Add dual-number mode (int/double) for the VM. Enabled for ARM.</li>
-<li>Improve narrowing of arithmetic operators and <tt>for</tt> loops.</li>
-<li>Tune loop unrolling heuristics and increase trace recorder limits.</li>
-<li>Eliminate dead slots in snapshots using bytecode data-flow analysis.</li>
-<li>Avoid phantom stores to proxy tables.</li>
-<li>Optimize lookups in empty proxy tables.</li>
-<li>Improve bytecode optimization of <tt>and</tt>/<tt>or</tt> operators.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta6">LuaJIT 2.0.0-beta6 — 2011-02-11</h2>
-<ul>
-<li>New features:
-<ul>
-<li>PowerPC/e500v2 port of the LuaJIT interpreter is complete.</li>
-<li>Various minor features from Lua 5.2: Hex escapes in literals,
-<tt>'\*'</tt> escape, reversible <tt>string.format("%q",s)</tt>,
-<tt>"%g"</tt> pattern, <tt>table.sort</tt> checks callbacks,
-<tt>os.exit(status|true|false[,close])</tt>.</li>
-<li>Lua 5.2 <tt>__pairs</tt> and <tt>__ipairs</tt> metamethods
-(disabled by default).</li>
-<li>Initial release of the FFI library.</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>Fix <tt>string.format()</tt> for non-finite numbers.</li>
-<li>Fix memory leak when compiled to use the built-in allocator.</li>
-<li>x86/x64: Fix unnecessary resize in <tt>TSETM</tt> bytecode.</li>
-<li>Fix various GC issues with traces and <tt>jit.flush()</tt>.</li>
-<li>x64: Fix fusion of indexes for array references.</li>
-<li>x86/x64: Fix stack overflow handling for coroutine results.</li>
-<li>Enable low-2GB memory allocation on FreeBSD/x64.</li>
-<li>Fix <tt>collectgarbage("count")</tt> result if more than 2GB is in use.</li>
-<li>Fix parsing of hex floats.</li>
-<li>x86/x64: Fix loop branch inversion with trailing
-<tt>HREF+NE/EQ</tt>.</li>
-<li>Add <tt>jit.os</tt> string.</li>
-<li><tt>coroutine.create()</tt> permits running C functions, too.</li>
-<li>Fix OSX build to work with newer ld64 versions.</li>
-<li>Fix bytecode optimization of <tt>and</tt>/<tt>or</tt> operators.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Emit specialized bytecode for <tt>pairs()</tt>/<tt>next()</tt>.</li>
-<li>Improve bytecode coalescing of <tt>nil</tt> constants.</li>
-<li>Compile calls to vararg functions.</li>
-<li>Compile <tt>select()</tt>.</li>
-<li>Improve alias analysis, esp. for loads from allocations.</li>
-<li>Tuning of various compiler heuristics.</li>
-<li>Refactor and extend IR conversion instructions.</li>
-<li>x86/x64: Various backend enhancements related to the FFI.</li>
-<li>Add SPLIT pass to split 64 bit IR instructions for 32 bit CPUs.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta5">LuaJIT 2.0.0-beta5 — 2010-08-24</h2>
-<ul>
-<li>Correctness and completeness:
-<ul>
-<li>Fix trace exit dispatch to function headers.</li>
-<li>Fix Windows and OSX builds with LUAJIT_DISABLE_JIT.</li>
-<li>Reorganize and fix placement of generated machine code on x64.</li>
-<li>Fix TNEW in x64 interpreter.</li>
-<li>Do not eliminate PHIs for values only referenced from side exits.</li>
-<li>OS-independent canonicalization of strings for non-finite numbers.</li>
-<li>Fix <tt>string.char()</tt> range check on x64.</li>
-<li>Fix <tt>tostring()</tt> resolving within <tt>print()</tt>.</li>
-<li>Fix error handling for <tt>next()</tt>.</li>
-<li>Fix passing of constant arguments to external calls on x64.</li>
-<li>Fix interpreter argument check for two-argument SSE math functions.</li>
-<li>Fix C frame chain corruption caused by <tt>lua_cpcall()</tt>.</li>
-<li>Fix return from <tt>pcall()</tt> within active hook.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Replace on-trace GC frame syncing with interpreter exit.</li>
-<li>Improve hash lookup specialization by not removing dead keys during GC.</li>
-<li>Turn traces into true GC objects.</li>
-<li>Avoid starting a GC cycle immediately after library init.</li>
-<li>Add weak guards to improve dead-code elimination.</li>
-<li>Speed up string interning.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta4">LuaJIT 2.0.0-beta4 — 2010-03-28</h2>
-<ul>
-<li>Correctness and completeness:
-<ul>
-<li>Fix precondition for on-trace creation of table keys.</li>
-<li>Fix <tt>{f()}</tt> on x64 when table is resized.</li>
-<li>Fix folding of ordered comparisons with same references.</li>
-<li>Fix snapshot restores for multi-result bytecodes.</li>
-<li>Fix potential hang when recording bytecode with nested closures.</li>
-<li>Fix recording of <tt>getmetatable()</tt>, <tt>tonumber()</tt> and bad argument types.</li>
-<li>Fix SLOAD fusion across returns to lower frames.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Add array bounds check elimination. <tt>-Oabc</tt> is enabled by default.</li>
-<li>More tuning for x64, e.g. smaller table objects.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta3">LuaJIT 2.0.0-beta3 — 2010-03-07</h2>
-<ul>
-<li>LuaJIT x64 port:
-<ul>
-<li>Port integrated memory allocator to Linux/x64, Windows/x64 and OSX/x64.</li>
-<li>Port interpreter and JIT compiler to x64.</li>
-<li>Port DynASM to x64.</li>
-<li>Many 32/64 bit cleanups in the VM.</li>
-<li>Allow building the interpreter with either x87 or SSE2 arithmetics.</li>
-<li>Add external unwinding and C++ exception interop (default on x64).</li>
-</ul></li>
-<li>Correctness and completeness:
-<ul>
-<li>Fix constructor bytecode generation for certain conditional values.</li>
-<li>Fix some cases of ordered string comparisons.</li>
-<li>Fix <tt>lua_tocfunction()</tt>.</li>
-<li>Fix cutoff register in JMP bytecode for some conditional expressions.</li>
-<li>Fix PHI marking algorithm for references from variant slots.</li>
-<li>Fix <tt>package.cpath</tt> for non-default PREFIX.</li>
-<li>Fix DWARF2 frame unwind information for interpreter on OSX.</li>
-<li>Drive the GC forward on string allocations in the parser.</li>
-<li>Implement call/return hooks (zero-cost if disabled).</li>
-<li>Implement yield from C hooks.</li>
-<li>Disable JIT compiler on older non-SSE2 CPUs instead of aborting.</li>
-</ul></li>
-<li>Structural and performance enhancements:
-<ul>
-<li>Compile recursive code (tail-, up- and down-recursion).</li>
-<li>Improve heuristics for bytecode penalties and blacklisting.</li>
-<li>Split CALL/FUNC recording and clean up fast function call semantics.</li>
-<li>Major redesign of internal function call handling.</li>
-<li>Improve FOR loop const specialization and integerness checks.</li>
-<li>Switch to pre-initialized stacks. Avoid frame-clearing.</li>
-<li>Colocation of prototypes and related data: bytecode, constants, debug info.</li>
-<li>Cleanup parser and streamline bytecode generation.</li>
-<li>Add support for weak IR references to register allocator.</li>
-<li>Switch to compressed, extensible snapshots.</li>
-<li>Compile returns to frames below the start frame.</li>
-<li>Improve alias analysis of upvalues using a disambiguation hash value.</li>
-<li>Compile floor/ceil/trunc to SSE2 helper calls or SSE4.1 instructions.</li>
-<li>Add generic C call handling to IR and backend.</li>
-<li>Improve KNUM fuse vs. load heuristics.</li>
-<li>Compile various <tt>io.*()</tt> functions.</li>
-<li>Compile <tt>math.sinh()</tt>, <tt>math.cosh()</tt>, <tt>math.tanh()</tt>
-and <tt>math.random()</tt>.</li>
-</ul></li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta2">LuaJIT 2.0.0-beta2 — 2009-11-09</h2>
-<ul>
-<li>Reorganize build system. Build static+shared library on POSIX.</li>
-<li>Allow C++ exception conversion on all platforms
-using a wrapper function.</li>
-<li>Automatically catch C++ exceptions and rethrow Lua error
-(DWARF2 only).</li>
-<li>Check for the correct x87 FPU precision at strategic points.</li>
-<li>Always use wrappers for libm functions.</li>
-<li>Resurrect metamethod name strings before copying them.</li>
-<li>Mark current trace, even if compiler is idle.</li>
-<li>Ensure FILE metatable is created only once.</li>
-<li>Fix type comparisons when different integer types are involved.</li>
-<li>Fix <tt>getmetatable()</tt> recording.</li>
-<li>Fix TDUP with dead keys in template table.</li>
-<li><tt>jit.flush(tr)</tt> returns status.
-Prevent manual flush of a trace that's still linked.</li>
-<li>Improve register allocation heuristics for invariant references.</li>
-<li>Compile the push/pop variants of <tt>table.insert()</tt> and
-<tt>table.remove()</tt>.</li>
-<li>Compatibility with MSVC <tt>link /debug</tt>.</li>
-<li>Fix <tt>lua_iscfunction()</tt>.</li>
-<li>Fix <tt>math.random()</tt> when compiled with <tt>-fpic</tt> (OSX).</li>
-<li>Fix <tt>table.maxn()</tt>.</li>
-<li>Bump <tt>MACOSX_DEPLOYMENT_TARGET</tt> to <tt>10.4</tt></li>
-<li><tt>luaL_check*()</tt> and <tt>luaL_opt*()</tt> now support
-negative arguments, too.<br>
-This matches the behavior of Lua 5.1, but not the specification.</li>
-</ul>
-
-<h2 id="LuaJIT-2.0.0-beta1">LuaJIT 2.0.0-beta1 — 2009-10-31</h2>
-<ul>
-<li>This is the first public release of LuaJIT 2.0.</li>
-<li>The whole VM has been rewritten from the ground up, so there's
-no point in listing differences over earlier versions.</li>
-</ul>
-</div>
-<br class="flush">
-</div>
-<div id="foot">
-<hr class="hide">
-Copyright © 2005-2017 Mike Pall
-<span class="noprint">
-·
-<a href="contact.html">Contact</a>
-</span>
-</div>
-</body>
-</html>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/contact.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/contact.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/contact.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Contact</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -11,7 +10,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Contact</h1>
@@ -20,7 +19,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -38,6 +37,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -45,18 +46,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -63,11 +57,11 @@
<p>
If you want to report bugs, propose fixes or suggest enhancements,
please use the
-<a href="https://github.com/LuaJIT/LuaJIT/issues">GitHub issue tracker</a>.
+<a href="https://github.com/LuaJIT/LuaJIT/issues"><span class="ext">»</span> GitHub issue tracker</a>.
</p>
<p>
Please send general questions to the
-<a href="http://luajit.org/list.html"><span class="ext">»</span> LuaJIT mailing list</a>.
+<a href="https://luajit.org/list.html"><span class="ext">»</span> LuaJIT mailing list</a>.
</p>
<p>
You can also send any questions you have directly to me:
@@ -90,10 +84,17 @@
</p>
</noscript>
+<p><i>
+Note: I cannot reply to GMail, Google Workplace, Outlook or Office365
+mail addresses, since they prefer to mindlessly filter out mails sent
+from small domains using independent mail servers, such as mine. If you
+don't like that, please complain to Google or Microsoft, not me.
+</i></p>
+
<h2>Copyright</h2>
<p>
All documentation is
-Copyright © 2005-2017 Mike Pall.
+Copyright © 2005-2025 Mike Pall.
</p>
@@ -101,7 +102,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Added: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_buffer.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_buffer.html (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_buffer.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,689 @@
+<!DOCTYPE html>
+<html>
+<head>
+<title>String Buffer Library</title>
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
+<meta name="Language" content="en">
+<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
+<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
+<style type="text/css">
+.lib {
+ vertical-align: middle;
+ margin-left: 5px;
+ padding: 0 5px;
+ font-size: 60%;
+ border-radius: 5px;
+ background: #c5d5ff;
+ color: #000;
+}
+</style>
+</head>
+<body>
+<div id="site">
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+</div>
+<div id="head">
+<h1>String Buffer Library</h1>
+</div>
+<div id="nav">
+<ul><li>
+<a href="luajit.html">LuaJIT</a>
+<ul><li>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
+</li><li>
+<a href="install.html">Installation</a>
+</li><li>
+<a href="running.html">Running</a>
+</li></ul>
+</li><li>
+<a href="extensions.html">Extensions</a>
+<ul><li>
+<a href="ext_ffi.html">FFI Library</a>
+<ul><li>
+<a href="ext_ffi_tutorial.html">FFI Tutorial</a>
+</li><li>
+<a href="ext_ffi_api.html">ffi.* API</a>
+</li><li>
+<a href="ext_ffi_semantics.html">FFI Semantics</a>
+</li></ul>
+</li><li>
+<a class="current" href="ext_buffer.html">String Buffers</a>
+</li><li>
+<a href="ext_jit.html">jit.* Library</a>
+</li><li>
+<a href="ext_c_api.html">Lua/C API</a>
+</li><li>
+<a href="ext_profiler.html">Profiler</a>
+</li></ul>
+</li><li>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
+</li><li>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
+</li><li>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+</li></ul>
+</div>
+<div id="main">
+<p>
+The string buffer library allows <b>high-performance manipulation of
+string-like data</b>.
+</p>
+<p>
+Unlike Lua strings, which are constants, string buffers are
+<b>mutable</b> sequences of 8-bit (binary-transparent) characters. Data
+can be stored, formatted and encoded into a string buffer and later
+converted, extracted or decoded.
+</p>
+<p>
+The convenient string buffer API simplifies common string manipulation
+tasks, that would otherwise require creating many intermediate strings.
+String buffers improve performance by eliminating redundant memory
+copies, object creation, string interning and garbage collection
+overhead. In conjunction with the FFI library, they allow zero-copy
+operations.
+</p>
+<p>
+The string buffer library also includes a high-performance
+<a href="#serialize">serializer</a> for Lua objects.
+</p>
+
+<h2 id="use">Using the String Buffer Library</h2>
+<p>
+The string buffer library is built into LuaJIT by default, but it's not
+loaded by default. Add this to the start of every Lua file that needs
+one of its functions:
+</p>
+<pre class="code">
+local buffer = require("string.buffer")
+</pre>
+<p>
+The convention for the syntax shown on this page is that <tt>buffer</tt>
+refers to the buffer library and <tt>buf</tt> refers to an individual
+buffer object.
+</p>
+<p>
+Please note the difference between a Lua function call, e.g.
+<tt>buffer.new()</tt> (with a dot) and a Lua method call, e.g.
+<tt>buf:reset()</tt> (with a colon).
+</p>
+
+<h3 id="buffer_object">Buffer Objects</h3>
+<p>
+A buffer object is a garbage-collected Lua object. After creation with
+<tt>buffer.new()</tt>, it can (and should) be reused for many operations.
+When the last reference to a buffer object is gone, it will eventually
+be freed by the garbage collector, along with the allocated buffer
+space.
+</p>
+<p>
+Buffers operate like a FIFO (first-in first-out) data structure. Data
+can be appended (written) to the end of the buffer and consumed (read)
+from the front of the buffer. These operations may be freely mixed.
+</p>
+<p>
+The buffer space that holds the characters is managed automatically
+— it grows as needed and already consumed space is recycled. Use
+<tt>buffer.new(size)</tt> and <tt>buf:free()</tt>, if you need more
+control.
+</p>
+<p>
+The maximum size of a single buffer is the same as the maximum size of a
+Lua string, which is slightly below two gigabytes. For huge data sizes,
+neither strings nor buffers are the right data structure — use the
+FFI library to directly map memory or files up to the virtual memory
+limit of your OS.
+</p>
+
+<h3 id="buffer_overview">Buffer Method Overview</h3>
+<ul>
+<li>
+The <tt>buf:put*()</tt>-like methods append (write) characters to the
+end of the buffer.
+</li>
+<li>
+The <tt>buf:get*()</tt>-like methods consume (read) characters from the
+front of the buffer.
+</li>
+<li>
+Other methods, like <tt>buf:tostring()</tt> only read the buffer
+contents, but don't change the buffer.
+</li>
+<li>
+The <tt>buf:set()</tt> method allows zero-copy consumption of a string
+or an FFI cdata object as a buffer.
+</li>
+<li>
+The FFI-specific methods allow zero-copy read/write-style operations or
+modifying the buffer contents in-place. Please check the
+<a href="#ffi_caveats">FFI caveats</a> below, too.
+</li>
+<li>
+Methods that don't need to return anything specific, return the buffer
+object itself as a convenience. This allows method chaining, e.g.:
+<tt>buf:reset():encode(obj)</tt> or <tt>buf:skip(len):get()</tt>
+</li>
+</ul>
+
+<h2 id="create">Buffer Creation and Management</h2>
+
+<h3 id="buffer_new"><tt>local buf = buffer.new([size [,options]])<br>
+local buf = buffer.new([options])</tt></h3>
+<p>
+Creates a new buffer object.
+</p>
+<p>
+The optional <tt>size</tt> argument ensures a minimum initial buffer
+size. This is strictly an optimization when the required buffer size is
+known beforehand. The buffer space will grow as needed, in any case.
+</p>
+<p>
+The optional table <tt>options</tt> sets various
+<a href="#serialize_options">serialization options</a>.
+</p>
+
+<h3 id="buffer_reset"><tt>buf = buf:reset()</tt></h3>
+<p>
+Reset (empty) the buffer. The allocated buffer space is not freed and
+may be reused.
+</p>
+
+<h3 id="buffer_free"><tt>buf = buf:free()</tt></h3>
+<p>
+The buffer space of the buffer object is freed. The object itself
+remains intact, empty and may be reused.
+</p>
+<p>
+Note: you normally don't need to use this method. The garbage collector
+automatically frees the buffer space, when the buffer object is
+collected. Use this method, if you need to free the associated memory
+immediately.
+</p>
+
+<h2 id="write">Buffer Writers</h2>
+
+<h3 id="buffer_put"><tt>buf = buf:put([str|num|obj] [,…])</tt></h3>
+<p>
+Appends a string <tt>str</tt>, a number <tt>num</tt> or any object
+<tt>obj</tt> with a <tt>__tostring</tt> metamethod to the buffer.
+Multiple arguments are appended in the given order.
+</p>
+<p>
+Appending a buffer to a buffer is possible and short-circuited
+internally. But it still involves a copy. Better combine the buffer
+writes to use a single buffer.
+</p>
+
+<h3 id="buffer_putf"><tt>buf = buf:putf(format, …)</tt></h3>
+<p>
+Appends the formatted arguments to the buffer. The <tt>format</tt>
+string supports the same options as <tt>string.format()</tt>.
+</p>
+
+<h3 id="buffer_putcdata"><tt>buf = buf:putcdata(cdata, len)</tt><span class="lib">FFI</span></h3>
+<p>
+Appends the given <tt>len</tt> number of bytes from the memory pointed
+to by the FFI <tt>cdata</tt> object to the buffer. The object needs to
+be convertible to a (constant) pointer.
+</p>
+
+<h3 id="buffer_set"><tt>buf = buf:set(str)<br>
+buf = buf:set(cdata, len)</tt><span class="lib">FFI</span></h3>
+<p>
+This method allows zero-copy consumption of a string or an FFI cdata
+object as a buffer. It stores a reference to the passed string
+<tt>str</tt> or the FFI <tt>cdata</tt> object in the buffer. Any buffer
+space originally allocated is freed. This is <i>not</i> an append
+operation, unlike the <tt>buf:put*()</tt> methods.
+</p>
+<p>
+After calling this method, the buffer behaves as if
+<tt>buf:free():put(str)</tt> or <tt>buf:free():put(cdata, len)</tt>
+had been called. However, the data is only referenced and not copied, as
+long as the buffer is only consumed.
+</p>
+<p>
+In case the buffer is written to later on, the referenced data is copied
+and the object reference is removed (copy-on-write semantics).
+</p>
+<p>
+The stored reference is an anchor for the garbage collector and keeps the
+originally passed string or FFI cdata object alive.
+</p>
+
+<h3 id="buffer_reserve"><tt>ptr, len = buf:reserve(size)</tt><span class="lib">FFI</span><br>
+<tt>buf = buf:commit(used)</tt><span class="lib">FFI</span></h3>
+<p>
+The <tt>reserve</tt> method reserves at least <tt>size</tt> bytes of
+write space in the buffer. It returns an <tt>uint8_t *</tt> FFI
+cdata pointer <tt>ptr</tt> that points to this space.
+</p>
+<p>
+The available length in bytes is returned in <tt>len</tt>. This is at
+least <tt>size</tt> bytes, but may be more to facilitate efficient
+buffer growth. You can either make use of the additional space or ignore
+<tt>len</tt> and only use <tt>size</tt> bytes.
+</p>
+<p>
+The <tt>commit</tt> method appends the <tt>used</tt> bytes of the
+previously returned write space to the buffer data.
+</p>
+<p>
+This pair of methods allows zero-copy use of C read-style APIs:
+</p>
+<pre class="code">
+local MIN_SIZE = 65536
+repeat
+ local ptr, len = buf:reserve(MIN_SIZE)
+ local n = C.read(fd, ptr, len)
+ if n == 0 then break end -- EOF.
+ if n < 0 then error("read error") end
+ buf:commit(n)
+until false
+</pre>
+<p>
+The reserved write space is <i>not</i> initialized. At least the
+<tt>used</tt> bytes <b>must</b> be written to before calling the
+<tt>commit</tt> method. There's no need to call the <tt>commit</tt>
+method, if nothing is added to the buffer (e.g. on error).
+</p>
+
+<h2 id="read">Buffer Readers</h2>
+
+<h3 id="buffer_length"><tt>len = #buf</tt></h3>
+<p>
+Returns the current length of the buffer data in bytes.
+</p>
+
+<h3 id="buffer_concat"><tt>res = str|num|buf .. str|num|buf […]</tt></h3>
+<p>
+The Lua concatenation operator <tt>..</tt> also accepts buffers, just
+like strings or numbers. It always returns a string and not a buffer.
+</p>
+<p>
+Note that although this is supported for convenience, this thwarts one
+of the main reasons to use buffers, which is to avoid string
+allocations. Rewrite it with <tt>buf:put()</tt> and <tt>buf:get()</tt>.
+</p>
+<p>
+Mixing this with unrelated objects that have a <tt>__concat</tt>
+metamethod may not work, since these probably only expect strings.
+</p>
+
+<h3 id="buffer_skip"><tt>buf = buf:skip(len)</tt></h3>
+<p>
+Skips (consumes) <tt>len</tt> bytes from the buffer up to the current
+length of the buffer data.
+</p>
+
+<h3 id="buffer_get"><tt>str, … = buf:get([len|nil] [,…])</tt></h3>
+<p>
+Consumes the buffer data and returns one or more strings. If called
+without arguments, the whole buffer data is consumed. If called with a
+number, up to <tt>len</tt> bytes are consumed. A <tt>nil</tt> argument
+consumes the remaining buffer space (this only makes sense as the last
+argument). Multiple arguments consume the buffer data in the given
+order.
+</p>
+<p>
+Note: a zero length or no remaining buffer data returns an empty string
+and not <tt>nil</tt>.
+</p>
+
+<h3 id="buffer_tostring"><tt>str = buf:tostring()<br>
+str = tostring(buf)</tt></h3>
+<p>
+Creates a string from the buffer data, but doesn't consume it. The
+buffer remains unchanged.
+</p>
+<p>
+Buffer objects also define a <tt>__tostring</tt> metamethod. This means
+buffers can be passed to the global <tt>tostring()</tt> function and
+many other functions that accept this in place of strings. The important
+internal uses in functions like <tt>io.write()</tt> are short-circuited
+to avoid the creation of an intermediate string object.
+</p>
+
+<h3 id="buffer_ref"><tt>ptr, len = buf:ref()</tt><span class="lib">FFI</span></h3>
+<p>
+Returns an <tt>uint8_t *</tt> FFI cdata pointer <tt>ptr</tt> that
+points to the buffer data. The length of the buffer data in bytes is
+returned in <tt>len</tt>.
+</p>
+<p>
+The returned pointer can be directly passed to C functions that expect a
+buffer and a length. You can also do bytewise reads
+(<tt>local x = ptr[i]</tt>) or writes
+(<tt>ptr[i] = 0x40</tt>) of the buffer data.
+</p>
+<p>
+In conjunction with the <tt>skip</tt> method, this allows zero-copy use
+of C write-style APIs:
+</p>
+<pre class="code">
+repeat
+ local ptr, len = buf:ref()
+ if len == 0 then break end
+ local n = C.write(fd, ptr, len)
+ if n < 0 then error("write error") end
+ buf:skip(n)
+until n >= len
+</pre>
+<p>
+Unlike Lua strings, buffer data is <i>not</i> implicitly
+zero-terminated. It's not safe to pass <tt>ptr</tt> to C functions that
+expect zero-terminated strings. If you're not using <tt>len</tt>, then
+you're doing something wrong.
+</p>
+
+<h2 id="serialize">Serialization of Lua Objects</h2>
+<p>
+The following functions and methods allow <b>high-speed serialization</b>
+(encoding) of a Lua object into a string and decoding it back to a Lua
+object. This allows convenient storage and transport of <b>structured
+data</b>.
+</p>
+<p>
+The encoded data is in an <a href="#serialize_format">internal binary
+format</a>. The data can be stored in files, binary-transparent
+databases or transmitted to other LuaJIT instances across threads,
+processes or networks.
+</p>
+<p>
+Encoding speed can reach up to 1 Gigabyte/second on a modern desktop- or
+server-class system, even when serializing many small objects. Decoding
+speed is mostly constrained by object creation cost.
+</p>
+<p>
+The serializer handles most Lua types, common FFI number types and
+nested structures. Functions, thread objects, other FFI cdata and full
+userdata cannot be serialized (yet).
+</p>
+<p>
+The encoder serializes nested structures as trees. Multiple references
+to a single object will be stored separately and create distinct objects
+after decoding. Circular references cause an error.
+</p>
+
+<h3 id="serialize_methods">Serialization Functions and Methods</h3>
+
+<h3 id="buffer_encode"><tt>str = buffer.encode(obj)<br>
+buf = buf:encode(obj)</tt></h3>
+<p>
+Serializes (encodes) the Lua object <tt>obj</tt>. The stand-alone
+function returns a string <tt>str</tt>. The buffer method appends the
+encoding to the buffer.
+</p>
+<p>
+<tt>obj</tt> can be any of the supported Lua types — it doesn't
+need to be a Lua table.
+</p>
+<p>
+This function may throw an error when attempting to serialize
+unsupported object types, circular references or deeply nested tables.
+</p>
+
+<h3 id="buffer_decode"><tt>obj = buffer.decode(str)<br>
+obj = buf:decode()</tt></h3>
+<p>
+The stand-alone function deserializes (decodes) the string
+<tt>str</tt>, the buffer method deserializes one object from the
+buffer. Both return a Lua object <tt>obj</tt>.
+</p>
+<p>
+The returned object may be any of the supported Lua types —
+even <tt>nil</tt>.
+</p>
+<p>
+This function may throw an error when fed with malformed or incomplete
+encoded data. The stand-alone function throws when there's left-over
+data after decoding a single top-level object. The buffer method leaves
+any left-over data in the buffer.
+</p>
+<p>
+Attempting to deserialize an FFI type will throw an error, if the FFI
+library is not built-in or has not been loaded, yet.
+</p>
+
+<h3 id="serialize_options">Serialization Options</h3>
+<p>
+The <tt>options</tt> table passed to <tt>buffer.new()</tt> may contain
+the following members (all optional):
+</p>
+<ul>
+<li>
+<tt>dict</tt> is a Lua table holding a <b>dictionary of strings</b> that
+commonly occur as table keys of objects you are serializing. These keys
+are compactly encoded as indexes during serialization. A well-chosen
+dictionary saves space and improves serialization performance.
+</li>
+<li>
+<tt>metatable</tt> is a Lua table holding a <b>dictionary of metatables</b>
+for the table objects you are serializing.
+</li>
+</ul>
+<p>
+<tt>dict</tt> needs to be an array of strings and <tt>metatable</tt> needs
+to be an array of tables. Both starting at index 1 and without holes (no
+<tt>nil</tt> in between). The tables are anchored in the buffer object and
+internally modified into a two-way index (don't do this yourself, just pass
+a plain array). The tables must not be modified after they have been passed
+to <tt>buffer.new()</tt>.
+</p>
+<p>
+The <tt>dict</tt> and <tt>metatable</tt> tables used by the encoder and
+decoder must be the same. Put the most common entries at the front. Extend
+at the end to ensure backwards-compatibility — older encodings can
+then still be read. You may also set some indexes to <tt>false</tt> to
+explicitly drop backwards-compatibility. Old encodings that use these
+indexes will throw an error when decoded.
+</p>
+<p>
+Metatables that are not found in the <tt>metatable</tt> dictionary are
+ignored when encoding. Decoding returns a table with a <tt>nil</tt>
+metatable.
+</p>
+<p>
+Note: parsing and preparation of the options table is somewhat
+expensive. Create a buffer object only once and recycle it for multiple
+uses. Avoid mixing encoder and decoder buffers, since the
+<tt>buf:set()</tt> method frees the already allocated buffer space:
+</p>
+<pre class="code">
+local options = {
+ dict = { "commonly", "used", "string", "keys" },
+}
+local buf_enc = buffer.new(options)
+local buf_dec = buffer.new(options)
+
+local function encode(obj)
+ return buf_enc:reset():encode(obj):get()
+end
+
+local function decode(str)
+ return buf_dec:set(str):decode()
+end
+</pre>
+
+<h3 id="serialize_stream">Streaming Serialization</h3>
+<p>
+In some contexts, it's desirable to do piecewise serialization of large
+datasets, also known as <i>streaming</i>.
+</p>
+<p>
+This serialization format can be safely concatenated and supports streaming.
+Multiple encodings can simply be appended to a buffer and later decoded
+individually:
+</p>
+<pre class="code">
+local buf = buffer.new()
+buf:encode(obj1)
+buf:encode(obj2)
+local copy1 = buf:decode()
+local copy2 = buf:decode()
+</pre>
+<p>
+Here's how to iterate over a stream:
+</p>
+<pre class="code">
+while #buf ~= 0 do
+ local obj = buf:decode()
+ -- Do something with obj.
+end
+</pre>
+<p>
+Since the serialization format doesn't prepend a length to its encoding,
+network applications may need to transmit the length, too.
+</p>
+
+<h3 id="serialize_format">Serialization Format Specification</h3>
+<p>
+This serialization format is designed for <b>internal use</b> by LuaJIT
+applications. Serialized data is upwards-compatible and portable across
+all supported LuaJIT platforms.
+</p>
+<p>
+It's an <b>8-bit binary format</b> and not human-readable. It uses e.g.
+embedded zeroes and stores embedded Lua string objects unmodified, which
+are 8-bit-clean, too. Encoded data can be safely concatenated for
+streaming and later decoded one top-level object at a time.
+</p>
+<p>
+The encoding is reasonably compact, but tuned for maximum performance,
+not for minimum space usage. It compresses well with any of the common
+byte-oriented data compression algorithms.
+</p>
+<p>
+Although documented here for reference, this format is explicitly
+<b>not</b> intended to be a 'public standard' for structured data
+interchange across computer languages (like JSON or MessagePack). Please
+do not use it as such.
+</p>
+<p>
+The specification is given below as a context-free grammar with a
+top-level <tt>object</tt> as the starting point. Alternatives are
+separated by the <tt>|</tt> symbol and <tt>*</tt> indicates repeats.
+Grouping is implicit or indicated by <tt>{…}</tt>. Terminals are
+either plain hex numbers, encoded as bytes, or have a <tt>.format</tt>
+suffix.
+</p>
+<pre>
+object → nil | false | true
+ | null | lightud32 | lightud64
+ | int | num | tab | tab_mt
+ | int64 | uint64 | complex
+ | string
+
+nil → 0x00
+false → 0x01
+true → 0x02
+
+null → 0x03 // NULL lightuserdata
+lightud32 → 0x04 data.I // 32 bit lightuserdata
+lightud64 → 0x05 data.L // 64 bit lightuserdata
+
+int → 0x06 int.I // int32_t
+num → 0x07 double.L
+
+tab → 0x08 // Empty table
+ | 0x09 h.U h*{object object} // Key/value hash
+ | 0x0a a.U a*object // 0-based array
+ | 0x0b a.U h.U a*object h*{object object} // Mixed
+ | 0x0c a.U (a-1)*object // 1-based array
+ | 0x0d a.U h.U (a-1)*object h*{object object} // Mixed
+tab_mt → 0x0e (index-1).U tab // Metatable dict entry
+
+int64 → 0x10 int.L // FFI int64_t
+uint64 → 0x11 uint.L // FFI uint64_t
+complex → 0x12 re.L im.L // FFI complex
+
+string → (0x20+len).U len*char.B
+ | 0x0f (index-1).U // String dict entry
+
+.B = 8 bit
+.I = 32 bit little-endian
+.L = 64 bit little-endian
+.U = prefix-encoded 32 bit unsigned number n:
+ 0x00..0xdf → n.B
+ 0xe0..0x1fdf → (0xe0|(((n-0xe0)>>8)&0x1f)).B ((n-0xe0)&0xff).B
+ 0x1fe0.. → 0xff n.I
+</pre>
+
+<h2 id="error">Error handling</h2>
+<p>
+Many of the buffer methods can throw an error. Out-of-memory or usage
+errors are best caught with an outer wrapper for larger parts of code.
+There's not much one can do after that, anyway.
+</p>
+<p>
+OTOH, you may want to catch some errors individually. Buffer methods need
+to receive the buffer object as the first argument. The Lua colon-syntax
+<tt>obj:method()</tt> does that implicitly. But to wrap a method with
+<tt>pcall()</tt>, the arguments need to be passed like this:
+</p>
+<pre class="code">
+local ok, err = pcall(buf.encode, buf, obj)
+if not ok then
+ -- Handle error in err.
+end
+</pre>
+
+<h2 id="ffi_caveats">FFI caveats</h2>
+<p>
+The string buffer library has been designed to work well together with
+the FFI library. But due to the low-level nature of the FFI library,
+some care needs to be taken:
+</p>
+<p>
+First, please remember that FFI pointers are zero-indexed. The space
+returned by <tt>buf:reserve()</tt> and <tt>buf:ref()</tt> starts at the
+returned pointer and ends before <tt>len</tt> bytes after that.
+</p>
+<p>
+I.e. the first valid index is <tt>ptr[0]</tt> and the last valid index
+is <tt>ptr[len-1]</tt>. If the returned length is zero, there's no valid
+index at all. The returned pointer may even be <tt>NULL</tt>.
+</p>
+<p>
+The space pointed to by the returned pointer is only valid as long as
+the buffer is not modified in any way (neither append, nor consume, nor
+reset, etc.). The pointer is also not a GC anchor for the buffer object
+itself.
+</p>
+<p>
+Buffer data is only guaranteed to be byte-aligned. Casting the returned
+pointer to a data type with higher alignment may cause unaligned
+accesses. It depends on the CPU architecture whether this is allowed or
+not (it's always OK on x86/x64 and mostly OK on other modern
+architectures).
+</p>
+<p>
+FFI pointers or references do not count as GC anchors for an underlying
+object. E.g. an <tt>array</tt> allocated with <tt>ffi.new()</tt> is
+anchored by <tt>buf:set(array, len)</tt>, but not by
+<tt>buf:set(array+offset, len)</tt>. The addition of the offset
+creates a new pointer, even when the offset is zero. In this case, you
+need to make sure there's still a reference to the original array as
+long as its contents are in use by the buffer.
+</p>
+<p>
+Even though each LuaJIT VM instance is single-threaded (but you can
+create multiple VMs), FFI data structures can be accessed concurrently.
+Be careful when reading/writing FFI cdata from/to buffers to avoid
+concurrent accesses or modifications. In particular, the memory
+referenced by <tt>buf:set(cdata, len)</tt> must not be modified
+while buffer readers are working on it. Shared, but read-only memory
+mappings of files are OK, but only if the file does not change.
+</p>
+<br class="flush">
+</div>
+<div id="foot">
+<hr class="hide">
+Copyright © 2005-2025
+<span class="noprint">
+·
+<a href="contact.html">Contact</a>
+</span>
+</div>
+</body>
+</html>
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_buffer.html
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_c_api.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_c_api.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_c_api.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Lua/C API Extensions</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -11,7 +10,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Lua/C API Extensions</h1>
@@ -20,7 +19,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -38,6 +37,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a class="current" href="ext_c_api.html">Lua/C API</a>
@@ -45,18 +46,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -91,8 +85,8 @@
</p>
<p>
The third argument specifies the mode, which is 'or'ed with a flag.
-The flag can be <tt>LUAJIT_MODE_OFF</tt> to turn a feature on,
-<tt>LUAJIT_MODE_ON</tt> to turn a feature off, or
+The flag can be <tt>LUAJIT_MODE_OFF</tt> to turn a feature off,
+<tt>LUAJIT_MODE_ON</tt> to turn a feature on, or
<tt>LUAJIT_MODE_FLUSH</tt> to flush cached code.
</p>
<p>
@@ -111,7 +105,7 @@
This sets the mode for the function at the stack index <tt>idx</tt> or
the parent of the calling function (<tt>idx = 0</tt>). It either
enables JIT compilation for a function, disables it and flushes any
-already compiled code or only flushes already compiled code. This
+already compiled code, or only flushes already compiled code. This
applies recursively to all sub-functions of the function with
<tt>LUAJIT_MODE_ALLFUNC</tt> or only to the sub-functions with
<tt>LUAJIT_MODE_ALLSUBFUNC</tt>.
@@ -130,7 +124,7 @@
This mode defines a wrapper function for calls to C functions. If
called with <tt>LUAJIT_MODE_ON</tt>, the stack index at <tt>idx</tt>
must be a <tt>lightuserdata</tt> object holding a pointer to the wrapper
-function. From now on all C functions are called through the wrapper
+function. From now on, all C functions are called through the wrapper
function. If called with <tt>LUAJIT_MODE_OFF</tt> this mode is turned
off and all C functions are directly called.
</p>
@@ -179,7 +173,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>FFI Library</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -11,7 +10,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>FFI Library</h1>
@@ -20,7 +19,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -38,6 +37,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -45,18 +46,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -161,7 +155,7 @@
<h2 id="cdata">Motivating Example: Using C Data Structures</h2>
<p>
The FFI library allows you to create and access C data
-structures. Of course the main use for this is for interfacing with
+structures. Of course, the main use for this is for interfacing with
C functions. But they can be used stand-alone, too.
</p>
<p>
@@ -173,7 +167,7 @@
both a substantial memory overhead as well as a performance overhead.
</p>
<p>
-Here's a sketch of a library that operates on color images plus a
+Here's a sketch of a library that operates on color images, plus a
simple benchmark. First, the plain Lua version:
</p>
<pre class="code">
@@ -188,7 +182,7 @@
return img
end
-local function image_to_grey(img, n)
+local function image_to_gray(img, n)
for i=1,n do
local y = floor(0.3*img[i].red + 0.59*img[i].green + 0.11*img[i].blue)
img[i].red = y; img[i].green = y; img[i].blue = y
@@ -198,14 +192,14 @@
local N = 400*400
local img = image_ramp_green(N)
for i=1,1000 do
- image_to_grey(img, N)
+ image_to_gray(img, N)
end
</pre>
<p>
This creates a table with 160.000 pixels, each of which is a table
-holding four number values in the range of 0-255. First an image with
+holding four number values in the range of 0-255. First, an image with
a green ramp is created (1D for simplicity), then the image is
-converted to greyscale 1000 times. Yes, that's silly, but I was in
+converted to grayscale 1000 times. Yes, that's silly, but I was in
need of a simple example ...
</p>
<p>
@@ -312,7 +306,7 @@
~1.7x). Switching to a struct-of-arrays would help, too.
</p>
<p style="font-size: 8pt;">
-However the resulting code would be less idiomatic and rather
+However, the resulting code would be less idiomatic and rather
error-prone. And it still doesn't get even close to the performance of
the FFI version of the code. Also, high-level data structures cannot
be easily passed to other C functions, especially I/O functions,
@@ -322,7 +316,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_api.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_api.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_api.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>ffi.* API Functions</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -16,7 +15,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1><tt>ffi.*</tt> API Functions</h1>
@@ -25,7 +24,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -43,6 +42,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -50,18 +51,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -125,7 +119,7 @@
declaration may be omitted.
</p>
<p>
-Please note that external symbols are only <em>declared</em>, but they
+Please note, that external symbols are only <em>declared</em>, but they
are <em>not bound</em> to any specific address, yet. Binding is
achieved with C library namespaces (see below).
</p>
@@ -213,7 +207,7 @@
<tt>ffi.typeof()</tt>. Then use the ctype as a constructor repeatedly.
</p>
<p style="font-size: 8pt;">
-Please note that an anonymous <tt>struct</tt> declaration implicitly
+Please note, that an anonymous <tt>struct</tt> declaration implicitly
creates a new and distinguished ctype every time you use it for
<tt>ffi.new()</tt>. This is probably <b>not</b> what you want,
especially if you create more than one cdata object. Different anonymous
@@ -260,12 +254,12 @@
contents of an <tt>__index</tt> table (if any) may be modified
afterwards. The associated metatable automatically applies to all uses
of this type, no matter how the objects are created or where they
-originate from. Note that pre-defined operations on types have
-precedence (e.g. declared field names cannot be overriden).
+originate from. Note that predefined operations on types have
+precedence (e.g. declared field names cannot be overridden).
</p>
<p>
All standard Lua metamethods are implemented. These are called directly,
-without shortcuts and on any mix of types. For binary operations, the
+without shortcuts, and on any mix of types. For binary operations, the
left operand is checked first for a valid ctype metamethod. The
<tt>__gc</tt> metamethod only applies to <tt>struct</tt>/<tt>union</tt>
types and performs an implicit <a href="#ffi_gc"><tt>ffi.gc()</tt></a>
@@ -469,6 +463,10 @@
<tr class="odd">
<td class="abiparam">win</td><td class="abidesc">Windows variant of the standard ABI</td></tr>
<tr class="even">
+<td class="abiparam">pauth</td><td class="abidesc">Pointer authentication ABI</td></tr>
+<tr class="odd">
+<td class="abiparam">uwp</td><td class="abidesc">Universal Windows Platform</td></tr>
+<tr class="even">
<td class="abiparam">gc64</td><td class="abidesc">64 bit GC references</td></tr>
</table>
@@ -494,7 +492,7 @@
<p>
Free the resources associated with a callback. The associated Lua
function is unanchored and may be garbage collected. The callback
-function pointer is no longer valid and must not be called anymore
+function pointer is no longer valid and must not be called again
(it may be reused by a subsequently created callback).
</p>
@@ -560,7 +558,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_semantics.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_semantics.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_semantics.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>FFI Semantics</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -16,7 +15,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>FFI Semantics</h1>
@@ -25,7 +24,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -43,6 +42,8 @@
<a class="current" href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -50,18 +51,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -92,7 +86,7 @@
functions</a> to declare C types or external symbols.
</p>
<p>
-It's only purpose is to parse C declarations, as found e.g. in
+Its only purpose is to parse C declarations, as found e.g. in
C header files. Although it does evaluate constant expressions,
it's <em>not</em> a C compiler. The body of <tt>inline</tt>
C function definitions is simply ignored.
@@ -169,7 +163,7 @@
</ul>
<p>
-The following C types are pre-defined by the C parser (like
+The following C types are predefined by the C parser (like
a <tt>typedef</tt>, except re-declarations will be ignored):
</p>
<ul>
@@ -446,6 +440,19 @@
conversion rule applies. A vararg C function expecting an integer
will see a garbled or uninitialized value.
</p>
+<p>
+Note: this is the only place where creating a boxed scalar number type is
+actually useful. <b>Never use <tt>ffi.new("int")</tt>, <tt>ffi.new("float")</tt>
+etc. anywhere else!</b>
+</p>
+<p style="font-size: 8pt;">
+Ditto for <tt>ffi.cast()</tt>. Explicitly boxing scalars <b>does not</b>
+improve performance or force <tt>int</tt> or <tt>float</tt> arithmetic! It
+just adds costly boxing, unboxing and conversions steps. And it may lead
+to surprise results, because
+<a href="#cdata_arith">cdata arithmetic on scalar numbers</a>
+is always performed on 64 bit integers.
+</p>
<h2 id="init">Initializers</h2>
<p>
@@ -587,9 +594,9 @@
<h2 id="cdata_ops">Operations on cdata Objects</h2>
<p>
-All of the standard Lua operators can be applied to cdata objects or a
+All standard Lua operators can be applied to cdata objects or a
mix of a cdata object and another Lua object. The following list shows
-the pre-defined operations.
+the predefined operations.
</p>
<p>
Reference types are dereferenced <em>before</em> performing each of
@@ -597,7 +604,7 @@
C type pointed to by the reference.
</p>
<p>
-The pre-defined operations are always tried first before deferring to a
+The predefined operations are always tried first before deferring to a
metamethod or index table (if any) for the corresponding ctype (except
for <tt>__new</tt>). An error is raised if the metamethod lookup or
index table lookup fails.
@@ -647,7 +654,7 @@
</ul>
<p>
A ctype object can be indexed with a string key, too. The only
-pre-defined operation is reading scoped constants of
+predefined operation is reading scoped constants of
<tt>struct</tt>/<tt>union</tt> types. All other accesses defer
to the corresponding metamethods or index tables (if any).
</p>
@@ -660,7 +667,7 @@
<p>
As a consequence, the <em>elements</em> of complex numbers and
vectors are immutable. But the elements of an aggregate holding these
-types <em>may</em> be modified of course. I.e. you cannot assign to
+types <em>may</em> be modified, of course. I.e. you cannot assign to
<tt>foo.c.im</tt>, but you can assign a (newly created) complex number
to <tt>foo.c</tt>.
</p>
@@ -678,9 +685,9 @@
<a href="ext_ffi_api.html#ffi_new">constructor</a>. This is equivalent
to <tt>ffi.new(ct, ...)</tt>, unless a <tt>__new</tt> metamethod is
defined. The <tt>__new</tt> metamethod is called with the ctype object
-plus any other arguments passed to the contructor. Note that you have to
-use <tt>ffi.new</tt> inside of it, since calling <tt>ct(...)</tt> would
-cause infinite recursion.</li>
+plus any other arguments passed to the constructor. Note that you have to
+use <tt>ffi.new</tt> inside the metamethod, since calling <tt>ct(...)</tt>
+would cause infinite recursion.</li>
<li><b>C function call</b>: a cdata function or cdata function
pointer can be called. The passed arguments are
@@ -691,7 +698,7 @@
C function is called and the return value (if any) is
<a href="#convert_tolua">converted to a Lua object</a>.<br>
On Windows/x86 systems, <tt>__stdcall</tt> functions are automatically
-detected and a function declared as <tt>__cdecl</tt> (the default) is
+detected, and a function declared as <tt>__cdecl</tt> (the default) is
silently fixed up after the first call.</li>
</ul>
@@ -701,7 +708,7 @@
<li><b>Pointer arithmetic</b>: a cdata pointer/array and a cdata
number or a Lua number can be added or subtracted. The number must be
-on the right hand side for a subtraction. The result is a pointer of
+on the right-hand side for a subtraction. The result is a pointer of
the same type with an address plus or minus the number value
multiplied by the element size in bytes. An error is raised if the
element size is undefined.</li>
@@ -716,7 +723,7 @@
minus) can be applied to two cdata numbers, or a cdata number and a
Lua number. If one of them is an <tt>uint64_t</tt>, the other side is
converted to an <tt>uint64_t</tt> and an unsigned arithmetic operation
-is performed. Otherwise both sides are converted to an
+is performed. Otherwise, both sides are converted to an
<tt>int64_t</tt> and a signed arithmetic operation is performed. The
result is a boxed 64 bit cdata object.<br>
@@ -763,7 +770,7 @@
<li><b>64 bit integer comparison</b>: two cdata numbers, or a
cdata number and a Lua number can be compared with each other. If one
of them is an <tt>uint64_t</tt>, the other side is converted to an
-<tt>uint64_t</tt> and an unsigned comparison is performed. Otherwise
+<tt>uint64_t</tt> and an unsigned comparison is performed. Otherwise,
both sides are converted to an <tt>int64_t</tt> and a signed
comparison is performed.<br>
@@ -788,9 +795,9 @@
A cdata object is treated like any other garbage-collected object and
is hashed and compared by its address for table indexing. Since
there's no interning for cdata value types, the same value may be
-boxed in different cdata objects with different addresses. Thus
+boxed in different cdata objects with different addresses. Thus,
<tt>t[1LL+1LL]</tt> and <tt>t[2LL]</tt> usually <b>do not</b> point to
-the same hash slot and they certainly <b>do not</b> point to the same
+the same hash slot, and they certainly <b>do not</b> point to the same
hash slot as <tt>t[2]</tt>.
</p>
<p>
@@ -812,7 +819,7 @@
One obvious benefit: <tt>t[tonumber(2LL)]</tt> <b>does</b> point to
the same slot as <tt>t[2]</tt>.</li>
-<li>Otherwise use either <tt>tostring()</tt> on 64 bit integers
+<li>Otherwise, use either <tt>tostring()</tt> on 64 bit integers
or complex numbers or combine multiple fields of a cdata aggregate to
a Lua string (e.g. with
<a href="ext_ffi_api.html#ffi_string"><tt>ffi.string()</tt></a>). Then
@@ -820,7 +827,7 @@
<li>Create your own specialized hash table implementation using the
C types provided by the FFI library, just like you would in
-C code. Ultimately this may give much better performance than the
+C code. Ultimately, this may give much better performance than the
other alternatives or what a generic by-value hash table could
possibly provide.</li>
@@ -864,7 +871,7 @@
<p>
The main use for parameterized types are libraries implementing abstract
data types
-(<a href="http://www.freelists.org/post/luajit/ffi-type-of-pointer-to,8"><span class="ext">»</span> example</a>),
+(<a href="https://www.freelists.org/post/luajit/ffi-type-of-pointer-to,8"><span class="ext">»</span> example</a>),
similar to what can be achieved with C++ template metaprogramming.
Another use case are derived types of anonymous structs, which avoids
pollution of the global struct namespace.
@@ -886,7 +893,7 @@
the end of the next GC cycle).
</p>
<p>
-Please note that pointers themselves are cdata objects, however they
+Please note, that pointers themselves are cdata objects, however they
are <b>not</b> followed by the garbage collector. So e.g. if you
assign a cdata array to a pointer, you must keep the cdata object
holding the array alive as long as the pointer is still in use:
@@ -935,18 +942,18 @@
</p>
<p>
This can happen implicitly due to the usual conversions, e.g. when
-passing a Lua function to a function pointer argument. Or you can use
+passing a Lua function to a function pointer argument. Or, you can use
<tt>ffi.cast()</tt> to explicitly cast a Lua function to a
C function pointer.
</p>
<p>
-Currently only certain C function types can be used as callback
+Currently, only certain C function types can be used as callback
functions. Neither C vararg functions nor functions with
pass-by-value aggregate argument or result types are supported. There
-are no restrictions for the kind of Lua functions that can be called
+are no restrictions on the kind of Lua functions that can be called
from the callback — no checks for the proper number of arguments
are made. The return value of the Lua function will be converted to the
-result type and an error will be thrown for invalid conversions.
+result type, and an error will be thrown for invalid conversions.
</p>
<p>
It's allowed to throw errors across a callback invocation, but it's not
@@ -1007,7 +1014,7 @@
<tt>__stdcall</tt> calls <em>to</em> Windows functions.
</p>
<p>
-For some use cases it's necessary to free up the resources or to
+For some use cases, it's necessary to free up the resources or to
dynamically redirect callbacks. Use an explicit cast to a
C function pointer and keep the resulting cdata object. Then use
the <a href="ext_ffi_api.html#callback_free"><tt>cb:free()</tt></a>
@@ -1060,7 +1067,7 @@
</p>
<p>
For new designs <b>avoid push-style APIs</b>: a C function repeatedly
-calling a callback for each result. Instead <b>use pull-style APIs</b>:
+calling a callback for each result. Instead, <b>use pull-style APIs</b>:
call a C function repeatedly to get a new result. Calls from Lua
to C via the FFI are much faster than the other way round. Most well-designed
libraries already use pull-style APIs (read/write, get/put).
@@ -1079,7 +1086,7 @@
</p>
<p>
Indexing a C library namespace object with a symbol name (a Lua
-string) automatically binds it to the library. First the symbol type
+string) automatically binds it to the library. First, the symbol type
is resolved — it must have been declared with
<a href="ext_ffi_api.html#ffi_cdef"><tt>ffi.cdef</tt></a>. Then the
symbol address is resolved by searching for the symbol name in the
@@ -1134,7 +1141,7 @@
namespace objects and to the strings used to index it. This
effectively turns function cdata objects into constants. It's not
useful and actually counter-productive to explicitly cache these
-function objects, e.g. <tt>local strlen = ffi.C.strlen</tt>. OTOH it
+function objects, e.g. <tt>local strlen = ffi.C.strlen</tt>. OTOH, it
<em>is</em> useful to cache the namespace itself, e.g. <tt>local C =
ffi.C</tt>.
</p>
@@ -1159,7 +1166,7 @@
interoperability with C code. Adding extra safety measures, like
bounds checks, would be futile. There's no way to detect
misdeclarations of C functions, since shared libraries only
-provide symbol names, but no type information. Likewise there's no way
+provide symbol names, but no type information. Likewise, there's no way
to infer the valid range of indexes for a returned pointer.
</p>
<p>
@@ -1166,7 +1173,7 @@
Again: the FFI library is a low-level library. This implies it needs
to be used with care, but it's flexibility and performance often
outweigh this concern. If you're a C or C++ developer, it'll be easy
-to apply your existing knowledge. OTOH writing code for the FFI
+to apply your existing knowledge. OTOH, writing code for the FFI
library is not for the faint of heart and probably shouldn't be the
first exercise for someone with little experience in Lua, C or C++.
</p>
@@ -1194,7 +1201,7 @@
<li>C declarations are not passed through a C pre-processor,
yet.</li>
<li>The C parser is able to evaluate most constant expressions
-commonly found in C header files. However it doesn't handle the
+commonly found in C header files. However, it doesn't handle the
full range of C expression semantics and may fail for some
obscure constructs.</li>
<li><tt>static const</tt> declarations only work for integer types
@@ -1225,7 +1232,7 @@
<li>Table initializers.</li>
<li>Initialization of nested <tt>struct</tt>/<tt>union</tt> types.</li>
<li>Non-default initialization of VLA/VLS or large C types
-(> 128 bytes or > 16 array elements.</li>
+(> 128 bytes or > 16 array elements).</li>
<li>Bitfield initializations.</li>
<li>Pointer differences for element sizes that are not a power of
two.</li>
@@ -1252,7 +1259,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_tutorial.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_tutorial.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_ffi_tutorial.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>FFI Tutorial</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -18,7 +17,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>FFI Tutorial</h1>
@@ -27,7 +26,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -45,6 +44,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -52,18 +53,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -89,7 +83,7 @@
local ffi = require("ffi")
</pre>
<p>
-Please note this doesn't define an <tt>ffi</tt> variable in the table
+Please note, this doesn't define an <tt>ffi</tt> variable in the table
of globals — you really need to use the local variable. The
<tt>require</tt> function ensures the library is only loaded once.
</p>
@@ -198,7 +192,7 @@
<span class="mark">⑤</span> The <tt>poll()</tt>
function takes a couple more arguments we're not going to use. You can
simply use <tt>nil</tt> to pass a <tt>NULL</tt> pointer and <tt>0</tt>
-for the <tt>nfds</tt> parameter. Please note that the
+for the <tt>nfds</tt> parameter. Please note, that the
number <tt>0</tt> <em>does not convert to a pointer value</em>,
unlike in C++. You really have to pass pointers to pointer arguments
and numbers to number arguments.
@@ -222,7 +216,7 @@
<h2 id="zlib">Accessing the zlib Compression Library</h2>
<p>
The following code shows how to access the <a
-href="http://zlib.net/">zlib</a> compression library from Lua code.
+href="https://zlib.net/"><span class="ext">»</span> zlib</a> compression library from Lua code.
We'll define two convenience wrapper functions that take a string and
compress or uncompress it to another string:
</p>
@@ -295,17 +289,17 @@
<p>
<span class="mark">①</span> This defines some of the
C functions provided by zlib. For the sake of this example, some
-type indirections have been reduced and it uses the pre-defined
+type indirections have been reduced and it uses the predefined
fixed-size integer types, while still adhering to the zlib API/ABI.
</p>
<p>
<span class="mark">②</span> This loads the zlib shared
-library. On POSIX systems it's named <tt>libz.so</tt> and usually
+library. On POSIX systems, it's named <tt>libz.so</tt> and usually
comes pre-installed. Since <tt>ffi.load()</tt> automatically adds any
missing standard prefixes/suffixes, we can simply load the
<tt>"z"</tt> library. On Windows it's named <tt>zlib1.dll</tt> and
you'll have to download it first from the
-<a href="http://zlib.net/"><span class="ext">»</span> zlib site</a>. The check for
+<a href="https://zlib.net/"><span class="ext">»</span> zlib site</a>. The check for
<tt>ffi.os</tt> makes sure we pass the right name to
<tt>ffi.load()</tt>.
</p>
@@ -328,7 +322,7 @@
<p>
In C you'd pass in the address of a local variable
(<tt>&buflen</tt>). But since there's no address-of operator in
-Lua, we'll just pass in a one-element array. Conveniently it can be
+Lua, we'll just pass in a one-element array. Conveniently, it can be
initialized with the maximum buffer size in one step. Calling the
actual <tt>zlib.compress2</tt> function is then straightforward.
</p>
@@ -352,7 +346,7 @@
<span class="mark">⑥</span> The <tt>uncompress</tt>
functions does the exact opposite of the <tt>compress</tt> function.
The compressed data doesn't include the size of the original string,
-so this needs to be passed in. Otherwise no surprises here.
+so this needs to be passed in. Otherwise, no surprises here.
</p>
<p>
<span class="mark">⑦</span> The code, that makes use
@@ -386,7 +380,7 @@
wherever you'd want to use a number. That's why we get a away with
passing <tt>n</tt> to <tt>ffi.string()</tt> above. But other Lua
library functions or modules don't know how to deal with this. So for
-maximum portability one needs to use <tt>tonumber()</tt> on returned
+maximum portability, one needs to use <tt>tonumber()</tt> on returned
<tt>long</tt> results before passing them on. Otherwise the
application might work on some systems, but would fail in a POSIX/x64
environment.
@@ -458,7 +452,7 @@
</p>
<p>
<span class="mark">④</span> If we run out of operators, we can
-define named methods, too. Here the <tt>__index</tt> table defines an
+define named methods, too. Here, the <tt>__index</tt> table defines an
<tt>area</tt> function. For custom indexing needs, one might want to
define <tt>__index</tt> and <tt>__newindex</tt> <em>functions</em> instead.
</p>
@@ -472,13 +466,13 @@
apply to any and all uses of this type.
</p>
<p>
-Please note that the association with a metatable is permanent and
+Please note, that the association with a metatable is permanent and
<b>the metatable must not be modified afterwards!</b> Ditto for the
<tt>__index</tt> table.
</p>
<p>
<span class="mark">⑥</span> Here are some simple usage examples
-for the point type and their expected results. The pre-defined
+for the point type and their expected results. The predefined
operations (such as <tt>a.x</tt>) can be freely mixed with the newly
defined metamethods. Note that <tt>area</tt> is a method and must be
called with the Lua syntax for methods: <tt>a:area()</tt>, not
@@ -487,7 +481,7 @@
<p>
The C type metamethod mechanism is most useful when used in
conjunction with C libraries that are written in an object-oriented
-style. Creators return a pointer to a new instance and methods take an
+style. Creators return a pointer to a new instance, and methods take an
instance pointer as the first argument. Sometimes you can just point
<tt>__index</tt> to the library namespace and <tt>__gc</tt> to the
destructor and you're done. But often enough you'll want to add
@@ -573,7 +567,7 @@
</pre>
<p>
This turns them into indirect calls and generates bigger and slower
-machine code. Instead you'll want to cache the namespace itself and
+machine code. Instead, you'll want to cache the namespace itself and
rely on the JIT compiler to eliminate the lookups:
</p>
<pre class="code">
@@ -593,7 +587,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_jit.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_jit.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_jit.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>jit.* Library</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -11,7 +10,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1><tt>jit.*</tt> Library</h1>
@@ -20,7 +19,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -38,6 +37,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a class="current" href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -45,18 +46,11 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -141,7 +135,9 @@
<h3 id="jit_version_num"><tt>jit.version_num</tt></h3>
<p>
Contains the version number of the LuaJIT core. Version xx.yy.zz
-is represented by the decimal number xxyyzz.
+is represented by the decimal number xxyyzz.<br>
+<b>DEPRECATED after the switch to
+<a href="https://luajit.org/status.html#release"><span class="ext">»</span> rolling releases</a>. zz is frozen at 99.</b>
</p>
<h3 id="jit_os"><tt>jit.os</tt></h3>
@@ -153,12 +149,12 @@
<h3 id="jit_arch"><tt>jit.arch</tt></h3>
<p>
Contains the target architecture name:
-"x86", "x64", "arm", "arm64", "ppc", "mips" or "mips64".
+"x86", "x64", "arm", "arm64", "arm64be", "ppc", "mips", "mipsel", "mips64", "mips64el", "mips64r6", "mips64r6el".
</p>
<h2 id="jit_opt"><tt>jit.opt.*</tt> — JIT compiler optimization control</h2>
<p>
-This sub-module provides the backend for the <tt>-O</tt> command line
+This submodule provides the backend for the <tt>-O</tt> command line
option.
</p>
<p>
@@ -178,7 +174,7 @@
<h2 id="jit_util"><tt>jit.util.*</tt> — JIT compiler introspection</h2>
<p>
-This sub-module holds functions to introspect the bytecode, generated
+This submodule holds functions to introspect the bytecode, generated
traces, the IR and the generated machine code. The functionality
provided by this module is still in flux and therefore undocumented.
</p>
@@ -191,7 +187,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_profiler.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_profiler.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/ext_profiler.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Profiler</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -11,7 +10,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Profiler</h1>
@@ -20,7 +19,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -38,6 +37,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -45,18 +46,11 @@
<a class="current" href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
@@ -162,7 +156,7 @@
Combinations of <tt>v/z</tt> with <tt>f/F/l</tt> produce two-level
views, e.g. <tt>-jp=vf</tt> or <tt>-jp=fv</tt>. This shows the time
spent in a VM state or zone vs. hotspots. This can be used to answer
-questions like "Which time consuming functions are only interpreted?" or
+questions like "Which time-consuming functions are only interpreted?" or
"What's the garbage collector overhead for a specific function?".
</p>
<p>
@@ -221,7 +215,7 @@
This module can be used to implement your own higher-level profiler.
A typical profiling run starts the profiler, captures stack dumps in
the profiler callback, adds them to a hash table to aggregate the number
-of samples, stops the profiler and then analyzes all of the captured
+of samples, stops the profiler and then analyzes all captured
stack dumps. Other parameters can be sampled in the profiler callback,
too. But it's important not to spend too much time in the callback,
since this may skew the statistics.
@@ -275,9 +269,9 @@
formatted according to the <tt>fmt</tt> argument:
</p>
<ul>
-<li><tt>p</tt> — Preserve the full path for module names. Otherwise
+<li><tt>p</tt> — Preserve the full path for module names. Otherwise,
only the file name is used.</li>
-<li><tt>f</tt> — Dump the function name if it can be derived. Otherwise
+<li><tt>f</tt> — Dump the function name if it can be derived. Otherwise,
use module:line.</li>
<li><tt>F</tt> — Ditto, but dump module:name.</li>
<li><tt>l</tt> — Dump module:line.</li>
@@ -355,7 +349,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/extensions.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/extensions.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/extensions.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Extensions</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -28,7 +27,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Extensions</h1>
@@ -37,7 +36,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -55,6 +54,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -62,26 +63,20 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
+
<p>
LuaJIT is fully upwards-compatible with Lua 5.1. It supports all
-<a href="http://www.lua.org/manual/5.1/manual.html#5"><span class="ext">»</span> standard Lua
+<a href="https://www.lua.org/manual/5.1/manual.html#5"><span class="ext">»</span> standard Lua
library functions</a> and the full set of
-<a href="http://www.lua.org/manual/5.1/manual.html#3"><span class="ext">»</span> Lua/C API
+<a href="https://www.lua.org/manual/5.1/manual.html#3"><span class="ext">»</span> Lua/C API
functions</a>.
</p>
<p>
@@ -92,7 +87,7 @@
</p>
<p>
LuaJIT extends the standard Lua VM with new functionality and adds
-several extension modules. Please note this page is only about
+several extension modules. Please note, this page is only about
<em>functional</em> enhancements and not about performance enhancements,
such as the optimized VM, the faster interpreter or the JIT compiler.
</p>
@@ -105,7 +100,7 @@
<h3 id="bit"><tt>bit.*</tt> — Bitwise operations</h3>
<p>
LuaJIT supports all bitwise operations as defined by
-<a href="http://bitop.luajit.org"><span class="ext">»</span> Lua BitOp</a>:
+<a href="https://bitop.luajit.org"><span class="ext">»</span> Lua BitOp</a>:
</p>
<pre class="code">
bit.tobit bit.tohex bit.bnot bit.band bit.bor bit.bxor
@@ -114,7 +109,7 @@
<p>
This module is a LuaJIT built-in — you don't need to download or
install Lua BitOp. The Lua BitOp site has full documentation for all
-<a href="http://bitop.luajit.org/api.html"><span class="ext">»</span> Lua BitOp API functions</a>.
+<a href="https://bitop.luajit.org/api.html"><span class="ext">»</span> Lua BitOp API functions</a>.
The FFI adds support for
<a href="ext_ffi_semantics.html#cdata_arith">64 bit bitwise operations</a>,
using the same API functions.
@@ -165,7 +160,7 @@
which is called in a protected context.
</p>
-<h3 id="load"><tt>loadfile()</tt> etc. handle UTF-8 source code</h3>
+<h3 id="load"><tt>load*()</tt> handle UTF-8 source code</h3>
<p>
Non-ASCII characters are handled transparently by the Lua source code parser.
This allows the use of UTF-8 characters in identifiers and strings.
@@ -172,6 +167,26 @@
A UTF-8 BOM is skipped at the start of the source code.
</p>
+<h3 id="load_mode"><tt>load*()</tt> add a mode parameter</h3>
+<p>
+As an extension from Lua 5.2, the functions <tt>loadstring()</tt>,
+<tt>loadfile()</tt> and (new) <tt>load()</tt> add an optional
+<tt>mode</tt> parameter.
+</p>
+<p>
+The default mode string is <tt>"bt"</tt>, which allows loading of both
+source code and bytecode. Use <tt>"t"</tt> to allow only source code
+or <tt>"b"</tt> to allow only bytecode to be loaded.
+</p>
+<p>
+By default, the <tt>load*</tt> functions generate the native bytecode format.
+For cross-compilation purposes, add <tt>W</tt> to the mode string to
+force the 32 bit format and <tt>X</tt> to force the 64 bit format.
+Add both to force the opposite format. Note that non-native bytecode
+generated by <tt>load*</tt> cannot be run, but can still be passed
+to <tt>string.dump</tt>.
+</p>
+
<h3 id="tostring"><tt>tostring()</tt> etc. canonicalize NaN and ±Inf</h3>
<p>
All number-to-string conversions consistently convert non-finite numbers
@@ -191,28 +206,34 @@
numbers (e.g. <tt>0x1.5p-3</tt>).
</p>
-<h3 id="string_dump"><tt>string.dump(f [,strip])</tt> generates portable bytecode</h3>
+<h3 id="string_dump"><tt>string.dump(f [,mode])</tt> generates portable bytecode</h3>
<p>
An extra argument has been added to <tt>string.dump()</tt>. If set to
-<tt>true</tt>, 'stripped' bytecode without debug information is
-generated. This speeds up later bytecode loading and reduces memory
-usage. See also the
+<tt>true</tt> or to a string which contains the character <tt>s</tt>,
+'stripped' bytecode without debug information is generated. This speeds
+up later bytecode loading and reduces memory usage. See also the
<a href="running.html#opt_b"><tt>-b</tt> command line option</a>.
</p>
<p>
The generated bytecode is portable and can be loaded on any architecture
-that LuaJIT supports, independent of word size or endianess. However the
-bytecode compatibility versions must match. Bytecode stays compatible
-for dot releases (x.y.0 → x.y.1), but may change with major or
-minor releases (2.0 → 2.1) or between any beta release. Foreign
-bytecode (e.g. from Lua 5.1) is incompatible and cannot be loaded.
+that LuaJIT supports. However, the bytecode compatibility versions must
+match. Bytecode only stays compatible within a major+minor version
+(x.y.aaa → x.y.bbb), except for development branches. Foreign bytecode
+(e.g. from Lua 5.1) is incompatible and cannot be loaded.
</p>
<p>
Note: <tt>LJ_GC64</tt> mode requires a different frame layout, which implies
-a different, incompatible bytecode format for ports that use this mode (e.g.
-ARM64 or MIPS64) or when explicitly enabled for x64. This may be rectified
-in the future.
+a different, incompatible bytecode format between 32 bit and 64 bit ports.
+This may be rectified in the future. In the meantime, use the <tt>W</tt>
+and </tt>X</tt> <a href="#load_mode">modes of the <tt>load*</tt> functions</a>
+for cross-compilation purposes.
</p>
+<p>
+Due to VM hardening, bytecode is not deterministic. Add <tt>d</tt> to the
+mode string to dump it in a deterministic manner: identical source code
+always gives a byte-for-byte identical bytecode dump. This feature is
+mainly useful for reproducible builds.
+</p>
<h3 id="table_new"><tt>table.new(narray, nhash)</tt> allocates a pre-sized table</h3>
<p>
@@ -234,7 +255,7 @@
incremental array/hash part growth.
</p>
<p>
-Please note this function is meant for very specific situations. In most
+Please note, this function is meant for very specific situations. In most
cases it's better to replace the (usually single) link with a new table
and let the GC do its work.
</p>
@@ -244,7 +265,7 @@
LuaJIT uses a Tausworthe PRNG with period 2^223 to implement
<tt>math.random()</tt> and <tt>math.randomseed()</tt>. The quality of
the PRNG results is much superior compared to the standard Lua
-implementation which uses the platform-specific ANSI rand().
+implementation, which uses the platform-specific ANSI <tt>rand()</tt>.
</p>
<p>
The PRNG generates the same sequences from the same seeds on all
@@ -254,11 +275,19 @@
It's correctly scaled up and rounded for <tt>math.random(n [,m])</tt> to
preserve uniformity.
</p>
+<p>
+Call <tt>math.randomseed()</tt> without any arguments to seed it from
+system entropy.
+</p>
+<p>
+Important: Neither this nor any other PRNG based on the simplistic
+<tt>math.random()</tt> API is suitable for cryptographic use.
+</p>
<h3 id="io"><tt>io.*</tt> functions handle 64 bit file offsets</h3>
<p>
The file I/O functions in the standard <tt>io.*</tt> library handle
-64 bit file offsets. In particular this means it's possible
+64 bit file offsets. In particular, this means it's possible
to open files larger than 2 Gigabytes and to reposition or obtain
the current file position for offsets beyond 2 GB
(<tt>fp:seek()</tt> method).
@@ -288,7 +317,7 @@
</p>
<ul>
<li><tt>goto</tt> and <tt>::labels::</tt>.</li>
-<li>Hex escapes <tt>'\x3F'</tt> and <tt>'\*'</tt> escape in strings.</li>
+<li>Hex escapes <tt>'\x3F'</tt> and <tt>'\z'</tt> escape in strings.</li>
<li><tt>load(string|reader [, chunkname [,mode [,env]]])</tt>.</li>
<li><tt>loadstring()</tt> is an alias for <tt>load()</tt>.</li>
<li><tt>loadfile(filename [,mode [,env]])</tt>.</li>
@@ -374,6 +403,7 @@
<li>Unicode escape <tt>'\u{XX...}'</tt> embeds the UTF-8 encoding in string literals.</li>
<li>The argument table <tt>arg</tt> can be read (and modified) by <tt>LUA_INIT</tt> and <tt>-e</tt> chunks.</li>
<li><tt>io.read()</tt> and <tt>file:read()</tt> accept formats with or without a leading <tt>*</tt>.</li>
+<li><tt>assert()</tt> accepts any type of error object.</li>
<li><tt>table.move(a1, f, e, t [,a2])</tt>.</li>
<li><tt>coroutine.isyieldable()</tt>.</li>
<li>Lua/C API extensions:
@@ -394,31 +424,21 @@
<td class="excinterop">Interoperability</td>
</tr>
<tr class="odd separate">
-<td class="excplatform">POSIX/x64, DWARF2 unwinding</td>
-<td class="exccompiler">GCC 4.3+, Clang</td>
+<td class="excplatform">External frame unwinding</td>
+<td class="exccompiler">GCC, Clang, MSVC</td>
<td class="excinterop"><b style="color: #00a000;">Full</b></td>
</tr>
<tr class="even">
-<td class="excplatform">ARM <tt>-DLUAJIT_UNWIND_EXTERNAL</tt></td>
+<td class="excplatform">Internal frame unwinding + DWARF2</td>
<td class="exccompiler">GCC, Clang</td>
-<td class="excinterop"><b style="color: #00a000;">Full</b></td>
+<td class="excinterop"><b style="color: #c06000;">Limited</b></td>
</tr>
<tr class="odd">
-<td class="excplatform">Other platforms, DWARF2 unwinding</td>
-<td class="exccompiler">GCC, Clang</td>
+<td class="excplatform">Windows 64 bit</td>
+<td class="exccompiler">non-MSVC</td>
<td class="excinterop"><b style="color: #c06000;">Limited</b></td>
</tr>
<tr class="even">
-<td class="excplatform">Windows/x64</td>
-<td class="exccompiler">MSVC or WinSDK</td>
-<td class="excinterop"><b style="color: #00a000;">Full</b></td>
-</tr>
-<tr class="odd">
-<td class="excplatform">Windows/x86</td>
-<td class="exccompiler">Any</td>
-<td class="excinterop"><b style="color: #00a000;">Full</b></td>
-</tr>
-<tr class="even">
<td class="excplatform">Other platforms</td>
<td class="exccompiler">Other compilers</td>
<td class="excinterop"><b style="color: #a00000;">No</b></td>
@@ -472,7 +492,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/doc/faq.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/faq.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/faq.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,186 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
-<html>
-<head>
-<title>Frequently Asked Questions (FAQ)</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
-<meta name="Language" content="en">
-<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
-<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
-<style type="text/css">
-dd { margin-left: 1.5em; }
-</style>
-</head>
-<body>
-<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
-</div>
-<div id="head">
-<h1>Frequently Asked Questions (FAQ)</h1>
-</div>
-<div id="nav">
-<ul><li>
-<a href="luajit.html">LuaJIT</a>
-<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
-</li><li>
-<a href="install.html">Installation</a>
-</li><li>
-<a href="running.html">Running</a>
-</li></ul>
-</li><li>
-<a href="extensions.html">Extensions</a>
-<ul><li>
-<a href="ext_ffi.html">FFI Library</a>
-<ul><li>
-<a href="ext_ffi_tutorial.html">FFI Tutorial</a>
-</li><li>
-<a href="ext_ffi_api.html">ffi.* API</a>
-</li><li>
-<a href="ext_ffi_semantics.html">FFI Semantics</a>
-</li></ul>
-</li><li>
-<a href="ext_jit.html">jit.* Library</a>
-</li><li>
-<a href="ext_c_api.html">Lua/C API</a>
-</li><li>
-<a href="ext_profiler.html">Profiler</a>
-</li></ul>
-</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
-</li><li>
-<a class="current" href="faq.html">FAQ</a>
-</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
-</li></ul>
-</div>
-<div id="main">
-<dl>
-<dt>Q: Where can I learn more about LuaJIT and Lua?</dt>
-<dd>
-<ul style="padding: 0;">
-<li>The <a href="http://luajit.org/list.html"><span class="ext">»</span> LuaJIT mailing list</a> focuses on topics
-related to LuaJIT.</li>
-<li>The <a href="http://wiki.luajit.org/"><span class="ext">»</span> LuaJIT wiki</a> gathers community
-resources about LuaJIT.</li>
-<li>News about Lua itself can be found at the
-<a href="http://www.lua.org/lua-l.html"><span class="ext">»</span> Lua mailing list</a>.
-The mailing list archives are worth checking out for older postings
-about LuaJIT.</li>
-<li>The <a href="http://lua.org"><span class="ext">»</span> main Lua.org site</a> has complete
-<a href="http://www.lua.org/docs.html"><span class="ext">»</span> documentation</a> of the language
-and links to books and papers about Lua.</li>
-<li>The community-managed <a href="http://lua-users.org/wiki/"><span class="ext">»</span> Lua Wiki</a>
-has information about diverse topics.</li>
-</ul>
-</dl>
-
-<dl>
-<dt>Q: Where can I learn more about the compiler technology used by LuaJIT?</dt>
-<dd>
-I'm planning to write more documentation about the internals of LuaJIT.
-In the meantime, please use the following Google Scholar searches
-to find relevant papers:<br>
-Search for: <a href="http://scholar.google.com/scholar?q=Trace+Compiler"><span class="ext">»</span> Trace Compiler</a><br>
-Search for: <a href="http://scholar.google.com/scholar?q=JIT+Compiler"><span class="ext">»</span> JIT Compiler</a><br>
-Search for: <a href="http://scholar.google.com/scholar?q=Dynamic+Language+Optimizations"><span class="ext">»</span> Dynamic Language Optimizations</a><br>
-Search for: <a href="http://scholar.google.com/scholar?q=SSA+Form"><span class="ext">»</span> SSA Form</a><br>
-Search for: <a href="http://scholar.google.com/scholar?q=Linear+Scan+Register+Allocation"><span class="ext">»</span> Linear Scan Register Allocation</a><br>
-Here is a list of the <a href="http://article.gmane.org/gmane.comp.lang.lua.general/58908"><span class="ext">»</span> innovative features in LuaJIT</a>.<br>
-And, you know, reading the source is of course the only way to enlightenment. :-)
-</dd>
-</dl>
-
-<dl>
-<dt>Q: Why do I get this error: "attempt to index global 'arg' (a nil value)"?<br>
-Q: My vararg functions fail after switching to LuaJIT!</dt>
-<dd>LuaJIT is compatible to the Lua 5.1 language standard. It doesn't
-support the implicit <tt>arg</tt> parameter for old-style vararg
-functions from Lua 5.0.<br>Please convert your code to the
-<a href="http://www.lua.org/manual/5.1/manual.html#2.5.9"><span class="ext">»</span> Lua 5.1
-vararg syntax</a>.</dd>
-</dl>
-
-<dl>
-<dt>Q: Why do I get this error: "bad FPU precision"?<br>
-<dt>Q: I get weird behavior after initializing Direct3D.<br>
-<dt>Q: Some FPU operations crash after I load a Delphi DLL.<br>
-</dt>
-<dd>
-
-DirectX/Direct3D (up to version 9) sets the x87 FPU to single-precision
-mode by default. This violates the Windows ABI and interferes with the
-operation of many programs — LuaJIT is affected, too. Please make
-sure you always use the <tt>D3DCREATE_FPU_PRESERVE</tt> flag when
-initializing Direct3D.<br>
-
-Direct3D version 10 or higher do not show this behavior anymore.
-Consider testing your application with older versions, too.<br>
-
-Similarly, the Borland/Delphi runtime modifies the FPU control word and
-enables FP exceptions. Of course this violates the Windows ABI, too.
-Please check the Delphi docs for the Set8087CW method.
-
-</dl>
-
-<dl>
-<dt>Q: Sometimes Ctrl-C fails to stop my Lua program. Why?</dt>
-<dd>The interrupt signal handler sets a Lua debug hook. But this is
-currently ignored by compiled code (this will eventually be fixed). If
-your program is running in a tight loop and never falls back to the
-interpreter, the debug hook never runs and can't throw the
-"interrupted!" error.<br> In the meantime you have to press Ctrl-C
-twice to get stop your program. That's similar to when it's stuck
-running inside a C function under the Lua interpreter.</dd>
-</dl>
-
-<dl>
-<dt>Q: Why doesn't my favorite power-patch for Lua apply against LuaJIT?</dt>
-<dd>Because it's a completely redesigned VM and has very little code
-in common with Lua anymore. Also, if the patch introduces changes to
-the Lua semantics, these would need to be reflected everywhere in the
-VM, from the interpreter up to all stages of the compiler.<br> Please
-use only standard Lua language constructs. For many common needs you
-can use source transformations or use wrapper or proxy functions.
-The compiler will happily optimize away such indirections.</dd>
-</dl>
-
-<dl>
-<dt>Q: Lua runs everywhere. Why doesn't LuaJIT support my CPU?</dt>
-<dd>Because it's a compiler — it needs to generate native
-machine code. This means the code generator must be ported to each
-architecture. And the fast interpreter is written in assembler and
-must be ported, too. This is quite an undertaking.<br>
-The <a href="install.html">install documentation</a> shows the supported
-architectures. Other architectures will follow based on sufficient user
-demand and/or sponsoring.</dd>
-</dl>
-
-<dl>
-<dt>Q: When will feature X be added? When will the next version be released?</dt>
-<dd>When it's ready.<br>
-C'mon, it's open source — I'm doing it on my own time and you're
-getting it for free. You can either contribute a patch or sponsor
-the development of certain features, if they are important to you.
-</dd>
-</dl>
-<br class="flush">
-</div>
-<div id="foot">
-<hr class="hide">
-Copyright © 2005-2017 Mike Pall
-<span class="noprint">
-·
-<a href="contact.html">Contact</a>
-</span>
-</div>
-</body>
-</html>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/install.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/install.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/install.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Installation</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -15,31 +14,26 @@
}
table.compat td {
border: 1px solid #bfcfff;
- height: 2.5em;
+ height: 1.5em;
}
table.compat tr.compathead td {
font-weight: bold;
border-bottom: 2px solid #bfcfff;
}
-tr.compathead td.compatos {
- vertical-align: top;
+td.compatname {
+ width: 10%;
}
-table.compat td.compatcpu {
- width: 18%;
- border-right: 2px solid #bfcfff;
+td.compatbits {
+ width: 5%;
}
-td.compatos {
+td.compatx {
width: 21%;
- vertical-align: middle;
}
-td.compatno {
- background-color: #d0d0d0;
-}
</style>
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Installation</h1>
@@ -48,7 +42,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a class="current" href="install.html">Installation</a>
</li><li>
@@ -66,6 +60,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -73,25 +69,18 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
<p>
-LuaJIT is only distributed as a source package. This page explains
-how to build and install LuaJIT with different operating systems
-and C compilers.
+LuaJIT is only distributed as source code — get it from the
+<a href="https://luajit.org/download.html"><span class="ext">»</span> git repository</a>. This page explains how to build
+and install the LuaJIT binary and library for different operating systems.
</p>
<p>
For the impatient (on POSIX systems):
@@ -99,62 +88,24 @@
<pre class="code">
make && sudo make install
</pre>
+
+<h2 id="req">Requirements</h2>
<p>
-LuaJIT currently builds out-of-the box on most systems.
-Here's the compatibility matrix for the supported combinations of
-operating systems, CPUs and compilers:
+LuaJIT currently builds out-of-the box on most systems. Please check the
+supported operating systems and CPU architectures on the
+<a href="https://luajit.org/status.html"><span class="ext">»</span> status page</a>.
</p>
-<table class="compat">
-<tr class="compathead">
-<td class="compatcpu">CPU / OS</td>
-<td class="compatos"><a href="#posix">Linux</a> or<br><a href="#android">Android</a></td>
-<td class="compatos"><a href="#posix">*BSD, Other</a></td>
-<td class="compatos"><a href="#posix">OSX 10.4+</a> or<br><a href="#ios">iOS 3.0+</a></td>
-<td class="compatos"><a href="#windows">Windows<br>XP/Vista/7</a></td>
-</tr>
-<tr class="odd separate">
-<td class="compatcpu">x86 (32 bit)</td>
-<td class="compatos">GCC 4.2+</td>
-<td class="compatos">GCC 4.2+</td>
-<td class="compatos">XCode 5.0+<br>Clang</td>
-<td class="compatos">MSVC, MSVC/EE<br>WinSDK<br>MinGW, Cygwin</td>
-</tr>
-<tr class="even">
-<td class="compatcpu">x64 (64 bit)</td>
-<td class="compatos">GCC 4.2+</td>
-<td class="compatos">GCC 4.2+<br>ORBIS (<a href="#ps4">PS4</a>)</td>
-<td class="compatos">XCode 5.0+<br>Clang</td>
-<td class="compatos">MSVC + SDK v7.0<br>WinSDK v7.0<br>Durango (<a href="#xboxone">Xbox One</a>)</td>
-</tr>
-<tr class="odd">
-<td class="compatcpu"><a href="#cross2">ARMv5+<br>ARM9E+</a></td>
-<td class="compatos">GCC 4.2+</td>
-<td class="compatos">GCC 4.2+<br>PSP2 (<a href="#psvita">PS VITA</a>)</td>
-<td class="compatos">XCode 5.0+<br>Clang</td>
-<td class="compatos compatno"> </td>
-</tr>
-<tr class="even">
-<td class="compatcpu"><a href="#cross2">ARM64</a></td>
-<td class="compatos">GCC 4.8+</td>
-<td class="compatos compatno"> </td>
-<td class="compatos">XCode 6.0+<br>Clang 3.5+</td>
-<td class="compatos compatno"> </td>
-</tr>
-<tr class="odd">
-<td class="compatcpu"><a href="#cross2">PPC</a></td>
-<td class="compatos">GCC 4.3+</td>
-<td class="compatos">GCC 4.3+<br>GCC 4.1 (<a href="#ps3">PS3</a>)</td>
-<td class="compatos compatno"> </td>
-<td class="compatos">XEDK (<a href="#xbox360">Xbox 360</a>)</td>
-</tr>
-<tr class="even">
-<td class="compatcpu"><a href="#cross2">MIPS32<br>MIPS64</a></td>
-<td class="compatos">GCC 4.3+</td>
-<td class="compatos">GCC 4.3+</td>
-<td class="compatos compatno"> </td>
-<td class="compatos compatno"> </td>
-</tr>
-</table>
+<p>
+Building LuaJIT requires a recent toolchain based on GCC, Clang/LLVM or
+MSVC++.
+</p>
+<p>
+The Makefile-based build system requires GNU Make and supports
+cross-builds.
+</p>
+<p>
+Batch files are provided for MSVC++ builds and console cross-builds.
+</p>
<h2>Configuring LuaJIT</h2>
<p>
@@ -163,13 +114,12 @@
hold all user-configurable settings:
</p>
<ul>
-<li><tt>src/luaconf.h</tt> sets some configuration variables.</li>
<li><tt>Makefile</tt> has settings for <b>installing</b> LuaJIT (POSIX
only).</li>
<li><tt>src/Makefile</tt> has settings for <b>compiling</b> LuaJIT
under POSIX, MinGW or Cygwin.</li>
<li><tt>src/msvcbuild.bat</tt> has settings for compiling LuaJIT with
-MSVC or WinSDK.</li>
+MSVC (Visual Studio).</li>
</ul>
<p>
Please read the instructions given in these files, before changing
@@ -176,30 +126,22 @@
any settings.
</p>
<p>
-LuaJIT on x64 currently uses 32 bit GC objects by default.
-<tt>LJ_GC64</tt> mode may be explicitly enabled:
-add <tt>XCFLAGS=-DLUAJIT_ENABLE_GC64</tt> to the make command or run
-<tt>msvcbuild gc64</tt> for MSVC/WinSDK. Please check the note
-about the <a href="extensions.html#string_dump">bytecode format</a>
-differences, too.
+All LuaJIT 64 bit ports use 64 bit GC objects by default (<tt>LJ_GC64</tt>).
+For x64, you can select the old 32-on-64 bit mode by adding
+<tt>XCFLAGS=-DLUAJIT_DISABLE_GC64</tt> to the make command.
+Please check the note about the
+<a href="extensions.html#string_dump">bytecode format</a> differences, too.
</p>
-<h2 id="posix">POSIX Systems (Linux, OSX, *BSD etc.)</h2>
+<h2 id="posix">POSIX Systems (Linux, macOS, *BSD etc.)</h2>
<h3>Prerequisites</h3>
<p>
-Depending on your distribution, you may need to install a package for
-GCC, the development headers and/or a complete SDK. E.g. on a current
-Debian/Ubuntu, install <tt>libc6-dev</tt> with the package manager.
+Depending on your distribution, you may need to install a package for a
+compiler (GCC or Clang/LLVM), the development headers and/or a complete SDK.
+E.g. on a current Debian/Ubuntu, install <tt>build-essential</tt> with the
+package manager.
</p>
-<p>
-Download the current source package of LuaJIT (pick the .tar.gz),
-if you haven't already done so. Move it to a directory of your choice,
-open a terminal window and change to this directory. Now unpack the archive
-and change to the newly created directory:
-</p>
-<pre class="code">
-tar zxf LuaJIT-2.0.5.tar.gz
-cd LuaJIT-2.0.5</pre>
+</pre>
<h3>Building LuaJIT</h3>
<p>
The supplied Makefiles try to auto-detect the settings needed for your
@@ -223,9 +165,12 @@
make PREFIX=/home/myself/lj2
</pre>
<p>
-Note for OSX: if the <tt>MACOSX_DEPLOYMENT_TARGET</tt> environment
-variable is not set, then it's forced to <tt>10.4</tt>.
+Note for macOS: you <b>must</b> set the <tt>MACOSX_DEPLOYMENT_TARGET</tt>
+environment variable to a value supported by your toolchain:
</p>
+<pre class="code">
+MACOSX_DEPLOYMENT_TARGET=XX.YY make
+</pre>
<h3>Installing LuaJIT</h3>
<p>
The top-level Makefile installs LuaJIT by default under
@@ -252,28 +197,14 @@
<p>
Either install one of the open source SDKs
(<a href="http://mingw.org/"><span class="ext">»</span> MinGW</a> or
-<a href="http://www.cygwin.com/"><span class="ext">»</span> Cygwin</a>), which come with a modified
+<a href="https://www.cygwin.com/"><span class="ext">»</span> Cygwin</a>), which come with a modified
GCC plus the required development headers.
+Or install Microsoft's Visual Studio (MSVC).
</p>
-<p>
-Or install Microsoft's Visual C++ (MSVC). The freely downloadable
-<a href="http://www.microsoft.com/Express/VC/"><span class="ext">»</span> Express Edition</a>
-works just fine, but only contains an x86 compiler.
-</p>
-<p>
-The freely downloadable
-<a href="http://msdn.microsoft.com/en-us/windowsserver/bb980924.aspx"><span class="ext">»</span> Windows SDK</a>
-only comes with command line tools, but this is all you need to build LuaJIT.
-It contains x86 and x64 compilers.
-</p>
-<p>
-Next, download the source package and unpack it using an archive manager
-(e.g. the Windows Explorer) to a directory of your choice.
-</p>
<h3>Building with MSVC</h3>
<p>
-Open a "Visual Studio .NET Command Prompt", <tt>cd</tt> to the
-directory where you've unpacked the sources and run these commands:
+Open a "Visual Studio Command Prompt" (x86, x64 or ARM64), <tt>cd</tt> to the
+directory with the source code and run these commands:
</p>
<pre class="code">
cd src
@@ -280,37 +211,17 @@
msvcbuild
</pre>
<p>
+Check the <tt>msvcbuild.bat</tt> file for more options.
Then follow the installation instructions below.
</p>
-<h3>Building with the Windows SDK</h3>
<p>
-Open a "Windows SDK Command Shell" and select the x86 compiler:
+For an x64 to ARM64 cross-build run this first: <tt>vcvarsall.bat x64_arm64</tt>
</p>
-<pre class="code">
-setenv /release /x86
-</pre>
-<p>
-Or select the x64 compiler:
-</p>
-<pre class="code">
-setenv /release /x64
-</pre>
-<p>
-Then <tt>cd</tt> to the directory where you've unpacked the sources
-and run these commands:
-</p>
-<pre class="code">
-cd src
-msvcbuild
-</pre>
-<p>
-Then follow the installation instructions below.
-</p>
<h3>Building with MinGW or Cygwin</h3>
<p>
Open a command prompt window and make sure the MinGW or Cygwin programs
-are in your path. Then <tt>cd</tt> to the directory where
-you've unpacked the sources and run this command for MinGW:
+are in your path. Then <tt>cd</tt> to the directory of the git repository.
+Then run this command for MinGW:
</p>
<pre class="code">
mingw32-make
@@ -358,6 +269,7 @@
<li>Yes, you need a toolchain for both your host <em>and</em> your target!</li>
<li>Both host and target architectures must have the same pointer size.</li>
<li>E.g. if you want to cross-compile to a 32 bit target on a 64 bit host, you need to install the multilib development package (e.g. <tt>libc6-dev-i386</tt> on Debian/Ubuntu) and build a 32 bit host part (<tt>HOST_CC="gcc -m32"</tt>).</li>
+<li>On some distro versions, multilib conflicts with cross-compilers. The workaround is to install the x86 cross-compiler package <tt>gcc-i686-linux-gnu</tt> and use it to build the host part (<tt>HOST_CC=i686-linux-gnu-gcc</tt>).</li>
<li>64 bit targets always require compilation on a 64 bit host.</li>
</ul>
<p>
@@ -365,7 +277,7 @@
target OS differ, or you'll get assembler or linker errors:
</p>
<ul>
-<li>E.g. if you're compiling on a Windows or OSX host for embedded Linux or Android, you need to add <tt>TARGET_SYS=Linux</tt> to the examples below.</li>
+<li>E.g. if you're compiling on a Windows or macOS host for embedded Linux or Android, you need to add <tt>TARGET_SYS=Linux</tt> to the examples below.</li>
<li>For a minimal target OS, you may need to disable the built-in allocator in <tt>src/Makefile</tt> and use <tt>TARGET_SYS=Other</tt>.</li>
<li>Don't forget to specify the same <tt>TARGET_SYS</tt> for the install step, too.</li>
</ul>
@@ -412,15 +324,15 @@
make HOST_CC="gcc -m32" CROSS=arm-linux-gnueabihf-
# ARM64
-make CROSS=aarch64-linux-
+make CROSS=aarch64-linux-gnu-
# PPC
make HOST_CC="gcc -m32" CROSS=powerpc-linux-gnu-
# MIPS32 big-endian
-make HOST_CC="gcc -m32" CROSS=mips-linux-
+make HOST_CC="gcc -m32" CROSS=mips-linux-gnu-
# MIPS32 little-endian
-make HOST_CC="gcc -m32" CROSS=mipsel-linux-
+make HOST_CC="gcc -m32" CROSS=mipsel-linux-gnu-
# MIPS64 big-endian
make CROSS=mips-linux- TARGET_CFLAGS="-mips64r2 -mabi=64"
@@ -428,52 +340,33 @@
make CROSS=mipsel-linux- TARGET_CFLAGS="-mips64r2 -mabi=64"
</pre>
<p>
-You can cross-compile for <b id="android">Android</b> using the <a href="https://developer.android.com/ndk/index.html">Android NDK</a>.
-The environment variables need to match the install locations and the
-desired target platform. E.g. Android 4.0 corresponds to ABI level 14.
-For details check the folder <tt>docs</tt> in the NDK directory.
+You can cross-compile for <b id="android">Android</b> using the <a href="https://developer.android.com/ndk/"><span class="ext">»</span> Android NDK</a>.
+Please adapt the environment variables to match the install locations and the
+desired target platform. E.g. Android 4.1 corresponds to ABI level 16.
</p>
-<p>
-Only a few common variations for the different CPUs, ABIs and platforms
-are listed. Please use your own judgement for which combination you want
-to build/deploy or which lowest common denominator you want to pick:
-</p>
<pre class="code">
-# Android/ARM, armeabi (ARMv5TE soft-float), Android 2.2+ (Froyo)
-NDK=/opt/android/ndk
-NDKABI=8
-NDKVER=$NDK/toolchains/arm-linux-androideabi-4.9
-NDKP=$NDKVER/prebuilt/linux-x86/bin/arm-linux-androideabi-
-NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-arm"
-make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
+# Android/ARM64, aarch64, Android 5.0+ (L)
+NDKDIR=/opt/android/ndk
+NDKBIN=$NDKDIR/toolchains/llvm/prebuilt/linux-x86_64/bin
+NDKCROSS=$NDKBIN/aarch64-linux-android-
+NDKCC=$NDKBIN/aarch64-linux-android21-clang
+make CROSS=$NDKCROSS \
+ STATIC_CC=$NDKCC DYNAMIC_CC="$NDKCC -fPIC" \
+ TARGET_LD=$NDKCC TARGET_AR="$NDKBIN/llvm-ar rcus" \
+ TARGET_STRIP=$NDKBIN/llvm-strip
-# Android/ARM, armeabi-v7a (ARMv7 VFP), Android 4.0+ (ICS)
-NDK=/opt/android/ndk
-NDKABI=14
-NDKVER=$NDK/toolchains/arm-linux-androideabi-4.9
-NDKP=$NDKVER/prebuilt/linux-x86/bin/arm-linux-androideabi-
-NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-arm"
-NDKARCH="-march=armv7-a -mfloat-abi=softfp -Wl,--fix-cortex-a8"
-make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF $NDKARCH"
-
-# Android/MIPS, mipsel (MIPS32R1 hard-float), Android 4.0+ (ICS)
-NDK=/opt/android/ndk
-NDKABI=14
-NDKVER=$NDK/toolchains/mipsel-linux-android-4.9
-NDKP=$NDKVER/prebuilt/linux-x86/bin/mipsel-linux-android-
-NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-mips"
-make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
-
-# Android/x86, x86 (i686 SSE3), Android 4.0+ (ICS)
-NDK=/opt/android/ndk
-NDKABI=14
-NDKVER=$NDK/toolchains/x86-4.9
-NDKP=$NDKVER/prebuilt/linux-x86/bin/i686-linux-android-
-NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-x86"
-make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
+# Android/ARM, armeabi-v7a (ARMv7 VFP), Android 4.1+ (JB)
+NDKDIR=/opt/android/ndk
+NDKBIN=$NDKDIR/toolchains/llvm/prebuilt/linux-x86_64/bin
+NDKCROSS=$NDKBIN/arm-linux-androideabi-
+NDKCC=$NDKBIN/armv7a-linux-androideabi16-clang
+make HOST_CC="gcc -m32" CROSS=$NDKCROSS \
+ STATIC_CC=$NDKCC DYNAMIC_CC="$NDKCC -fPIC" \
+ TARGET_LD=$NDKCC TARGET_AR="$NDKBIN/llvm-ar rcus" \
+ TARGET_STRIP=$NDKBIN/llvm-strip
</pre>
<p>
-You can cross-compile for <b id="ios">iOS 3.0+</b> (iPhone/iPad) using the <a href="http://developer.apple.com/devcenter/ios/index.action"><span class="ext">»</span> iOS SDK</a>:
+You can cross-compile for <b id="ios">iOS 3.0+</b> (iPhone/iPad) using the <a href="https://developer.apple.com/ios/"><span class="ext">»</span> iOS SDK</a>:
</p>
<p style="font-size: 8pt;">
Note: <b>the JIT compiler is disabled for iOS</b>, because regular iOS Apps
@@ -483,13 +376,6 @@
Or use Android. :-p
</p>
<pre class="code">
-# iOS/ARM (32 bit)
-ISDKP=$(xcrun --sdk iphoneos --show-sdk-path)
-ICC=$(xcrun --sdk iphoneos --find clang)
-ISDKF="-arch armv7 -isysroot $ISDKP"
-make DEFAULT_CC=clang HOST_CC="clang -m32 -arch i386" \
- CROSS="$(dirname $ICC)/" TARGET_FLAGS="$ISDKF" TARGET_SYS=iOS
-
# iOS/ARM64
ISDKP=$(xcrun --sdk iphoneos --show-sdk-path)
ICC=$(xcrun --sdk iphoneos --find clang)
@@ -501,8 +387,7 @@
<h3 id="consoles">Cross-compiling for consoles</h3>
<p>
Building LuaJIT for consoles requires both a supported host compiler
-(x86 or x64) and a cross-compiler (to PPC or ARM) from the official
-console SDK.
+(x86 or x64) and a cross-compiler from the official console SDK.
</p>
<p>
Due to restrictions on consoles, the JIT compiler is disabled and only
@@ -523,45 +408,58 @@
make HOST_CC="gcc -m32" CROSS=ppu-lv2-
</pre>
<p>
-To cross-compile for <b id="ps4">PS4</b> from a Windows host,
-open a "Visual Studio .NET Command Prompt" (64 bit host compiler),
-<tt>cd</tt> to the directory where you've unpacked the sources and
-run the following commands:
+To cross-compile for the other consoles from a Windows host, open a
+"Native Tools Command Prompt for VS". You need to choose either the 32
+or the 64 bit version of the host compiler to match the target.
+Then <tt>cd</tt> to the <tt>src</tt> directory below the source code
+and run the build command given in the table:
</p>
-<pre class="code">
-cd src
-ps4build
-</pre>
+<table class="compat">
+<tr class="compathead">
+<td class="compatname">Console</td>
+<td class="compatbits">Bits</td>
+<td class="compatx">Build Command</td>
+</tr>
+<tr class="odd separate">
+<td class="compatname"><b id="ps4">PS4</b></td>
+<td class="compatbits">64</td>
+<td class="compatx"><tt>ps4build</tt></td>
+</tr>
+<tr class="even">
+<td class="compatname"><b id="ps5">PS5</b></td>
+<td class="compatbits">64</td>
+<td class="compatx"><tt>ps5build</tt></td>
+</tr>
+<tr class="odd">
+<td class="compatname"><b id="psvita">PS Vita</b></td>
+<td class="compatbits">32</td>
+<td class="compatx"><tt>psvitabuild</tt></td>
+</tr>
+<tr class="even">
+<td class="compatname"><b id="xbox360">Xbox 360</b></td>
+<td class="compatbits">32</td>
+<td class="compatx"><tt>xedkbuild</tt></td>
+</tr>
+<tr class="odd">
+<td class="compatname"><b id="xboxone">Xbox One</b></td>
+<td class="compatbits">64</td>
+<td class="compatx"><tt>xb1build</tt></td>
+</tr>
+<tr class="even">
+<td class="compatname"><b id="nx32">Nintendo Switch NX32</b></td>
+<td class="compatbits">32</td>
+<td class="compatx"><tt>nxbuild</tt></td>
+</tr>
+<tr class="odd">
+<td class="compatname"><b id="nx64">Nintendo Switch NX64</b></td>
+<td class="compatbits">64</td>
+<td class="compatx"><tt>nxbuild</tt></td>
+</tr>
+</table>
<p>
-To cross-compile for <b id="psvita">PS Vita</b> from a Windows host,
-open a "Visual Studio .NET Command Prompt" (32 bit host compiler),
-<tt>cd</tt> to the directory where you've unpacked the sources and
-run the following commands:
+Please check out the comments in the corresponding <tt>*.bat</tt>
+file for more options.
</p>
-<pre class="code">
-cd src
-psvitabuild
-</pre>
-<p>
-To cross-compile for <b id="xbox360">Xbox 360</b> from a Windows host,
-open a "Visual Studio .NET Command Prompt" (32 bit host compiler),
-<tt>cd</tt> to the directory where you've unpacked the sources and run
-the following commands:
-</p>
-<pre class="code">
-cd src
-xedkbuild
-</pre>
-<p>
-To cross-compile for <b id="xboxone">Xbox One</b> from a Windows host,
-open a "Visual Studio .NET Command Prompt" (64 bit host compiler),
-<tt>cd</tt> to the directory where you've unpacked the sources and run
-the following commands:
-</p>
-<pre class="code">
-cd src
-xb1build
-</pre>
<h2 id="embed">Embedding LuaJIT</h2>
<p>
@@ -590,14 +488,6 @@
intend to load Lua/C modules at runtime.
</li></ul>
</li>
-<li>
-If you're building a 64 bit application on OSX which links directly or
-indirectly against LuaJIT which is not built for <tt>LJ_GC64</tt> mode,
-you need to link your main executable with these flags:
-<pre class="code">
--pagezero_size 10000 -image_base 100000000
-</pre>
-</li>
</ul>
<p>Additional hints for initializing LuaJIT using the C API functions:</p>
<ul>
@@ -606,16 +496,16 @@
for embedding Lua or LuaJIT into your application.</li>
<li>Make sure you use <tt>luaL_newstate</tt>. Avoid using
<tt>lua_newstate</tt>, since this uses the (slower) default memory
-allocator from your system (no support for this on x64).</li>
+allocator from your system (no support for this on 64 bit architectures).</li>
<li>Make sure you use <tt>luaL_openlibs</tt> and not the old Lua 5.0 style
of calling <tt>luaopen_base</tt> etc. directly.</li>
<li>To change or extend the list of standard libraries to load, copy
<tt>src/lib_init.c</tt> to your project and modify it accordingly.
-Make sure the <tt>jit</tt> library is loaded or the JIT compiler
+Make sure the <tt>jit</tt> library is loaded, or the JIT compiler
will not be activated.</li>
<li>The <tt>bit.*</tt> module for bitwise operations
is already built-in. There's no need to statically link
-<a href="http://bitop.luajit.org/"><span class="ext">»</span> Lua BitOp</a> to your application.</li>
+<a href="https://bitop.luajit.org/"><span class="ext">»</span> Lua BitOp</a> to your application.</li>
</ul>
<h2 id="distro">Hints for Distribution Maintainers</h2>
@@ -630,7 +520,7 @@
There should be absolutely no need to patch <tt>luaconf.h</tt> or any
of the Makefiles. And please do not hand-pick files for your packages —
simply use whatever <tt>make install</tt> creates. There's a reason
-for all of the files <em>and</em> directories it creates.
+for all the files <em>and</em> directories it creates.
</p>
<p>
The build system uses GNU make and auto-detects most settings based on
@@ -682,7 +572,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/luajit.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/luajit.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/luajit.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>LuaJIT</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -68,35 +67,11 @@
background-image: -o-linear-gradient(#41bfbf 10%, #b0ffff 95%);
background-image: -ms-linear-gradient(#41bfbf 10%, #b0ffff 95%);
}
-table.stats td {
- color: #ffffff;
- background: #a0a0a0;
- background-image: linear-gradient(#808080 10%, #d0d0d0 95%);
- background-image: -moz-linear-gradient(#808080 10%, #d0d0d0 95%);
- background-image: -webkit-linear-gradient(#808080 10%, #d0d0d0 95%);
- background-image: -o-linear-gradient(#808080 10%, #d0d0d0 95%);
- background-image: -ms-linear-gradient(#808080 10%, #d0d0d0 95%);
-}
-table.stats td.speed {
- color: #ff4020;
-}
-table.stats td.kb {
- color: #ffff80;
- background: #808080;
- background-image: linear-gradient(#606060 10%, #c0c0c0 95%);
- background-image: -moz-linear-gradient(#606060 10%, #c0c0c0 95%);
- background-image: -webkit-linear-gradient(#606060 10%, #c0c0c0 95%);
- background-image: -o-linear-gradient(#606060 10%, #c0c0c0 95%);
- background-image: -ms-linear-gradient(#606060 10%, #c0c0c0 95%);
-}
-table.feature small {
- font-size: 50%;
-}
</style>
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>LuaJIT</h1>
@@ -105,7 +80,7 @@
<ul><li>
<a class="current" href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -123,6 +98,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -130,30 +107,23 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
<p>
LuaJIT is a <b>Just-In-Time Compiler</b> (JIT) for the
-<a href="http://www.lua.org/"><span class="ext">»</span> Lua</a> programming language.
+<a href="https://www.lua.org/"><span class="ext">»</span> Lua</a> programming language.
Lua is a powerful, dynamic and light-weight programming language.
It may be embedded or used as a general-purpose, stand-alone language.
</p>
<p>
-LuaJIT is Copyright © 2005-2017 Mike Pall, released under the
-<a href="http://www.opensource.org/licenses/mit-license.php"><span class="ext">»</span> MIT open source license</a>.
+LuaJIT is Copyright © 2005-2025 Mike Pall, released under the
+<a href="https://www.opensource.org/licenses/mit-license.php"><span class="ext">»</span> MIT open source license</a>.
</p>
<p>
</p>
@@ -160,13 +130,13 @@
<h2>Compatibility</h2>
<table class="feature os os1">
-<tr><td>Windows</td><td>Linux</td><td>BSD</td><td>OSX</td><td>POSIX</td></tr>
+<tr><td>Windows</td><td>Linux</td><td>BSD</td><td>macOS</td><td>POSIX</td></tr>
</table>
<table class="feature os os2">
<tr><td><span style="font-size:90%;">Embedded</span></td><td>Android</td><td>iOS</td></tr>
</table>
<table class="feature os os3">
-<tr><td>PS3</td><td>PS4</td><td>PS Vita</td><td>Xbox 360</td><td>Xbox One</td></tr>
+<tr><td>PS3</td><td>PS4<br>PS5</td><td>PS Vita</td><td>Xbox 360</td><td>Xbox One</td><td>Nintendo<br>Switch</td></tr>
</table>
<table class="feature compiler">
<tr><td>GCC</td><td>Clang<br>LLVM</td><td>MSVC</td></tr>
@@ -179,25 +149,22 @@
</table>
<h2>Overview</h2>
-<table class="feature stats">
-<tr>
-<td class="speed">3x<br>- 100x</td>
-<td class="kb">115 <small>KB</small><br>VM</td>
-<td class="kb">90 <small>KB</small><br>JIT</td>
-<td class="kloc">63 <small>KLOC</small><br>C</td>
-<td class="kloc">24 <small>KLOC</small><br>ASM</td>
-<td class="kloc">11 <small>KLOC</small><br>Lua</td>
-</tr>
-</table>
<p style="margin-top: 1em;">
LuaJIT has been successfully used as a <b>scripting middleware</b> in
games, appliances, network and graphics apps, numerical simulations,
-trading platforms and many other specialty applications. It scales from
-embedded devices, smartphones, desktops up to server farms. It combines
-high flexibility with <a href="http://luajit.org/performance.html"><span class="ext">»</span> high performance</a>
-and an unmatched <b>low memory footprint</b>.
+trading platforms and many other specialty applications.
</p>
<p>
+LuaJIT is part of a hundred million web sites, huge SaaS installations,
+network switches, set-top boxes and other embedded devices. You've probably
+already used LuaJIT without knowing about it.
+</p>
+<p>
+LuaJIT scales from embedded devices, smartphones, desktops up to server
+farms. It combines high flexibility with high performance and an unmatched
+<b>low memory footprint</b>.
+</p>
+<p>
LuaJIT has been in continuous development since 2005. It's widely
considered to be <b>one of the fastest dynamic language
implementations</b>. It has outperformed other dynamic languages on many
@@ -226,7 +193,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/doc/running.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/running.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/running.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,9 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!DOCTYPE html>
<html>
<head>
<title>Running LuaJIT</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
+<meta charset="utf-8">
+<meta name="Copyright" content="Copyright (C) 2005-2025">
<meta name="Language" content="en">
<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
@@ -33,7 +32,7 @@
</head>
<body>
<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
+<a href="https://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
</div>
<div id="head">
<h1>Running LuaJIT</h1>
@@ -42,7 +41,7 @@
<ul><li>
<a href="luajit.html">LuaJIT</a>
<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
+<a href="https://luajit.org/download.html">Download <span class="ext">»</span></a>
</li><li>
<a href="install.html">Installation</a>
</li><li>
@@ -60,6 +59,8 @@
<a href="ext_ffi_semantics.html">FFI Semantics</a>
</li></ul>
</li><li>
+<a href="ext_buffer.html">String Buffers</a>
+</li><li>
<a href="ext_jit.html">jit.* Library</a>
</li><li>
<a href="ext_c_api.html">Lua/C API</a>
@@ -67,21 +68,15 @@
<a href="ext_profiler.html">Profiler</a>
</li></ul>
</li><li>
-<a href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
+<a href="https://luajit.org/status.html">Status <span class="ext">»</span></a>
</li><li>
-<a href="faq.html">FAQ</a>
+<a href="https://luajit.org/faq.html">FAQ <span class="ext">»</span></a>
</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
+<a href="https://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
</li></ul>
</div>
<div id="main">
+
<p>
LuaJIT has only a single stand-alone executable, called <tt>luajit</tt> on
POSIX systems or <tt>luajit.exe</tt> on Windows. It can be used to run simple
@@ -95,7 +90,7 @@
version of the regular <tt>lua</tt> stand-alone executable.
It supports the same basic options, too. <tt>luajit -h</tt>
prints a short list of the available options. Please have a look at the
-<a href="http://www.lua.org/manual/5.1/manual.html#6"><span class="ext">»</span> Lua manual</a>
+<a href="https://www.lua.org/manual/5.1/manual.html#6"><span class="ext">»</span> Lua manual</a>
for details.
</p>
<p>
@@ -111,10 +106,14 @@
<li><tt>-l</tt> — Only list bytecode.</li>
<li><tt>-s</tt> — Strip debug info (this is the default).</li>
<li><tt>-g</tt> — Keep debug info.</li>
+<li><tt>-W</tt> — Generate 32 bit (non-GC64) bytecode.</li>
+<li><tt>-X</tt> — Generate 64 bit (GC64) bytecode.</li>
+<li><tt>-d</tt> — Generate bytecode in deterministic manner.</li>
<li><tt>-n name</tt> — Set module name (default: auto-detect from input name)</li>
<li><tt>-t type</tt> — Set output file type (default: auto-detect from output name).</li>
<li><tt>-a arch</tt> — Override architecture for object files (default: native).</li>
<li><tt>-o os</tt> — Override OS for object files (default: native).</li>
+<li><tt>-F name</tt> — Override filename (default: input filename).</li>
<li><tt>-e chunk</tt> — Use chunk string as input.</li>
<li><tt>-</tt> (a single minus sign) — Use stdin as input and/or stdout as output.</li>
</ul>
@@ -124,7 +123,8 @@
</p>
<ul>
<li><tt>c</tt> — C source file, exported bytecode data.</li>
-<li><tt>h</tt> — C header file, static bytecode data.</li>
+<li><tt>cc</tt> — C++ source file, exported bytecode data.</li>
+<li><tt>h</tt> — C/C++ header file, static bytecode data.</li>
<li><tt>obj</tt> or <tt>o</tt> — Object file, exported bytecode data
(OS- and architecture-specific).</li>
<li><tt>raw</tt> or any other extension — Raw bytecode file (portable).
@@ -188,9 +188,9 @@
itself. For a description of their options and output format, please
read the comment block at the start of their source.
They can be found in the <tt>lib</tt> directory of the source
-distribution or installed under the <tt>jit</tt> directory. By default
-this is <tt>/usr/local/share/luajit-2.0.5/jit</tt> on POSIX
-systems.
+distribution or installed under the <tt>jit</tt> directory. By default,
+this is <tt>/usr/local/share/luajit-XX.YY.ZZ>/jit</tt> on POSIX
+systems (replace XX.YY.ZZ by the installed version).
</p>
<h3 id="opt_O"><tt>-O[level]</tt><br>
@@ -220,11 +220,17 @@
You can either use this option multiple times (like <tt>-Ocse
-O-dce -Ohotloop=10</tt>) or separate several settings with a comma
(like <tt>-O+cse,-dce,hotloop=10</tt>). The settings are applied from
-left to right and later settings override earlier ones. You can freely
+left to right, and later settings override earlier ones. You can freely
mix the three forms, but note that setting an optimization level
overrides all earlier flags.
</p>
<p>
+Note that <tt>-Ofma</tt> is not enabled by default at any level,
+because it affects floating-point result accuracy. Only enable this,
+if you fully understand the trade-offs of FMA for performance (higher),
+determinism (lower) and numerical accuracy (higher).
+</p>
+<p>
Here are the available flags and at what optimization levels they
are enabled:
</p>
@@ -256,6 +262,8 @@
<td class="flag_name">sink</td><td class="flag_level"> </td><td class="flag_level"> </td><td class="flag_level">•</td><td class="flag_desc">Allocation/Store Sinking</td></tr>
<tr class="even">
<td class="flag_name">fuse</td><td class="flag_level"> </td><td class="flag_level"> </td><td class="flag_level">•</td><td class="flag_desc">Fusion of operands into instructions</td></tr>
+<tr class="odd">
+<td class="flag_name">fma </td><td class="flag_level"> </td><td class="flag_level"> </td><td class="flag_level"> </td><td class="flag_desc">Fused multiply-add</td></tr>
</table>
<p>
Here are the parameters and their default settings:
@@ -299,7 +307,7 @@
</div>
<div id="foot">
<hr class="hide">
-Copyright © 2005-2017 Mike Pall
+Copyright © 2005-2025
<span class="noprint">
·
<a href="contact.html">Contact</a>
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/doc/status.html
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/doc/status.html 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/doc/status.html 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,123 +0,0 @@
-<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
-<html>
-<head>
-<title>Status</title>
-<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
-<meta name="Author" content="Mike Pall">
-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
-<meta name="Language" content="en">
-<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
-<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
-<style type="text/css">
-ul li { padding-bottom: 0.3em; }
-</style>
-</head>
-<body>
-<div id="site">
-<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
-</div>
-<div id="head">
-<h1>Status</h1>
-</div>
-<div id="nav">
-<ul><li>
-<a href="luajit.html">LuaJIT</a>
-<ul><li>
-<a href="http://luajit.org/download.html">Download <span class="ext">»</span></a>
-</li><li>
-<a href="install.html">Installation</a>
-</li><li>
-<a href="running.html">Running</a>
-</li></ul>
-</li><li>
-<a href="extensions.html">Extensions</a>
-<ul><li>
-<a href="ext_ffi.html">FFI Library</a>
-<ul><li>
-<a href="ext_ffi_tutorial.html">FFI Tutorial</a>
-</li><li>
-<a href="ext_ffi_api.html">ffi.* API</a>
-</li><li>
-<a href="ext_ffi_semantics.html">FFI Semantics</a>
-</li></ul>
-</li><li>
-<a href="ext_jit.html">jit.* Library</a>
-</li><li>
-<a href="ext_c_api.html">Lua/C API</a>
-</li><li>
-<a href="ext_profiler.html">Profiler</a>
-</li></ul>
-</li><li>
-<a class="current" href="status.html">Status</a>
-<ul><li>
-<a href="changes.html">Changes</a>
-</li></ul>
-</li><li>
-<a href="faq.html">FAQ</a>
-</li><li>
-<a href="http://luajit.org/performance.html">Performance <span class="ext">»</span></a>
-</li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">»</span></a>
-</li><li>
-<a href="http://luajit.org/list.html">Mailing List <span class="ext">»</span></a>
-</li></ul>
-</div>
-<div id="main">
-<p>
-<span style="color: #0000c0;">LuaJIT 2.0</span> is the current
-<span style="color: #0000c0;">stable branch</span>. This branch is in
-feature-freeze — new features will only be added to LuaJIT 2.1.
-</p>
-
-<h2>Current Status</h2>
-<p>
-LuaJIT ought to run all Lua 5.1-compatible source code just fine.
-It's considered a serious bug if the VM crashes or produces unexpected
-results — please report this.
-</p>
-<p>
-Known incompatibilities and issues in LuaJIT 2.0:
-</p>
-<ul>
-<li>
-There are some differences in <b>implementation-defined</b> behavior.
-These either have a good reason, are arbitrary design choices
-or are due to quirks in the VM. The latter cases may get fixed if a
-demonstrable need is shown.
-</li>
-<li>
-The Lua <b>debug API</b> is missing a couple of features (return
-hooks for non-Lua functions) and shows slightly different behavior
-in LuaJIT (no per-coroutine hooks, no tail call counting).
-</li>
-<li>
-Currently some <b>out-of-memory</b> errors from <b>on-trace code</b> are not
-handled correctly. The error may fall through an on-trace
-<tt>pcall</tt> or it may be passed on to the function set with
-<tt>lua_atpanic</tt> on x64. This issue will be fixed with the new
-garbage collector.
-</li>
-<li>
-LuaJIT on 64 bit systems provides a <b>limited range</b> of 47 bits for the
-<b>legacy <tt>lightuserdata</tt></b> data type.
-This is only relevant on x64 systems which use the negative part of the
-virtual address space in user mode, e.g. Solaris/x64, and on ARM64 systems
-configured with a 48 bit or 52 bit VA.
-Avoid using <tt>lightuserdata</tt> to hold pointers that may point outside
-of that range, e.g. variables on the stack. In general, avoid this data
-type for new code and replace it with (much more performant) FFI bindings.
-FFI cdata pointers can address the full 64 bit range.
-</li>
-</ul>
-<br class="flush">
-</div>
-<div id="foot">
-<hr class="hide">
-Copyright © 2005-2017 Mike Pall
-<span class="noprint">
-·
-<a href="contact.html">Contact</a>
-</span>
-</div>
-</body>
-</html>
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM ARM encoding engine.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -70,7 +70,7 @@
size_t lgsize;
int *pclabels; /* PC label chains/pos ptrs. */
size_t pcsize;
- void **globals; /* Array of globals (bias -10). */
+ void **globals; /* Array of globals. */
dasm_Section *section; /* Pointer to active section. */
size_t codesize; /* Total size of all code sections. */
int maxsection; /* 0 <= sectionidx < maxsection. */
@@ -87,7 +87,6 @@
{
dasm_State *D;
size_t psz = 0;
- int i;
Dst_REF = NULL;
DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
D = Dst_REF;
@@ -98,12 +97,7 @@
D->pcsize = 0;
D->globals = NULL;
D->maxsection = maxsection;
- for (i = 0; i < maxsection; i++) {
- D->sections[i].buf = NULL; /* Need this for pass3. */
- D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
- D->sections[i].bsize = 0;
- D->sections[i].epos = 0; /* Wrong, but is recalculated after resize. */
- }
+ memset((void *)D->sections, 0, maxsection * sizeof(dasm_Section));
}
/* Free DynASM state. */
@@ -123,7 +117,7 @@
void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
{
dasm_State *D = Dst_REF;
- D->globals = gl - 10; /* Negative bias to compensate for locals. */
+ D->globals = gl;
DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
}
@@ -148,6 +142,7 @@
if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
for (i = 0; i < D->maxsection; i++) {
D->sections[i].pos = DASM_SEC2POS(i);
+ D->sections[i].rbuf = D->sections[i].buf - D->sections[i].pos;
D->sections[i].ofs = 0;
}
}
@@ -254,6 +249,7 @@
case DASM_IMMV8:
CK((n & 3) == 0, RANGE_I);
n >>= 2;
+ /* fallthrough */
case DASM_IMML8:
case DASM_IMML12:
CK(n >= 0 ? ((n>>((ins>>5)&31)) == 0) :
@@ -293,7 +289,7 @@
{ /* Handle globals not defined in this translation unit. */
int idx;
- for (idx = 20; idx*sizeof(int) < D->lgsize; idx++) {
+ for (idx = 10; idx*sizeof(int) < D->lgsize; idx++) {
int n = D->lglabels[idx];
/* Undefined label: Collapse rel chain and replace with marker (< 0). */
while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = -idx; }
@@ -370,7 +366,11 @@
ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0xe1a00000;
break;
case DASM_REL_LG:
- CK(n >= 0, UNDEF_LG);
+ if (n < 0) {
+ n = (int)((ptrdiff_t)D->globals[-n-10] - (ptrdiff_t)cp - 4);
+ goto patchrel;
+ }
+ /* fallthrough */
case DASM_REL_PC:
CK(n >= 0, UNDEF_PC);
n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base) - 4;
@@ -391,7 +391,7 @@
}
break;
case DASM_LABEL_LG:
- ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
+ ins &= 2047; if (ins >= 20) D->globals[ins-20] = (void *)(base + n);
break;
case DASM_LABEL_PC: break;
case DASM_IMM:
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM ARM module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
@@ -9,9 +9,9 @@
local _info = {
arch = "arm",
description = "DynASM ARM module",
- version = "1.4.0",
- vernum = 10400,
- release = "2015-10-18",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
license = "MIT",
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM ARM64 encoding engine.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -21,8 +21,9 @@
/* The following actions need a buffer position. */
DASM_ALIGN, DASM_REL_LG, DASM_LABEL_LG,
/* The following actions also have an argument. */
- DASM_REL_PC, DASM_LABEL_PC,
+ DASM_REL_PC, DASM_LABEL_PC, DASM_REL_A,
DASM_IMM, DASM_IMM6, DASM_IMM12, DASM_IMM13W, DASM_IMM13X, DASM_IMML,
+ DASM_IMMV, DASM_VREG,
DASM__MAX
};
@@ -39,6 +40,7 @@
#define DASM_S_RANGE_LG 0x13000000
#define DASM_S_RANGE_PC 0x14000000
#define DASM_S_RANGE_REL 0x15000000
+#define DASM_S_RANGE_VREG 0x16000000
#define DASM_S_UNDEF_LG 0x21000000
#define DASM_S_UNDEF_PC 0x22000000
@@ -70,7 +72,7 @@
size_t lgsize;
int *pclabels; /* PC label chains/pos ptrs. */
size_t pcsize;
- void **globals; /* Array of globals (bias -10). */
+ void **globals; /* Array of globals. */
dasm_Section *section; /* Pointer to active section. */
size_t codesize; /* Total size of all code sections. */
int maxsection; /* 0 <= sectionidx < maxsection. */
@@ -87,7 +89,6 @@
{
dasm_State *D;
size_t psz = 0;
- int i;
Dst_REF = NULL;
DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
D = Dst_REF;
@@ -98,12 +99,7 @@
D->pcsize = 0;
D->globals = NULL;
D->maxsection = maxsection;
- for (i = 0; i < maxsection; i++) {
- D->sections[i].buf = NULL; /* Need this for pass3. */
- D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
- D->sections[i].bsize = 0;
- D->sections[i].epos = 0; /* Wrong, but is recalculated after resize. */
- }
+ memset((void *)D->sections, 0, maxsection * sizeof(dasm_Section));
}
/* Free DynASM state. */
@@ -123,7 +119,7 @@
void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
{
dasm_State *D = Dst_REF;
- D->globals = gl - 10; /* Negative bias to compensate for locals. */
+ D->globals = gl;
DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
}
@@ -148,6 +144,7 @@
if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
for (i = 0; i < D->maxsection; i++) {
D->sections[i].pos = DASM_SEC2POS(i);
+ D->sections[i].rbuf = D->sections[i].buf - D->sections[i].pos;
D->sections[i].ofs = 0;
}
}
@@ -156,10 +153,10 @@
#ifdef DASM_CHECKS
#define CK(x, st) \
do { if (!(x)) { \
- D->status = DASM_S_##st|(p-D->actionlist-1); return; } } while (0)
+ D->status = DASM_S_##st|(int)(p-D->actionlist-1); return; } } while (0)
#define CKPL(kind, st) \
do { if ((size_t)((char *)pl-(char *)D->kind##labels) >= D->kind##size) { \
- D->status = DASM_S_RANGE_##st|(p-D->actionlist-1); return; } } while (0)
+ D->status = DASM_S_RANGE_##st|(int)(p-D->actionlist-1); return; } } while (0)
#else
#define CK(x, st) ((void)0)
#define CKPL(kind, st) ((void)0)
@@ -188,7 +185,9 @@
unsigned long long n = (((unsigned long long)hi) << 32) | (unsigned int)lo;
unsigned long long m = 1ULL, a, b, c;
if (n & 1) { n = ~n; inv = 1; }
- a = n & -n; b = (n+a)&-(n+a); c = (n+a-b)&-(n+a-b);
+ a = n & (unsigned long long)-(long long)n;
+ b = (n+a)&(unsigned long long)-(long long)(n+a);
+ c = (n+a-b)&(unsigned long long)-(long long)(n+a-b);
xa = dasm_ffs(a); xb = dasm_ffs(b);
if (c) {
w = dasm_ffs(c) - xa;
@@ -247,7 +246,7 @@
n = (ins & 255); CK(n < D->maxsection, RANGE_SEC);
D->section = &D->sections[n]; goto stop;
case DASM_ESC: p++; ofs += 4; break;
- case DASM_REL_EXT: break;
+ case DASM_REL_EXT: if ((ins & 0x8000)) ofs += 8; break;
case DASM_ALIGN: ofs += (ins & 255); b[pos++] = ofs; break;
case DASM_REL_LG:
n = (ins & 2047) - 10; pl = D->lglabels + n;
@@ -268,7 +267,12 @@
*pl = pos;
}
pos++;
+ if ((ins & 0x8000)) ofs += 8;
break;
+ case DASM_REL_A:
+ b[pos++] = n;
+ b[pos++] = va_arg(ap, int);
+ break;
case DASM_LABEL_LG:
pl = D->lglabels + (ins & 2047) - 10; CKPL(lg, LG); goto putlabel;
case DASM_LABEL_PC:
@@ -312,7 +316,7 @@
}
case DASM_IMML: {
#ifdef DASM_CHECKS
- int scale = (p[-2] >> 30);
+ int scale = (ins & 3);
CK((!(n & ((1<<scale)-1)) && (unsigned int)(n>>scale) < 4096) ||
(unsigned int)(n+256) < 512, RANGE_I);
#endif
@@ -319,6 +323,14 @@
b[pos++] = n;
break;
}
+ case DASM_IMMV:
+ ofs += 4;
+ b[pos++] = n;
+ break;
+ case DASM_VREG:
+ CK(n < 32, RANGE_VREG);
+ b[pos++] = n;
+ break;
}
}
}
@@ -348,7 +360,7 @@
{ /* Handle globals not defined in this translation unit. */
int idx;
- for (idx = 20; idx*sizeof(int) < D->lgsize; idx++) {
+ for (idx = 10; idx*sizeof(int) < D->lgsize; idx++) {
int n = D->lglabels[idx];
/* Undefined label: Collapse rel chain and replace with marker (< 0). */
while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = -idx; }
@@ -375,8 +387,8 @@
case DASM_REL_LG: case DASM_REL_PC: pos++; break;
case DASM_LABEL_LG: case DASM_LABEL_PC: b[pos++] += ofs; break;
case DASM_IMM: case DASM_IMM6: case DASM_IMM12: case DASM_IMM13W:
- case DASM_IMML: pos++; break;
- case DASM_IMM13X: pos += 2; break;
+ case DASM_IMML: case DASM_IMMV: case DASM_VREG: pos++; break;
+ case DASM_IMM13X: case DASM_REL_A: pos += 2; break;
}
}
stop: (void)0;
@@ -391,7 +403,7 @@
#ifdef DASM_CHECKS
#define CK(x, st) \
- do { if (!(x)) return DASM_S_##st|(p-D->actionlist-1); } while (0)
+ do { if (!(x)) return DASM_S_##st|(int)(p-D->actionlist-1); } while (0)
#else
#define CK(x, st) ((void)0)
#endif
@@ -423,10 +435,16 @@
n = DASM_EXTERN(Dst, (unsigned char *)cp, (ins&2047), !(ins&2048));
goto patchrel;
case DASM_ALIGN:
- ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0xe1a00000;
+ ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0xd503201f;
break;
case DASM_REL_LG:
- CK(n >= 0, UNDEF_LG);
+ if (n < 0) {
+ ptrdiff_t na = (ptrdiff_t)D->globals[-n-10] - (ptrdiff_t)cp + 4;
+ n = (int)na;
+ CK((ptrdiff_t)n == na, RANGE_REL);
+ goto patchrel;
+ }
+ /* fallthrough */
case DASM_REL_PC:
CK(n >= 0, UNDEF_PC);
n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base) + 4;
@@ -445,10 +463,26 @@
} else if ((ins & 0x1000)) { /* TBZ, TBNZ */
CK((n & 3) == 0 && ((n+0x00008000) >> 16) == 0, RANGE_REL);
cp[-1] |= ((n << 3) & 0x0007ffe0);
+ } else if ((ins & 0x8000)) { /* absolute */
+ cp[0] = (unsigned int)((ptrdiff_t)cp - 4 + n);
+ cp[1] = (unsigned int)(((ptrdiff_t)cp - 4 + n) >> 32);
+ cp += 2;
}
break;
+ case DASM_REL_A: {
+ ptrdiff_t na = (((ptrdiff_t)(*b++) << 32) | (unsigned int)n);
+ if ((ins & 0x3000) == 0x3000) { /* ADRP */
+ ins &= ~0x1000;
+ na = (na >> 12) - (((ptrdiff_t)cp - 4) >> 12);
+ } else {
+ na = na - (ptrdiff_t)cp + 4;
+ }
+ n = (int)na;
+ CK((ptrdiff_t)n == na, RANGE_REL);
+ goto patchrel;
+ }
case DASM_LABEL_LG:
- ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
+ ins &= 2047; if (ins >= 20) D->globals[ins-20] = (void *)(base + n);
break;
case DASM_LABEL_PC: break;
case DASM_IMM:
@@ -467,11 +501,17 @@
cp[-1] |= (dasm_imm13(n, *b++) << 10);
break;
case DASM_IMML: {
- int scale = (p[-2] >> 30);
+ int scale = (ins & 3);
cp[-1] |= (!(n & ((1<<scale)-1)) && (unsigned int)(n>>scale) < 4096) ?
((n << (10-scale)) | 0x01000000) : ((n & 511) << 12);
break;
}
+ case DASM_IMMV:
+ *cp++ = n;
+ break;
+ case DASM_VREG:
+ cp[-1] |= (n & 0x1f) << (ins & 0x1f);
+ break;
default: *cp++ = ins; break;
}
}
@@ -511,7 +551,7 @@
}
if (D->status == DASM_S_OK && secmatch >= 0 &&
D->section != &D->sections[secmatch])
- D->status = DASM_S_MATCH_SEC|(D->section-D->sections);
+ D->status = DASM_S_MATCH_SEC|(int)(D->section-D->sections);
return D->status;
}
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_arm64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM ARM64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
@@ -9,9 +9,9 @@
local _info = {
arch = "arm",
description = "DynASM ARM64 module",
- version = "1.4.0",
- vernum = 10400,
- release = "2015-10-18",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
license = "MIT",
}
@@ -23,12 +23,12 @@
local type, tonumber, pairs, ipairs = type, tonumber, pairs, ipairs
local assert, setmetatable, rawget = assert, setmetatable, rawget
local _s = string
-local sub, format, byte, char = _s.sub, _s.format, _s.byte, _s.char
+local format, byte, char = _s.format, _s.byte, _s.char
local match, gmatch, gsub = _s.match, _s.gmatch, _s.gsub
local concat, sort, insert = table.concat, table.sort, table.insert
local bit = bit or require("bit")
local band, shl, shr, sar = bit.band, bit.lshift, bit.rshift, bit.arshift
-local ror, tohex = bit.ror, bit.tohex
+local ror, tohex, tobit = bit.ror, bit.tohex, bit.tobit
-- Inherited tables and callbacks.
local g_opt, g_arch
@@ -39,7 +39,9 @@
local action_names = {
"STOP", "SECTION", "ESC", "REL_EXT",
"ALIGN", "REL_LG", "LABEL_LG",
- "REL_PC", "LABEL_PC", "IMM", "IMM6", "IMM12", "IMM13W", "IMM13X", "IMML",
+ "REL_PC", "LABEL_PC", "REL_A",
+ "IMM", "IMM6", "IMM12", "IMM13W", "IMM13X", "IMML", "IMMV",
+ "VREG",
}
-- Maximum number of section buffer positions for dasm_put().
@@ -246,9 +248,12 @@
local parse_reg_type
-local function parse_reg(expr)
+local function parse_reg(expr, shift, no_vreg)
if not expr then werror("expected register name") end
local tname, ovreg = match(expr, "^([%w_]+):(@?%l%d+)$")
+ if not tname then
+ tname, ovreg = match(expr, "^([%w_]+):(R[xwqdshb]%b())$")
+ end
local tp = map_type[tname or expr]
if tp then
local reg = ovreg or tp.reg
@@ -266,18 +271,28 @@
elseif parse_reg_type ~= rt then
werror("register size mismatch")
end
- return r, tp
+ return shl(r, shift), tp
end
end
+ local vrt, vreg = match(expr, "^R([xwqdshb])(%b())$")
+ if vreg then
+ if not parse_reg_type then
+ parse_reg_type = vrt
+ elseif parse_reg_type ~= vrt then
+ werror("register size mismatch")
+ end
+ if not no_vreg then waction("VREG", shift, vreg) end
+ return 0
+ end
werror("bad register name `"..expr.."'")
end
local function parse_reg_base(expr)
if expr == "sp" then return 0x3e0 end
- local base, tp = parse_reg(expr)
+ local base, tp = parse_reg(expr, 5)
if parse_reg_type ~= "x" then werror("bad register type") end
parse_reg_type = false
- return shl(base, 5), tp
+ return base, tp
end
local parse_ctx = {}
@@ -297,7 +312,7 @@
local code = loadenv("return "..n)
if code then
local ok, y = pcall(code)
- if ok then return y end
+ if ok and type(y) == "number" then return y end
end
return nil
end
@@ -403,7 +418,7 @@
end
werror("out of range immediate `"..imm.."'")
else
- waction("IMML", 0, imm)
+ waction("IMML", scale, imm)
return 0
end
end
@@ -462,6 +477,7 @@
local function parse_load(params, nparams, n, op)
if params[n+2] then werror("too many operands") end
+ local scale = shr(op, 30)
local pn, p2 = params[n], params[n+1]
local p1, wb = match(pn, "^%[%s*(.-)%s*%](!?)$")
if not p1 then
@@ -470,7 +486,7 @@
if reg and tailr ~= "" then
local base, tp = parse_reg_base(reg)
if tp then
- waction("IMML", 0, format(tp.ctypefmt, tailr))
+ waction("IMML", scale, format(tp.ctypefmt, tailr))
return op + base
end
end
@@ -477,7 +493,6 @@
end
werror("expected address operand")
end
- local scale = shr(op, 30)
if p2 then
if wb == "!" then werror("bad use of '!'") end
op = op + parse_reg_base(p1) + parse_imm(p2, 9, 12, 0, true) + 0x400
@@ -494,7 +509,7 @@
op = op + parse_imm_load(imm, scale)
else
local p2b, p3b, p3s = match(p2a, "^,%s*([^,%s]*)%s*,?%s*(%S*)%s*(.*)$")
- op = op + shl(parse_reg(p2b), 16) + 0x00200800
+ op = op + parse_reg(p2b, 16) + 0x00200800
if parse_reg_type ~= "x" and parse_reg_type ~= "w" then
werror("bad index register type")
end
@@ -534,7 +549,7 @@
local function parse_load_pair(params, nparams, n, op)
if params[n+2] then werror("too many operands") end
local pn, p2 = params[n], params[n+1]
- local scale = shr(op, 30) == 0 and 2 or 3
+ local scale = 2 + shr(op, 31 - band(shr(op, 26), 1))
local p1, wb = match(pn, "^%[%s*(.-)%s*%](!?)$")
if not p1 then
if not p2 then
@@ -561,14 +576,14 @@
end
local function parse_label(label, def)
- local prefix = sub(label, 1, 2)
+ local prefix = label:sub(1, 2)
-- =>label (pc label reference)
if prefix == "=>" then
- return "PC", 0, sub(label, 3)
+ return "PC", 0, label:sub(3)
end
-- ->name (global label reference)
if prefix == "->" then
- return "LG", map_global[sub(label, 3)]
+ return "LG", map_global[label:sub(3)]
end
if def then
-- [1-9] (local label definition)
@@ -586,8 +601,11 @@
if extname then
return "EXT", map_extern[extname]
end
+ -- &expr (pointer)
+ if label:sub(1, 1) == "&" then
+ return "A", 0, format("(ptrdiff_t)(%s)", label:sub(2))
+ end
end
- werror("bad label `"..label.."'")
end
local function branch_type(op)
@@ -620,24 +638,24 @@
end
local function alias_bfiz(p)
- parse_reg(p[1])
+ parse_reg(p[1], 0, true)
if parse_reg_type == "w" then
- p[3] = "#-("..p[3]:sub(2)..")%32"
+ p[3] = "#(32-("..p[3]:sub(2).."))%32"
p[4] = "#("..p[4]:sub(2)..")-1"
else
- p[3] = "#-("..p[3]:sub(2)..")%64"
+ p[3] = "#(64-("..p[3]:sub(2).."))%64"
p[4] = "#("..p[4]:sub(2)..")-1"
end
end
local alias_lslimm = op_alias("ubfm_4", function(p)
- parse_reg(p[1])
+ parse_reg(p[1], 0, true)
local sh = p[3]:sub(2)
if parse_reg_type == "w" then
- p[3] = "#-("..sh..")%32"
+ p[3] = "#(32-("..sh.."))%32"
p[4] = "#31-("..sh..")"
else
- p[3] = "#-("..sh..")%64"
+ p[3] = "#(64-("..sh.."))%64"
p[4] = "#63-("..sh..")"
end
end)
@@ -788,8 +806,8 @@
["ldrsw_*"] = "98000000DxB|b8800000DxL",
-- NOTE: ldur etc. are handled by ldr et al.
- ["stp_*"] = "28000000DAwP|a8000000DAxP|2c000000DAsP|6c000000DAdP",
- ["ldp_*"] = "28400000DAwP|a8400000DAxP|2c400000DAsP|6c400000DAdP",
+ ["stp_*"] = "28000000DAwP|a8000000DAxP|2c000000DAsP|6c000000DAdP|ac000000DAqP",
+ ["ldp_*"] = "28400000DAwP|a8400000DAxP|2c400000DAsP|6c400000DAdP|ac400000DAqP",
["ldpsw_*"] = "68400000DAxP",
-- Branches.
@@ -805,6 +823,13 @@
tbz_3 = "36000000DTBw|36000000DTBx",
tbnz_3 = "37000000DTBw|37000000DTBx",
+ -- ARM64e: Pointer authentication codes (PAC).
+ blraaz_1 = "d63f081fNx",
+ braa_2 = "d71f0800NDx",
+ braaz_1 = "d61f081fNx",
+ pacibsp_0 = "d503237f",
+ retab_0 = "d65f0fff",
+
-- Miscellaneous instructions.
-- TODO: hlt, hvc, smc, svc, eret, dcps[123], drps, mrs, msr
-- TODO: sys, sysl, ic, dc, at, tlbi
@@ -881,7 +906,7 @@
-- Handle opcodes defined with template strings.
local function parse_template(params, template, nparams, pos)
- local op = tonumber(sub(template, 1, 8), 16)
+ local op = tonumber(template:sub(1, 8), 16)
local n = 1
local rtt = {}
@@ -888,18 +913,18 @@
parse_reg_type = false
-- Process each character.
- for p in gmatch(sub(template, 9), ".") do
+ for p in gmatch(template:sub(9), ".") do
local q = params[n]
if p == "D" then
- op = op + parse_reg(q); n = n + 1
+ op = op + parse_reg(q, 0); n = n + 1
elseif p == "N" then
- op = op + shl(parse_reg(q), 5); n = n + 1
+ op = op + parse_reg(q, 5); n = n + 1
elseif p == "M" then
- op = op + shl(parse_reg(q), 16); n = n + 1
+ op = op + parse_reg(q, 16); n = n + 1
elseif p == "A" then
- op = op + shl(parse_reg(q), 10); n = n + 1
+ op = op + parse_reg(q, 10); n = n + 1
elseif p == "m" then
- op = op + shl(parse_reg(params[n-1]), 16)
+ op = op + parse_reg(params[n-1], 16)
elseif p == "p" then
if q == "sp" then params[n] = "@x31" end
@@ -917,7 +942,7 @@
werror("bad register type")
end
parse_reg_type = false
- elseif p == "x" or p == "w" or p == "d" or p == "s" then
+ elseif p == "x" or p == "w" or p == "d" or p == "s" or p == "q" then
if parse_reg_type ~= p then
werror("register size mismatch")
end
@@ -930,8 +955,14 @@
elseif p == "B" then
local mode, v, s = parse_label(q, false); n = n + 1
+ if not mode then werror("bad label `"..q.."'") end
local m = branch_type(op)
- waction("REL_"..mode, v+m, s, 1)
+ if mode == "A" then
+ waction("REL_"..mode, v+m, format("(unsigned int)(%s)", s))
+ actargs[#actargs+1] = format("(unsigned int)((%s)>>32)", s)
+ else
+ waction("REL_"..mode, v+m, s, 1)
+ end
elseif p == "I" then
op = op + parse_imm12(q); n = n + 1
@@ -977,8 +1008,8 @@
if not params then return template:gsub("%x%x%x%x%x%x%x%x", "") end
-- Limit number of section buffer positions used by a single dasm_put().
- -- A single opcode needs a maximum of 3 positions.
- if secpos+3 > maxsecpos then wflush() end
+ -- A single opcode needs a maximum of 4 positions.
+ if secpos+4 > maxsecpos then wflush() end
local pos = wpos()
local lpos, apos, spos = #actlist, #actargs, secpos
@@ -990,9 +1021,11 @@
actlist[lpos+1] = nil
actlist[lpos+2] = nil
actlist[lpos+3] = nil
+ actlist[lpos+4] = nil
actargs[apos+1] = nil
actargs[apos+2] = nil
actargs[apos+3] = nil
+ actargs[apos+4] = nil
end
error(err, 0)
end
@@ -1036,7 +1069,7 @@
if not params then return "[1-9] | ->global | =>pcexpr" end
if secpos+1 > maxsecpos then wflush() end
local mode, n, s = parse_label(params[1], true)
- if mode == "EXT" then werror("bad label definition") end
+ if not mode or mode == "EXT" then werror("bad label definition") end
waction("LABEL_"..mode, n, s, 1)
end
@@ -1043,16 +1076,43 @@
------------------------------------------------------------------------------
-- Pseudo-opcodes for data storage.
-map_op[".long_*"] = function(params)
+local function op_data(params)
if not params then return "imm..." end
+ local sz = params.op == ".long" and 4 or 8
for _,p in ipairs(params) do
- local n = tonumber(p)
- if not n then werror("bad immediate `"..p.."'") end
- if n < 0 then n = n + 2^32 end
- wputw(n)
+ local imm = parse_number(p)
+ if imm then
+ local n = tobit(imm)
+ if n == imm or (n < 0 and n + 2^32 == imm) then
+ wputw(n < 0 and n + 2^32 or n)
+ if sz == 8 then
+ wputw(imm < 0 and 0xffffffff or 0)
+ end
+ elseif sz == 4 then
+ werror("bad immediate `"..p.."'")
+ else
+ imm = nil
+ end
+ end
+ if not imm then
+ local mode, v, s = parse_label(p, false)
+ if sz == 4 then
+ if mode then werror("label does not fit into .long") end
+ waction("IMMV", 0, p)
+ elseif mode and mode ~= "A" then
+ waction("REL_"..mode, v+0x8000, s, 1)
+ else
+ if mode == "A" then p = s end
+ waction("IMMV", 0, format("(unsigned int)(%s)", p))
+ waction("IMMV", 0, format("(unsigned int)((unsigned long long)(%s)>>32)", p))
+ end
+ end
if secpos+2 > maxsecpos then wflush() end
end
end
+map_op[".long_*"] = op_data
+map_op[".quad_*"] = op_data
+map_op[".addr_*"] = op_data
-- Alignment pseudo-opcode.
map_op[".align_1"] = function(params)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM MIPS encoding engine.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -69,7 +69,7 @@
size_t lgsize;
int *pclabels; /* PC label chains/pos ptrs. */
size_t pcsize;
- void **globals; /* Array of globals (bias -10). */
+ void **globals; /* Array of globals. */
dasm_Section *section; /* Pointer to active section. */
size_t codesize; /* Total size of all code sections. */
int maxsection; /* 0 <= sectionidx < maxsection. */
@@ -86,7 +86,6 @@
{
dasm_State *D;
size_t psz = 0;
- int i;
Dst_REF = NULL;
DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
D = Dst_REF;
@@ -97,12 +96,7 @@
D->pcsize = 0;
D->globals = NULL;
D->maxsection = maxsection;
- for (i = 0; i < maxsection; i++) {
- D->sections[i].buf = NULL; /* Need this for pass3. */
- D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
- D->sections[i].bsize = 0;
- D->sections[i].epos = 0; /* Wrong, but is recalculated after resize. */
- }
+ memset((void *)D->sections, 0, maxsection * sizeof(dasm_Section));
}
/* Free DynASM state. */
@@ -122,7 +116,7 @@
void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
{
dasm_State *D = Dst_REF;
- D->globals = gl - 10; /* Negative bias to compensate for locals. */
+ D->globals = gl;
DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
}
@@ -147,6 +141,7 @@
if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
for (i = 0; i < D->maxsection; i++) {
D->sections[i].pos = DASM_SEC2POS(i);
+ D->sections[i].rbuf = D->sections[i].buf - D->sections[i].pos;
D->sections[i].ofs = 0;
}
}
@@ -155,10 +150,10 @@
#ifdef DASM_CHECKS
#define CK(x, st) \
do { if (!(x)) { \
- D->status = DASM_S_##st|(p-D->actionlist-1); return; } } while (0)
+ D->status = DASM_S_##st|(int)(p-D->actionlist-1); return; } } while (0)
#define CKPL(kind, st) \
do { if ((size_t)((char *)pl-(char *)D->kind##labels) >= D->kind##size) { \
- D->status = DASM_S_RANGE_##st|(p-D->actionlist-1); return; } } while (0)
+ D->status = DASM_S_RANGE_##st|(int)(p-D->actionlist-1); return; } } while (0)
#else
#define CK(x, st) ((void)0)
#define CKPL(kind, st) ((void)0)
@@ -273,7 +268,7 @@
{ /* Handle globals not defined in this translation unit. */
int idx;
- for (idx = 20; idx*sizeof(int) < D->lgsize; idx++) {
+ for (idx = 10; idx*sizeof(int) < D->lgsize; idx++) {
int n = D->lglabels[idx];
/* Undefined label: Collapse rel chain and replace with marker (< 0). */
while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = -idx; }
@@ -314,7 +309,7 @@
#ifdef DASM_CHECKS
#define CK(x, st) \
- do { if (!(x)) return DASM_S_##st|(p-D->actionlist-1); } while (0)
+ do { if (!(x)) return DASM_S_##st|(int)(p-D->actionlist-1); } while (0)
#else
#define CK(x, st) ((void)0)
#endif
@@ -349,22 +344,27 @@
ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0x60000000;
break;
case DASM_REL_LG:
- CK(n >= 0, UNDEF_LG);
+ if (n < 0) {
+ n = (int)((ptrdiff_t)D->globals[-n-10] - (ptrdiff_t)cp);
+ goto patchrel;
+ }
+ /* fallthrough */
case DASM_REL_PC:
CK(n >= 0, UNDEF_PC);
n = *DASM_POS2PTR(D, n);
if (ins & 2048)
+ n = (n + (int)(size_t)base) & 0x0fffffff;
+ else
n = n - (int)((char *)cp - base);
- else
- n = (n + (int)(size_t)base) & 0x0fffffff;
- patchrel:
+ patchrel: {
+ unsigned int e = 16 + ((ins >> 12) & 15);
CK((n & 3) == 0 &&
- ((n + ((ins & 2048) ? 0x00020000 : 0)) >>
- ((ins & 2048) ? 18 : 28)) == 0, RANGE_REL);
- cp[-1] |= ((n>>2) & ((ins & 2048) ? 0x0000ffff: 0x03ffffff));
+ ((n + ((ins & 2048) ? 0 : (1<<(e+1)))) >> (e+2)) == 0, RANGE_REL);
+ cp[-1] |= ((n>>2) & ((1<<e)-1));
+ }
break;
case DASM_LABEL_LG:
- ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
+ ins &= 2047; if (ins >= 20) D->globals[ins-20] = (void *)(base + n);
break;
case DASM_LABEL_PC: break;
case DASM_IMMS:
@@ -412,7 +412,7 @@
}
if (D->status == DASM_S_OK && secmatch >= 0 &&
D->section != &D->sections[secmatch])
- D->status = DASM_S_MATCH_SEC|(D->section-D->sections);
+ D->status = DASM_S_MATCH_SEC|(int)(D->section-D->sections);
return D->status;
}
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,19 +1,20 @@
------------------------------------------------------------------------------
-- DynASM MIPS32/MIPS64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
local mips64 = mips64
+local mipsr6 = _map_def.MIPSR6
-- Module information:
local _info = {
arch = mips64 and "mips64" or "mips",
description = "DynASM MIPS32/MIPS64 module",
- version = "1.4.0",
- vernum = 10400,
- release = "2016-05-24",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
license = "MIT",
}
@@ -238,7 +239,6 @@
bne_3 = "14000000STB",
blez_2 = "18000000SB",
bgtz_2 = "1c000000SB",
- addi_3 = "20000000TSI",
li_2 = "24000000TI",
addiu_3 = "24000000TSI",
slti_3 = "28000000TSI",
@@ -248,40 +248,22 @@
ori_3 = "34000000TSU",
xori_3 = "38000000TSU",
lui_2 = "3c000000TU",
- beqzl_2 = "50000000SB",
- beql_3 = "50000000STB",
- bnezl_2 = "54000000SB",
- bnel_3 = "54000000STB",
- blezl_2 = "58000000SB",
- bgtzl_2 = "5c000000SB",
- daddi_3 = mips64 and "60000000TSI",
daddiu_3 = mips64 and "64000000TSI",
ldl_2 = mips64 and "68000000TO",
ldr_2 = mips64 and "6c000000TO",
lb_2 = "80000000TO",
lh_2 = "84000000TO",
- lwl_2 = "88000000TO",
lw_2 = "8c000000TO",
lbu_2 = "90000000TO",
lhu_2 = "94000000TO",
- lwr_2 = "98000000TO",
lwu_2 = mips64 and "9c000000TO",
sb_2 = "a0000000TO",
sh_2 = "a4000000TO",
- swl_2 = "a8000000TO",
sw_2 = "ac000000TO",
- sdl_2 = mips64 and "b0000000TO",
- sdr_2 = mips64 and "b1000000TO",
- swr_2 = "b8000000TO",
- cache_2 = "bc000000NO",
- ll_2 = "c0000000TO",
lwc1_2 = "c4000000HO",
- pref_2 = "cc000000NO",
ldc1_2 = "d4000000HO",
ld_2 = mips64 and "dc000000TO",
- sc_2 = "e0000000TO",
swc1_2 = "e4000000HO",
- scd_2 = mips64 and "f0000000TO",
sdc1_2 = "f4000000HO",
sd_2 = mips64 and "fc000000TO",
@@ -289,10 +271,6 @@
nop_0 = "00000000",
sll_3 = "00000000DTA",
sextw_2 = "00000000DT",
- movf_2 = "00000001DS",
- movf_3 = "00000001DSC",
- movt_2 = "00010001DS",
- movt_3 = "00010001DSC",
srl_3 = "00000002DTA",
rotr_3 = "00200002DTA",
sra_3 = "00000003DTA",
@@ -301,31 +279,16 @@
rotrv_3 = "00000046DTS",
drotrv_3 = mips64 and "00000056DTS",
srav_3 = "00000007DTS",
- jr_1 = "00000008S",
jalr_1 = "0000f809S",
jalr_2 = "00000009DS",
- movz_3 = "0000000aDST",
- movn_3 = "0000000bDST",
syscall_0 = "0000000c",
syscall_1 = "0000000cY",
break_0 = "0000000d",
break_1 = "0000000dY",
sync_0 = "0000000f",
- mfhi_1 = "00000010D",
- mthi_1 = "00000011S",
- mflo_1 = "00000012D",
- mtlo_1 = "00000013S",
dsllv_3 = mips64 and "00000014DTS",
dsrlv_3 = mips64 and "00000016DTS",
dsrav_3 = mips64 and "00000017DTS",
- mult_2 = "00000018ST",
- multu_2 = "00000019ST",
- div_2 = "0000001aST",
- divu_2 = "0000001bST",
- dmult_2 = mips64 and "0000001cST",
- dmultu_2 = mips64 and "0000001dST",
- ddiv_2 = mips64 and "0000001eST",
- ddivu_2 = mips64 and "0000001fST",
add_3 = "00000020DST",
move_2 = mips64 and "00000025DS" or "00000021DS",
addu_3 = "00000021DST",
@@ -369,32 +332,9 @@
bgez_2 = "04010000SB",
bltzl_2 = "04020000SB",
bgezl_2 = "04030000SB",
- tgei_2 = "04080000SI",
- tgeiu_2 = "04090000SI",
- tlti_2 = "040a0000SI",
- tltiu_2 = "040b0000SI",
- teqi_2 = "040c0000SI",
- tnei_2 = "040e0000SI",
- bltzal_2 = "04100000SB",
bal_1 = "04110000B",
- bgezal_2 = "04110000SB",
- bltzall_2 = "04120000SB",
- bgezall_2 = "04130000SB",
synci_1 = "041f0000O",
- -- Opcode SPECIAL2.
- madd_2 = "70000000ST",
- maddu_2 = "70000001ST",
- mul_3 = "70000002DST",
- msub_2 = "70000004ST",
- msubu_2 = "70000005ST",
- clz_2 = "70000020DS=",
- clo_2 = "70000021DS=",
- dclz_2 = mips64 and "70000024DS=",
- dclo_2 = mips64 and "70000025DS=",
- sdbbp_0 = "7000003f",
- sdbbp_1 = "7000003fY",
-
-- Opcode SPECIAL3.
ext_4 = "7c000000TSAM", -- Note: last arg is msbd = size-1
dextm_4 = mips64 and "7c000001TSAM", -- Args: pos | size-1-32
@@ -445,15 +385,6 @@
ctc1_2 = "44c00000TG",
mthc1_2 = "44e00000TG",
- bc1f_1 = "45000000B",
- bc1f_2 = "45000000CB",
- bc1t_1 = "45010000B",
- bc1t_2 = "45010000CB",
- bc1fl_1 = "45020000B",
- bc1fl_2 = "45020000CB",
- bc1tl_1 = "45030000B",
- bc1tl_2 = "45030000CB",
-
["add.s_3"] = "46000000FGH",
["sub.s_3"] = "46000001FGH",
["mul.s_3"] = "46000002FGH",
@@ -470,51 +401,11 @@
["trunc.w.s_2"] = "4600000dFG",
["ceil.w.s_2"] = "4600000eFG",
["floor.w.s_2"] = "4600000fFG",
- ["movf.s_2"] = "46000011FG",
- ["movf.s_3"] = "46000011FGC",
- ["movt.s_2"] = "46010011FG",
- ["movt.s_3"] = "46010011FGC",
- ["movz.s_3"] = "46000012FGT",
- ["movn.s_3"] = "46000013FGT",
["recip.s_2"] = "46000015FG",
["rsqrt.s_2"] = "46000016FG",
["cvt.d.s_2"] = "46000021FG",
["cvt.w.s_2"] = "46000024FG",
["cvt.l.s_2"] = "46000025FG",
- ["cvt.ps.s_3"] = "46000026FGH",
- ["c.f.s_2"] = "46000030GH",
- ["c.f.s_3"] = "46000030VGH",
- ["c.un.s_2"] = "46000031GH",
- ["c.un.s_3"] = "46000031VGH",
- ["c.eq.s_2"] = "46000032GH",
- ["c.eq.s_3"] = "46000032VGH",
- ["c.ueq.s_2"] = "46000033GH",
- ["c.ueq.s_3"] = "46000033VGH",
- ["c.olt.s_2"] = "46000034GH",
- ["c.olt.s_3"] = "46000034VGH",
- ["c.ult.s_2"] = "46000035GH",
- ["c.ult.s_3"] = "46000035VGH",
- ["c.ole.s_2"] = "46000036GH",
- ["c.ole.s_3"] = "46000036VGH",
- ["c.ule.s_2"] = "46000037GH",
- ["c.ule.s_3"] = "46000037VGH",
- ["c.sf.s_2"] = "46000038GH",
- ["c.sf.s_3"] = "46000038VGH",
- ["c.ngle.s_2"] = "46000039GH",
- ["c.ngle.s_3"] = "46000039VGH",
- ["c.seq.s_2"] = "4600003aGH",
- ["c.seq.s_3"] = "4600003aVGH",
- ["c.ngl.s_2"] = "4600003bGH",
- ["c.ngl.s_3"] = "4600003bVGH",
- ["c.lt.s_2"] = "4600003cGH",
- ["c.lt.s_3"] = "4600003cVGH",
- ["c.nge.s_2"] = "4600003dGH",
- ["c.nge.s_3"] = "4600003dVGH",
- ["c.le.s_2"] = "4600003eGH",
- ["c.le.s_3"] = "4600003eVGH",
- ["c.ngt.s_2"] = "4600003fGH",
- ["c.ngt.s_3"] = "4600003fVGH",
-
["add.d_3"] = "46200000FGH",
["sub.d_3"] = "46200001FGH",
["mul.d_3"] = "46200002FGH",
@@ -531,130 +422,410 @@
["trunc.w.d_2"] = "4620000dFG",
["ceil.w.d_2"] = "4620000eFG",
["floor.w.d_2"] = "4620000fFG",
- ["movf.d_2"] = "46200011FG",
- ["movf.d_3"] = "46200011FGC",
- ["movt.d_2"] = "46210011FG",
- ["movt.d_3"] = "46210011FGC",
- ["movz.d_3"] = "46200012FGT",
- ["movn.d_3"] = "46200013FGT",
["recip.d_2"] = "46200015FG",
["rsqrt.d_2"] = "46200016FG",
["cvt.s.d_2"] = "46200020FG",
["cvt.w.d_2"] = "46200024FG",
["cvt.l.d_2"] = "46200025FG",
- ["c.f.d_2"] = "46200030GH",
- ["c.f.d_3"] = "46200030VGH",
- ["c.un.d_2"] = "46200031GH",
- ["c.un.d_3"] = "46200031VGH",
- ["c.eq.d_2"] = "46200032GH",
- ["c.eq.d_3"] = "46200032VGH",
- ["c.ueq.d_2"] = "46200033GH",
- ["c.ueq.d_3"] = "46200033VGH",
- ["c.olt.d_2"] = "46200034GH",
- ["c.olt.d_3"] = "46200034VGH",
- ["c.ult.d_2"] = "46200035GH",
- ["c.ult.d_3"] = "46200035VGH",
- ["c.ole.d_2"] = "46200036GH",
- ["c.ole.d_3"] = "46200036VGH",
- ["c.ule.d_2"] = "46200037GH",
- ["c.ule.d_3"] = "46200037VGH",
- ["c.sf.d_2"] = "46200038GH",
- ["c.sf.d_3"] = "46200038VGH",
- ["c.ngle.d_2"] = "46200039GH",
- ["c.ngle.d_3"] = "46200039VGH",
- ["c.seq.d_2"] = "4620003aGH",
- ["c.seq.d_3"] = "4620003aVGH",
- ["c.ngl.d_2"] = "4620003bGH",
- ["c.ngl.d_3"] = "4620003bVGH",
- ["c.lt.d_2"] = "4620003cGH",
- ["c.lt.d_3"] = "4620003cVGH",
- ["c.nge.d_2"] = "4620003dGH",
- ["c.nge.d_3"] = "4620003dVGH",
- ["c.le.d_2"] = "4620003eGH",
- ["c.le.d_3"] = "4620003eVGH",
- ["c.ngt.d_2"] = "4620003fGH",
- ["c.ngt.d_3"] = "4620003fVGH",
-
- ["add.ps_3"] = "46c00000FGH",
- ["sub.ps_3"] = "46c00001FGH",
- ["mul.ps_3"] = "46c00002FGH",
- ["abs.ps_2"] = "46c00005FG",
- ["mov.ps_2"] = "46c00006FG",
- ["neg.ps_2"] = "46c00007FG",
- ["movf.ps_2"] = "46c00011FG",
- ["movf.ps_3"] = "46c00011FGC",
- ["movt.ps_2"] = "46c10011FG",
- ["movt.ps_3"] = "46c10011FGC",
- ["movz.ps_3"] = "46c00012FGT",
- ["movn.ps_3"] = "46c00013FGT",
- ["cvt.s.pu_2"] = "46c00020FG",
- ["cvt.s.pl_2"] = "46c00028FG",
- ["pll.ps_3"] = "46c0002cFGH",
- ["plu.ps_3"] = "46c0002dFGH",
- ["pul.ps_3"] = "46c0002eFGH",
- ["puu.ps_3"] = "46c0002fFGH",
- ["c.f.ps_2"] = "46c00030GH",
- ["c.f.ps_3"] = "46c00030VGH",
- ["c.un.ps_2"] = "46c00031GH",
- ["c.un.ps_3"] = "46c00031VGH",
- ["c.eq.ps_2"] = "46c00032GH",
- ["c.eq.ps_3"] = "46c00032VGH",
- ["c.ueq.ps_2"] = "46c00033GH",
- ["c.ueq.ps_3"] = "46c00033VGH",
- ["c.olt.ps_2"] = "46c00034GH",
- ["c.olt.ps_3"] = "46c00034VGH",
- ["c.ult.ps_2"] = "46c00035GH",
- ["c.ult.ps_3"] = "46c00035VGH",
- ["c.ole.ps_2"] = "46c00036GH",
- ["c.ole.ps_3"] = "46c00036VGH",
- ["c.ule.ps_2"] = "46c00037GH",
- ["c.ule.ps_3"] = "46c00037VGH",
- ["c.sf.ps_2"] = "46c00038GH",
- ["c.sf.ps_3"] = "46c00038VGH",
- ["c.ngle.ps_2"] = "46c00039GH",
- ["c.ngle.ps_3"] = "46c00039VGH",
- ["c.seq.ps_2"] = "46c0003aGH",
- ["c.seq.ps_3"] = "46c0003aVGH",
- ["c.ngl.ps_2"] = "46c0003bGH",
- ["c.ngl.ps_3"] = "46c0003bVGH",
- ["c.lt.ps_2"] = "46c0003cGH",
- ["c.lt.ps_3"] = "46c0003cVGH",
- ["c.nge.ps_2"] = "46c0003dGH",
- ["c.nge.ps_3"] = "46c0003dVGH",
- ["c.le.ps_2"] = "46c0003eGH",
- ["c.le.ps_3"] = "46c0003eVGH",
- ["c.ngt.ps_2"] = "46c0003fGH",
- ["c.ngt.ps_3"] = "46c0003fVGH",
-
["cvt.s.w_2"] = "46800020FG",
["cvt.d.w_2"] = "46800021FG",
-
["cvt.s.l_2"] = "46a00020FG",
["cvt.d.l_2"] = "46a00021FG",
-
- -- Opcode COP1X.
- lwxc1_2 = "4c000000FX",
- ldxc1_2 = "4c000001FX",
- luxc1_2 = "4c000005FX",
- swxc1_2 = "4c000008FX",
- sdxc1_2 = "4c000009FX",
- suxc1_2 = "4c00000dFX",
- prefx_2 = "4c00000fMX",
- ["alnv.ps_4"] = "4c00001eFGHS",
- ["madd.s_4"] = "4c000020FRGH",
- ["madd.d_4"] = "4c000021FRGH",
- ["madd.ps_4"] = "4c000026FRGH",
- ["msub.s_4"] = "4c000028FRGH",
- ["msub.d_4"] = "4c000029FRGH",
- ["msub.ps_4"] = "4c00002eFRGH",
- ["nmadd.s_4"] = "4c000030FRGH",
- ["nmadd.d_4"] = "4c000031FRGH",
- ["nmadd.ps_4"] = "4c000036FRGH",
- ["nmsub.s_4"] = "4c000038FRGH",
- ["nmsub.d_4"] = "4c000039FRGH",
- ["nmsub.ps_4"] = "4c00003eFRGH",
}
+if mipsr6 then -- Instructions added with MIPSR6.
+
+ for k,v in pairs({
+
+ -- Add immediate to upper bits.
+ aui_3 = "3c000000TSI",
+ daui_3 = mips64 and "74000000TSI",
+ dahi_2 = mips64 and "04060000SI",
+ dati_2 = mips64 and "041e0000SI",
+
+ -- TODO: addiupc, auipc, aluipc, lwpc, lwupc, ldpc.
+
+ -- Compact branches.
+ blezalc_2 = "18000000TB", -- rt != 0.
+ bgezalc_2 = "18000000T=SB", -- rt != 0.
+ bgtzalc_2 = "1c000000TB", -- rt != 0.
+ bltzalc_2 = "1c000000T=SB", -- rt != 0.
+
+ blezc_2 = "58000000TB", -- rt != 0.
+ bgezc_2 = "58000000T=SB", -- rt != 0.
+ bgec_3 = "58000000STB", -- rs != rt.
+ blec_3 = "58000000TSB", -- rt != rs.
+
+ bgtzc_2 = "5c000000TB", -- rt != 0.
+ bltzc_2 = "5c000000T=SB", -- rt != 0.
+ bltc_3 = "5c000000STB", -- rs != rt.
+ bgtc_3 = "5c000000TSB", -- rt != rs.
+
+ bgeuc_3 = "18000000STB", -- rs != rt.
+ bleuc_3 = "18000000TSB", -- rt != rs.
+ bltuc_3 = "1c000000STB", -- rs != rt.
+ bgtuc_3 = "1c000000TSB", -- rt != rs.
+
+ beqzalc_2 = "20000000TB", -- rt != 0.
+ bnezalc_2 = "60000000TB", -- rt != 0.
+ beqc_3 = "20000000STB", -- rs < rt.
+ bnec_3 = "60000000STB", -- rs < rt.
+ bovc_3 = "20000000STB", -- rs >= rt.
+ bnvc_3 = "60000000STB", -- rs >= rt.
+
+ beqzc_2 = "d8000000SK", -- rs != 0.
+ bnezc_2 = "f8000000SK", -- rs != 0.
+ jic_2 = "d8000000TI",
+ jialc_2 = "f8000000TI",
+ bc_1 = "c8000000L",
+ balc_1 = "e8000000L",
+
+ -- Opcode SPECIAL.
+ jr_1 = "00000009S",
+ sdbbp_0 = "0000000e",
+ sdbbp_1 = "0000000eY",
+ lsa_4 = "00000005DSTA",
+ dlsa_4 = mips64 and "00000015DSTA",
+ seleqz_3 = "00000035DST",
+ selnez_3 = "00000037DST",
+ clz_2 = "00000050DS",
+ clo_2 = "00000051DS",
+ dclz_2 = mips64 and "00000052DS",
+ dclo_2 = mips64 and "00000053DS",
+ mul_3 = "00000098DST",
+ muh_3 = "000000d8DST",
+ mulu_3 = "00000099DST",
+ muhu_3 = "000000d9DST",
+ div_3 = "0000009aDST",
+ mod_3 = "000000daDST",
+ divu_3 = "0000009bDST",
+ modu_3 = "000000dbDST",
+ dmul_3 = mips64 and "0000009cDST",
+ dmuh_3 = mips64 and "000000dcDST",
+ dmulu_3 = mips64 and "0000009dDST",
+ dmuhu_3 = mips64 and "000000ddDST",
+ ddiv_3 = mips64 and "0000009eDST",
+ dmod_3 = mips64 and "000000deDST",
+ ddivu_3 = mips64 and "0000009fDST",
+ dmodu_3 = mips64 and "000000dfDST",
+
+ -- Opcode SPECIAL3.
+ align_4 = "7c000220DSTA",
+ dalign_4 = mips64 and "7c000224DSTA",
+ bitswap_2 = "7c000020DT",
+ dbitswap_2 = mips64 and "7c000024DT",
+
+ -- Opcode COP1.
+ bc1eqz_2 = "45200000HB",
+ bc1nez_2 = "45a00000HB",
+
+ ["sel.s_3"] = "46000010FGH",
+ ["seleqz.s_3"] = "46000014FGH",
+ ["selnez.s_3"] = "46000017FGH",
+ ["maddf.s_3"] = "46000018FGH",
+ ["msubf.s_3"] = "46000019FGH",
+ ["rint.s_2"] = "4600001aFG",
+ ["class.s_2"] = "4600001bFG",
+ ["min.s_3"] = "4600001cFGH",
+ ["mina.s_3"] = "4600001dFGH",
+ ["max.s_3"] = "4600001eFGH",
+ ["maxa.s_3"] = "4600001fFGH",
+ ["cmp.af.s_3"] = "46800000FGH",
+ ["cmp.un.s_3"] = "46800001FGH",
+ ["cmp.or.s_3"] = "46800011FGH",
+ ["cmp.eq.s_3"] = "46800002FGH",
+ ["cmp.une.s_3"] = "46800012FGH",
+ ["cmp.ueq.s_3"] = "46800003FGH",
+ ["cmp.ne.s_3"] = "46800013FGH",
+ ["cmp.lt.s_3"] = "46800004FGH",
+ ["cmp.ult.s_3"] = "46800005FGH",
+ ["cmp.le.s_3"] = "46800006FGH",
+ ["cmp.ule.s_3"] = "46800007FGH",
+ ["cmp.saf.s_3"] = "46800008FGH",
+ ["cmp.sun.s_3"] = "46800009FGH",
+ ["cmp.sor.s_3"] = "46800019FGH",
+ ["cmp.seq.s_3"] = "4680000aFGH",
+ ["cmp.sune.s_3"] = "4680001aFGH",
+ ["cmp.sueq.s_3"] = "4680000bFGH",
+ ["cmp.sne.s_3"] = "4680001bFGH",
+ ["cmp.slt.s_3"] = "4680000cFGH",
+ ["cmp.sult.s_3"] = "4680000dFGH",
+ ["cmp.sle.s_3"] = "4680000eFGH",
+ ["cmp.sule.s_3"] = "4680000fFGH",
+
+ ["sel.d_3"] = "46200010FGH",
+ ["seleqz.d_3"] = "46200014FGH",
+ ["selnez.d_3"] = "46200017FGH",
+ ["maddf.d_3"] = "46200018FGH",
+ ["msubf.d_3"] = "46200019FGH",
+ ["rint.d_2"] = "4620001aFG",
+ ["class.d_2"] = "4620001bFG",
+ ["min.d_3"] = "4620001cFGH",
+ ["mina.d_3"] = "4620001dFGH",
+ ["max.d_3"] = "4620001eFGH",
+ ["maxa.d_3"] = "4620001fFGH",
+ ["cmp.af.d_3"] = "46a00000FGH",
+ ["cmp.un.d_3"] = "46a00001FGH",
+ ["cmp.or.d_3"] = "46a00011FGH",
+ ["cmp.eq.d_3"] = "46a00002FGH",
+ ["cmp.une.d_3"] = "46a00012FGH",
+ ["cmp.ueq.d_3"] = "46a00003FGH",
+ ["cmp.ne.d_3"] = "46a00013FGH",
+ ["cmp.lt.d_3"] = "46a00004FGH",
+ ["cmp.ult.d_3"] = "46a00005FGH",
+ ["cmp.le.d_3"] = "46a00006FGH",
+ ["cmp.ule.d_3"] = "46a00007FGH",
+ ["cmp.saf.d_3"] = "46a00008FGH",
+ ["cmp.sun.d_3"] = "46a00009FGH",
+ ["cmp.sor.d_3"] = "46a00019FGH",
+ ["cmp.seq.d_3"] = "46a0000aFGH",
+ ["cmp.sune.d_3"] = "46a0001aFGH",
+ ["cmp.sueq.d_3"] = "46a0000bFGH",
+ ["cmp.sne.d_3"] = "46a0001bFGH",
+ ["cmp.slt.d_3"] = "46a0000cFGH",
+ ["cmp.sult.d_3"] = "46a0000dFGH",
+ ["cmp.sle.d_3"] = "46a0000eFGH",
+ ["cmp.sule.d_3"] = "46a0000fFGH",
+
+ }) do map_op[k] = v end
+
+else -- Instructions removed by MIPSR6.
+
+ for k,v in pairs({
+ -- Traps, don't use.
+ addi_3 = "20000000TSI",
+ daddi_3 = mips64 and "60000000TSI",
+
+ -- Branch on likely, don't use.
+ beqzl_2 = "50000000SB",
+ beql_3 = "50000000STB",
+ bnezl_2 = "54000000SB",
+ bnel_3 = "54000000STB",
+ blezl_2 = "58000000SB",
+ bgtzl_2 = "5c000000SB",
+
+ lwl_2 = "88000000TO",
+ lwr_2 = "98000000TO",
+ swl_2 = "a8000000TO",
+ sdl_2 = mips64 and "b0000000TO",
+ sdr_2 = mips64 and "b1000000TO",
+ swr_2 = "b8000000TO",
+ cache_2 = "bc000000NO",
+ ll_2 = "c0000000TO",
+ pref_2 = "cc000000NO",
+ sc_2 = "e0000000TO",
+ scd_2 = mips64 and "f0000000TO",
+
+ -- Opcode SPECIAL.
+ movf_2 = "00000001DS",
+ movf_3 = "00000001DSC",
+ movt_2 = "00010001DS",
+ movt_3 = "00010001DSC",
+ jr_1 = "00000008S",
+ movz_3 = "0000000aDST",
+ movn_3 = "0000000bDST",
+ mfhi_1 = "00000010D",
+ mthi_1 = "00000011S",
+ mflo_1 = "00000012D",
+ mtlo_1 = "00000013S",
+ mult_2 = "00000018ST",
+ multu_2 = "00000019ST",
+ div_3 = "0000001aST",
+ divu_3 = "0000001bST",
+ ddiv_3 = mips64 and "0000001eST",
+ ddivu_3 = mips64 and "0000001fST",
+ dmult_2 = mips64 and "0000001cST",
+ dmultu_2 = mips64 and "0000001dST",
+
+ -- Opcode REGIMM.
+ tgei_2 = "04080000SI",
+ tgeiu_2 = "04090000SI",
+ tlti_2 = "040a0000SI",
+ tltiu_2 = "040b0000SI",
+ teqi_2 = "040c0000SI",
+ tnei_2 = "040e0000SI",
+ bltzal_2 = "04100000SB",
+ bgezal_2 = "04110000SB",
+ bltzall_2 = "04120000SB",
+ bgezall_2 = "04130000SB",
+
+ -- Opcode SPECIAL2.
+ madd_2 = "70000000ST",
+ maddu_2 = "70000001ST",
+ mul_3 = "70000002DST",
+ msub_2 = "70000004ST",
+ msubu_2 = "70000005ST",
+ clz_2 = "70000020D=TS",
+ clo_2 = "70000021D=TS",
+ dclz_2 = mips64 and "70000024D=TS",
+ dclo_2 = mips64 and "70000025D=TS",
+ sdbbp_0 = "7000003f",
+ sdbbp_1 = "7000003fY",
+
+ -- Opcode COP1.
+ bc1f_1 = "45000000B",
+ bc1f_2 = "45000000CB",
+ bc1t_1 = "45010000B",
+ bc1t_2 = "45010000CB",
+ bc1fl_1 = "45020000B",
+ bc1fl_2 = "45020000CB",
+ bc1tl_1 = "45030000B",
+ bc1tl_2 = "45030000CB",
+
+ ["movf.s_2"] = "46000011FG",
+ ["movf.s_3"] = "46000011FGC",
+ ["movt.s_2"] = "46010011FG",
+ ["movt.s_3"] = "46010011FGC",
+ ["movz.s_3"] = "46000012FGT",
+ ["movn.s_3"] = "46000013FGT",
+ ["cvt.ps.s_3"] = "46000026FGH",
+ ["c.f.s_2"] = "46000030GH",
+ ["c.f.s_3"] = "46000030VGH",
+ ["c.un.s_2"] = "46000031GH",
+ ["c.un.s_3"] = "46000031VGH",
+ ["c.eq.s_2"] = "46000032GH",
+ ["c.eq.s_3"] = "46000032VGH",
+ ["c.ueq.s_2"] = "46000033GH",
+ ["c.ueq.s_3"] = "46000033VGH",
+ ["c.olt.s_2"] = "46000034GH",
+ ["c.olt.s_3"] = "46000034VGH",
+ ["c.ult.s_2"] = "46000035GH",
+ ["c.ult.s_3"] = "46000035VGH",
+ ["c.ole.s_2"] = "46000036GH",
+ ["c.ole.s_3"] = "46000036VGH",
+ ["c.ule.s_2"] = "46000037GH",
+ ["c.ule.s_3"] = "46000037VGH",
+ ["c.sf.s_2"] = "46000038GH",
+ ["c.sf.s_3"] = "46000038VGH",
+ ["c.ngle.s_2"] = "46000039GH",
+ ["c.ngle.s_3"] = "46000039VGH",
+ ["c.seq.s_2"] = "4600003aGH",
+ ["c.seq.s_3"] = "4600003aVGH",
+ ["c.ngl.s_2"] = "4600003bGH",
+ ["c.ngl.s_3"] = "4600003bVGH",
+ ["c.lt.s_2"] = "4600003cGH",
+ ["c.lt.s_3"] = "4600003cVGH",
+ ["c.nge.s_2"] = "4600003dGH",
+ ["c.nge.s_3"] = "4600003dVGH",
+ ["c.le.s_2"] = "4600003eGH",
+ ["c.le.s_3"] = "4600003eVGH",
+ ["c.ngt.s_2"] = "4600003fGH",
+ ["c.ngt.s_3"] = "4600003fVGH",
+ ["movf.d_2"] = "46200011FG",
+ ["movf.d_3"] = "46200011FGC",
+ ["movt.d_2"] = "46210011FG",
+ ["movt.d_3"] = "46210011FGC",
+ ["movz.d_3"] = "46200012FGT",
+ ["movn.d_3"] = "46200013FGT",
+ ["c.f.d_2"] = "46200030GH",
+ ["c.f.d_3"] = "46200030VGH",
+ ["c.un.d_2"] = "46200031GH",
+ ["c.un.d_3"] = "46200031VGH",
+ ["c.eq.d_2"] = "46200032GH",
+ ["c.eq.d_3"] = "46200032VGH",
+ ["c.ueq.d_2"] = "46200033GH",
+ ["c.ueq.d_3"] = "46200033VGH",
+ ["c.olt.d_2"] = "46200034GH",
+ ["c.olt.d_3"] = "46200034VGH",
+ ["c.ult.d_2"] = "46200035GH",
+ ["c.ult.d_3"] = "46200035VGH",
+ ["c.ole.d_2"] = "46200036GH",
+ ["c.ole.d_3"] = "46200036VGH",
+ ["c.ule.d_2"] = "46200037GH",
+ ["c.ule.d_3"] = "46200037VGH",
+ ["c.sf.d_2"] = "46200038GH",
+ ["c.sf.d_3"] = "46200038VGH",
+ ["c.ngle.d_2"] = "46200039GH",
+ ["c.ngle.d_3"] = "46200039VGH",
+ ["c.seq.d_2"] = "4620003aGH",
+ ["c.seq.d_3"] = "4620003aVGH",
+ ["c.ngl.d_2"] = "4620003bGH",
+ ["c.ngl.d_3"] = "4620003bVGH",
+ ["c.lt.d_2"] = "4620003cGH",
+ ["c.lt.d_3"] = "4620003cVGH",
+ ["c.nge.d_2"] = "4620003dGH",
+ ["c.nge.d_3"] = "4620003dVGH",
+ ["c.le.d_2"] = "4620003eGH",
+ ["c.le.d_3"] = "4620003eVGH",
+ ["c.ngt.d_2"] = "4620003fGH",
+ ["c.ngt.d_3"] = "4620003fVGH",
+ ["add.ps_3"] = "46c00000FGH",
+ ["sub.ps_3"] = "46c00001FGH",
+ ["mul.ps_3"] = "46c00002FGH",
+ ["abs.ps_2"] = "46c00005FG",
+ ["mov.ps_2"] = "46c00006FG",
+ ["neg.ps_2"] = "46c00007FG",
+ ["movf.ps_2"] = "46c00011FG",
+ ["movf.ps_3"] = "46c00011FGC",
+ ["movt.ps_2"] = "46c10011FG",
+ ["movt.ps_3"] = "46c10011FGC",
+ ["movz.ps_3"] = "46c00012FGT",
+ ["movn.ps_3"] = "46c00013FGT",
+ ["cvt.s.pu_2"] = "46c00020FG",
+ ["cvt.s.pl_2"] = "46c00028FG",
+ ["pll.ps_3"] = "46c0002cFGH",
+ ["plu.ps_3"] = "46c0002dFGH",
+ ["pul.ps_3"] = "46c0002eFGH",
+ ["puu.ps_3"] = "46c0002fFGH",
+ ["c.f.ps_2"] = "46c00030GH",
+ ["c.f.ps_3"] = "46c00030VGH",
+ ["c.un.ps_2"] = "46c00031GH",
+ ["c.un.ps_3"] = "46c00031VGH",
+ ["c.eq.ps_2"] = "46c00032GH",
+ ["c.eq.ps_3"] = "46c00032VGH",
+ ["c.ueq.ps_2"] = "46c00033GH",
+ ["c.ueq.ps_3"] = "46c00033VGH",
+ ["c.olt.ps_2"] = "46c00034GH",
+ ["c.olt.ps_3"] = "46c00034VGH",
+ ["c.ult.ps_2"] = "46c00035GH",
+ ["c.ult.ps_3"] = "46c00035VGH",
+ ["c.ole.ps_2"] = "46c00036GH",
+ ["c.ole.ps_3"] = "46c00036VGH",
+ ["c.ule.ps_2"] = "46c00037GH",
+ ["c.ule.ps_3"] = "46c00037VGH",
+ ["c.sf.ps_2"] = "46c00038GH",
+ ["c.sf.ps_3"] = "46c00038VGH",
+ ["c.ngle.ps_2"] = "46c00039GH",
+ ["c.ngle.ps_3"] = "46c00039VGH",
+ ["c.seq.ps_2"] = "46c0003aGH",
+ ["c.seq.ps_3"] = "46c0003aVGH",
+ ["c.ngl.ps_2"] = "46c0003bGH",
+ ["c.ngl.ps_3"] = "46c0003bVGH",
+ ["c.lt.ps_2"] = "46c0003cGH",
+ ["c.lt.ps_3"] = "46c0003cVGH",
+ ["c.nge.ps_2"] = "46c0003dGH",
+ ["c.nge.ps_3"] = "46c0003dVGH",
+ ["c.le.ps_2"] = "46c0003eGH",
+ ["c.le.ps_3"] = "46c0003eVGH",
+ ["c.ngt.ps_2"] = "46c0003fGH",
+ ["c.ngt.ps_3"] = "46c0003fVGH",
+
+ -- Opcode COP1X.
+ lwxc1_2 = "4c000000FX",
+ ldxc1_2 = "4c000001FX",
+ luxc1_2 = "4c000005FX",
+ swxc1_2 = "4c000008FX",
+ sdxc1_2 = "4c000009FX",
+ suxc1_2 = "4c00000dFX",
+ prefx_2 = "4c00000fMX",
+ ["alnv.ps_4"] = "4c00001eFGHS",
+ ["madd.s_4"] = "4c000020FRGH",
+ ["madd.d_4"] = "4c000021FRGH",
+ ["madd.ps_4"] = "4c000026FRGH",
+ ["msub.s_4"] = "4c000028FRGH",
+ ["msub.d_4"] = "4c000029FRGH",
+ ["msub.ps_4"] = "4c00002eFRGH",
+ ["nmadd.s_4"] = "4c000030FRGH",
+ ["nmadd.d_4"] = "4c000031FRGH",
+ ["nmadd.ps_4"] = "4c000036FRGH",
+ ["nmsub.s_4"] = "4c000038FRGH",
+ ["nmsub.d_4"] = "4c000039FRGH",
+ ["nmsub.ps_4"] = "4c00003eFRGH",
+
+ }) do map_op[k] = v end
+
+end
+
------------------------------------------------------------------------------
local function parse_gpr(expr)
@@ -808,10 +979,12 @@
op = op + parse_disp(params[n]); n = n + 1
elseif p == "X" then
op = op + parse_index(params[n]); n = n + 1
- elseif p == "B" or p == "J" then
- local mode, n, s = parse_label(params[n], false)
- if p == "B" then n = n + 2048 end
- waction("REL_"..mode, n, s, 1)
+ elseif p == "B" or p == "J" or p == "K" or p == "L" then
+ local mode, m, s = parse_label(params[n], false)
+ if p == "J" then m = m + 0xa800
+ elseif p == "K" then m = m + 0x5000
+ elseif p == "L" then m = m + 0xa000 end
+ waction("REL_"..mode, m, s, 1)
n = n + 1
elseif p == "A" then
op = op + parse_imm(params[n], 5, 6, 0, false); n = n + 1
@@ -833,7 +1006,7 @@
elseif p == "Z" then
op = op + parse_imm(params[n], 10, 6, 0, false); n = n + 1
elseif p == "=" then
- op = op + shl(band(op, 0xf800), 5) -- Copy D to T for clz, clo.
+ n = n - 1 -- Re-use previous parameter for next template char.
else
assert(false)
end
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_mips64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM MIPS64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
-- This module just sets 64 bit mode for the combined MIPS/MIPS64 module.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM PPC/PPC64 encoding engine.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -69,7 +69,7 @@
size_t lgsize;
int *pclabels; /* PC label chains/pos ptrs. */
size_t pcsize;
- void **globals; /* Array of globals (bias -10). */
+ void **globals; /* Array of globals. */
dasm_Section *section; /* Pointer to active section. */
size_t codesize; /* Total size of all code sections. */
int maxsection; /* 0 <= sectionidx < maxsection. */
@@ -86,7 +86,6 @@
{
dasm_State *D;
size_t psz = 0;
- int i;
Dst_REF = NULL;
DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
D = Dst_REF;
@@ -97,12 +96,7 @@
D->pcsize = 0;
D->globals = NULL;
D->maxsection = maxsection;
- for (i = 0; i < maxsection; i++) {
- D->sections[i].buf = NULL; /* Need this for pass3. */
- D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
- D->sections[i].bsize = 0;
- D->sections[i].epos = 0; /* Wrong, but is recalculated after resize. */
- }
+ memset((void *)D->sections, 0, maxsection * sizeof(dasm_Section));
}
/* Free DynASM state. */
@@ -122,7 +116,7 @@
void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
{
dasm_State *D = Dst_REF;
- D->globals = gl - 10; /* Negative bias to compensate for locals. */
+ D->globals = gl;
DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
}
@@ -147,6 +141,7 @@
if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
for (i = 0; i < D->maxsection; i++) {
D->sections[i].pos = DASM_SEC2POS(i);
+ D->sections[i].rbuf = D->sections[i].buf - D->sections[i].pos;
D->sections[i].ofs = 0;
}
}
@@ -277,7 +272,7 @@
{ /* Handle globals not defined in this translation unit. */
int idx;
- for (idx = 20; idx*sizeof(int) < D->lgsize; idx++) {
+ for (idx = 10; idx*sizeof(int) < D->lgsize; idx++) {
int n = D->lglabels[idx];
/* Undefined label: Collapse rel chain and replace with marker (< 0). */
while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = -idx; }
@@ -353,7 +348,11 @@
ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0x60000000;
break;
case DASM_REL_LG:
- CK(n >= 0, UNDEF_LG);
+ if (n < 0) {
+ n = (int)((ptrdiff_t)D->globals[-n-10] - (ptrdiff_t)cp);
+ goto patchrel;
+ }
+ /* fallthrough */
case DASM_REL_PC:
CK(n >= 0, UNDEF_PC);
n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base);
@@ -364,7 +363,7 @@
cp[-1] |= ((n+4) & ((ins & 2048) ? 0x0000fffc: 0x03fffffc));
break;
case DASM_LABEL_LG:
- ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
+ ins &= 2047; if (ins >= 20) D->globals[ins-20] = (void *)(base + n);
break;
case DASM_LABEL_PC: break;
case DASM_IMM:
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_ppc.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM PPC/PPC64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
--
-- Support for various extensions contributed by Caio Souza Oliveira.
@@ -11,9 +11,9 @@
local _info = {
arch = "ppc",
description = "DynASM PPC module",
- version = "1.4.0",
- vernum = 10400,
- release = "2015-10-18",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
license = "MIT",
}
@@ -1722,9 +1722,9 @@
elseif p == "M" then
op = op + parse_shiftmask(params[n], false); n = n + 1
elseif p == "J" or p == "K" then
- local mode, n, s = parse_label(params[n], false)
- if p == "K" then n = n + 2048 end
- waction("REL_"..mode, n, s, 1)
+ local mode, m, s = parse_label(params[n], false)
+ if p == "K" then m = m + 2048 end
+ waction("REL_"..mode, m, s, 1)
n = n + 1
elseif p == "0" then
if band(shr(op, rs), 31) == 0 then werror("cannot use r0") end
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_proto.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_proto.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_proto.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM encoding engine prototypes.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -10,8 +10,8 @@
#include <stddef.h>
#include <stdarg.h>
-#define DASM_IDENT "DynASM 1.4.0"
-#define DASM_VERSION 10400 /* 1.4.0 */
+#define DASM_IDENT "DynASM 1.5.0"
+#define DASM_VERSION 10500 /* 1.5.0 */
#ifndef Dst_DECL
#define Dst_DECL dasm_State **Dst
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM x64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
-- This module just sets 64 bit mode for the combined x86/x64 module.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DynASM x86 encoding engine.
-** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+** Copyright (C) 2005-2025 Mike Pall. All rights reserved.
** Released under the MIT license. See dynasm.lua for full copyright notice.
*/
@@ -68,7 +68,7 @@
size_t lgsize;
int *pclabels; /* PC label chains/pos ptrs. */
size_t pcsize;
- void **globals; /* Array of globals (bias -10). */
+ void **globals; /* Array of globals. */
dasm_Section *section; /* Pointer to active section. */
size_t codesize; /* Total size of all code sections. */
int maxsection; /* 0 <= sectionidx < maxsection. */
@@ -85,7 +85,6 @@
{
dasm_State *D;
size_t psz = 0;
- int i;
Dst_REF = NULL;
DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
D = Dst_REF;
@@ -96,12 +95,7 @@
D->pcsize = 0;
D->globals = NULL;
D->maxsection = maxsection;
- for (i = 0; i < maxsection; i++) {
- D->sections[i].buf = NULL; /* Need this for pass3. */
- D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
- D->sections[i].bsize = 0;
- D->sections[i].epos = 0; /* Wrong, but is recalculated after resize. */
- }
+ memset((void *)D->sections, 0, maxsection * sizeof(dasm_Section));
}
/* Free DynASM state. */
@@ -121,7 +115,7 @@
void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
{
dasm_State *D = Dst_REF;
- D->globals = gl - 10; /* Negative bias to compensate for locals. */
+ D->globals = gl;
DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
}
@@ -146,6 +140,7 @@
if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
for (i = 0; i < D->maxsection; i++) {
D->sections[i].pos = DASM_SEC2POS(i);
+ D->sections[i].rbuf = D->sections[i].buf - D->sections[i].pos;
D->sections[i].ofs = 0;
}
}
@@ -194,12 +189,13 @@
switch (action) {
case DASM_DISP:
if (n == 0) { if (mrm < 0) mrm = p[-2]; if ((mrm&7) != 5) break; }
- case DASM_IMM_DB: if (((n+128)&-256) == 0) goto ob;
+ /* fallthrough */
+ case DASM_IMM_DB: if (((n+128)&-256) == 0) goto ob; /* fallthrough */
case DASM_REL_A: /* Assumes ptrdiff_t is int. !x64 */
case DASM_IMM_D: ofs += 4; break;
case DASM_IMM_S: CK(((n+128)&-256) == 0, RANGE_I); goto ob;
case DASM_IMM_B: CK((n&-256) == 0, RANGE_I); ob: ofs++; break;
- case DASM_IMM_WB: if (((n+128)&-256) == 0) goto ob;
+ case DASM_IMM_WB: if (((n+128)&-256) == 0) goto ob; /* fallthrough */
case DASM_IMM_W: CK((n&-65536) == 0, RANGE_I); ofs += 2; break;
case DASM_SPACE: p++; ofs += n; break;
case DASM_SETLABEL: b[pos-2] = -0x40000000; break; /* Neg. label ofs. */
@@ -207,8 +203,8 @@
if (*p < 0x40 && p[1] == DASM_DISP) mrm = n;
if (*p < 0x20 && (n&7) == 4) ofs++;
switch ((*p++ >> 3) & 3) {
- case 3: n |= b[pos-3];
- case 2: n |= b[pos-2];
+ case 3: n |= b[pos-3]; /* fallthrough */
+ case 2: n |= b[pos-2]; /* fallthrough */
case 1: if (n <= 7) { b[pos-1] |= 0x10; ofs--; }
}
continue;
@@ -238,8 +234,11 @@
}
pos++;
ofs += 4; /* Maximum offset needed. */
- if (action == DASM_REL_LG || action == DASM_REL_PC)
+ if (action == DASM_REL_LG || action == DASM_REL_PC) {
b[pos++] = ofs; /* Store pass1 offset estimate. */
+ } else if (sizeof(ptrdiff_t) == 8) {
+ ofs += 4;
+ }
break;
case DASM_LABEL_LG: pl = D->lglabels + *p++; CKPL(lg, LG); goto putlabel;
case DASM_LABEL_PC: pl = D->pclabels + va_arg(ap, int); CKPL(pc, PC);
@@ -304,11 +303,13 @@
while (pos != lastpos) {
dasm_ActList p = D->actionlist + b[pos++];
+ int op = 0;
while (1) {
- int op, action = *p++;
+ int action = *p++;
switch (action) {
- case DASM_REL_LG: p++; op = p[-3]; goto rel_pc;
- case DASM_REL_PC: op = p[-2]; rel_pc: {
+ case DASM_REL_LG: p++;
+ /* fallthrough */
+ case DASM_REL_PC: {
int shrink = op == 0xe9 ? 3 : ((op&0xf0) == 0x80 ? 4 : 0);
if (shrink) { /* Shrinkable branch opcode? */
int lofs, lpos = b[pos];
@@ -329,17 +330,21 @@
pos += 2;
break;
}
+ /* fallthrough */
case DASM_SPACE: case DASM_IMM_LG: case DASM_VREG: p++;
+ /* fallthrough */
case DASM_DISP: case DASM_IMM_S: case DASM_IMM_B: case DASM_IMM_W:
case DASM_IMM_D: case DASM_IMM_WB: case DASM_IMM_DB:
case DASM_SETLABEL: case DASM_REL_A: case DASM_IMM_PC: pos++; break;
case DASM_LABEL_LG: p++;
+ /* fallthrough */
case DASM_LABEL_PC: b[pos++] += ofs; break; /* Fix label offset. */
case DASM_ALIGN: ofs -= (b[pos++]+ofs)&*p++; break; /* Adjust ofs. */
case DASM_EXTERN: p += 2; break;
- case DASM_ESC: p++; break;
+ case DASM_ESC: op = *p++; break;
case DASM_MARK: break;
case DASM_SECTION: case DASM_STOP: goto stop;
+ default: op = action; break;
}
}
stop: (void)0;
@@ -358,10 +363,22 @@
do { *((unsigned short *)cp) = (unsigned short)(x); cp+=2; } while (0)
#define dasmd(x) \
do { *((unsigned int *)cp) = (unsigned int)(x); cp+=4; } while (0)
+#define dasmq(x) \
+ do { *((unsigned long long *)cp) = (unsigned long long)(x); cp+=8; } while (0)
#else
#define dasmw(x) do { dasmb(x); dasmb((x)>>8); } while (0)
#define dasmd(x) do { dasmw(x); dasmw((x)>>16); } while (0)
+#define dasmq(x) do { dasmd(x); dasmd((x)>>32); } while (0)
#endif
+static unsigned char *dasma_(unsigned char *cp, ptrdiff_t x)
+{
+ if (sizeof(ptrdiff_t) == 8)
+ dasmq((unsigned long long)x);
+ else
+ dasmd((unsigned int)x);
+ return cp;
+}
+#define dasma(x) (cp = dasma_(cp, (x)))
/* Pass 3: Encode sections. */
int dasm_encode(Dst_DECL, void *buffer)
@@ -391,12 +408,15 @@
if (mrm != 5) { mm[-1] -= 0x80; break; } }
if (((n+128) & -256) != 0) goto wd; else mm[-1] -= 0x40;
}
+ /* fallthrough */
case DASM_IMM_S: case DASM_IMM_B: wb: dasmb(n); break;
case DASM_IMM_DB: if (((n+128)&-256) == 0) {
db: if (!mark) mark = cp; mark[-2] += 2; mark = NULL; goto wb;
} else mark = NULL;
+ /* fallthrough */
case DASM_IMM_D: wd: dasmd(n); break;
case DASM_IMM_WB: if (((n+128)&-256) == 0) goto db; else mark = NULL;
+ /* fallthrough */
case DASM_IMM_W: dasmw(n); break;
case DASM_VREG: {
int t = *p++;
@@ -420,8 +440,10 @@
break;
}
case DASM_REL_LG: p++; if (n >= 0) goto rel_pc;
- b++; n = (int)(ptrdiff_t)D->globals[-n];
- case DASM_REL_A: rel_a: n -= (int)(ptrdiff_t)(cp+4); goto wd; /* !x64 */
+ b++; n = (int)(ptrdiff_t)D->globals[-n-10];
+ /* fallthrough */
+ case DASM_REL_A: rel_a:
+ n -= (unsigned int)(ptrdiff_t)(cp+4); goto wd; /* !x64 */
case DASM_REL_PC: rel_pc: {
int shrink = *b++;
int *pb = DASM_POS2PTR(D, n); if (*pb < 0) { n = pb[1]; goto rel_a; }
@@ -431,16 +453,18 @@
goto wb;
}
case DASM_IMM_LG:
- p++; if (n < 0) { n = (int)(ptrdiff_t)D->globals[-n]; goto wd; }
+ p++;
+ if (n < 0) { dasma((ptrdiff_t)D->globals[-n-10]); break; }
+ /* fallthrough */
case DASM_IMM_PC: {
int *pb = DASM_POS2PTR(D, n);
- n = *pb < 0 ? pb[1] : (*pb + (int)(ptrdiff_t)base);
- goto wd;
+ dasma(*pb < 0 ? (ptrdiff_t)pb[1] : (*pb + (ptrdiff_t)base));
+ break;
}
case DASM_LABEL_LG: {
int idx = *p++;
if (idx >= 10)
- D->globals[idx] = (void *)(base + (*p == DASM_SETLABEL ? *b : n));
+ D->globals[idx-10] = (void *)(base + (*p == DASM_SETLABEL ? *b : n));
break;
}
case DASM_LABEL_PC: case DASM_SETLABEL: break;
@@ -452,6 +476,7 @@
case DASM_EXTERN: n = DASM_EXTERN(Dst, cp, p[1], *p); p += 2; goto wd;
case DASM_MARK: mark = cp; break;
case DASM_ESC: action = *p++;
+ /* fallthrough */
default: *cp++ = action; break;
case DASM_SECTION: case DASM_STOP: goto stop;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dasm_x86.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
------------------------------------------------------------------------------
-- DynASM x86/x64 module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See dynasm.lua for full copyright notice.
------------------------------------------------------------------------------
@@ -11,9 +11,9 @@
local _info = {
arch = x64 and "x64" or "x86",
description = "DynASM x86/x64 module",
- version = "1.4.0",
- vernum = 10400,
- release = "2015-10-18",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
license = "MIT",
}
@@ -484,6 +484,22 @@
end
end
+-- Put signed or unsigned qword or arg.
+local function wputqarg(n)
+ local tn = type(n)
+ if tn == "number" then -- This is only used for numbers from -2^31..2^32-1.
+ wputb(band(n, 255))
+ wputb(band(shr(n, 8), 255))
+ wputb(band(shr(n, 16), 255))
+ wputb(shr(n, 24))
+ local sign = n < 0 and 255 or 0
+ wputb(sign); wputb(sign); wputb(sign); wputb(sign)
+ else
+ waction("IMM_D", format("(unsigned int)(%s)", n))
+ waction("IMM_D", format("(unsigned int)((unsigned long long)(%s)>>32)", n))
+ end
+end
+
-- Put operand-size dependent number or arg (defaults to dword).
local function wputszarg(sz, n)
if not sz or sz == "d" or sz == "q" then wputdarg(n)
@@ -611,7 +627,11 @@
werror("NYI: rip-relative displacement followed by immediate")
end
-- The previous byte in the action buffer cannot be 0xe9 or 0x80-0x8f.
- wputlabel("REL_", disp[1], 2)
+ if disp[2] == "iPJ" then
+ waction("REL_A", disp[1])
+ else
+ wputlabel("REL_", disp[1], 2)
+ end
else
wputdarg(disp)
end
@@ -663,10 +683,16 @@
end
-- Convert number to valid integer or nil.
-local function toint(expr)
+local function toint(expr, isqword)
local n = tonumber(expr)
if n then
- if n % 1 ~= 0 or n < -2147483648 or n > 4294967295 then
+ if n % 1 ~= 0 then
+ werror("not an integer number `"..expr.."'")
+ elseif isqword then
+ if n < -2147483648 or n > 2147483647 then
+ n = nil -- Handle it as an expression to avoid precision loss.
+ end
+ elseif n < -2147483648 or n > 4294967295 then
werror("bad integer number `"..expr.."'")
end
return n
@@ -722,9 +748,9 @@
return imm*map_opsizenum[ops]
end
local mode, iexpr = immexpr(dispt)
- if mode == "iJ" then
+ if mode == "iJ" or mode == "iPJ" then
if c == "-" then werror("cannot invert label reference") end
- return { iexpr }
+ return { iexpr, mode }
end
return expr -- Need to return original signed expression.
end
@@ -749,7 +775,7 @@
end
-- Parse operand and return { mode, opsize, reg, xreg, xsc, disp, imm }.
-local function parseoperand(param)
+local function parseoperand(param, isqword)
local t = {}
local expr = param
@@ -810,7 +836,7 @@
if t.disp then break end
-- [reg+xreg...]
- local xreg, tailx = match(tailr, "^+%s*([@%w_:]+)%s*(.*)$")
+ local xreg, tailx = match(tailr, "^%+%s*([@%w_:]+)%s*(.*)$")
xreg, t.xreg, tp = rtexpr(xreg)
if not t.xreg then
-- [reg+-expr]
@@ -837,7 +863,7 @@
t.disp = dispexpr(tailx)
else
-- imm or opsize*imm
- local imm = toint(expr)
+ local imm = toint(expr, isqword)
if not imm and sub(expr, 1, 1) == "*" and t.opsize then
imm = toint(sub(expr, 2))
if imm then
@@ -955,6 +981,7 @@
-- "u" Use VEX encoding, vvvv unused.
-- "v"/"V" Use VEX encoding, vvvv from 1st/2nd operand (the operand is
-- removed from the list used by future characters).
+-- "w" Use VEX encoding, vvvv from 3rd operand.
-- "L" Force VEX.L
--
-- All of the following characters force a flush of the opcode:
@@ -1124,6 +1151,8 @@
rep_0 = "F3",
repe_0 = "F3",
repz_0 = "F3",
+ endbr32_0 = "F30F1EFB",
+ endbr64_0 = "F30F1EFA",
-- F4: *hlt
cmc_0 = "F5",
-- F6: test... mb,i; div... mb
@@ -1536,8 +1565,8 @@
vrcpss_3 = "rrro:F30FV53rM|rrx/ood:",
vrsqrtps_2 = "rmoy:0Fu52rM",
vrsqrtss_3 = "rrro:F30FV52rM|rrx/ood:",
- vroundpd_3 = "rmioy:660F3AV09rMU",
- vroundps_3 = "rmioy:660F3AV08rMU",
+ vroundpd_3 = "rmioy:660F3Au09rMU",
+ vroundps_3 = "rmioy:660F3Au08rMU",
vroundsd_4 = "rrrio:660F3AV0BrMU|rrxi/ooq:",
vroundss_4 = "rrrio:660F3AV0ArMU|rrxi/ood:",
vshufpd_4 = "rrmioy:660FVC6rMU",
@@ -1677,6 +1706,91 @@
-- Intel ADX
adcx_2 = "rmqd:660F38F6rM",
adox_2 = "rmqd:F30F38F6rM",
+
+ -- BMI1
+ andn_3 = "rrmqd:0F38VF2rM",
+ bextr_3 = "rmrqd:0F38wF7rM",
+ blsi_2 = "rmqd:0F38vF33m",
+ blsmsk_2 = "rmqd:0F38vF32m",
+ blsr_2 = "rmqd:0F38vF31m",
+ tzcnt_2 = "rmqdw:F30FBCrM",
+
+ -- BMI2
+ bzhi_3 = "rmrqd:0F38wF5rM",
+ mulx_3 = "rrmqd:F20F38VF6rM",
+ pdep_3 = "rrmqd:F20F38VF5rM",
+ pext_3 = "rrmqd:F30F38VF5rM",
+ rorx_3 = "rmSqd:F20F3AuF0rMS",
+ sarx_3 = "rmrqd:F30F38wF7rM",
+ shrx_3 = "rmrqd:F20F38wF7rM",
+ shlx_3 = "rmrqd:660F38wF7rM",
+
+ -- FMA3
+ vfmaddsub132pd_3 = "rrmoy:660F38VX96rM",
+ vfmaddsub132ps_3 = "rrmoy:660F38V96rM",
+ vfmaddsub213pd_3 = "rrmoy:660F38VXA6rM",
+ vfmaddsub213ps_3 = "rrmoy:660F38VA6rM",
+ vfmaddsub231pd_3 = "rrmoy:660F38VXB6rM",
+ vfmaddsub231ps_3 = "rrmoy:660F38VB6rM",
+
+ vfmsubadd132pd_3 = "rrmoy:660F38VX97rM",
+ vfmsubadd132ps_3 = "rrmoy:660F38V97rM",
+ vfmsubadd213pd_3 = "rrmoy:660F38VXA7rM",
+ vfmsubadd213ps_3 = "rrmoy:660F38VA7rM",
+ vfmsubadd231pd_3 = "rrmoy:660F38VXB7rM",
+ vfmsubadd231ps_3 = "rrmoy:660F38VB7rM",
+
+ vfmadd132pd_3 = "rrmoy:660F38VX98rM",
+ vfmadd132ps_3 = "rrmoy:660F38V98rM",
+ vfmadd132sd_3 = "rrro:660F38VX99rM|rrx/ooq:",
+ vfmadd132ss_3 = "rrro:660F38V99rM|rrx/ood:",
+ vfmadd213pd_3 = "rrmoy:660F38VXA8rM",
+ vfmadd213ps_3 = "rrmoy:660F38VA8rM",
+ vfmadd213sd_3 = "rrro:660F38VXA9rM|rrx/ooq:",
+ vfmadd213ss_3 = "rrro:660F38VA9rM|rrx/ood:",
+ vfmadd231pd_3 = "rrmoy:660F38VXB8rM",
+ vfmadd231ps_3 = "rrmoy:660F38VB8rM",
+ vfmadd231sd_3 = "rrro:660F38VXB9rM|rrx/ooq:",
+ vfmadd231ss_3 = "rrro:660F38VB9rM|rrx/ood:",
+
+ vfmsub132pd_3 = "rrmoy:660F38VX9ArM",
+ vfmsub132ps_3 = "rrmoy:660F38V9ArM",
+ vfmsub132sd_3 = "rrro:660F38VX9BrM|rrx/ooq:",
+ vfmsub132ss_3 = "rrro:660F38V9BrM|rrx/ood:",
+ vfmsub213pd_3 = "rrmoy:660F38VXAArM",
+ vfmsub213ps_3 = "rrmoy:660F38VAArM",
+ vfmsub213sd_3 = "rrro:660F38VXABrM|rrx/ooq:",
+ vfmsub213ss_3 = "rrro:660F38VABrM|rrx/ood:",
+ vfmsub231pd_3 = "rrmoy:660F38VXBArM",
+ vfmsub231ps_3 = "rrmoy:660F38VBArM",
+ vfmsub231sd_3 = "rrro:660F38VXBBrM|rrx/ooq:",
+ vfmsub231ss_3 = "rrro:660F38VBBrM|rrx/ood:",
+
+ vfnmadd132pd_3 = "rrmoy:660F38VX9CrM",
+ vfnmadd132ps_3 = "rrmoy:660F38V9CrM",
+ vfnmadd132sd_3 = "rrro:660F38VX9DrM|rrx/ooq:",
+ vfnmadd132ss_3 = "rrro:660F38V9DrM|rrx/ood:",
+ vfnmadd213pd_3 = "rrmoy:660F38VXACrM",
+ vfnmadd213ps_3 = "rrmoy:660F38VACrM",
+ vfnmadd213sd_3 = "rrro:660F38VXADrM|rrx/ooq:",
+ vfnmadd213ss_3 = "rrro:660F38VADrM|rrx/ood:",
+ vfnmadd231pd_3 = "rrmoy:660F38VXBCrM",
+ vfnmadd231ps_3 = "rrmoy:660F38VBCrM",
+ vfnmadd231sd_3 = "rrro:660F38VXBDrM|rrx/ooq:",
+ vfnmadd231ss_3 = "rrro:660F38VBDrM|rrx/ood:",
+
+ vfnmsub132pd_3 = "rrmoy:660F38VX9ErM",
+ vfnmsub132ps_3 = "rrmoy:660F38V9ErM",
+ vfnmsub132sd_3 = "rrro:660F38VX9FrM|rrx/ooq:",
+ vfnmsub132ss_3 = "rrro:660F38V9FrM|rrx/ood:",
+ vfnmsub213pd_3 = "rrmoy:660F38VXAErM",
+ vfnmsub213ps_3 = "rrmoy:660F38VAErM",
+ vfnmsub213sd_3 = "rrro:660F38VXAFrM|rrx/ooq:",
+ vfnmsub213ss_3 = "rrro:660F38VAFrM|rrx/ood:",
+ vfnmsub231pd_3 = "rrmoy:660F38VXBErM",
+ vfnmsub231ps_3 = "rrmoy:660F38VBErM",
+ vfnmsub231sd_3 = "rrro:660F38VXBFrM|rrx/ooq:",
+ vfnmsub231ss_3 = "rrro:660F38VBFrM|rrx/ood:",
}
------------------------------------------------------------------------------
@@ -1766,7 +1880,7 @@
------------------------------------------------------------------------------
-local map_vexarg = { u = false, v = 1, V = 2 }
+local map_vexarg = { u = false, v = 1, V = 2, w = 3 }
-- Process pattern string.
local function dopattern(pat, args, sz, op, needrex)
@@ -1866,7 +1980,7 @@
local a = args[narg]
narg = narg + 1
local mode, imm = a.mode, a.imm
- if mode == "iJ" and not match("iIJ", c) then
+ if mode == "iJ" and not match(x64 and "J" or "iIJ", c) then
werror("bad operand size for label")
end
if c == "S" then
@@ -2058,14 +2172,16 @@
local function op_data(params)
if not params then return "imm..." end
local sz = sub(params.op, 2, 2)
- if sz == "a" then sz = addrsize end
+ if sz == "l" then sz = "d" elseif sz == "a" then sz = addrsize end
for _,p in ipairs(params) do
- local a = parseoperand(p)
+ local a = parseoperand(p, sz == "q")
if sub(a.mode, 1, 1) ~= "i" or (a.opsize and a.opsize ~= sz) then
werror("bad mode or size in `"..p.."'")
end
if a.mode == "iJ" then
wputlabel("IMM_", a.imm, 1)
+ elseif sz == "q" then
+ wputqarg(a.imm)
else
wputszarg(sz, a.imm)
end
@@ -2077,7 +2193,11 @@
map_op[".sbyte_*"] = op_data
map_op[".word_*"] = op_data
map_op[".dword_*"] = op_data
+map_op[".qword_*"] = op_data
map_op[".aword_*"] = op_data
+map_op[".long_*"] = op_data
+map_op[".quad_*"] = op_data
+map_op[".addr_*"] = op_data
------------------------------------------------------------------------------
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dynasm.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dynasm.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/dynasm/dynasm.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -2,7 +2,7 @@
-- DynASM. A dynamic assembler for code generation engines.
-- Originally designed and implemented for LuaJIT.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- See below for full copyright notice.
------------------------------------------------------------------------------
@@ -10,14 +10,14 @@
local _info = {
name = "DynASM",
description = "A dynamic assembler for code generation engines",
- version = "1.4.0",
- vernum = 10400,
- release = "2015-10-18",
+ version = "1.5.0",
+ vernum = 10500,
+ release = "2021-05-02",
author = "Mike Pall",
- url = "http://luajit.org/dynasm.html",
+ url = "https://luajit.org/dynasm.html",
license = "MIT",
copyright = [[
-Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+Copyright (C) 2005-2025 Mike Pall. All rights reserved.
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
@@ -38,7 +38,7 @@
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-[ MIT license: http://www.opensource.org/licenses/mit-license.php ]
+[ MIT license: https://www.opensource.org/licenses/mit-license.php ]
]],
}
@@ -75,7 +75,7 @@
g_synclineno = g_synclineno + 1
end
--- Write assembler line as a comment, if requestd.
+-- Write assembler line as a comment, if requested.
local function wcomment(aline)
if g_opt.comment then
wline(g_opt.comment..aline..g_opt.endcomment, true)
@@ -630,6 +630,7 @@
-- Load architecture-specific module.
local function loadarch(arch)
if not match(arch, "^[%w_]+$") then return "bad arch name" end
+ _G._map_def = map_def
local ok, m_arch = pcall(require, "dasm_"..arch)
if not ok then return "cannot load module: "..m_arch end
g_arch = m_arch
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.1
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.1 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.1 2025-02-02 17:39:28 UTC (rev 73695)
@@ -6,7 +6,7 @@
.B luajit
[\fIoptions\fR]... [\fIscript\fR [\fIargs\fR]...]
.SH "WEB SITE"
-.IR http://luajit.org
+.IR https://luajit.org
.SH DESCRIPTION
.PP
This is the command-line program to run Lua programs with \fBLuaJIT\fR.
@@ -74,15 +74,15 @@
Runs some nested loops and shows the resulting traces.
.SH COPYRIGHT
.PP
-\fBLuaJIT\fR is Copyright \(co 2005-2017 Mike Pall.
+\fBLuaJIT\fR is Copyright \(co 2005-2025 Mike Pall.
.br
\fBLuaJIT\fR is open source software, released under the MIT license.
.SH SEE ALSO
.PP
More details in the provided HTML docs or at:
-.IR http://luajit.org
+.IR https://luajit.org
.br
More about the Lua language can be found at:
-.IR http://lua.org/docs.html
+.IR https://lua.org/docs.html
.PP
lua(1)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.pc
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.pc 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/etc/luajit.pc 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,8 +1,8 @@
# Package information for LuaJIT to be used by pkg-config.
majver=2
minver=1
-relver=0
-version=${majver}.${minver}.${relver}-beta3
+relver=ROLLING
+version=${majver}.${minver}.${relver}
abiver=5.1
prefix=/usr/local
@@ -17,7 +17,7 @@
Name: LuaJIT
Description: Just-in-time compiler for Lua
-URL: http://luajit.org
+URL: https://luajit.org
Version: ${version}
Requires:
Libs: -L${libdir} -l${libname}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile 2025-02-02 17:39:28 UTC (rev 73695)
@@ -7,12 +7,11 @@
# Also works with MinGW and Cygwin on Windows.
# Please check msvcbuild.bat for building with MSVC on Windows.
#
-# Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+# Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
##############################################################################
MAJVER= 2
MINVER= 1
-RELVER= 0
ABIVER= 5.1
NODOTABIVER= 51
@@ -110,8 +109,8 @@
#XCFLAGS+= -DLUAJIT_NUMMODE=1
#XCFLAGS+= -DLUAJIT_NUMMODE=2
#
-# Enable GC64 mode for x64.
-#XCFLAGS+= -DLUAJIT_ENABLE_GC64
+# Disable LJ_GC64 mode for x64.
+#XCFLAGS+= -DLUAJIT_DISABLE_GC64
#
##############################################################################
@@ -132,7 +131,6 @@
#
# This define is required to run LuaJIT under Valgrind. The Valgrind
# header files must be installed. You should enable debug information, too.
-# Use --suppressions=lj.supp to avoid some false positives.
#XCFLAGS+= -DLUAJIT_USE_VALGRIND
#
# This is the client for the GDB JIT API. GDB 7.0 or higher is required
@@ -158,7 +156,6 @@
ifeq (Windows,$(findstring Windows,$(OS))$(MSYSTEM)$(TERM))
HOST_SYS= Windows
- HOST_RM= del
else
HOST_SYS:= $(shell uname -s)
ifneq (,$(findstring MINGW,$(HOST_SYS)))
@@ -165,6 +162,10 @@
HOST_SYS= Windows
HOST_MSYS= mingw
endif
+ ifneq (,$(findstring MSYS,$(HOST_SYS)))
+ HOST_SYS= Windows
+ HOST_MSYS= mingw
+ endif
ifneq (,$(findstring CYGWIN,$(HOST_SYS)))
HOST_SYS= Windows
HOST_MSYS= cygwin
@@ -191,7 +192,7 @@
LDOPTIONS= $(CCDEBUG) $(LDFLAGS)
HOST_CC= $(CC)
-HOST_RM= rm -f
+HOST_RM?= rm -f
# If left blank, minilua is built and used. You can supply an installed
# copy of (plain) Lua 5.1 or 5.2, plus Lua BitOp. E.g. with: HOST_LUA=lua
HOST_LUA=
@@ -209,7 +210,7 @@
TARGET_STCC= $(STATIC_CC)
TARGET_DYNCC= $(DYNAMIC_CC)
TARGET_LD= $(CROSS)$(CC)
-TARGET_AR= $(CROSS)ar rcus 2>/dev/null
+TARGET_AR= $(CROSS)ar rcus
TARGET_STRIP= $(CROSS)strip
TARGET_LIBPATH= $(or $(PREFIX),/usr/local)/$(or $(MULTILIB),lib)
@@ -217,6 +218,7 @@
TARGET_DYLIBNAME= libluajit-$(ABIVER).$(MAJVER).dylib
TARGET_DYLIBPATH= $(TARGET_LIBPATH)/$(TARGET_DYLIBNAME)
TARGET_DLLNAME= lua$(NODOTABIVER).dll
+TARGET_DLLDOTANAME= libluajit-$(ABIVER).dll.a
TARGET_XSHLDFLAGS= -shared -fPIC -Wl,-soname,$(TARGET_SONAME)
TARGET_DYNXLDOPTS=
@@ -231,7 +233,7 @@
TARGET_ASHLDFLAGS= $(LDOPTIONS) $(TARGET_XSHLDFLAGS) $(TARGET_FLAGS) $(TARGET_SHLDFLAGS)
TARGET_ALIBS= $(TARGET_XLIBS) $(LIBS) $(TARGET_LIBS)
-TARGET_TESTARCH=$(shell $(TARGET_CC) $(TARGET_TCFLAGS) -E lj_arch.h -dM)
+TARGET_TESTARCH:=$(shell $(TARGET_CC) $(TARGET_TCFLAGS) -E lj_arch.h -dM)
ifneq (,$(findstring LJ_TARGET_X64 ,$(TARGET_TESTARCH)))
TARGET_LJARCH= x64
else
@@ -297,6 +299,9 @@
ifneq (,$(LMULTILIB))
TARGET_XCFLAGS+= -DLUA_LMULTILIB=\"$(LMULTILIB)\"
endif
+ifneq (,$(INSTALL_LJLIBD))
+ TARGET_XCFLAGS+= -DLUA_LJDIR=\"$(INSTALL_LJLIBD)\"
+endif
##############################################################################
# Target system detection.
@@ -305,34 +310,39 @@
TARGET_SYS?= $(HOST_SYS)
ifeq (Windows,$(TARGET_SYS))
TARGET_STRIP+= --strip-unneeded
- TARGET_XSHLDFLAGS= -shared
+ TARGET_XSHLDFLAGS= -shared -Wl,--out-implib,$(TARGET_DLLDOTANAME)
TARGET_DYNXLDOPTS=
else
+ TARGET_AR+= 2>/dev/null
ifeq (,$(shell $(TARGET_CC) -o /dev/null -c -x c /dev/null -fno-stack-protector 2>/dev/null || echo 1))
TARGET_XCFLAGS+= -fno-stack-protector
endif
ifeq (Darwin,$(TARGET_SYS))
ifeq (,$(MACOSX_DEPLOYMENT_TARGET))
- export MACOSX_DEPLOYMENT_TARGET=10.4
+ $(error missing: export MACOSX_DEPLOYMENT_TARGET=XX.YY)
endif
TARGET_STRIP+= -x
- TARGET_XSHLDFLAGS= -dynamiclib -single_module -undefined dynamic_lookup -fPIC
+ TARGET_XCFLAGS+= -DLUAJIT_UNWIND_EXTERNAL
+ TARGET_XSHLDFLAGS= -dynamiclib -undefined dynamic_lookup -fPIC
TARGET_DYNXLDOPTS=
- TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).$(RELVER)
- ifeq (x64,$(TARGET_LJARCH))
- TARGET_XLDFLAGS+= -pagezero_size 10000 -image_base 100000000
- TARGET_XSHLDFLAGS+= -image_base 7fff04c4a000
- endif
+ TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).255
else
ifeq (iOS,$(TARGET_SYS))
TARGET_STRIP+= -x
- TARGET_XSHLDFLAGS= -dynamiclib -single_module -undefined dynamic_lookup -fPIC
+ TARGET_XSHLDFLAGS= -dynamiclib -undefined dynamic_lookup -fPIC
TARGET_DYNXLDOPTS=
- TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).$(RELVER)
+ TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).255
ifeq (arm64,$(TARGET_LJARCH))
TARGET_XCFLAGS+= -fno-omit-frame-pointer
endif
else
+ ifeq (,$(findstring LJ_NO_UNWIND 1,$(TARGET_TESTARCH)))
+ # Find out whether the target toolchain always generates unwind tables.
+ TARGET_TESTUNWIND=$(shell exec 2>/dev/null; echo 'extern void b(void);int a(void){b();return 0;}' | $(TARGET_CC) -c -x c - -o tmpunwind.o && { grep -qa -e eh_frame -e __unwind_info tmpunwind.o || grep -qU -e eh_frame -e __unwind_info tmpunwind.o; } && echo E; rm -f tmpunwind.o)
+ ifneq (,$(findstring E,$(TARGET_TESTUNWIND)))
+ TARGET_XCFLAGS+= -DLUAJIT_UNWIND_EXTERNAL
+ endif
+ endif
ifneq (SunOS,$(TARGET_SYS))
ifneq (PS3,$(TARGET_SYS))
TARGET_XLDFLAGS+= -Wl,-E
@@ -359,7 +369,7 @@
HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OSX
else
ifeq (iOS,$(TARGET_SYS))
- HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OSX
+ HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OSX -DTARGET_OS_IPHONE=1
else
HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OTHER
endif
@@ -380,10 +390,11 @@
MINILUA_LIBS= -lm
MINILUA_T= host/minilua
MINILUA_X= $(MINILUA_T)
+MINILUA_DEP=
ifeq (,$(HOST_LUA))
HOST_LUA= $(MINILUA_X)
- DASM_DEP= $(MINILUA_T)
+ MINILUA_DEP= $(MINILUA_T)
endif
DASM_DIR= ../dynasm
@@ -425,6 +436,10 @@
DASM_AFLAGS+= -D NO_UNWIND
TARGET_ARCH+= -DLUAJIT_NO_UNWIND
endif
+ifneq (,$(findstring LJ_ABI_PAUTH 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D PAUTH
+ TARGET_ARCH+= -DLJ_ABI_PAUTH=1
+endif
DASM_AFLAGS+= -D VER=$(subst LJ_ARCH_VERSION_,,$(filter LJ_ARCH_VERSION_%,$(subst LJ_ARCH_VERSION ,LJ_ARCH_VERSION_,$(TARGET_TESTARCH))))
ifeq (Windows,$(TARGET_SYS))
DASM_AFLAGS+= -D WIN
@@ -439,6 +454,9 @@
DASM_AFLAGS+= -D IOS
endif
else
+ifneq (,$(findstring LJ_TARGET_MIPSR6 ,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D MIPSR6
+endif
ifeq (ppc,$(TARGET_LJARCH))
ifneq (,$(findstring LJ_ARCH_SQRT 1,$(TARGET_TESTARCH)))
DASM_AFLAGS+= -D SQRT
@@ -452,9 +470,6 @@
ifeq (PS3,$(TARGET_SYS))
DASM_AFLAGS+= -D PPE -D TOC
endif
- ifneq (,$(findstring LJ_ARCH_PPC64 ,$(TARGET_TESTARCH)))
- DASM_ARCH= ppc64
- endif
endif
endif
endif
@@ -462,6 +477,14 @@
DASM_FLAGS= $(DASM_XFLAGS) $(DASM_AFLAGS)
DASM_DASC= vm_$(DASM_ARCH).dasc
+GIT= git
+ifeq (Windows,$(HOST_SYS)$(HOST_MSYS))
+ GIT_RELVER= if exist ..\.git ( $(GIT) show -s --format=%%ct >luajit_relver.txt ) else ( type ..\.relver >luajit_relver.txt )
+else
+ GIT_RELVER= [ -e ../.git ] && $(GIT) show -s --format=%ct >luajit_relver.txt 2>/dev/null || cat ../.relver >luajit_relver.txt 2>/dev/null || :
+endif
+GIT_DEP= $(wildcard ../.git/HEAD ../.git/refs/heads/*)
+
BUILDVM_O= host/buildvm.o host/buildvm_asm.o host/buildvm_peobj.o \
host/buildvm_lib.o host/buildvm_fold.o
BUILDVM_T= host/buildvm
@@ -476,13 +499,15 @@
LJVM_MODE= elfasm
LJLIB_O= lib_base.o lib_math.o lbitlib.o lib_bit.o lib_string.o lib_table.o \
- lib_io.o lib_os.o lib_package.o lib_debug.o lib_jit.o lib_ffi.o
+ lib_io.o lib_os.o lib_package.o lib_debug.o lib_jit.o lib_ffi.o \
+ lib_buffer.o
LJLIB_C= $(LJLIB_O:.o=.c)
-LJCORE_O= lj_gc.o lj_err.o lj_char.o lj_bc.o lj_obj.o lj_buf.o \
+LJCORE_O= lj_assert.o lj_gc.o lj_err.o lj_char.o lj_bc.o lj_obj.o lj_buf.o \
lj_str.o lj_tab.o lj_func.o lj_udata.o lj_meta.o lj_debug.o \
- lj_state.o lj_dispatch.o lj_vmevent.o lj_vmmath.o lj_strscan.o \
- lj_strfmt.o lj_strfmt_num.o lj_api.o lj_profile.o \
+ lj_prng.o lj_state.o lj_dispatch.o lj_vmevent.o lj_vmmath.o \
+ lj_strscan.o lj_strfmt.o lj_strfmt_num.o lj_serialize.o \
+ lj_api.o lj_profile.o \
lj_lex.o lj_parse.o lj_bcread.o lj_bcwrite.o lj_load.o \
lj_ir.o lj_opt_mem.o lj_opt_fold.o lj_opt_narrow.o \
lj_opt_dce.o lj_opt_loop.o lj_opt_split.o lj_opt_sink.o \
@@ -506,8 +531,8 @@
ALL_T= $(LUAJIT_T) $(LUAJIT_A) $(LUAJIT_SO) $(HOST_T)
ALL_HDRGEN= lj_bcdef.h lj_ffdef.h lj_libdef.h lj_recdef.h lj_folddef.h \
- host/buildvm_arch.h
-ALL_GEN= $(LJVM_S) $(ALL_HDRGEN) $(LIB_VMDEFP)
+ host/buildvm_arch.h luajit.h
+ALL_GEN= $(LJVM_S) $(ALL_HDRGEN) luajit_relver.txt $(LIB_VMDEFP)
WIN_RM= *.obj *.lib *.exp *.dll *.exe *.manifest *.pdb *.ilk
ALL_RM= $(ALL_T) $(ALL_GEN) *.o host/*.o $(WIN_RM)
@@ -557,6 +582,7 @@
MINILUA_X= host\minilua
BUILDVM_X= host\buildvm
ALL_RM:= $(subst /,\,$(ALL_RM))
+ HOST_RM= del
endif
endif
@@ -597,7 +623,6 @@
default all: $(TARGET_T)
amalg:
- @grep "^[+|]" ljamalg.c
$(MAKE) all "LJCORE_O=ljamalg.o"
clean:
@@ -631,7 +656,12 @@
$(E) "HOSTLINK $@"
$(Q)$(HOST_CC) $(HOST_ALDFLAGS) -o $@ $(MINILUA_O) $(MINILUA_LIBS) $(HOST_ALIBS)
-host/buildvm_arch.h: $(DASM_DASC) $(DASM_DEP) $(DASM_DIR)/*.lua
+luajit.h: $(MINILUA_DEP) $(GIT_DEP) luajit_rolling.h
+ $(E) "VERSION $@"
+ $(Q)$(GIT_RELVER)
+ $(Q)$(HOST_LUA) host/genversion.lua
+
+host/buildvm_arch.h: $(DASM_DASC) $(MINILUA_DEP) lj_arch.h lua.h luaconf.h
$(E) "DYNASM $@"
$(Q)$(DASM) $(DASM_FLAGS) -o $@ $(DASM_DASC)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.dep
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.dep 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.dep 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,16 +1,20 @@
lib_aux.o: lib_aux.c lua.h luaconf.h lauxlib.h lj_obj.h lj_def.h \
lj_arch.h lj_err.h lj_errmsg.h lj_state.h lj_trace.h lj_jit.h lj_ir.h \
- lj_dispatch.h lj_bc.h lj_traceerr.h lj_lib.h lj_alloc.h
+ lj_dispatch.h lj_bc.h lj_traceerr.h lj_lib.h lj_vmevent.h
lib_base.o: lib_base.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
- lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h \
- lj_tab.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cconv.h \
- lj_ff.h lj_ffdef.h lj_dispatch.h lj_jit.h lj_ir.h lj_char.h lj_strscan.h \
- lj_strfmt.h lj_lib.h lj_libdef.h
+ lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_buf.h \
+ lj_str.h lj_tab.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h \
+ lj_cconv.h lj_ff.h lj_ffdef.h lj_dispatch.h lj_jit.h lj_ir.h lj_char.h \
+ lj_strscan.h lj_strfmt.h lj_lib.h lj_libdef.h
lbitlib.o: lbitlib.c lua.h luaconf.h lauxlib.h lualib.h
lib_bit.o: lib_bit.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
lj_arch.h lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_strscan.h \
lj_strfmt.h lj_ctype.h lj_cdata.h lj_cconv.h lj_carith.h lj_ff.h \
lj_ffdef.h lj_lib.h lj_libdef.h
+lib_buffer.o: lib_buffer.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
+ lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h \
+ lj_tab.h lj_udata.h lj_meta.h lj_ctype.h lj_cdata.h lj_cconv.h \
+ lj_strfmt.h lj_serialize.h lj_lib.h lj_libdef.h
lib_debug.o: lib_debug.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_lib.h \
lj_libdef.h
@@ -29,7 +33,8 @@
lj_target.h lj_target_*.h lj_trace.h lj_dispatch.h lj_traceerr.h \
lj_vm.h lj_vmevent.h lj_lib.h luajit.h lj_libdef.h
lib_math.o: lib_math.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
- lj_def.h lj_arch.h lj_lib.h lj_vm.h lj_libdef.h
+ lj_def.h lj_arch.h lj_err.h lj_errmsg.h lj_lib.h lj_vm.h lj_prng.h \
+ lj_libdef.h
lib_os.o: lib_os.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_lib.h \
lj_libdef.h
@@ -42,16 +47,18 @@
lib_table.o: lib_table.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h \
lj_tab.h lj_ff.h lj_ffdef.h lj_lib.h lj_libdef.h
-lj_alloc.o: lj_alloc.c lj_def.h lua.h luaconf.h lj_arch.h lj_alloc.h
+lj_alloc.o: lj_alloc.c lj_def.h lua.h luaconf.h lj_arch.h lj_alloc.h \
+ lj_prng.h
lj_api.o: lj_api.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h lj_func.h lj_udata.h \
lj_meta.h lj_state.h lj_bc.h lj_frame.h lj_trace.h lj_jit.h lj_ir.h \
lj_dispatch.h lj_traceerr.h lj_vm.h lj_strscan.h lj_strfmt.h
lj_asm.o: lj_asm.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
- lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ctype.h lj_ir.h lj_jit.h \
- lj_ircall.h lj_iropt.h lj_mcode.h lj_trace.h lj_dispatch.h lj_traceerr.h \
- lj_snap.h lj_asm.h lj_vm.h lj_target.h lj_target_*.h lj_emit_*.h \
- lj_asm_*.h
+ lj_buf.h lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ctype.h lj_ir.h \
+ lj_jit.h lj_ircall.h lj_iropt.h lj_mcode.h lj_trace.h lj_dispatch.h \
+ lj_traceerr.h lj_snap.h lj_asm.h lj_vm.h lj_target.h lj_target_*.h \
+ lj_prng.h lj_emit_*.h lj_asm_*.h
+lj_assert.o: lj_assert.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h
lj_bc.o: lj_bc.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_bc.h \
lj_bcdef.h
lj_bcread.o: lj_bcread.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
@@ -76,8 +83,8 @@
lj_target_*.h lj_mcode.h lj_jit.h lj_ir.h lj_trace.h lj_dispatch.h \
lj_traceerr.h lj_vm.h
lj_cconv.o: lj_cconv.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
- lj_err.h lj_errmsg.h lj_tab.h lj_ctype.h lj_gc.h lj_cdata.h lj_cconv.h \
- lj_ccallback.h
+ lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_tab.h lj_ctype.h \
+ lj_cdata.h lj_cconv.h lj_ccallback.h
lj_cdata.o: lj_cdata.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_ctype.h lj_cconv.h lj_cdata.h
lj_char.o: lj_char.c lj_char.h lj_def.h lua.h luaconf.h
@@ -92,7 +99,7 @@
lj_cdata.h lj_cparse.h lj_cconv.h lj_carith.h lj_clib.h lj_ccall.h \
lj_ff.h lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
lj_dispatch.h lj_traceerr.h lj_record.h lj_ffrecord.h lj_snap.h \
- lj_crecord.h lj_strfmt.h
+ lj_crecord.h lj_strfmt.h lj_strscan.h
lj_ctype.o: lj_ctype.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_strfmt.h lj_ctype.h \
lj_ccallback.h lj_buf.h
@@ -109,10 +116,10 @@
lj_ff.h lj_ffdef.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h \
lj_traceerr.h lj_vm.h lj_strfmt.h
lj_ffrecord.o: lj_ffrecord.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
- lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ff.h \
- lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
- lj_dispatch.h lj_traceerr.h lj_record.h lj_ffrecord.h lj_crecord.h \
- lj_vm.h lj_strscan.h lj_strfmt.h lj_recdef.h
+ lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_tab.h lj_frame.h \
+ lj_bc.h lj_ff.h lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h \
+ lj_trace.h lj_dispatch.h lj_traceerr.h lj_record.h lj_ffrecord.h \
+ lj_crecord.h lj_vm.h lj_strscan.h lj_strfmt.h lj_serialize.h lj_recdef.h
lj_func.o: lj_func.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_func.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_bc.h \
lj_traceerr.h lj_vm.h
@@ -119,7 +126,7 @@
lj_gc.o: lj_gc.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_func.h lj_udata.h \
lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cdata.h lj_trace.h \
- lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h
+ lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h lj_vmevent.h
lj_gdbjit.o: lj_gdbjit.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_frame.h lj_bc.h lj_buf.h \
lj_str.h lj_strfmt.h lj_jit.h lj_ir.h lj_dispatch.h
@@ -126,7 +133,7 @@
lj_ir.o: lj_ir.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_buf.h lj_str.h lj_tab.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h \
lj_trace.h lj_dispatch.h lj_bc.h lj_traceerr.h lj_ctype.h lj_cdata.h \
- lj_carith.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lib.h
+ lj_carith.h lj_vm.h lj_strscan.h lj_serialize.h lj_strfmt.h lj_prng.h
lj_lex.o: lj_lex.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_ctype.h lj_cdata.h \
lualib.h lj_state.h lj_lex.h lj_parse.h lj_char.h lj_strscan.h \
@@ -133,14 +140,14 @@
lj_strfmt.h
lj_lib.o: lj_lib.c lauxlib.h lua.h luaconf.h lj_obj.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_func.h lj_bc.h \
- lj_dispatch.h lj_jit.h lj_ir.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lex.h \
- lj_bcdump.h lj_lib.h
+ lj_dispatch.h lj_jit.h lj_ir.h lj_ctype.h lj_vm.h lj_strscan.h \
+ lj_strfmt.h lj_lex.h lj_bcdump.h lj_lib.h
lj_load.o: lj_load.c lua.h luaconf.h lauxlib.h lj_obj.h lj_def.h \
lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_func.h \
lj_frame.h lj_bc.h lj_vm.h lj_lex.h lj_bcdump.h lj_parse.h
lj_mcode.o: lj_mcode.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_jit.h lj_ir.h lj_mcode.h lj_trace.h \
- lj_dispatch.h lj_bc.h lj_traceerr.h lj_vm.h
+ lj_dispatch.h lj_bc.h lj_traceerr.h lj_prng.h lj_vm.h
lj_meta.o: lj_meta.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_meta.h lj_frame.h \
lj_bc.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lib.h
@@ -156,7 +163,7 @@
lj_iropt.h lj_trace.h lj_dispatch.h lj_bc.h lj_traceerr.h lj_snap.h \
lj_vm.h
lj_opt_mem.o: lj_opt_mem.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
- lj_tab.h lj_ir.h lj_jit.h lj_iropt.h lj_ircall.h
+ lj_tab.h lj_ir.h lj_jit.h lj_iropt.h lj_ircall.h lj_dispatch.h lj_bc.h
lj_opt_narrow.o: lj_opt_narrow.c lj_obj.h lua.h luaconf.h lj_def.h \
lj_arch.h lj_bc.h lj_ir.h lj_jit.h lj_iropt.h lj_trace.h lj_dispatch.h \
lj_traceerr.h lj_vm.h lj_strscan.h
@@ -169,6 +176,7 @@
lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_buf.h lj_str.h lj_tab.h \
lj_func.h lj_state.h lj_bc.h lj_ctype.h lj_strfmt.h lj_lex.h lj_parse.h \
lj_vm.h lj_vmevent.h
+lj_prng.o: lj_prng.c lj_def.h lua.h luaconf.h lj_arch.h lj_prng.h
lj_profile.o: lj_profile.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_buf.h lj_gc.h lj_str.h lj_frame.h lj_bc.h lj_debug.h lj_dispatch.h \
lj_jit.h lj_ir.h lj_trace.h lj_traceerr.h lj_profile.h luajit.h
@@ -176,7 +184,10 @@
lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_meta.h lj_frame.h lj_bc.h \
lj_ctype.h lj_gc.h lj_ff.h lj_ffdef.h lj_debug.h lj_ir.h lj_jit.h \
lj_ircall.h lj_iropt.h lj_trace.h lj_dispatch.h lj_traceerr.h \
- lj_record.h lj_ffrecord.h lj_snap.h lj_vm.h
+ lj_record.h lj_ffrecord.h lj_snap.h lj_vm.h lj_prng.h
+lj_serialize.o: lj_serialize.c lj_obj.h lua.h luaconf.h lj_def.h \
+ lj_arch.h lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_tab.h \
+ lj_udata.h lj_ctype.h lj_cdata.h lj_ir.h lj_serialize.h
lj_snap.o: lj_snap.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
lj_tab.h lj_state.h lj_frame.h lj_bc.h lj_ir.h lj_jit.h lj_iropt.h \
lj_trace.h lj_dispatch.h lj_traceerr.h lj_snap.h lj_target.h \
@@ -184,11 +195,13 @@
lj_state.o: lj_state.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_func.h \
lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_trace.h lj_jit.h \
- lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h lj_lex.h lj_alloc.h luajit.h
+ lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h lj_prng.h lj_lex.h \
+ lj_alloc.h luajit.h
lj_str.o: lj_str.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
- lj_err.h lj_errmsg.h lj_str.h lj_char.h
+ lj_err.h lj_errmsg.h lj_str.h lj_char.h lj_prng.h
lj_strfmt.o: lj_strfmt.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
- lj_buf.h lj_gc.h lj_str.h lj_state.h lj_char.h lj_strfmt.h
+ lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_meta.h lj_state.h \
+ lj_char.h lj_strfmt.h lj_ctype.h lj_lib.h
lj_strfmt_num.o: lj_strfmt_num.c lj_obj.h lua.h luaconf.h lj_def.h \
lj_arch.h lj_buf.h lj_gc.h lj_str.h lj_strfmt.h
lj_strscan.o: lj_strscan.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
@@ -199,24 +212,25 @@
lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_frame.h lj_bc.h \
lj_state.h lj_ir.h lj_jit.h lj_iropt.h lj_mcode.h lj_trace.h \
lj_dispatch.h lj_traceerr.h lj_snap.h lj_gdbjit.h lj_record.h lj_asm.h \
- lj_vm.h lj_vmevent.h lj_target.h lj_target_*.h
+ lj_vm.h lj_vmevent.h lj_target.h lj_target_*.h lj_prng.h
lj_udata.o: lj_udata.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
- lj_gc.h lj_udata.h
+ lj_gc.h lj_err.h lj_errmsg.h lj_udata.h
lj_vmevent.o: lj_vmevent.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_str.h lj_tab.h lj_state.h lj_dispatch.h lj_bc.h lj_jit.h lj_ir.h \
lj_vm.h lj_vmevent.h
lj_vmmath.o: lj_vmmath.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
lj_ir.h lj_vm.h
-ljamalg.o: ljamalg.c lua.h luaconf.h lauxlib.h lj_gc.c lj_obj.h lj_def.h \
- lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h \
- lj_func.h lj_udata.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h \
- lj_cdata.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h \
- lj_vm.h lj_err.c lj_debug.h lj_ff.h lj_ffdef.h lj_strfmt.h lj_char.c \
- lj_char.h lj_bc.c lj_bcdef.h lj_obj.c lj_buf.c lj_str.c lj_tab.c \
- lj_func.c lj_udata.c lj_meta.c lj_strscan.h lj_lib.h lj_debug.c \
- lj_state.c lj_lex.h lj_alloc.h luajit.h lj_dispatch.c lj_ccallback.h \
- lj_profile.h lj_vmevent.c lj_vmevent.h lj_vmmath.c lj_strscan.c \
- lj_strfmt.c lj_strfmt_num.c lj_api.c lj_profile.c lj_lex.c lualib.h \
+ljamalg.o: ljamalg.c lua.h luaconf.h lauxlib.h lj_assert.c lj_obj.h \
+ lj_def.h lj_arch.h lj_gc.c lj_gc.h lj_err.h lj_errmsg.h lj_buf.h \
+ lj_str.h lj_tab.h lj_func.h lj_udata.h lj_meta.h lj_state.h lj_frame.h \
+ lj_bc.h lj_ctype.h lj_cdata.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h \
+ lj_traceerr.h lj_vm.h lj_vmevent.h lj_err.c lj_debug.h lj_ff.h \
+ lj_ffdef.h lj_strfmt.h lj_char.c lj_char.h lj_bc.c lj_bcdef.h lj_obj.c \
+ lj_buf.c lj_str.c lj_prng.h lj_tab.c lj_func.c lj_udata.c lj_meta.c \
+ lj_strscan.h lj_lib.h lj_debug.c lj_prng.c lj_state.c lj_lex.h \
+ lj_alloc.h luajit.h lj_dispatch.c lj_ccallback.h lj_profile.h \
+ lj_vmevent.c lj_vmmath.c lj_strscan.c lj_strfmt.c lj_strfmt_num.c \
+ lj_serialize.c lj_serialize.h lj_api.c lj_profile.c lj_lex.c lualib.h \
lj_parse.h lj_parse.c lj_bcread.c lj_bcdump.h lj_bcwrite.c lj_load.c \
lj_ctype.c lj_cdata.c lj_cconv.h lj_cconv.c lj_ccall.c lj_ccall.h \
lj_ccallback.c lj_target.h lj_target_*.h lj_mcode.h lj_carith.c \
@@ -228,7 +242,7 @@
lj_emit_*.h lj_asm_*.h lj_trace.c lj_gdbjit.h lj_gdbjit.c lj_alloc.c \
lib_aux.c lib_base.c lj_libdef.h lib_math.c lib_string.c lib_table.c \
lib_io.c lib_os.c lib_package.c lib_debug.c lib_bit.c lib_jit.c \
- lib_ffi.c lib_init.c
+ lib_ffi.c lib_buffer.c lib_init.c
luajit.o: luajit.c lua.h luaconf.h lauxlib.h lualib.h luajit.h lj_arch.h
host/buildvm.o: host/buildvm.c host/buildvm.h lj_def.h lua.h luaconf.h \
lj_arch.h lj_obj.h lj_def.h lj_arch.h lj_gc.h lj_obj.h lj_bc.h lj_ir.h \
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.std
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.std (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.std 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,746 @@
+##############################################################################
+# LuaJIT Makefile. Requires GNU Make.
+#
+# Please read doc/install.html before changing any variables!
+#
+# Suitable for POSIX platforms (Linux, *BSD, OSX etc.).
+# Also works with MinGW and Cygwin on Windows.
+# Please check msvcbuild.bat for building with MSVC on Windows.
+#
+# Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
+##############################################################################
+
+MAJVER= 2
+MINVER= 1
+ABIVER= 5.1
+NODOTABIVER= 51
+
+##############################################################################
+############################# COMPILER OPTIONS #############################
+##############################################################################
+# These options mainly affect the speed of the JIT compiler itself, not the
+# speed of the JIT-compiled code. Turn any of the optional settings on by
+# removing the '#' in front of them. Make sure you force a full recompile
+# with "make clean", followed by "make" if you change any options.
+#
+DEFAULT_CC = gcc
+#
+# LuaJIT builds as a native 32 or 64 bit binary by default.
+CC= $(DEFAULT_CC)
+#
+# Use this if you want to force a 32 bit build on a 64 bit multilib OS.
+#CC= $(DEFAULT_CC) -m32
+#
+# Since the assembler part does NOT maintain a frame pointer, it's pointless
+# to slow down the C part by not omitting it. Debugging, tracebacks and
+# unwinding are not affected -- the assembler part has frame unwind
+# information and GCC emits it where needed (x64) or with -g (see CCDEBUG).
+CCOPT= -O2 -fomit-frame-pointer
+# Use this if you want to generate a smaller binary (but it's slower):
+#CCOPT= -Os -fomit-frame-pointer
+# Note: it's no longer recommended to use -O3 with GCC 4.x.
+# The I-Cache bloat usually outweighs the benefits from aggressive inlining.
+#
+# Target-specific compiler options:
+#
+# x86/x64 only: For GCC 4.2 or higher and if you don't intend to distribute
+# the binaries to a different machine you could also use: -march=native
+#
+CCOPT_x86= -march=i686 -msse -msse2 -mfpmath=sse
+CCOPT_x64=
+CCOPT_arm=
+CCOPT_arm64=
+CCOPT_ppc=
+CCOPT_mips=
+#
+CCDEBUG=
+# Uncomment the next line to generate debug information:
+#CCDEBUG= -g
+#
+CCWARN= -Wall
+# Uncomment the next line to enable more warnings:
+#CCWARN+= -Wextra -Wdeclaration-after-statement -Wredundant-decls -Wshadow -Wpointer-arith
+#
+##############################################################################
+
+##############################################################################
+################################ BUILD MODE ################################
+##############################################################################
+# The default build mode is mixed mode on POSIX. On Windows this is the same
+# as dynamic mode.
+#
+# Mixed mode creates a static + dynamic library and a statically linked luajit.
+BUILDMODE= mixed
+#
+# Static mode creates a static library and a statically linked luajit.
+#BUILDMODE= static
+#
+# Dynamic mode creates a dynamic library and a dynamically linked luajit.
+# Note: this executable will only run when the library is installed!
+#BUILDMODE= dynamic
+#
+##############################################################################
+
+##############################################################################
+################################# FEATURES #################################
+##############################################################################
+# Enable/disable these features as needed, but make sure you force a full
+# recompile with "make clean", followed by "make".
+XCFLAGS=
+#
+# Permanently disable the FFI extension to reduce the size of the LuaJIT
+# executable. But please consider that the FFI library is compiled-in,
+# but NOT loaded by default. It only allocates any memory, if you actually
+# make use of it.
+#XCFLAGS+= -DLUAJIT_DISABLE_FFI
+#
+# Features from Lua 5.2 that are unlikely to break existing code are
+# enabled by default. Some other features that *might* break some existing
+# code (e.g. __pairs or os.execute() return values) can be enabled here.
+# Note: this does not provide full compatibility with Lua 5.2 at this time.
+#XCFLAGS+= -DLUAJIT_ENABLE_LUA52COMPAT
+#
+# Disable the JIT compiler, i.e. turn LuaJIT into a pure interpreter.
+#XCFLAGS+= -DLUAJIT_DISABLE_JIT
+#
+# Some architectures (e.g. PPC) can use either single-number (1) or
+# dual-number (2) mode. Uncomment one of these lines to override the
+# default mode. Please see LJ_ARCH_NUMMODE in lj_arch.h for details.
+#XCFLAGS+= -DLUAJIT_NUMMODE=1
+#XCFLAGS+= -DLUAJIT_NUMMODE=2
+#
+# Disable LJ_GC64 mode for x64.
+#XCFLAGS+= -DLUAJIT_DISABLE_GC64
+#
+##############################################################################
+
+##############################################################################
+############################ DEBUGGING SUPPORT #############################
+##############################################################################
+# Enable these options as needed, but make sure you force a full recompile
+# with "make clean", followed by "make".
+# Note that most of these are NOT suitable for benchmarking or release mode!
+#
+# Use the system provided memory allocator (realloc) instead of the
+# bundled memory allocator. This is slower, but sometimes helpful for
+# debugging. This option cannot be enabled on x64 without GC64, since
+# realloc usually doesn't return addresses in the right address range.
+# OTOH this option is mandatory for Valgrind's memcheck tool on x64 and
+# the only way to get useful results from it for all other architectures.
+#XCFLAGS+= -DLUAJIT_USE_SYSMALLOC
+#
+# This define is required to run LuaJIT under Valgrind. The Valgrind
+# header files must be installed. You should enable debug information, too.
+#XCFLAGS+= -DLUAJIT_USE_VALGRIND
+#
+# This is the client for the GDB JIT API. GDB 7.0 or higher is required
+# to make use of it. See lj_gdbjit.c for details. Enabling this causes
+# a non-negligible overhead, even when not running under GDB.
+#XCFLAGS+= -DLUAJIT_USE_GDBJIT
+#
+# Turn on assertions for the Lua/C API to debug problems with lua_* calls.
+# This is rather slow -- use only while developing C libraries/embeddings.
+#XCFLAGS+= -DLUA_USE_APICHECK
+#
+# Turn on assertions for the whole LuaJIT VM. This significantly slows down
+# everything. Use only if you suspect a problem with LuaJIT itself.
+#XCFLAGS+= -DLUA_USE_ASSERT
+#
+##############################################################################
+# You probably don't need to change anything below this line!
+##############################################################################
+
+##############################################################################
+# Host system detection.
+##############################################################################
+
+ifeq (Windows,$(findstring Windows,$(OS))$(MSYSTEM)$(TERM))
+ HOST_SYS= Windows
+else
+ HOST_SYS:= $(shell uname -s)
+ ifneq (,$(findstring MINGW,$(HOST_SYS)))
+ HOST_SYS= Windows
+ HOST_MSYS= mingw
+ endif
+ ifneq (,$(findstring MSYS,$(HOST_SYS)))
+ HOST_SYS= Windows
+ HOST_MSYS= mingw
+ endif
+ ifneq (,$(findstring CYGWIN,$(HOST_SYS)))
+ HOST_SYS= Windows
+ HOST_MSYS= cygwin
+ endif
+endif
+
+##############################################################################
+# Flags and options for host and target.
+##############################################################################
+
+# You can override the following variables at the make command line:
+# CC HOST_CC STATIC_CC DYNAMIC_CC
+# CFLAGS HOST_CFLAGS TARGET_CFLAGS
+# LDFLAGS HOST_LDFLAGS TARGET_LDFLAGS TARGET_SHLDFLAGS
+# LIBS HOST_LIBS TARGET_LIBS
+# CROSS HOST_SYS TARGET_SYS TARGET_FLAGS
+#
+# Cross-compilation examples:
+# make HOST_CC="gcc -m32" CROSS=i586-mingw32msvc- TARGET_SYS=Windows
+# make HOST_CC="gcc -m32" CROSS=powerpc-linux-gnu-
+
+ASOPTIONS= $(CCOPT) $(CCWARN) $(XCFLAGS) $(CFLAGS)
+CCOPTIONS= $(CCDEBUG) $(ASOPTIONS)
+LDOPTIONS= $(CCDEBUG) $(LDFLAGS)
+
+HOST_CC= $(CC)
+HOST_RM?= rm -f
+# If left blank, minilua is built and used. You can supply an installed
+# copy of (plain) Lua 5.1 or 5.2, plus Lua BitOp. E.g. with: HOST_LUA=lua
+HOST_LUA=
+
+HOST_XCFLAGS= -I.
+HOST_XLDFLAGS=
+HOST_XLIBS=
+HOST_ACFLAGS= $(CCOPTIONS) $(HOST_XCFLAGS) $(TARGET_ARCH) $(HOST_CFLAGS)
+HOST_ALDFLAGS= $(LDOPTIONS) $(HOST_XLDFLAGS) $(HOST_LDFLAGS)
+HOST_ALIBS= $(HOST_XLIBS) $(LIBS) $(HOST_LIBS)
+
+STATIC_CC = $(CROSS)$(CC)
+DYNAMIC_CC = $(CROSS)$(CC) -fPIC
+TARGET_CC= $(STATIC_CC)
+TARGET_STCC= $(STATIC_CC)
+TARGET_DYNCC= $(DYNAMIC_CC)
+TARGET_LD= $(CROSS)$(CC)
+TARGET_AR= $(CROSS)ar rcus
+TARGET_STRIP= $(CROSS)strip
+
+TARGET_LIBPATH= $(or $(PREFIX),/usr/local)/$(or $(MULTILIB),lib)
+TARGET_SONAME= libluajit-$(ABIVER).so.$(MAJVER)
+TARGET_DYLIBNAME= libluajit-$(ABIVER).$(MAJVER).dylib
+TARGET_DYLIBPATH= $(TARGET_LIBPATH)/$(TARGET_DYLIBNAME)
+TARGET_DLLNAME= lua$(NODOTABIVER).dll
+TARGET_DLLDOTANAME= libluajit-$(ABIVER).dll.a
+TARGET_XSHLDFLAGS= -shared -fPIC -Wl,-soname,$(TARGET_SONAME)
+TARGET_DYNXLDOPTS=
+
+TARGET_LFSFLAGS= -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
+TARGET_XCFLAGS= $(TARGET_LFSFLAGS) -U_FORTIFY_SOURCE
+TARGET_XLDFLAGS=
+TARGET_XLIBS= -lm
+TARGET_TCFLAGS= $(CCOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
+TARGET_ACFLAGS= $(CCOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
+TARGET_ASFLAGS= $(ASOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
+TARGET_ALDFLAGS= $(LDOPTIONS) $(TARGET_XLDFLAGS) $(TARGET_FLAGS) $(TARGET_LDFLAGS)
+TARGET_ASHLDFLAGS= $(LDOPTIONS) $(TARGET_XSHLDFLAGS) $(TARGET_FLAGS) $(TARGET_SHLDFLAGS)
+TARGET_ALIBS= $(TARGET_XLIBS) $(LIBS) $(TARGET_LIBS)
+
+TARGET_TESTARCH:=$(shell $(TARGET_CC) $(TARGET_TCFLAGS) -E lj_arch.h -dM)
+ifneq (,$(findstring LJ_TARGET_X64 ,$(TARGET_TESTARCH)))
+ TARGET_LJARCH= x64
+else
+ifneq (,$(findstring LJ_TARGET_X86 ,$(TARGET_TESTARCH)))
+ TARGET_LJARCH= x86
+else
+ifneq (,$(findstring LJ_TARGET_ARM ,$(TARGET_TESTARCH)))
+ TARGET_LJARCH= arm
+else
+ifneq (,$(findstring LJ_TARGET_ARM64 ,$(TARGET_TESTARCH)))
+ ifneq (,$(findstring __AARCH64EB__ ,$(TARGET_TESTARCH)))
+ TARGET_ARCH= -D__AARCH64EB__=1
+ endif
+ TARGET_LJARCH= arm64
+else
+ifneq (,$(findstring LJ_TARGET_PPC ,$(TARGET_TESTARCH)))
+ ifneq (,$(findstring LJ_LE 1,$(TARGET_TESTARCH)))
+ TARGET_ARCH= -DLJ_ARCH_ENDIAN=LUAJIT_LE
+ else
+ TARGET_ARCH= -DLJ_ARCH_ENDIAN=LUAJIT_BE
+ endif
+ TARGET_LJARCH= ppc
+else
+ifneq (,$(findstring LJ_TARGET_MIPS ,$(TARGET_TESTARCH)))
+ ifneq (,$(findstring MIPSEL ,$(TARGET_TESTARCH)))
+ TARGET_ARCH= -D__MIPSEL__=1
+ endif
+ ifneq (,$(findstring LJ_TARGET_MIPS64 ,$(TARGET_TESTARCH)))
+ TARGET_LJARCH= mips64
+ else
+ TARGET_LJARCH= mips
+ endif
+else
+ $(error Unsupported target architecture)
+endif
+endif
+endif
+endif
+endif
+endif
+
+ifneq (,$(findstring LJ_TARGET_PS3 1,$(TARGET_TESTARCH)))
+ TARGET_SYS= PS3
+ TARGET_ARCH+= -D__CELLOS_LV2__
+ TARGET_XCFLAGS+= -DLUAJIT_USE_SYSMALLOC
+ TARGET_XLIBS+= -lpthread
+endif
+
+TARGET_XCFLAGS+= $(CCOPT_$(TARGET_LJARCH))
+TARGET_ARCH+= $(patsubst %,-DLUAJIT_TARGET=LUAJIT_ARCH_%,$(TARGET_LJARCH))
+
+ifneq (,$(PREFIX))
+ifneq (/usr/local,$(PREFIX))
+ TARGET_XCFLAGS+= -DLUA_ROOT=\"$(PREFIX)\"
+ ifneq (/usr,$(PREFIX))
+ TARGET_DYNXLDOPTS= -Wl,-rpath,$(TARGET_LIBPATH)
+ endif
+endif
+endif
+ifneq (,$(MULTILIB))
+ TARGET_XCFLAGS+= -DLUA_MULTILIB=\"$(MULTILIB)\"
+endif
+ifneq (,$(LMULTILIB))
+ TARGET_XCFLAGS+= -DLUA_LMULTILIB=\"$(LMULTILIB)\"
+endif
+ifneq (,$(INSTALL_LJLIBD))
+ TARGET_XCFLAGS+= -DLUA_LJDIR=\"$(INSTALL_LJLIBD)\"
+endif
+
+##############################################################################
+# Target system detection.
+##############################################################################
+
+TARGET_SYS?= $(HOST_SYS)
+ifeq (Windows,$(TARGET_SYS))
+ TARGET_STRIP+= --strip-unneeded
+ TARGET_XSHLDFLAGS= -shared -Wl,--out-implib,$(TARGET_DLLDOTANAME)
+ TARGET_DYNXLDOPTS=
+else
+ TARGET_AR+= 2>/dev/null
+ifeq (,$(shell $(TARGET_CC) -o /dev/null -c -x c /dev/null -fno-stack-protector 2>/dev/null || echo 1))
+ TARGET_XCFLAGS+= -fno-stack-protector
+endif
+ifeq (Darwin,$(TARGET_SYS))
+ ifeq (,$(MACOSX_DEPLOYMENT_TARGET))
+ $(error missing: export MACOSX_DEPLOYMENT_TARGET=XX.YY)
+ endif
+ TARGET_STRIP+= -x
+ TARGET_XCFLAGS+= -DLUAJIT_UNWIND_EXTERNAL
+ TARGET_XSHLDFLAGS= -dynamiclib -undefined dynamic_lookup -fPIC
+ TARGET_DYNXLDOPTS=
+ TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).255
+else
+ifeq (iOS,$(TARGET_SYS))
+ TARGET_STRIP+= -x
+ TARGET_XSHLDFLAGS= -dynamiclib -undefined dynamic_lookup -fPIC
+ TARGET_DYNXLDOPTS=
+ TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).255
+ ifeq (arm64,$(TARGET_LJARCH))
+ TARGET_XCFLAGS+= -fno-omit-frame-pointer
+ endif
+else
+ ifeq (,$(findstring LJ_NO_UNWIND 1,$(TARGET_TESTARCH)))
+ # Find out whether the target toolchain always generates unwind tables.
+ TARGET_TESTUNWIND=$(shell exec 2>/dev/null; echo 'extern void b(void);int a(void){b();return 0;}' | $(TARGET_CC) -c -x c - -o tmpunwind.o && { grep -qa -e eh_frame -e __unwind_info tmpunwind.o || grep -qU -e eh_frame -e __unwind_info tmpunwind.o; } && echo E; rm -f tmpunwind.o)
+ ifneq (,$(findstring E,$(TARGET_TESTUNWIND)))
+ TARGET_XCFLAGS+= -DLUAJIT_UNWIND_EXTERNAL
+ endif
+ endif
+ ifneq (SunOS,$(TARGET_SYS))
+ ifneq (PS3,$(TARGET_SYS))
+ TARGET_XLDFLAGS+= -Wl,-E
+ endif
+ endif
+ ifeq (Linux,$(TARGET_SYS))
+ TARGET_XLIBS+= -ldl
+ endif
+ ifeq (GNU/kFreeBSD,$(TARGET_SYS))
+ TARGET_XLIBS+= -ldl
+ endif
+endif
+endif
+endif
+
+ifneq ($(HOST_SYS),$(TARGET_SYS))
+ ifeq (Windows,$(TARGET_SYS))
+ HOST_XCFLAGS+= -malign-double -DLUAJIT_OS=LUAJIT_OS_WINDOWS
+ else
+ ifeq (Linux,$(TARGET_SYS))
+ HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_LINUX
+ else
+ ifeq (Darwin,$(TARGET_SYS))
+ HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OSX
+ else
+ ifeq (iOS,$(TARGET_SYS))
+ HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OSX -DTARGET_OS_IPHONE=1
+ else
+ HOST_XCFLAGS+= -DLUAJIT_OS=LUAJIT_OS_OTHER
+ endif
+ endif
+ endif
+ endif
+endif
+
+ifneq (,$(CCDEBUG))
+ TARGET_STRIP= @:
+endif
+
+##############################################################################
+# Files and pathnames.
+##############################################################################
+
+MINILUA_O= host/minilua.o
+MINILUA_LIBS= -lm
+MINILUA_T= host/minilua
+MINILUA_X= $(MINILUA_T)
+MINILUA_DEP=
+
+ifeq (,$(HOST_LUA))
+ HOST_LUA= $(MINILUA_X)
+ MINILUA_DEP= $(MINILUA_T)
+endif
+
+DASM_DIR= ../dynasm
+DASM= $(HOST_LUA) $(DASM_DIR)/dynasm.lua
+DASM_XFLAGS=
+DASM_AFLAGS=
+DASM_ARCH= $(TARGET_LJARCH)
+
+ifneq (,$(findstring LJ_LE 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D ENDIAN_LE
+else
+ DASM_AFLAGS+= -D ENDIAN_BE
+endif
+ifneq (,$(findstring LJ_ARCH_BITS 64,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D P64
+endif
+ifneq (,$(findstring LJ_HASJIT 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D JIT
+endif
+ifneq (,$(findstring LJ_HASFFI 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D FFI
+endif
+ifneq (,$(findstring LJ_DUALNUM 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D DUALNUM
+endif
+ifneq (,$(findstring LJ_ARCH_HASFPU 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D FPU
+ TARGET_ARCH+= -DLJ_ARCH_HASFPU=1
+else
+ TARGET_ARCH+= -DLJ_ARCH_HASFPU=0
+endif
+ifeq (,$(findstring LJ_ABI_SOFTFP 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D HFABI
+ TARGET_ARCH+= -DLJ_ABI_SOFTFP=0
+else
+ TARGET_ARCH+= -DLJ_ABI_SOFTFP=1
+endif
+ifneq (,$(findstring LJ_NO_UNWIND 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D NO_UNWIND
+ TARGET_ARCH+= -DLUAJIT_NO_UNWIND
+endif
+ifneq (,$(findstring LJ_ABI_PAUTH 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D PAUTH
+ TARGET_ARCH+= -DLJ_ABI_PAUTH=1
+endif
+DASM_AFLAGS+= -D VER=$(subst LJ_ARCH_VERSION_,,$(filter LJ_ARCH_VERSION_%,$(subst LJ_ARCH_VERSION ,LJ_ARCH_VERSION_,$(TARGET_TESTARCH))))
+ifeq (Windows,$(TARGET_SYS))
+ DASM_AFLAGS+= -D WIN
+endif
+ifeq (x64,$(TARGET_LJARCH))
+ ifeq (,$(findstring LJ_FR2 1,$(TARGET_TESTARCH)))
+ DASM_ARCH= x86
+ endif
+else
+ifeq (arm,$(TARGET_LJARCH))
+ ifeq (iOS,$(TARGET_SYS))
+ DASM_AFLAGS+= -D IOS
+ endif
+else
+ifneq (,$(findstring LJ_TARGET_MIPSR6 ,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D MIPSR6
+endif
+ifeq (ppc,$(TARGET_LJARCH))
+ ifneq (,$(findstring LJ_ARCH_SQRT 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D SQRT
+ endif
+ ifneq (,$(findstring LJ_ARCH_ROUND 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D ROUND
+ endif
+ ifneq (,$(findstring LJ_ARCH_PPC32ON64 1,$(TARGET_TESTARCH)))
+ DASM_AFLAGS+= -D GPR64
+ endif
+ ifeq (PS3,$(TARGET_SYS))
+ DASM_AFLAGS+= -D PPE -D TOC
+ endif
+endif
+endif
+endif
+
+DASM_FLAGS= $(DASM_XFLAGS) $(DASM_AFLAGS)
+DASM_DASC= vm_$(DASM_ARCH).dasc
+
+GIT= git
+ifeq (Windows,$(HOST_SYS)$(HOST_MSYS))
+ GIT_RELVER= if exist ..\.git ( $(GIT) show -s --format=%%ct >luajit_relver.txt ) else ( type ..\.relver >luajit_relver.txt )
+else
+ GIT_RELVER= [ -e ../.git ] && $(GIT) show -s --format=%ct >luajit_relver.txt 2>/dev/null || cat ../.relver >luajit_relver.txt 2>/dev/null || :
+endif
+GIT_DEP= $(wildcard ../.git/HEAD ../.git/refs/heads/*)
+
+BUILDVM_O= host/buildvm.o host/buildvm_asm.o host/buildvm_peobj.o \
+ host/buildvm_lib.o host/buildvm_fold.o
+BUILDVM_T= host/buildvm
+BUILDVM_X= $(BUILDVM_T)
+
+HOST_O= $(MINILUA_O) $(BUILDVM_O)
+HOST_T= $(MINILUA_T) $(BUILDVM_T)
+
+LJVM_S= lj_vm.S
+LJVM_O= lj_vm.o
+LJVM_BOUT= $(LJVM_S)
+LJVM_MODE= elfasm
+
+LJLIB_O= lib_base.o lib_math.o lib_bit.o lib_string.o lib_table.o \
+ lib_io.o lib_os.o lib_package.o lib_debug.o lib_jit.o lib_ffi.o \
+ lib_buffer.o
+LJLIB_C= $(LJLIB_O:.o=.c)
+
+LJCORE_O= lj_assert.o lj_gc.o lj_err.o lj_char.o lj_bc.o lj_obj.o lj_buf.o \
+ lj_str.o lj_tab.o lj_func.o lj_udata.o lj_meta.o lj_debug.o \
+ lj_prng.o lj_state.o lj_dispatch.o lj_vmevent.o lj_vmmath.o \
+ lj_strscan.o lj_strfmt.o lj_strfmt_num.o lj_serialize.o \
+ lj_api.o lj_profile.o \
+ lj_lex.o lj_parse.o lj_bcread.o lj_bcwrite.o lj_load.o \
+ lj_ir.o lj_opt_mem.o lj_opt_fold.o lj_opt_narrow.o \
+ lj_opt_dce.o lj_opt_loop.o lj_opt_split.o lj_opt_sink.o \
+ lj_mcode.o lj_snap.o lj_record.o lj_crecord.o lj_ffrecord.o \
+ lj_asm.o lj_trace.o lj_gdbjit.o \
+ lj_ctype.o lj_cdata.o lj_cconv.o lj_ccall.o lj_ccallback.o \
+ lj_carith.o lj_clib.o lj_cparse.o \
+ lj_lib.o lj_alloc.o lib_aux.o \
+ $(LJLIB_O) lib_init.o
+
+LJVMCORE_O= $(LJVM_O) $(LJCORE_O)
+LJVMCORE_DYNO= $(LJVMCORE_O:.o=_dyn.o)
+
+LIB_VMDEF= jit/vmdef.lua
+LIB_VMDEFP= $(LIB_VMDEF)
+
+LUAJIT_O= luajit.o
+LUAJIT_A= libluajit.a
+LUAJIT_SO= libluajit.so
+LUAJIT_T= luajit
+
+ALL_T= $(LUAJIT_T) $(LUAJIT_A) $(LUAJIT_SO) $(HOST_T)
+ALL_HDRGEN= lj_bcdef.h lj_ffdef.h lj_libdef.h lj_recdef.h lj_folddef.h \
+ host/buildvm_arch.h luajit.h
+ALL_GEN= $(LJVM_S) $(ALL_HDRGEN) luajit_relver.txt $(LIB_VMDEFP)
+WIN_RM= *.obj *.lib *.exp *.dll *.exe *.manifest *.pdb *.ilk
+ALL_RM= $(ALL_T) $(ALL_GEN) *.o host/*.o $(WIN_RM)
+
+##############################################################################
+# Build mode handling.
+##############################################################################
+
+# Mixed mode defaults.
+TARGET_O= $(LUAJIT_A)
+TARGET_T= $(LUAJIT_T) $(LUAJIT_SO)
+TARGET_DEP= $(LIB_VMDEF) $(LUAJIT_SO)
+
+ifeq (Windows,$(TARGET_SYS))
+ TARGET_DYNCC= $(STATIC_CC)
+ LJVM_MODE= peobj
+ LJVM_BOUT= $(LJVM_O)
+ LUAJIT_T= luajit.exe
+ ifeq (cygwin,$(HOST_MSYS))
+ LUAJIT_SO= cyg$(TARGET_DLLNAME)
+ else
+ LUAJIT_SO= $(TARGET_DLLNAME)
+ endif
+ # Mixed mode is not supported on Windows. And static mode doesn't work well.
+ # C modules cannot be loaded, because they bind to lua51.dll.
+ ifneq (static,$(BUILDMODE))
+ BUILDMODE= dynamic
+ TARGET_XCFLAGS+= -DLUA_BUILD_AS_DLL
+ endif
+endif
+ifeq (Darwin,$(TARGET_SYS))
+ LJVM_MODE= machasm
+endif
+ifeq (iOS,$(TARGET_SYS))
+ LJVM_MODE= machasm
+endif
+ifeq (SunOS,$(TARGET_SYS))
+ BUILDMODE= static
+endif
+ifeq (PS3,$(TARGET_SYS))
+ BUILDMODE= static
+endif
+
+ifeq (Windows,$(HOST_SYS))
+ MINILUA_T= host/minilua.exe
+ BUILDVM_T= host/buildvm.exe
+ ifeq (,$(HOST_MSYS))
+ MINILUA_X= host\minilua
+ BUILDVM_X= host\buildvm
+ ALL_RM:= $(subst /,\,$(ALL_RM))
+ HOST_RM= del
+ endif
+endif
+
+ifeq (static,$(BUILDMODE))
+ TARGET_DYNCC= @:
+ TARGET_T= $(LUAJIT_T)
+ TARGET_DEP= $(LIB_VMDEF)
+else
+ifeq (dynamic,$(BUILDMODE))
+ ifneq (Windows,$(TARGET_SYS))
+ TARGET_CC= $(DYNAMIC_CC)
+ endif
+ TARGET_DYNCC= @:
+ LJVMCORE_DYNO= $(LJVMCORE_O)
+ TARGET_O= $(LUAJIT_SO)
+ TARGET_XLDFLAGS+= $(TARGET_DYNXLDOPTS)
+else
+ifeq (Darwin,$(TARGET_SYS))
+ TARGET_DYNCC= @:
+ LJVMCORE_DYNO= $(LJVMCORE_O)
+endif
+ifeq (iOS,$(TARGET_SYS))
+ TARGET_DYNCC= @:
+ LJVMCORE_DYNO= $(LJVMCORE_O)
+endif
+endif
+endif
+
+Q= @
+E= @echo
+#Q=
+#E= @:
+
+##############################################################################
+# Make targets.
+##############################################################################
+
+default all: $(TARGET_T)
+
+amalg:
+ $(MAKE) all "LJCORE_O=ljamalg.o"
+
+clean:
+ $(HOST_RM) $(ALL_RM)
+
+libbc:
+ ./$(LUAJIT_T) host/genlibbc.lua -o host/buildvm_libbc.h $(LJLIB_C)
+ $(MAKE) all
+
+depend:
+ @for file in $(ALL_HDRGEN); do \
+ test -f $$file || touch $$file; \
+ done
+ @$(HOST_CC) $(HOST_ACFLAGS) -MM *.c host/*.c | \
+ sed -e "s| [^ ]*/dasm_\S*\.h||g" \
+ -e "s|^\([^l ]\)|host/\1|" \
+ -e "s| lj_target_\S*\.h| lj_target_*.h|g" \
+ -e "s| lj_emit_\S*\.h| lj_emit_*.h|g" \
+ -e "s| lj_asm_\S*\.h| lj_asm_*.h|g" >Makefile.dep
+ @for file in $(ALL_HDRGEN); do \
+ test -s $$file || $(HOST_RM) $$file; \
+ done
+
+.PHONY: default all amalg clean libbc depend
+
+##############################################################################
+# Rules for generated files.
+##############################################################################
+
+$(MINILUA_T): $(MINILUA_O)
+ $(E) "HOSTLINK $@"
+ $(Q)$(HOST_CC) $(HOST_ALDFLAGS) -o $@ $(MINILUA_O) $(MINILUA_LIBS) $(HOST_ALIBS)
+
+luajit.h: $(MINILUA_DEP) $(GIT_DEP) luajit_rolling.h
+ $(E) "VERSION $@"
+ $(Q)$(GIT_RELVER)
+ $(Q)$(HOST_LUA) host/genversion.lua
+
+host/buildvm_arch.h: $(DASM_DASC) $(MINILUA_DEP) lj_arch.h lua.h luaconf.h
+ $(E) "DYNASM $@"
+ $(Q)$(DASM) $(DASM_FLAGS) -o $@ $(DASM_DASC)
+
+host/buildvm.o: $(DASM_DIR)/dasm_*.h
+
+$(BUILDVM_T): $(BUILDVM_O)
+ $(E) "HOSTLINK $@"
+ $(Q)$(HOST_CC) $(HOST_ALDFLAGS) -o $@ $(BUILDVM_O) $(HOST_ALIBS)
+
+$(LJVM_BOUT): $(BUILDVM_T)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m $(LJVM_MODE) -o $@
+
+lj_bcdef.h: $(BUILDVM_T) $(LJLIB_C)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m bcdef -o $@ $(LJLIB_C)
+
+lj_ffdef.h: $(BUILDVM_T) $(LJLIB_C)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m ffdef -o $@ $(LJLIB_C)
+
+lj_libdef.h: $(BUILDVM_T) $(LJLIB_C)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m libdef -o $@ $(LJLIB_C)
+
+lj_recdef.h: $(BUILDVM_T) $(LJLIB_C)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m recdef -o $@ $(LJLIB_C)
+
+$(LIB_VMDEF): $(BUILDVM_T) $(LJLIB_C)
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m vmdef -o $(LIB_VMDEFP) $(LJLIB_C)
+
+lj_folddef.h: $(BUILDVM_T) lj_opt_fold.c
+ $(E) "BUILDVM $@"
+ $(Q)$(BUILDVM_X) -m folddef -o $@ lj_opt_fold.c
+
+##############################################################################
+# Object file rules.
+##############################################################################
+
+%.o: %.c
+ $(E) "CC $@"
+ $(Q)$(TARGET_DYNCC) $(TARGET_ACFLAGS) -c -o $(@:.o=_dyn.o) $<
+ $(Q)$(TARGET_CC) $(TARGET_ACFLAGS) -c -o $@ $<
+
+%.o: %.S
+ $(E) "ASM $@"
+ $(Q)$(TARGET_DYNCC) $(TARGET_ASFLAGS) -c -o $(@:.o=_dyn.o) $<
+ $(Q)$(TARGET_CC) $(TARGET_ASFLAGS) -c -o $@ $<
+
+$(LUAJIT_O):
+ $(E) "CC $@"
+ $(Q)$(TARGET_STCC) $(TARGET_ACFLAGS) -c -o $@ $<
+
+$(HOST_O): %.o: %.c
+ $(E) "HOSTCC $@"
+ $(Q)$(HOST_CC) $(HOST_ACFLAGS) -c -o $@ $<
+
+include Makefile.dep
+
+##############################################################################
+# Target file rules.
+##############################################################################
+
+$(LUAJIT_A): $(LJVMCORE_O)
+ $(E) "AR $@"
+ $(Q)$(TARGET_AR) $@ $(LJVMCORE_O)
+
+# The dependency on _O, but linking with _DYNO is intentional.
+$(LUAJIT_SO): $(LJVMCORE_O)
+ $(E) "DYNLINK $@"
+ $(Q)$(TARGET_LD) $(TARGET_ASHLDFLAGS) -o $@ $(LJVMCORE_DYNO) $(TARGET_ALIBS)
+ $(Q)$(TARGET_STRIP) $@
+
+$(LUAJIT_T): $(TARGET_O) $(LUAJIT_O) $(TARGET_DEP)
+ $(E) "LINK $@"
+ $(Q)$(TARGET_LD) $(TARGET_ALDFLAGS) -o $@ $(LUAJIT_O) $(TARGET_O) $(TARGET_ALIBS)
+ $(Q)$(TARGET_STRIP) $@
+ $(E) "OK Successfully built LuaJIT"
+
+##############################################################################
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/Makefile.std
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_bcdef.h_
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_bcdef.h_ 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_bcdef.h_ 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,226 +0,0 @@
-/* This is a generated file. DO NOT EDIT! */
-
-LJ_DATADEF const uint16_t lj_bc_ofs[] = {
-0,
-72,
-144,
-216,
-288,
-427,
-569,
-632,
-695,
-764,
-833,
-886,
-938,
-989,
-1040,
-1081,
-1122,
-1148,
-1180,
-1240,
-1314,
-1368,
-1422,
-1476,
-1530,
-1589,
-1643,
-1697,
-1751,
-1805,
-1841,
-1908,
-1975,
-2042,
-2109,
-2158,
-2230,
-2306,
-2342,
-2378,
-2408,
-2437,
-2462,
-2505,
-2541,
-2628,
-2710,
-2748,
-2782,
-2833,
-2897,
-3006,
-3099,
-3117,
-3135,
-3283,
-3407,
-3506,
-3679,
-3908,
-4032,
-4174,
-4220,
-4262,
-4266,
-4414,
-4482,
-4647,
-4838,
-4926,
-4930,
-5066,
-5158,
-5263,
-5360,
-5465,
-5485,
-5555,
-5622,
-5642,
-5686,
-5725,
-5745,
-5763,
-5810,
-5835,
-5855,
-5918,
-5972,
-5972,
-6097,
-6098,
-6177,
-7841,
-7908,
-8419,
-8522,
-8579,
-8710,
-7974,
-8136,
-8228,
-8280,
-8311,
-8768,
-8809,
-9417,
-8864,
-9167,
-9469,
-9596,
-9620,
-9647,
-9711,
-9744,
-9778,
-9809,
-9840,
-9873,
-9914,
-9957,
-9990,
-10030,
-10070,
-10245,
-10393,
-10110,
-10110,
-9678,
-10149,
-10549,
-10492,
-10196,
-10603,
-10662,
-11596,
-11994,
-11941,
-12063,
-12142,
-12224,
-12306,
-12388,
-11650,
-11747,
-11844,
-10721,
-10752,
-10799,
-10921,
-11090,
-11217,
-11327,
-11442,
-11557
-};
-
-LJ_DATADEF const uint16_t lj_bc_mode[] = {
-BCDEF(BCMODE)
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF,
-BCMODE_FF
-};
-
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_ffdef.h_
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_ffdef.h_ 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_ffdef.h_ 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,210 +0,0 @@
-/* This is a generated file. DO NOT EDIT! */
-
-FFDEF(assert)
-FFDEF(type)
-FFDEF(next)
-FFDEF(pairs)
-FFDEF(ipairs_aux)
-FFDEF(ipairs)
-FFDEF(getmetatable)
-FFDEF(setmetatable)
-FFDEF(getfenv)
-FFDEF(setfenv)
-FFDEF(rawget)
-FFDEF(rawset)
-FFDEF(rawequal)
-FFDEF(unpack)
-FFDEF(select)
-FFDEF(tonumber)
-FFDEF(tostring)
-FFDEF(error)
-FFDEF(pcall)
-FFDEF(xpcall)
-FFDEF(loadfile)
-FFDEF(load)
-FFDEF(loadstring)
-FFDEF(dofile)
-FFDEF(gcinfo)
-FFDEF(collectgarbage)
-FFDEF(newproxy)
-FFDEF(print)
-FFDEF(coroutine_status)
-FFDEF(coroutine_running)
-FFDEF(coroutine_create)
-FFDEF(coroutine_yield)
-FFDEF(coroutine_resume)
-FFDEF(coroutine_wrap_aux)
-FFDEF(coroutine_wrap)
-FFDEF(math_abs)
-FFDEF(math_floor)
-FFDEF(math_ceil)
-FFDEF(math_sqrt)
-FFDEF(math_log10)
-FFDEF(math_exp)
-FFDEF(math_sin)
-FFDEF(math_cos)
-FFDEF(math_tan)
-FFDEF(math_asin)
-FFDEF(math_acos)
-FFDEF(math_atan)
-FFDEF(math_sinh)
-FFDEF(math_cosh)
-FFDEF(math_tanh)
-FFDEF(math_frexp)
-FFDEF(math_modf)
-FFDEF(math_deg)
-FFDEF(math_rad)
-FFDEF(math_log)
-FFDEF(math_atan2)
-FFDEF(math_pow)
-FFDEF(math_fmod)
-FFDEF(math_ldexp)
-FFDEF(math_min)
-FFDEF(math_max)
-FFDEF(math_random)
-FFDEF(math_randomseed)
-FFDEF(bit_tobit)
-FFDEF(bit_bnot)
-FFDEF(bit_bswap)
-FFDEF(bit_lshift)
-FFDEF(bit_rshift)
-FFDEF(bit_arshift)
-FFDEF(bit_rol)
-FFDEF(bit_ror)
-FFDEF(bit_band)
-FFDEF(bit_bor)
-FFDEF(bit_bxor)
-FFDEF(bit_tohex)
-FFDEF(string_len)
-FFDEF(string_byte)
-FFDEF(string_char)
-FFDEF(string_sub)
-FFDEF(string_rep)
-FFDEF(string_reverse)
-FFDEF(string_lower)
-FFDEF(string_upper)
-FFDEF(string_dump)
-FFDEF(string_find)
-FFDEF(string_match)
-FFDEF(string_gmatch_aux)
-FFDEF(string_gmatch)
-FFDEF(string_gsub)
-FFDEF(string_format)
-FFDEF(table_foreachi)
-FFDEF(table_foreach)
-FFDEF(table_getn)
-FFDEF(table_maxn)
-FFDEF(table_insert)
-FFDEF(table_remove)
-FFDEF(table_concat)
-FFDEF(table_sort)
-FFDEF(io_method_close)
-FFDEF(io_method_read)
-FFDEF(io_method_write)
-FFDEF(io_method_flush)
-FFDEF(io_method_seek)
-FFDEF(io_method_setvbuf)
-FFDEF(io_method_lines)
-FFDEF(io_method___gc)
-FFDEF(io_method___tostring)
-FFDEF(io_open)
-FFDEF(io_popen)
-FFDEF(io_tmpfile)
-FFDEF(io_close)
-FFDEF(io_read)
-FFDEF(io_write)
-FFDEF(io_flush)
-FFDEF(io_input)
-FFDEF(io_output)
-FFDEF(io_lines)
-FFDEF(io_type)
-FFDEF(os_execute)
-FFDEF(os_remove)
-FFDEF(os_rename)
-FFDEF(os_tmpname)
-FFDEF(os_getenv)
-FFDEF(os_exit)
-FFDEF(os_clock)
-FFDEF(os_date)
-FFDEF(os_time)
-FFDEF(os_difftime)
-FFDEF(os_setlocale)
-FFDEF(debug_getregistry)
-FFDEF(debug_getmetatable)
-FFDEF(debug_setmetatable)
-FFDEF(debug_getfenv)
-FFDEF(debug_setfenv)
-FFDEF(debug_getinfo)
-FFDEF(debug_getlocal)
-FFDEF(debug_setlocal)
-FFDEF(debug_getupvalue)
-FFDEF(debug_setupvalue)
-FFDEF(debug_upvalueid)
-FFDEF(debug_upvaluejoin)
-FFDEF(debug_sethook)
-FFDEF(debug_gethook)
-FFDEF(debug_debug)
-FFDEF(debug_traceback)
-FFDEF(jit_on)
-FFDEF(jit_off)
-FFDEF(jit_flush)
-FFDEF(jit_status)
-FFDEF(jit_attach)
-FFDEF(jit_util_funcinfo)
-FFDEF(jit_util_funcbc)
-FFDEF(jit_util_funck)
-FFDEF(jit_util_funcuvname)
-FFDEF(jit_util_traceinfo)
-FFDEF(jit_util_traceir)
-FFDEF(jit_util_tracek)
-FFDEF(jit_util_tracesnap)
-FFDEF(jit_util_tracemc)
-FFDEF(jit_util_traceexitstub)
-FFDEF(jit_util_ircalladdr)
-FFDEF(jit_opt_start)
-FFDEF(ffi_meta___index)
-FFDEF(ffi_meta___newindex)
-FFDEF(ffi_meta___eq)
-FFDEF(ffi_meta___len)
-FFDEF(ffi_meta___lt)
-FFDEF(ffi_meta___le)
-FFDEF(ffi_meta___concat)
-FFDEF(ffi_meta___call)
-FFDEF(ffi_meta___add)
-FFDEF(ffi_meta___sub)
-FFDEF(ffi_meta___mul)
-FFDEF(ffi_meta___div)
-FFDEF(ffi_meta___mod)
-FFDEF(ffi_meta___pow)
-FFDEF(ffi_meta___unm)
-FFDEF(ffi_meta___tostring)
-FFDEF(ffi_meta___pairs)
-FFDEF(ffi_meta___ipairs)
-FFDEF(ffi_clib___index)
-FFDEF(ffi_clib___newindex)
-FFDEF(ffi_clib___gc)
-FFDEF(ffi_callback_free)
-FFDEF(ffi_callback_set)
-FFDEF(ffi_cdef)
-FFDEF(ffi_new)
-FFDEF(ffi_cast)
-FFDEF(ffi_typeof)
-FFDEF(ffi_istype)
-FFDEF(ffi_sizeof)
-FFDEF(ffi_alignof)
-FFDEF(ffi_offsetof)
-FFDEF(ffi_errno)
-FFDEF(ffi_string)
-FFDEF(ffi_copy)
-FFDEF(ffi_fill)
-FFDEF(ffi_abi)
-FFDEF(ffi_metatype)
-FFDEF(ffi_gc)
-FFDEF(ffi_load)
-
-#undef FFDEF
-
-#ifndef FF_NUM_ASMFUNC
-#define FF_NUM_ASMFUNC 62
-#endif
-
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_folddef.h_
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_folddef.h_ 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_folddef.h_ 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,1068 +0,0 @@
-/* This is a generated file. DO NOT EDIT! */
-
-static const FoldFunc fold_func[] = {
- fold_kfold_numarith,
- fold_kfold_ldexp,
- fold_kfold_fpmath,
- fold_kfold_numpow,
- fold_kfold_numcomp,
- fold_kfold_intarith,
- fold_kfold_intovarith,
- fold_kfold_bnot,
- fold_kfold_bswap,
- fold_kfold_intcomp,
- fold_kfold_intcomp0,
- fold_kfold_int64arith,
- fold_kfold_int64arith2,
- fold_kfold_int64shift,
- fold_kfold_bnot64,
- fold_kfold_bswap64,
- fold_kfold_int64comp,
- fold_kfold_int64comp0,
- fold_kfold_snew_kptr,
- fold_kfold_snew_empty,
- fold_kfold_strref,
- fold_kfold_strref_snew,
- fold_kfold_strcmp,
- fold_kfold_add_kgc,
- fold_kfold_add_kptr,
- fold_kfold_add_kright,
- fold_kfold_tobit,
- fold_kfold_conv_kint_num,
- fold_kfold_conv_kintu32_num,
- fold_kfold_conv_kint_ext,
- fold_kfold_conv_kint_i64,
- fold_kfold_conv_kint64_num_i64,
- fold_kfold_conv_kint64_num_u64,
- fold_kfold_conv_kint64_int_i64,
- fold_kfold_conv_knum_int_num,
- fold_kfold_conv_knum_u32_num,
- fold_kfold_conv_knum_i64_num,
- fold_kfold_conv_knum_u64_num,
- fold_kfold_tostr_knum,
- fold_kfold_tostr_kint,
- fold_kfold_strto,
- lj_opt_cse,
- fold_kfold_kref,
- fold_shortcut_round,
- fold_shortcut_left,
- fold_shortcut_dropleft,
- fold_shortcut_leftleft,
- fold_simplify_numadd_negx,
- fold_simplify_numadd_xneg,
- fold_simplify_numsub_k,
- fold_simplify_numsub_negk,
- fold_simplify_numsub_xneg,
- fold_simplify_nummuldiv_k,
- fold_simplify_nummuldiv_negk,
- fold_simplify_nummuldiv_negneg,
- fold_simplify_numpow_xk,
- fold_simplify_numpow_kx,
- fold_shortcut_conv_num_int,
- fold_simplify_conv_int_num,
- fold_simplify_conv_i64_num,
- fold_simplify_conv_int_i64,
- fold_simplify_conv_flt_num,
- fold_simplify_tobit_conv,
- fold_simplify_floor_conv,
- fold_simplify_conv_sext,
- fold_simplify_conv_narrow,
- fold_cse_conv,
- fold_narrow_convert,
- fold_simplify_intadd_k,
- fold_simplify_intmul_k,
- fold_simplify_intsub_k,
- fold_simplify_intsub_kleft,
- fold_simplify_intadd_k64,
- fold_simplify_intsub_k64,
- fold_simplify_intmul_k32,
- fold_simplify_intmul_k64,
- fold_simplify_intmod_k,
- fold_simplify_intmod_kleft,
- fold_simplify_intsub,
- fold_simplify_intsubadd_leftcancel,
- fold_simplify_intsubsub_leftcancel,
- fold_simplify_intsubsub_rightcancel,
- fold_simplify_intsubadd_rightcancel,
- fold_simplify_intsubaddadd_cancel,
- fold_simplify_band_k,
- fold_simplify_bor_k,
- fold_simplify_bxor_k,
- fold_simplify_shift_ik,
- fold_simplify_shift_andk,
- fold_simplify_shift1_ki,
- fold_simplify_shift2_ki,
- fold_simplify_shiftk_andk,
- fold_simplify_andk_shiftk,
- fold_reassoc_intarith_k,
- fold_reassoc_intarith_k64,
- fold_reassoc_dup,
- fold_reassoc_bxor,
- fold_reassoc_shift,
- fold_reassoc_minmax_k,
- fold_reassoc_minmax_left,
- fold_reassoc_minmax_right,
- fold_abc_fwd,
- fold_abc_k,
- fold_abc_invar,
- fold_comm_swap,
- fold_comm_equal,
- fold_comm_comp,
- fold_comm_dup,
- fold_comm_bxor,
- fold_merge_eqne_snew_kgc,
- lj_opt_fwd_aload,
- fold_kfold_hload_kkptr,
- lj_opt_fwd_hload,
- lj_opt_fwd_uload,
- lj_opt_fwd_tab_len,
- fold_cse_uref,
- lj_opt_fwd_hrefk,
- fold_fwd_href_tnew,
- fold_fwd_href_tdup,
- fold_fload_tab_tnew_asize,
- fold_fload_tab_tnew_hmask,
- fold_fload_tab_tdup_asize,
- fold_fload_tab_tdup_hmask,
- fold_fload_tab_ah,
- fold_fload_str_len_kgc,
- fold_fload_str_len_snew,
- fold_fload_cdata_typeid_kgc,
- fold_fload_cdata_int64_kgc,
- fold_fload_cdata_typeid_cnew,
- fold_fload_cdata_ptr_int64_cnew,
- lj_opt_cse,
- lj_opt_fwd_fload,
- fold_fwd_sload,
- fold_xload_kptr,
- lj_opt_fwd_xload,
- fold_barrier_tab,
- fold_barrier_tnew_tdup,
- lj_opt_dse_ahstore,
- lj_opt_dse_ustore,
- lj_opt_dse_fstore,
- lj_opt_dse_xstore,
- lj_ir_emit
-};
-
-static const uint32_t fold_hash[916] = {
-0xffffffff,
-0xffffffff,
-0x5b4c8016,
-0x0d4e7016,
-0xffffffff,
-0x1000701c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x29110c1a,
-0xffffffff,
-0xffffffff,
-0x5b488016,
-0x0d4a7016,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x7b87fc07,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x0d467016,
-0xffffffff,
-0x5a4c73ff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5153fc29,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5d408016,
-0xffffffff,
-0x594873ff,
-0x8187440f,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8287fc0f,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6715ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a11fc1a,
-0xffffffff,
-0x1daa5a70,
-0xffffffff,
-0xffffffff,
-0x0a0bfc16,
-0x5c408c16,
-0x6911ffff,
-0x8db7ffff,
-0xffffffff,
-0xffffffff,
-0x1caa59d4,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6a0dffff,
-0x2b68d002,
-0xffffffff,
-0x3cab5695,
-0xffffffff,
-0x41aaa675,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x27ae5800,
-0xffffffff,
-0x6a09ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x7f865c0f,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6a05ffff,
-0x42abffff,
-0x5e44881c,
-0x5d50a016,
-0x066c5816,
-0x00646c1b,
-0x75753bff,
-0x1951fc18,
-0x6264c81b,
-0x1850641c,
-0xffffffff,
-0x6a01ffff,
-0x87a7ffff,
-0x4953fc1c,
-0x8da80000,
-0x4f52a3ff,
-0x00606c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5d428416,
-0x88a53800,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x05645816,
-0xffffffff,
-0x005c6c1b,
-0x20aa71d6,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x1399fc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x157f33ff,
-0xffffffff,
-0xffffffff,
-0x584dfc20,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8d9bffff,
-0xffffffff,
-0x055c5816,
-0xffffffff,
-0x00546c1b,
-0xffffffff,
-0xffffffff,
-0x5849fc20,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8c97ffff,
-0x5543fc1c,
-0x05585816,
-0xffffffff,
-0x00506c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8a93ffff,
-0x26ae6c00,
-0x05545816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x76753c17,
-0x41aaa695,
-0xffffffff,
-0x898fffff,
-0xffffffff,
-0x05505816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x858867ff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x848bffff,
-0xffffffff,
-0x054c5816,
-0x79873c06,
-0x47525bff,
-0xffffffff,
-0x3f695401,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8387ffff,
-0xffffffff,
-0x05485816,
-0xffffffff,
-0x5a4e5bff,
-0xffffffff,
-0xffffffff,
-0x6264c816,
-0x43aaa26e,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x05445816,
-0xffffffff,
-0x5a4a5bff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3455fc1b,
-0x0c5a701c,
-0x6366cbff,
-0x0e3c7000,
-0xffffffff,
-0x05405816,
-0xffffffff,
-0x59465bff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x41aaa276,
-0x0c56701c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x21aa7275,
-0x0b52701c,
-0x61489016,
-0x6465fc33,
-0x8d77ffff,
-0xffffffff,
-0x7b87fc05,
-0xffffffff,
-0xffffffff,
-0x2a126bff,
-0x385a6fff,
-0xffffffff,
-0x446dfc16,
-0xffffffff,
-0x7473ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x7d873000,
-0xffffffff,
-0x5c409016,
-0x686fffff,
-0x8187440d,
-0xffffffff,
-0xffffffff,
-0x3554b81b,
-0x8287fc0d,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x686bffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8d9ffc00,
-0x737a5fff,
-0x41aaaa75,
-0xffffffff,
-0xffffffff,
-0x5e40801c,
-0x0b42701c,
-0x6b67ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2b68d000,
-0xffffffff,
-0xffffffff,
-0x6d133017,
-0xffffffff,
-0xffffffff,
-0x4c59fc16,
-0xffffffff,
-0xffffffff,
-0x110bfc1c,
-0x3aab566e,
-0xffffffff,
-0x5052a7ff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6515fc28,
-0x4a55fc16,
-0x7f865c0d,
-0x88a53c00,
-0x41aaa296,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x4451fc16,
-0xffffffff,
-0x60448bff,
-0x21aa7295,
-0xffffffff,
-0x3cab5676,
-0x04106c1b,
-0xffffffff,
-0x78873807,
-0xffffffff,
-0xffffffff,
-0x574dfc16,
-0xffffffff,
-0x4e53ffff,
-0xffffffff,
-0x09145816,
-0xffffffff,
-0x040c6c1b,
-0x8287fc00,
-0x5e50a01c,
-0x6467fc32,
-0xffffffff,
-0x5749fc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a105816,
-0x2e3e7c00,
-0x04086c1b,
-0x7083fc00,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5645fc16,
-0xffffffff,
-0x22aa6e6e,
-0x5e42841c,
-0x614e9c16,
-0x090c5816,
-0x04046c1b,
-0x1eaa5ab3,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5441fc16,
-0x41aaaa95,
-0xffffffff,
-0x5352a028,
-0x09085816,
-0x17505c16,
-0x04006c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6b43ffff,
-0xffffffff,
-0x09045816,
-0xffffffff,
-0x43aaa2ae,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x083e5800,
-0x7c865c00,
-0xffffffff,
-0x76753c15,
-0x3051fc2e,
-0x09005816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3aab568e,
-0xffffffff,
-0x43aaa66e,
-0xffffffff,
-0x1daa5a71,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a12701c,
-0x5f66cfff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3cab5696,
-0xffffffff,
-0x100e701c,
-0x41aaa676,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a125c17,
-0x3654b82e,
-0x100a701c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x1006701c,
-0xffffffff,
-0x1951fc19,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x23aa6e8e,
-0xffffffff,
-0x5b4e8016,
-0xffffffff,
-0x1eaa5ad3,
-0x1002701c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x29130c1a,
-0xffffffff,
-0xffffffff,
-0x0d4c7016,
-0xffffffff,
-0x475273ff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5b468016,
-0x0d487016,
-0x5a4e73ff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5d54a816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5a4a73ff,
-0x6615fc16,
-0x3bab56ae,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x594673ff,
-0xffffffff,
-0x61468c16,
-0x8d17ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a13fc1a,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6913ffff,
-0x40abfeb3,
-0x8db9ffff,
-0x41aaa696,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6a0fffff,
-0x8db5ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x7a873c07,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6a0bffff,
-0x3f695402,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x24aa6eae,
-0xffffffff,
-0xffffffff,
-0x6a07ffff,
-0xffffffff,
-0xffffffff,
-0x066e5816,
-0xffffffff,
-0x00666c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6a03ffff,
-0xffffffff,
-0x4b55fc1c,
-0x066a5816,
-0xffffffff,
-0x00626c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x4851fc1c,
-0x05665816,
-0x18506016,
-0x005e6c1b,
-0x12986416,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8da1ffff,
-0xffffffff,
-0x3bab56ce,
-0xffffffff,
-0x43aaa6ae,
-0xffffffff,
-0xffffffff,
-0x584ffc20,
-0x7b87fc06,
-0xffffffff,
-0x5f4287ff,
-0x8d9dffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x00566c1b,
-0xffffffff,
-0xffffffff,
-0x584bfc20,
-0x5253fc28,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5645fc1c,
-0xffffffff,
-0x40abfed3,
-0x00526c1b,
-0x8187440e,
-0xffffffff,
-0x5847fc20,
-0x8287fc0e,
-0xffffffff,
-0xffffffff,
-0x8b95ffff,
-0x2e3c7800,
-0x5441fc1c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x17505c1c,
-0xffffffff,
-0xffffffff,
-0x41aaaa76,
-0xffffffff,
-0x614c9816,
-0x8991ffff,
-0x1daa5a6f,
-0x05525816,
-0x4d585bff,
-0xffffffff,
-0x8087400c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x1baa59d3,
-0x828dffff,
-0x25aa6ece,
-0x054e5816,
-0x76753c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2b68d001,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x8689ffff,
-0xffffffff,
-0x054a5816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x43aca01b,
-0x05465816,
-0x7f865c0e,
-0x5a4c5bff,
-0x39ab55d3,
-0x01626c16,
-0x02686fff,
-0x3457fc1b,
-0xffffffff,
-0xffffffff,
-0x0f3e7000,
-0x3dab55ae,
-0x05425816,
-0x1951fc17,
-0x59485bff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3153fc1b,
-0x0c58701c,
-0x5f64cbff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x035a6c16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x0b54701c,
-0xffffffff,
-0x8779ffff,
-0x1faa71d5,
-0xffffffff,
-0xffffffff,
-0x2d5eb81b,
-0x72b5fc08,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x0b50701c,
-0x456ffc16,
-0x7b75ffff,
-0xffffffff,
-0xffffffff,
-0x147e5c16,
-0xffffffff,
-0xffffffff,
-0x2a106bff,
-0xffffffff,
-0x1eaa5ab4,
-0x446bfc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x41aaaa96,
-0xffffffff,
-0x3556b81b,
-0x87a5fc00,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x4e6dffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3252b81b,
-0xffffffff,
-0x5e54a81c,
-0xffffffff,
-0xffffffff,
-0x0b44701c,
-0x28b05c00,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x73785fff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x0b40701c,
-0xffffffff,
-0x6b65ffff,
-0xffffffff,
-0xffffffff,
-0x1daa5a72,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6266cc1b,
-0xffffffff,
-0x375bfc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x3f695400,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6d113017,
-0x3ead541b,
-0xffffffff,
-0x5d448816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x18506416,
-0xffffffff,
-0xffffffff,
-0x16b37400,
-0xffffffff,
-0x4653fc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x04126c1b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x574ffc16,
-0xffffffff,
-0x6855ffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x040e6c1b,
-0x41aaa275,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x574bfc16,
-0x6f826400,
-0x6851ffff,
-0x1eaa5ad4,
-0x2a125816,
-0xffffffff,
-0x040a6c1b,
-0x7185fc00,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x5747fc16,
-0x7b87fc04,
-0xffffffff,
-0xffffffff,
-0x090e5816,
-0xffffffff,
-0x04066c1b,
-0x6e81fc00,
-0x1aac6c1b,
-0x1850601c,
-0x2e5cbbff,
-0x5543fc16,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x090a5816,
-0xffffffff,
-0x04026c1b,
-0xffffffff,
-0xffffffff,
-0x8087440c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6c45ffff,
-0x8287fc0c,
-0x09065816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x6b41ffff,
-0x3353fc2e,
-0x09025816,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2f50bbff,
-0x073c5800,
-0x6266cc16,
-0x5f4083ff,
-0xffffffff,
-0xffffffff,
-0x43aca41b,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2a10701c,
-0x6364cfff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x7e865c0c,
-0xffffffff,
-0xffffffff,
-0x3656b82e,
-0x41aaa295,
-0x100c701c,
-0x614a9416,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x2c5ebc1b,
-0xffffffff,
-0x2a105c17,
-0xffffffff,
-0x1008701c,
-0x3cab5675,
-0xffffffff,
-0xffffffff,
-0x77873806,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0x1004701c,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff,
-0xffffffff
-};
-
-#define fold_hashkey(k) (lj_rol(lj_rol((k),17)-(k),16)%915)
-
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_libdef.h_
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_libdef.h_ 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_libdef.h_ 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,393 +0,0 @@
-/* This is a generated file. DO NOT EDIT! */
-
-#ifdef LJLIB_MODULE_base
-#undef LJLIB_MODULE_base
-static const lua_CFunction lj_lib_cf_base[] = {
- lj_ffh_assert,
- lj_ffh_next,
- lj_ffh_pairs,
- lj_ffh_ipairs_aux,
- lj_ffh_ipairs,
- lj_ffh_setmetatable,
- lj_cf_getfenv,
- lj_cf_setfenv,
- lj_ffh_rawget,
- lj_cf_rawset,
- lj_cf_rawequal,
- lj_cf_unpack,
- lj_cf_select,
- lj_ffh_tonumber,
- lj_ffh_tostring,
- lj_cf_error,
- lj_ffh_pcall,
- lj_cf_loadfile,
- lj_cf_load,
- lj_cf_loadstring,
- lj_cf_dofile,
- lj_cf_gcinfo,
- lj_cf_collectgarbage,
- lj_cf_newproxy,
- lj_cf_print
-};
-static const uint8_t lj_lib_init_base[] = {
-2,0,28,70,97,115,115,101,114,116,195,110,105,108,199,98,111,111,108,101,97,
-110,252,1,200,117,115,101,114,100,97,116,97,198,115,116,114,105,110,103,197,
-117,112,118,97,108,198,116,104,114,101,97,100,197,112,114,111,116,111,200,102,
-117,110,99,116,105,111,110,197,116,114,97,99,101,197,99,100,97,116,97,197,116,
-97,98,108,101,252,9,198,110,117,109,98,101,114,132,116,121,112,101,68,110,101,
-120,116,253,69,112,97,105,114,115,64,253,70,105,112,97,105,114,115,140,103,
-101,116,109,101,116,97,116,97,98,108,101,76,115,101,116,109,101,116,97,116,
-97,98,108,101,7,103,101,116,102,101,110,118,7,115,101,116,102,101,110,118,70,
-114,97,119,103,101,116,6,114,97,119,115,101,116,8,114,97,119,101,113,117,97,
-108,6,117,110,112,97,99,107,6,115,101,108,101,99,116,72,116,111,110,117,109,
-98,101,114,195,110,105,108,197,102,97,108,115,101,196,116,114,117,101,72,116,
-111,115,116,114,105,110,103,5,101,114,114,111,114,69,112,99,97,108,108,134,
-120,112,99,97,108,108,8,108,111,97,100,102,105,108,101,4,108,111,97,100,10,
-108,111,97,100,115,116,114,105,110,103,6,100,111,102,105,108,101,6,103,99,105,
-110,102,111,14,99,111,108,108,101,99,116,103,97,114,98,97,103,101,252,2,8,110,
-101,119,112,114,111,120,121,200,116,111,115,116,114,105,110,103,5,112,114,105,
-110,116,252,3,200,95,86,69,82,83,73,79,78,250,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_coroutine
-#undef LJLIB_MODULE_coroutine
-static const lua_CFunction lj_lib_cf_coroutine[] = {
- lj_cf_coroutine_status,
- lj_cf_coroutine_running,
- lj_cf_coroutine_create,
- lj_ffh_coroutine_yield,
- lj_ffh_coroutine_resume,
- lj_cf_coroutine_wrap
-};
-static const uint8_t lj_lib_init_coroutine[] = {
-30,13,6,6,115,116,97,116,117,115,7,114,117,110,110,105,110,103,6,99,114,101,
-97,116,101,69,121,105,101,108,100,70,114,101,115,117,109,101,254,4,119,114,
-97,112,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_math
-#undef LJLIB_MODULE_math
-static const lua_CFunction lj_lib_cf_math[] = {
- lj_ffh_math_abs,
- lj_ffh_math_sqrt,
- lj_ffh_math_log,
- lj_ffh_math_atan2,
- lj_ffh_math_ldexp,
- lj_ffh_math_min,
- lj_cf_math_random,
- lj_cf_math_randomseed
-};
-static const uint8_t lj_lib_init_math[] = {
-37,16,30,67,97,98,115,133,102,108,111,111,114,132,99,101,105,108,68,115,113,
-114,116,133,108,111,103,49,48,131,101,120,112,131,115,105,110,131,99,111,115,
-131,116,97,110,132,97,115,105,110,132,97,99,111,115,132,97,116,97,110,132,115,
-105,110,104,132,99,111,115,104,132,116,97,110,104,133,102,114,101,120,112,132,
-109,111,100,102,251,248,193,99,26,220,165,76,64,131,100,101,103,251,57,157,
-82,162,70,223,145,63,131,114,97,100,67,108,111,103,69,97,116,97,110,50,131,
-112,111,119,132,102,109,111,100,69,108,100,101,120,112,67,109,105,110,131,109,
-97,120,251,24,45,68,84,251,33,9,64,194,112,105,250,251,0,0,0,0,0,0,240,127,
-196,104,117,103,101,250,252,2,6,114,97,110,100,111,109,252,2,10,114,97,110,
-100,111,109,115,101,101,100,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_bit
-#undef LJLIB_MODULE_bit
-static const lua_CFunction lj_lib_cf_bit[] = {
- lj_ffh_bit_tobit,
- lj_ffh_bit_lshift,
- lj_ffh_bit_band,
- lj_cf_bit_tohex
-};
-static const uint8_t lj_lib_init_bit[] = {
-65,42,12,69,116,111,98,105,116,132,98,110,111,116,133,98,115,119,97,112,70,
-108,115,104,105,102,116,134,114,115,104,105,102,116,135,97,114,115,104,105,
-102,116,131,114,111,108,131,114,111,114,68,98,97,110,100,131,98,111,114,132,
-98,120,111,114,5,116,111,104,101,120,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_string
-#undef LJLIB_MODULE_string
-static const lua_CFunction lj_lib_cf_string[] = {
- lj_ffh_string_len,
- lj_ffh_string_byte,
- lj_ffh_string_char,
- lj_ffh_string_sub,
- lj_ffh_string_rep,
- lj_ffh_string_reverse,
- lj_cf_string_dump,
- lj_cf_string_find,
- lj_cf_string_match,
- lj_cf_string_gmatch,
- lj_cf_string_gsub,
- lj_cf_string_format
-};
-static const uint8_t lj_lib_init_string[] = {
-77,53,14,67,108,101,110,68,98,121,116,101,68,99,104,97,114,67,115,117,98,67,
-114,101,112,71,114,101,118,101,114,115,101,133,108,111,119,101,114,133,117,
-112,112,101,114,4,100,117,109,112,4,102,105,110,100,5,109,97,116,99,104,254,
-6,103,109,97,116,99,104,4,103,115,117,98,6,102,111,114,109,97,116,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_table
-#undef LJLIB_MODULE_table
-static const lua_CFunction lj_lib_cf_table[] = {
- lj_cf_table_foreachi,
- lj_cf_table_foreach,
- lj_ffh_table_getn,
- lj_cf_table_maxn,
- lj_cf_table_insert,
- lj_cf_table_remove,
- lj_cf_table_concat,
- lj_cf_table_sort
-};
-static const uint8_t lj_lib_init_table[] = {
-92,61,8,8,102,111,114,101,97,99,104,105,7,102,111,114,101,97,99,104,68,103,
-101,116,110,4,109,97,120,110,6,105,110,115,101,114,116,6,114,101,109,111,118,
-101,6,99,111,110,99,97,116,4,115,111,114,116,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_io_method
-#undef LJLIB_MODULE_io_method
-static const lua_CFunction lj_lib_cf_io_method[] = {
- lj_cf_io_method_close,
- lj_cf_io_method_read,
- lj_cf_io_method_write,
- lj_cf_io_method_flush,
- lj_cf_io_method_seek,
- lj_cf_io_method_setvbuf,
- lj_cf_io_method_lines,
- lj_cf_io_method___gc,
- lj_cf_io_method___tostring
-};
-static const uint8_t lj_lib_init_io_method[] = {
-100,62,10,5,99,108,111,115,101,4,114,101,97,100,5,119,114,105,116,101,5,102,
-108,117,115,104,4,115,101,101,107,7,115,101,116,118,98,117,102,5,108,105,110,
-101,115,4,95,95,103,99,10,95,95,116,111,115,116,114,105,110,103,252,1,199,95,
-95,105,110,100,101,120,250,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_io
-#undef LJLIB_MODULE_io
-static const lua_CFunction lj_lib_cf_io[] = {
- lj_cf_io_open,
- lj_cf_io_popen,
- lj_cf_io_tmpfile,
- lj_cf_io_close,
- lj_cf_io_read,
- lj_cf_io_write,
- lj_cf_io_flush,
- lj_cf_io_input,
- lj_cf_io_output,
- lj_cf_io_lines,
- lj_cf_io_type
-};
-static const uint8_t lj_lib_init_io[] = {
-109,62,12,252,2,192,250,4,111,112,101,110,5,112,111,112,101,110,7,116,109,112,
-102,105,108,101,5,99,108,111,115,101,4,114,101,97,100,5,119,114,105,116,101,
-5,102,108,117,115,104,5,105,110,112,117,116,6,111,117,116,112,117,116,5,108,
-105,110,101,115,4,116,121,112,101,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_os
-#undef LJLIB_MODULE_os
-static const lua_CFunction lj_lib_cf_os[] = {
- lj_cf_os_execute,
- lj_cf_os_remove,
- lj_cf_os_rename,
- lj_cf_os_tmpname,
- lj_cf_os_getenv,
- lj_cf_os_exit,
- lj_cf_os_clock,
- lj_cf_os_date,
- lj_cf_os_time,
- lj_cf_os_difftime,
- lj_cf_os_setlocale
-};
-static const uint8_t lj_lib_init_os[] = {
-120,62,11,7,101,120,101,99,117,116,101,6,114,101,109,111,118,101,6,114,101,
-110,97,109,101,7,116,109,112,110,97,109,101,6,103,101,116,101,110,118,4,101,
-120,105,116,5,99,108,111,99,107,4,100,97,116,101,4,116,105,109,101,8,100,105,
-102,102,116,105,109,101,9,115,101,116,108,111,99,97,108,101,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_debug
-#undef LJLIB_MODULE_debug
-static const lua_CFunction lj_lib_cf_debug[] = {
- lj_cf_debug_getregistry,
- lj_cf_debug_getmetatable,
- lj_cf_debug_setmetatable,
- lj_cf_debug_getfenv,
- lj_cf_debug_setfenv,
- lj_cf_debug_getinfo,
- lj_cf_debug_getlocal,
- lj_cf_debug_setlocal,
- lj_cf_debug_getupvalue,
- lj_cf_debug_setupvalue,
- lj_cf_debug_upvalueid,
- lj_cf_debug_upvaluejoin,
- lj_cf_debug_sethook,
- lj_cf_debug_gethook,
- lj_cf_debug_debug,
- lj_cf_debug_traceback
-};
-static const uint8_t lj_lib_init_debug[] = {
-131,62,16,11,103,101,116,114,101,103,105,115,116,114,121,12,103,101,116,109,
-101,116,97,116,97,98,108,101,12,115,101,116,109,101,116,97,116,97,98,108,101,
-7,103,101,116,102,101,110,118,7,115,101,116,102,101,110,118,7,103,101,116,105,
-110,102,111,8,103,101,116,108,111,99,97,108,8,115,101,116,108,111,99,97,108,
-10,103,101,116,117,112,118,97,108,117,101,10,115,101,116,117,112,118,97,108,
-117,101,9,117,112,118,97,108,117,101,105,100,11,117,112,118,97,108,117,101,
-106,111,105,110,7,115,101,116,104,111,111,107,7,103,101,116,104,111,111,107,
-5,100,101,98,117,103,9,116,114,97,99,101,98,97,99,107,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_jit
-#undef LJLIB_MODULE_jit
-static const lua_CFunction lj_lib_cf_jit[] = {
- lj_cf_jit_on,
- lj_cf_jit_off,
- lj_cf_jit_flush,
- lj_cf_jit_status,
- lj_cf_jit_attach
-};
-static const uint8_t lj_lib_init_jit[] = {
-147,62,9,2,111,110,3,111,102,102,5,102,108,117,115,104,6,115,116,97,116,117,
-115,6,97,116,116,97,99,104,252,5,194,111,115,250,252,4,196,97,114,99,104,250,
-252,3,203,118,101,114,115,105,111,110,95,110,117,109,250,252,2,199,118,101,
-114,115,105,111,110,250,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_jit_util
-#undef LJLIB_MODULE_jit_util
-static const lua_CFunction lj_lib_cf_jit_util[] = {
- lj_cf_jit_util_funcinfo,
- lj_cf_jit_util_funcbc,
- lj_cf_jit_util_funck,
- lj_cf_jit_util_funcuvname,
- lj_cf_jit_util_traceinfo,
- lj_cf_jit_util_traceir,
- lj_cf_jit_util_tracek,
- lj_cf_jit_util_tracesnap,
- lj_cf_jit_util_tracemc,
- lj_cf_jit_util_traceexitstub,
- lj_cf_jit_util_ircalladdr
-};
-static const uint8_t lj_lib_init_jit_util[] = {
-152,62,11,8,102,117,110,99,105,110,102,111,6,102,117,110,99,98,99,5,102,117,
-110,99,107,10,102,117,110,99,117,118,110,97,109,101,9,116,114,97,99,101,105,
-110,102,111,7,116,114,97,99,101,105,114,6,116,114,97,99,101,107,9,116,114,97,
-99,101,115,110,97,112,7,116,114,97,99,101,109,99,13,116,114,97,99,101,101,120,
-105,116,115,116,117,98,10,105,114,99,97,108,108,97,100,100,114,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_jit_opt
-#undef LJLIB_MODULE_jit_opt
-static const lua_CFunction lj_lib_cf_jit_opt[] = {
- lj_cf_jit_opt_start
-};
-static const uint8_t lj_lib_init_jit_opt[] = {
-163,62,1,5,115,116,97,114,116,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_ffi_meta
-#undef LJLIB_MODULE_ffi_meta
-static const lua_CFunction lj_lib_cf_ffi_meta[] = {
- lj_cf_ffi_meta___index,
- lj_cf_ffi_meta___newindex,
- lj_cf_ffi_meta___eq,
- lj_cf_ffi_meta___len,
- lj_cf_ffi_meta___lt,
- lj_cf_ffi_meta___le,
- lj_cf_ffi_meta___concat,
- lj_cf_ffi_meta___call,
- lj_cf_ffi_meta___add,
- lj_cf_ffi_meta___sub,
- lj_cf_ffi_meta___mul,
- lj_cf_ffi_meta___div,
- lj_cf_ffi_meta___mod,
- lj_cf_ffi_meta___pow,
- lj_cf_ffi_meta___unm,
- lj_cf_ffi_meta___tostring,
- lj_cf_ffi_meta___pairs,
- lj_cf_ffi_meta___ipairs
-};
-static const uint8_t lj_lib_init_ffi_meta[] = {
-164,62,19,7,95,95,105,110,100,101,120,10,95,95,110,101,119,105,110,100,101,
-120,4,95,95,101,113,5,95,95,108,101,110,4,95,95,108,116,4,95,95,108,101,8,95,
-95,99,111,110,99,97,116,6,95,95,99,97,108,108,5,95,95,97,100,100,5,95,95,115,
-117,98,5,95,95,109,117,108,5,95,95,100,105,118,5,95,95,109,111,100,5,95,95,
-112,111,119,5,95,95,117,110,109,10,95,95,116,111,115,116,114,105,110,103,7,
-95,95,112,97,105,114,115,8,95,95,105,112,97,105,114,115,195,102,102,105,203,
-95,95,109,101,116,97,116,97,98,108,101,250,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_ffi_clib
-#undef LJLIB_MODULE_ffi_clib
-static const lua_CFunction lj_lib_cf_ffi_clib[] = {
- lj_cf_ffi_clib___index,
- lj_cf_ffi_clib___newindex,
- lj_cf_ffi_clib___gc
-};
-static const uint8_t lj_lib_init_ffi_clib[] = {
-182,62,3,7,95,95,105,110,100,101,120,10,95,95,110,101,119,105,110,100,101,120,
-4,95,95,103,99,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_ffi_callback
-#undef LJLIB_MODULE_ffi_callback
-static const lua_CFunction lj_lib_cf_ffi_callback[] = {
- lj_cf_ffi_callback_free,
- lj_cf_ffi_callback_set
-};
-static const uint8_t lj_lib_init_ffi_callback[] = {
-185,62,3,4,102,114,101,101,3,115,101,116,252,1,199,95,95,105,110,100,101,120,
-250,255
-};
-#endif
-
-#ifdef LJLIB_MODULE_ffi
-#undef LJLIB_MODULE_ffi
-static const lua_CFunction lj_lib_cf_ffi[] = {
- lj_cf_ffi_cdef,
- lj_cf_ffi_new,
- lj_cf_ffi_cast,
- lj_cf_ffi_typeof,
- lj_cf_ffi_istype,
- lj_cf_ffi_sizeof,
- lj_cf_ffi_alignof,
- lj_cf_ffi_offsetof,
- lj_cf_ffi_errno,
- lj_cf_ffi_string,
- lj_cf_ffi_copy,
- lj_cf_ffi_fill,
- lj_cf_ffi_abi,
- lj_cf_ffi_metatype,
- lj_cf_ffi_gc,
- lj_cf_ffi_load
-};
-static const uint8_t lj_lib_init_ffi[] = {
-187,62,22,4,99,100,101,102,3,110,101,119,4,99,97,115,116,6,116,121,112,101,
-111,102,6,105,115,116,121,112,101,6,115,105,122,101,111,102,7,97,108,105,103,
-110,111,102,8,111,102,102,115,101,116,111,102,5,101,114,114,110,111,6,115,116,
-114,105,110,103,4,99,111,112,121,4,102,105,108,108,3,97,98,105,252,8,192,250,
-8,109,101,116,97,116,121,112,101,252,7,192,250,2,103,99,252,5,192,250,4,108,
-111,97,100,252,4,193,67,250,252,3,194,111,115,250,252,2,196,97,114,99,104,250,
-255
-};
-#endif
-
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_recdef.h_
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_recdef.h_ 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/_lj_recdef.h_ 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,263 +0,0 @@
-/* This is a generated file. DO NOT EDIT! */
-
-static const uint16_t recff_idmap[] = {
-0,
-0x0100,
-0x0200,
-0x0300,
-0,
-0,
-0x0400,
-0x0500,
-0x0600,
-0x0700,
-0,
-0,
-0x0800,
-0x0900,
-0x0a00,
-0,
-0x0b00,
-0x0c00,
-0x0d00,
-0,
-0x0e00,
-0x0f00,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0x1000,
-0x1100+(IRFPM_FLOOR),
-0x1100+(IRFPM_CEIL),
-0x1200+(IRFPM_SQRT),
-0x1200+(IRFPM_LOG10),
-0x1200+(IRFPM_EXP),
-0x1200+(IRFPM_SIN),
-0x1200+(IRFPM_COS),
-0x1200+(IRFPM_TAN),
-0x1300+(FF_math_asin),
-0x1300+(FF_math_acos),
-0x1300+(FF_math_atan),
-0x1400+(IRCALL_sinh),
-0x1400+(IRCALL_cosh),
-0x1400+(IRCALL_tanh),
-0,
-0x1500,
-0x1600,
-0x1600,
-0x1700,
-0x1800,
-0x1900,
-0,
-0x1a00,
-0x1b00+(IR_MIN),
-0x1b00+(IR_MAX),
-0x1c00,
-0,
-0x1d00+(IR_TOBIT),
-0x1d00+(IR_BNOT),
-0x1d00+(IR_BSWAP),
-0x1e00+(IR_BSHL),
-0x1e00+(IR_BSHR),
-0x1e00+(IR_BSAR),
-0x1e00+(IR_BROL),
-0x1e00+(IR_BROR),
-0x1f00+(IR_BAND),
-0x1f00+(IR_BOR),
-0x1f00+(IR_BXOR),
-0,
-0x2000,
-0x2100+(0),
-0,
-0x2100+(1),
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0x2200,
-0,
-0x2300,
-0x2400,
-0,
-0,
-0,
-0,
-0x2500+(0),
-0x2600+(0),
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0x2500+(GCROOT_IO_OUTPUT),
-0x2600+(GCROOT_IO_OUTPUT),
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0,
-0x2700+(0),
-0x2700+(1),
-0x2800+(MM_eq),
-0x2800+(MM_len),
-0x2800+(MM_lt),
-0x2800+(MM_le),
-0x2800+(MM_concat),
-0x2900,
-0x2800+(MM_add),
-0x2800+(MM_sub),
-0x2800+(MM_mul),
-0x2800+(MM_div),
-0x2800+(MM_mod),
-0x2800+(MM_pow),
-0x2800+(MM_unm),
-0,
-0,
-0,
-0x2a00+(1),
-0x2a00+(0),
-0,
-0,
-0,
-0,
-0x2b00,
-0x2b00,
-0x2c00,
-0x2d00,
-0x2e00+(FF_ffi_sizeof),
-0x2e00+(FF_ffi_alignof),
-0x2e00+(FF_ffi_offsetof),
-0x2f00,
-0x3000,
-0x3100,
-0x3200,
-0x3300,
-0,
-0x3400
-};
-
-static const RecordFunc recff_func[] = {
-recff_nyi,
-recff_c,
-recff_assert,
-recff_type,
-recff_ipairs_aux,
-recff_ipairs,
-recff_getmetatable,
-recff_setmetatable,
-recff_rawget,
-recff_rawset,
-recff_rawequal,
-recff_select,
-recff_tonumber,
-recff_tostring,
-recff_pcall,
-recff_xpcall,
-recff_math_abs,
-recff_math_round,
-recff_math_unary,
-recff_math_atrig,
-recff_math_htrig,
-recff_math_modf,
-recff_math_degrad,
-recff_math_log,
-recff_math_atan2,
-recff_math_pow,
-recff_math_ldexp,
-recff_math_minmax,
-recff_math_random,
-recff_bit_unary,
-recff_bit_shift,
-recff_bit_nary,
-recff_string_len,
-recff_string_range,
-recff_table_getn,
-recff_table_insert,
-recff_table_remove,
-recff_io_write,
-recff_io_flush,
-recff_cdata_index,
-recff_cdata_arith,
-recff_cdata_call,
-recff_clib_index,
-recff_ffi_new,
-recff_ffi_typeof,
-recff_ffi_istype,
-recff_ffi_xof,
-recff_ffi_errno,
-recff_ffi_string,
-recff_ffi_copy,
-recff_ffi_fill,
-recff_ffi_abi,
-recff_ffi_gc
-};
-
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/.gitignore
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/.gitignore (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/.gitignore 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,3 @@
+minilua
+buildvm
+buildvm_arch.h
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM builder.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** This is a tool to build the hand-tuned assembler code required for
** LuaJIT's bytecode interpreter. It supports a variety of output formats
@@ -18,8 +18,10 @@
#include "lj_obj.h"
#include "lj_gc.h"
#include "lj_bc.h"
+#if LJ_HASJIT
#include "lj_ir.h"
#include "lj_ircall.h"
+#endif
#include "lj_frame.h"
#include "lj_dispatch.h"
#if LJ_HASFFI
@@ -113,7 +115,7 @@
name[0] = name[1] == 'R' ? '_' : '@'; /* Just for _RtlUnwind at 16. */
else
*p = '\0';
-#elif LJ_TARGET_PPC && !LJ_TARGET_OSX && !LJ_TARGET_CONSOLE
+#elif LJ_TARGET_PPC && !LJ_TARGET_CONSOLE
/* Keep @plt etc. */
#else
*p = '\0';
@@ -250,6 +252,7 @@
NULL
};
+#if LJ_HASJIT
const char *const ir_names[] = {
#define IRNAME(name, m, m1, m2) #name,
IRDEF(IRNAME)
@@ -290,7 +293,9 @@
#include "lj_traceerr.h"
NULL
};
+#endif
+#if LJ_HASJIT
static const char *lower(char *buf, const char *s)
{
char *p = buf;
@@ -301,6 +306,7 @@
*p = '\0';
return buf;
}
+#endif
/* Emit C source code for bytecode-related definitions. */
static void emit_bcdef(BuildCtx *ctx)
@@ -318,9 +324,12 @@
/* Emit VM definitions as Lua code for debug modules. */
static void emit_vmdef(BuildCtx *ctx)
{
+#if LJ_HASJIT
char buf[80];
+#endif
int i;
fprintf(ctx->fp, "-- This is a generated file. DO NOT EDIT!\n\n");
+ fprintf(ctx->fp, "assert(require(\"jit\").version == \"%s\", \"LuaJIT core/library version mismatch\")\n\n", LUAJIT_VERSION);
fprintf(ctx->fp, "return {\n\n");
fprintf(ctx->fp, "bcnames = \"");
@@ -327,6 +336,7 @@
for (i = 0; bc_names[i]; i++) fprintf(ctx->fp, "%-6s", bc_names[i]);
fprintf(ctx->fp, "\",\n\n");
+#if LJ_HASJIT
fprintf(ctx->fp, "irnames = \"");
for (i = 0; ir_names[i]; i++) fprintf(ctx->fp, "%-6s", ir_names[i]);
fprintf(ctx->fp, "\",\n\n");
@@ -355,6 +365,7 @@
for (i = 0; trace_errors[i]; i++)
fprintf(ctx->fp, "\"%s\",\n", trace_errors[i]);
fprintf(ctx->fp, "},\n\n");
+#endif
}
/* -- Argument parsing ---------------------------------------------------- */
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM builder.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _BUILDVM_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_asm.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_asm.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_asm.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM builder: Assembler source code emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "buildvm.h"
@@ -144,14 +144,6 @@
fprintf(ctx->fp, "\t%s %d, %d, " TOCPREFIX "%s\n",
(ins & 1) ? "bcl" : "bc", (ins >> 21) & 31, (ins >> 16) & 31, sym);
} else if ((ins >> 26) == 18) {
-#if LJ_ARCH_PPC64
- const char *suffix = strchr(sym, '@');
- if (suffix && suffix[1] == 'h') {
- fprintf(ctx->fp, "\taddis 11, 2, %s\n", sym);
- } else if (suffix && suffix[1] == 'l') {
- fprintf(ctx->fp, "\tld 12, %s\n", sym);
- } else
-#endif
fprintf(ctx->fp, "\t%s " TOCPREFIX "%s\n", (ins & 1) ? "bl" : "b", sym);
} else {
fprintf(stderr,
@@ -250,10 +242,13 @@
int i, rel;
fprintf(ctx->fp, "\t.file \"buildvm_%s.dasc\"\n", ctx->dasm_arch);
-#if LJ_ARCH_PPC64
- fprintf(ctx->fp, "\t.abiversion 2\n");
+ fprintf(ctx->fp, "\t.text\n");
+#if LJ_TARGET_MIPS32 && !LJ_ABI_SOFTFP
+ fprintf(ctx->fp, "\t.module fp=32\n");
#endif
- fprintf(ctx->fp, "\t.text\n");
+#if LJ_TARGET_MIPS
+ fprintf(ctx->fp, "\t.set nomips16\n\t.abicalls\n\t.set noreorder\n\t.set nomacro\n");
+#endif
emit_asm_align(ctx, 4);
#if LJ_TARGET_PS3
@@ -280,9 +275,6 @@
".pad #28\n");
#endif
#endif
-#if LJ_TARGET_MIPS
- fprintf(ctx->fp, ".set nomips16\n.abicalls\n.set noreorder\n.set nomacro\n");
-#endif
for (i = rel = 0; i < ctx->nsym; i++) {
int32_t ofs = ctx->sym[i].ofs;
@@ -338,7 +330,7 @@
#if !(LJ_TARGET_PS3 || LJ_TARGET_PSVITA)
fprintf(ctx->fp, "\t.section .note.GNU-stack,\"\"," ELFASM_PX "progbits\n");
#endif
-#if LJ_TARGET_PPC && !LJ_TARGET_PS3
+#if LJ_TARGET_PPC && !LJ_TARGET_PS3 && !LJ_ABI_SOFTFP
/* Hard-float ABI. */
fprintf(ctx->fp, "\t.gnu_attribute 4, 1\n");
#endif
@@ -347,6 +339,10 @@
fprintf(ctx->fp, "\t.ident \"%s\"\n", ctx->dasm_ident);
break;
case BUILD_machasm:
+#if defined(__apple_build_version__) && __apple_build_version__ >= 15000000 && __apple_build_version__ < 15000300
+ /* Workaround for XCode 15.0 - 15.2. */
+ fprintf(ctx->fp, "\t.subsections_via_symbols\n");
+#endif
fprintf(ctx->fp,
"\t.cstring\n"
"\t.ascii \"%s\\0\"\n", ctx->dasm_ident);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_fold.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_fold.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_fold.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,10 +1,11 @@
/*
** LuaJIT VM builder: IR folding hash table generator.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "buildvm.h"
#include "lj_obj.h"
+#if LJ_HASJIT
#include "lj_ir.h"
/* Context for the folding hash table generator. */
@@ -226,4 +227,10 @@
makehash(ctx);
}
+#else
+void emit_fold(BuildCtx *ctx)
+{
+ UNUSED(ctx);
+}
+#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_lib.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_lib.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_lib.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM builder: library definition compiler.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "buildvm.h"
@@ -379,12 +379,21 @@
/* Simplistic pre-processor. Only handles top-level #if/#endif. */
if (buf[0] == '#' && buf[1] == 'i' && buf[2] == 'f') {
int ok = 1;
- if (!strcmp(buf, "#if LJ_52\n"))
+ size_t len = strlen(buf);
+ if (buf[len-1] == '\n') {
+ buf[len-1] = 0;
+ if (buf[len-2] == '\r') {
+ buf[len-2] = 0;
+ }
+ }
+ if (!strcmp(buf, "#if LJ_52"))
ok = LJ_52;
- else if (!strcmp(buf, "#if LJ_HASJIT\n"))
+ else if (!strcmp(buf, "#if LJ_HASJIT"))
ok = LJ_HASJIT;
- else if (!strcmp(buf, "#if LJ_HASFFI\n"))
+ else if (!strcmp(buf, "#if LJ_HASFFI"))
ok = LJ_HASFFI;
+ else if (!strcmp(buf, "#if LJ_HASBUFFER"))
+ ok = LJ_HASBUFFER;
if (!ok) {
int lvl = 1;
while (fgets(buf, sizeof(buf), fp) != NULL) {
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_libbc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_libbc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_libbc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -4,42 +4,67 @@
static const uint8_t libbc_code[] = {
#if LJ_FR2
-0,1,2,0,0,1,2,24,1,0,0,76,1,2,0,241,135,158,166,3,220,203,178,130,4,0,1,2,0,
-0,1,2,24,1,0,0,76,1,2,0,243,244,148,165,20,198,190,199,252,3,0,1,2,0,0,0,3,
-16,0,5,0,21,1,0,0,76,1,2,0,0,2,10,0,0,0,15,16,0,12,0,16,1,9,0,41,2,1,0,21,3,
-0,0,41,4,1,0,77,2,8,128,18,6,1,0,18,8,5,0,59,9,5,0,66,6,3,2,10,6,0,0,88,7,1,
-128,76,6,2,0,79,2,248,127,75,0,1,0,0,2,11,0,0,0,16,16,0,12,0,16,1,9,0,43,2,
-0,0,18,3,0,0,41,4,0,0,88,5,7,128,18,7,1,0,18,9,5,0,18,10,6,0,66,7,3,2,10,7,
-0,0,88,8,1,128,76,7,2,0,70,5,3,3,82,5,247,127,75,0,1,0,0,1,2,0,0,0,3,16,0,12,
-0,21,1,0,0,76,1,2,0,0,2,10,0,0,2,30,16,0,12,0,21,2,0,0,11,1,0,0,88,3,7,128,
-8,2,0,0,88,3,23,128,59,3,2,0,43,4,0,0,64,4,2,0,76,3,2,0,88,3,18,128,16,1,14,
-0,41,3,1,0,3,3,1,0,88,3,14,128,3,1,2,0,88,3,12,128,59,3,1,0,22,4,1,1,18,5,2,
-0,41,6,1,0,77,4,4,128,23,8,1,7,59,9,7,0,64,9,8,0,79,4,252,127,43,4,0,0,64,4,
-2,0,76,3,2,0,75,0,1,0,0,2,0,5,12,0,0,0,35,16,0,12,0,16,1,14,0,16,2,14,0,16,
-3,14,0,11,4,0,0,88,5,1,128,18,4,0,0,16,4,12,0,3,1,2,0,88,5,24,128,33,5,1,3,
-0,2,3,0,88,6,4,128,2,3,1,0,88,6,2,128,4,4,0,0,88,6,9,128,18,6,1,0,18,7,2,0,
-41,8,1,0,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,6,252,127,88,6,8,128,
-18,6,2,0,18,7,1,0,41,8,255,255,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,
-6,252,127,76,4,2,0,0
+/* math.deg */ 0,1,2,0,0,1,2,BC_MULVN,1,0,0,BC_RET1,1,2,0,241,135,158,166,3,
+220,203,178,130,4,
+/* math.rad */ 0,1,2,0,0,1,2,BC_MULVN,1,0,0,BC_RET1,1,2,0,243,244,148,165,20,
+198,190,199,252,3,
+/* string.len */ 0,1,2,0,0,0,3,BC_ISTYPE,0,5,0,BC_LEN,1,0,0,BC_RET1,1,2,0,
+/* table.foreachi */ 0,2,10,0,0,0,15,BC_ISTYPE,0,12,0,BC_ISTYPE,1,9,0,
+BC_KSHORT,2,1,0,BC_LEN,3,0,0,BC_KSHORT,4,1,0,BC_FORI,2,8,128,BC_MOV,6,1,0,
+BC_MOV,8,5,0,BC_TGETR,9,5,0,BC_CALL,6,3,2,BC_ISEQP,6,0,0,BC_JMP,7,1,128,
+BC_RET1,6,2,0,BC_FORL,2,248,127,BC_RET0,0,1,0,
+/* table.foreach */ 0,2,11,0,0,1,16,BC_ISTYPE,0,12,0,BC_ISTYPE,1,9,0,BC_KPRI,
+2,0,0,BC_MOV,3,0,0,BC_KNUM,4,0,0,BC_JMP,5,7,128,BC_MOV,7,1,0,BC_MOV,9,5,0,
+BC_MOV,10,6,0,BC_CALL,7,3,2,BC_ISEQP,7,0,0,BC_JMP,8,1,128,BC_RET1,7,2,0,
+BC_ITERN,5,3,3,BC_ITERL,5,247,127,BC_RET0,0,1,0,1,255,255,249,255,15,
+/* table.getn */ 0,1,2,0,0,0,3,BC_ISTYPE,0,12,0,BC_LEN,1,0,0,BC_RET1,1,2,0,
+/* table.remove */ 0,2,10,0,0,2,30,BC_ISTYPE,0,12,0,BC_LEN,2,0,0,BC_ISNEP,1,0,
+0,BC_JMP,3,7,128,BC_ISEQN,2,0,0,BC_JMP,3,23,128,BC_TGETR,3,2,0,BC_KPRI,4,0,0,
+BC_TSETR,4,2,0,BC_RET1,3,2,0,BC_JMP,3,18,128,BC_ISTYPE,1,14,0,BC_KSHORT,3,1,0,
+BC_ISGT,3,1,0,BC_JMP,3,14,128,BC_ISGT,1,2,0,BC_JMP,3,12,128,BC_TGETR,3,1,0,
+BC_ADDVN,4,1,1,BC_MOV,5,2,0,BC_KSHORT,6,1,0,BC_FORI,4,4,128,BC_SUBVN,8,1,7,
+BC_TGETR,9,7,0,BC_TSETR,9,8,0,BC_FORL,4,252,127,BC_KPRI,4,0,0,BC_TSETR,4,2,0,
+BC_RET1,3,2,0,BC_RET0,0,1,0,0,2,
+/* table.move */ 0,5,12,0,0,0,35,BC_ISTYPE,0,12,0,BC_ISTYPE,1,14,0,BC_ISTYPE,
+2,14,0,BC_ISTYPE,3,14,0,BC_ISNEP,4,0,0,BC_JMP,5,1,128,BC_MOV,4,0,0,BC_ISTYPE,
+4,12,0,BC_ISGT,1,2,0,BC_JMP,5,24,128,BC_SUBVV,5,1,3,BC_ISLT,2,3,0,BC_JMP,6,4,
+128,BC_ISLE,3,1,0,BC_JMP,6,2,128,BC_ISEQV,4,0,0,BC_JMP,6,9,128,BC_MOV,6,1,0,
+BC_MOV,7,2,0,BC_KSHORT,8,1,0,BC_FORI,6,4,128,BC_ADDVV,10,5,9,BC_TGETR,11,9,0,
+BC_TSETR,11,10,4,BC_FORL,6,252,127,BC_JMP,6,8,128,BC_MOV,6,2,0,BC_MOV,7,1,0,
+BC_KSHORT,8,255,255,BC_FORI,6,4,128,BC_ADDVV,10,5,9,BC_TGETR,11,9,0,BC_TSETR,
+11,10,4,BC_FORL,6,252,127,BC_RET1,4,2,0,
#else
-0,1,2,0,0,1,2,24,1,0,0,76,1,2,0,241,135,158,166,3,220,203,178,130,4,0,1,2,0,
-0,1,2,24,1,0,0,76,1,2,0,243,244,148,165,20,198,190,199,252,3,0,1,2,0,0,0,3,
-16,0,5,0,21,1,0,0,76,1,2,0,0,2,9,0,0,0,15,16,0,12,0,16,1,9,0,41,2,1,0,21,3,
-0,0,41,4,1,0,77,2,8,128,18,6,1,0,18,7,5,0,59,8,5,0,66,6,3,2,10,6,0,0,88,7,1,
-128,76,6,2,0,79,2,248,127,75,0,1,0,0,2,10,0,0,0,16,16,0,12,0,16,1,9,0,43,2,
-0,0,18,3,0,0,41,4,0,0,88,5,7,128,18,7,1,0,18,8,5,0,18,9,6,0,66,7,3,2,10,7,0,
-0,88,8,1,128,76,7,2,0,70,5,3,3,82,5,247,127,75,0,1,0,0,1,2,0,0,0,3,16,0,12,
-0,21,1,0,0,76,1,2,0,0,2,10,0,0,2,30,16,0,12,0,21,2,0,0,11,1,0,0,88,3,7,128,
-8,2,0,0,88,3,23,128,59,3,2,0,43,4,0,0,64,4,2,0,76,3,2,0,88,3,18,128,16,1,14,
-0,41,3,1,0,3,3,1,0,88,3,14,128,3,1,2,0,88,3,12,128,59,3,1,0,22,4,1,1,18,5,2,
-0,41,6,1,0,77,4,4,128,23,8,1,7,59,9,7,0,64,9,8,0,79,4,252,127,43,4,0,0,64,4,
-2,0,76,3,2,0,75,0,1,0,0,2,0,5,12,0,0,0,35,16,0,12,0,16,1,14,0,16,2,14,0,16,
-3,14,0,11,4,0,0,88,5,1,128,18,4,0,0,16,4,12,0,3,1,2,0,88,5,24,128,33,5,1,3,
-0,2,3,0,88,6,4,128,2,3,1,0,88,6,2,128,4,4,0,0,88,6,9,128,18,6,1,0,18,7,2,0,
-41,8,1,0,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,6,252,127,88,6,8,128,
-18,6,2,0,18,7,1,0,41,8,255,255,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,
-6,252,127,76,4,2,0,0
+/* math.deg */ 0,1,2,0,0,1,2,BC_MULVN,1,0,0,BC_RET1,1,2,0,241,135,158,166,3,
+220,203,178,130,4,
+/* math.rad */ 0,1,2,0,0,1,2,BC_MULVN,1,0,0,BC_RET1,1,2,0,243,244,148,165,20,
+198,190,199,252,3,
+/* string.len */ 0,1,2,0,0,0,3,BC_ISTYPE,0,5,0,BC_LEN,1,0,0,BC_RET1,1,2,0,
+/* table.foreachi */ 0,2,9,0,0,0,15,BC_ISTYPE,0,12,0,BC_ISTYPE,1,9,0,
+BC_KSHORT,2,1,0,BC_LEN,3,0,0,BC_KSHORT,4,1,0,BC_FORI,2,8,128,BC_MOV,6,1,0,
+BC_MOV,7,5,0,BC_TGETR,8,5,0,BC_CALL,6,3,2,BC_ISEQP,6,0,0,BC_JMP,7,1,128,
+BC_RET1,6,2,0,BC_FORL,2,248,127,BC_RET0,0,1,0,
+/* table.foreach */ 0,2,10,0,0,1,16,BC_ISTYPE,0,12,0,BC_ISTYPE,1,9,0,BC_KPRI,
+2,0,0,BC_MOV,3,0,0,BC_KNUM,4,0,0,BC_JMP,5,7,128,BC_MOV,7,1,0,BC_MOV,8,5,0,
+BC_MOV,9,6,0,BC_CALL,7,3,2,BC_ISEQP,7,0,0,BC_JMP,8,1,128,BC_RET1,7,2,0,
+BC_ITERN,5,3,3,BC_ITERL,5,247,127,BC_RET0,0,1,0,1,255,255,249,255,15,
+/* table.getn */ 0,1,2,0,0,0,3,BC_ISTYPE,0,12,0,BC_LEN,1,0,0,BC_RET1,1,2,0,
+/* table.remove */ 0,2,10,0,0,2,30,BC_ISTYPE,0,12,0,BC_LEN,2,0,0,BC_ISNEP,1,0,
+0,BC_JMP,3,7,128,BC_ISEQN,2,0,0,BC_JMP,3,23,128,BC_TGETR,3,2,0,BC_KPRI,4,0,0,
+BC_TSETR,4,2,0,BC_RET1,3,2,0,BC_JMP,3,18,128,BC_ISTYPE,1,14,0,BC_KSHORT,3,1,0,
+BC_ISGT,3,1,0,BC_JMP,3,14,128,BC_ISGT,1,2,0,BC_JMP,3,12,128,BC_TGETR,3,1,0,
+BC_ADDVN,4,1,1,BC_MOV,5,2,0,BC_KSHORT,6,1,0,BC_FORI,4,4,128,BC_SUBVN,8,1,7,
+BC_TGETR,9,7,0,BC_TSETR,9,8,0,BC_FORL,4,252,127,BC_KPRI,4,0,0,BC_TSETR,4,2,0,
+BC_RET1,3,2,0,BC_RET0,0,1,0,0,2,
+/* table.move */ 0,5,12,0,0,0,35,BC_ISTYPE,0,12,0,BC_ISTYPE,1,14,0,BC_ISTYPE,
+2,14,0,BC_ISTYPE,3,14,0,BC_ISNEP,4,0,0,BC_JMP,5,1,128,BC_MOV,4,0,0,BC_ISTYPE,
+4,12,0,BC_ISGT,1,2,0,BC_JMP,5,24,128,BC_SUBVV,5,1,3,BC_ISLT,2,3,0,BC_JMP,6,4,
+128,BC_ISLE,3,1,0,BC_JMP,6,2,128,BC_ISEQV,4,0,0,BC_JMP,6,9,128,BC_MOV,6,1,0,
+BC_MOV,7,2,0,BC_KSHORT,8,1,0,BC_FORI,6,4,128,BC_ADDVV,10,5,9,BC_TGETR,11,9,0,
+BC_TSETR,11,10,4,BC_FORL,6,252,127,BC_JMP,6,8,128,BC_MOV,6,2,0,BC_MOV,7,1,0,
+BC_KSHORT,8,255,255,BC_FORI,6,4,128,BC_ADDVV,10,5,9,BC_TGETR,11,9,0,BC_TSETR,
+11,10,4,BC_FORL,6,252,127,BC_RET1,4,2,0,
#endif
+0
};
static const struct { const char *name; int ofs; } libbc_map[] = {
@@ -48,9 +73,9 @@
{"string_len",50},
{"table_foreachi",69},
{"table_foreach",136},
-{"table_getn",207},
-{"table_remove",226},
-{"table_move",355},
-{NULL,502}
+{"table_getn",213},
+{"table_remove",232},
+{"table_move",361},
+{NULL,508}
};
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_peobj.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_peobj.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/buildvm_peobj.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM builder: PE object emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Only used for building on Windows, since we cannot assume the presence
** of a suitable assembler. The host and target byte order must match.
@@ -9,7 +9,7 @@
#include "buildvm.h"
#include "lj_bc.h"
-#if LJ_TARGET_X86ORX64 || LJ_TARGET_PPC
+#if LJ_TARGET_WINDOWS || LJ_TARGET_CYGWIN
/* Context for PE object emitter. */
static char *strtab;
@@ -93,12 +93,17 @@
#define PEOBJ_RELOC_ADDR32NB 0x03
#define PEOBJ_RELOC_OFS 0
#define PEOBJ_TEXT_FLAGS 0x60500020 /* 60=r+x, 50=align16, 20=code. */
-#elif LJ_TARGET_PPC
-#define PEOBJ_ARCH_TARGET 0x01f2
-#define PEOBJ_RELOC_REL32 0x06
-#define PEOBJ_RELOC_DIR32 0x02
+#define PEOBJ_PDATA_NRELOC 6
+#define PEOBJ_XDATA_SIZE (8*2+4+6*2)
+#elif LJ_TARGET_ARM64
+#define PEOBJ_ARCH_TARGET 0xaa64
+#define PEOBJ_RELOC_REL32 0x03 /* MS: BRANCH26. */
+#define PEOBJ_RELOC_DIR32 0x01
+#define PEOBJ_RELOC_ADDR32NB 0x02
#define PEOBJ_RELOC_OFS (-4)
-#define PEOBJ_TEXT_FLAGS 0x60400020 /* 60=r+x, 40=align8, 20=code. */
+#define PEOBJ_TEXT_FLAGS 0x60500020 /* 60=r+x, 50=align16, 20=code. */
+#define PEOBJ_PDATA_NRELOC 4
+#define PEOBJ_XDATA_SIZE (4+24+4 +4+8)
#endif
/* Section numbers (0-based). */
@@ -106,7 +111,7 @@
PEOBJ_SECT_ABS = -2,
PEOBJ_SECT_UNDEF = -1,
PEOBJ_SECT_TEXT,
-#if LJ_TARGET_X64
+#ifdef PEOBJ_PDATA_NRELOC
PEOBJ_SECT_PDATA,
PEOBJ_SECT_XDATA,
#elif LJ_TARGET_X86
@@ -181,6 +186,9 @@
uint32_t sofs;
int i, nrsym;
union { uint8_t b; uint32_t u; } host_endian;
+#ifdef PEOBJ_PDATA_NRELOC
+ uint32_t fcofs = (uint32_t)ctx->sym[ctx->nsym-1].ofs;
+#endif
sofs = sizeof(PEheader) + PEOBJ_NSECTIONS*sizeof(PEsection);
@@ -194,18 +202,18 @@
/* Flags: 60 = read+execute, 50 = align16, 20 = code. */
pesect[PEOBJ_SECT_TEXT].flags = PEOBJ_TEXT_FLAGS;
-#if LJ_TARGET_X64
+#ifdef PEOBJ_PDATA_NRELOC
memcpy(pesect[PEOBJ_SECT_PDATA].name, ".pdata", sizeof(".pdata")-1);
pesect[PEOBJ_SECT_PDATA].ofs = sofs;
- sofs += (pesect[PEOBJ_SECT_PDATA].size = 6*4);
+ sofs += (pesect[PEOBJ_SECT_PDATA].size = PEOBJ_PDATA_NRELOC*4);
pesect[PEOBJ_SECT_PDATA].relocofs = sofs;
- sofs += (pesect[PEOBJ_SECT_PDATA].nreloc = 6) * PEOBJ_RELOC_SIZE;
+ sofs += (pesect[PEOBJ_SECT_PDATA].nreloc = PEOBJ_PDATA_NRELOC) * PEOBJ_RELOC_SIZE;
/* Flags: 40 = read, 30 = align4, 40 = initialized data. */
pesect[PEOBJ_SECT_PDATA].flags = 0x40300040;
memcpy(pesect[PEOBJ_SECT_XDATA].name, ".xdata", sizeof(".xdata")-1);
pesect[PEOBJ_SECT_XDATA].ofs = sofs;
- sofs += (pesect[PEOBJ_SECT_XDATA].size = 8*2+4+6*2); /* See below. */
+ sofs += (pesect[PEOBJ_SECT_XDATA].size = PEOBJ_XDATA_SIZE); /* See below. */
pesect[PEOBJ_SECT_XDATA].relocofs = sofs;
sofs += (pesect[PEOBJ_SECT_XDATA].nreloc = 1) * PEOBJ_RELOC_SIZE;
/* Flags: 40 = read, 30 = align4, 40 = initialized data. */
@@ -240,7 +248,7 @@
*/
nrsym = ctx->nrelocsym;
pehdr.nsyms = 1+PEOBJ_NSECTIONS*2 + 1+ctx->nsym + nrsym;
-#if LJ_TARGET_X64
+#ifdef PEOBJ_PDATA_NRELOC
pehdr.nsyms += 1; /* Symbol for lj_err_unwind_win. */
#endif
@@ -251,15 +259,8 @@
/* Write .text section. */
host_endian.u = 1;
if (host_endian.b != LJ_ENDIAN_SELECT(1, 0)) {
-#if LJ_TARGET_PPC
- uint32_t *p = (uint32_t *)ctx->code;
- int n = (int)(ctx->codesz >> 2);
- for (i = 0; i < n; i++, p++)
- *p = lj_bswap(*p); /* Byteswap .text section. */
-#else
fprintf(stderr, "Error: different byte order for host and target\n");
exit(1);
-#endif
}
owrite(ctx, ctx->code, ctx->codesz);
for (i = 0; i < ctx->nreloc; i++) {
@@ -272,7 +273,6 @@
#if LJ_TARGET_X64
{ /* Write .pdata section. */
- uint32_t fcofs = (uint32_t)ctx->sym[ctx->nsym-1].ofs;
uint32_t pdata[3]; /* Start of .text, end of .text and .xdata. */
PEreloc reloc;
pdata[0] = 0; pdata[1] = fcofs; pdata[2] = 0;
@@ -321,6 +321,87 @@
reloc.type = PEOBJ_RELOC_ADDR32NB;
owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
}
+#elif LJ_TARGET_ARM64
+ /* https://learn.microsoft.com/en-us/cpp/build/arm64-exception-handling */
+ { /* Write .pdata section. */
+ uint32_t pdata[4];
+ PEreloc reloc;
+ pdata[0] = 0;
+ pdata[1] = 0;
+ pdata[2] = fcofs;
+ pdata[3] = 4+24+4;
+ owrite(ctx, &pdata, sizeof(pdata));
+ /* Start of .text and start of .xdata. */
+ reloc.vaddr = 0; reloc.symidx = 1+2+nrsym+2+2+1;
+ reloc.type = PEOBJ_RELOC_ADDR32NB;
+ owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
+ reloc.vaddr = 4; reloc.symidx = 1+2+nrsym+2;
+ reloc.type = PEOBJ_RELOC_ADDR32NB;
+ owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
+ /* Start of vm_ffi_call and start of second part of .xdata. */
+ reloc.vaddr = 8; reloc.symidx = 1+2+nrsym+2+2+1;
+ reloc.type = PEOBJ_RELOC_ADDR32NB;
+ owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
+ reloc.vaddr = 12; reloc.symidx = 1+2+nrsym+2;
+ reloc.type = PEOBJ_RELOC_ADDR32NB;
+ owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
+ }
+ { /* Write .xdata section. */
+ uint32_t u32;
+ uint8_t *p, uwc[24];
+ PEreloc reloc;
+
+#define CBE16(x) (*p = ((x) >> 8) & 0xff, p[1] = (x) & 0xff, p += 2)
+#define CALLOC_S(s) (*p++ = ((s) >> 4)) /* s < 512 */
+#define CSAVE_FPLR(o) (*p++ = 0x40 | ((o) >> 3)) /* o <= 504 */
+#define CSAVE_REGP(r,o) CBE16(0xc800 | (((r) - 19) << 6) | ((o) >> 3))
+#define CSAVE_REGS(r1,r2,o1) do { \
+ int r, o; for (r = r1, o = o1; r <= r2; r += 2, o -= 16) CSAVE_REGP(r, o); \
+} while (0)
+#define CSAVE_REGPX(r,o) CBE16(0xcc00 | (((r) - 19) << 6) | (~(o) >> 3))
+#define CSAVE_FREGP(r,o) CBE16(0xd800 | (((r) - 8) << 6) | ((o) >> 3))
+#define CSAVE_FREGS(r1,r2,o1) do { \
+ int r, o; for (r = r1, o = o1; r <= r2; r += 2, o -= 16) CSAVE_FREGP(r, o); \
+} while (0)
+#define CADD_FP(s) CBE16(0xe200 | ((s) >> 3)) /* s < 8*256 */
+#define CODE_NOP 0xe3
+#define CODE_END 0xe4
+#define CEND_ALIGN do { \
+ *p++ = CODE_END; \
+ while ((p - uwc) & 3) *p++ = CODE_NOP; \
+} while (0)
+
+ /* Unwind codes for .text section with handler. */
+ p = uwc;
+ CADD_FP(192); /* +2 */
+ CSAVE_REGS(19, 28, 176); /* +5*2 */
+ CSAVE_FREGS(8, 15, 96); /* +4*2 */
+ CSAVE_FPLR(192); /* +1 */
+ CALLOC_S(208); /* +1 */
+ CEND_ALIGN; /* +1 +1 -> 24 */
+
+ u32 = ((24u >> 2) << 27) | (1u << 20) | (fcofs >> 2);
+ owrite(ctx, &u32, 4);
+ owrite(ctx, &uwc, 24);
+
+ u32 = 0; /* Handler RVA to be relocated at 4 + 24. */
+ owrite(ctx, &u32, 4);
+
+ /* Unwind codes for vm_ffi_call without handler. */
+ p = uwc;
+ CADD_FP(16); /* +2 */
+ CSAVE_FPLR(16); /* +1 */
+ CSAVE_REGPX(19, -32); /* +2 */
+ CEND_ALIGN; /* +1 +2 -> 8 */
+
+ u32 = ((8u >> 2) << 27) | (((uint32_t)ctx->codesz - fcofs) >> 2);
+ owrite(ctx, &u32, 4);
+ owrite(ctx, &uwc, 8);
+
+ reloc.vaddr = 4 + 24; reloc.symidx = 1+2+nrsym+2+2;
+ reloc.type = PEOBJ_RELOC_ADDR32NB;
+ owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
+ }
#elif LJ_TARGET_X86
/* Write .sxdata section. */
for (i = 0; i < nrsym; i++) {
@@ -352,7 +433,7 @@
emit_peobj_sym(ctx, ctx->relocsym[i], 0,
PEOBJ_SECT_UNDEF, PEOBJ_TYPE_FUNC, PEOBJ_SCL_EXTERN);
-#if LJ_TARGET_X64
+#ifdef PEOBJ_PDATA_NRELOC
emit_peobj_sym_sect(ctx, pesect, PEOBJ_SECT_PDATA);
emit_peobj_sym_sect(ctx, pesect, PEOBJ_SECT_XDATA);
emit_peobj_sym(ctx, "lj_err_unwind_win", 0,
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genlibbc.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genlibbc.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genlibbc.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -2,7 +2,7 @@
-- Lua script to dump the bytecode of the library functions written in Lua.
-- The resulting 'buildvm_libbc.h' is used for the build process of LuaJIT.
----------------------------------------------------------------------------
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
@@ -55,7 +55,7 @@
end)
code = string.gsub(code, "PAIRS%((.-)%)", function(var)
fixup.PAIRS = true
- return format("nil, %s, 0", var)
+ return format("nil, %s, 0x4dp80", var)
end)
return "return "..code, fixup
end
@@ -79,9 +79,11 @@
str = 5, func = 9, tab = 12, int = 14, num = 15
}
-local BC = {}
+local BC, BCN = {}, {}
for i=0,#bcnames/6-1 do
- BC[string.gsub(string.sub(bcnames, i*6+1, i*6+6), " ", "")] = i
+ local name = bcnames:sub(i*6+1, i*6+6):gsub(" ", "")
+ BC[name] = i
+ BCN[i] = name
end
local xop, xra = isbe and 3 or 0, isbe and 2 or 1
local xrc, xrb = isbe and 1 or 2, isbe and 0 or 3
@@ -96,6 +98,7 @@
p = read_uleb128(p)
p = read_uleb128(p)
p, sizebc = read_uleb128(p)
+ local startbc = tonumber(p - start)
local rawtab = {}
for i=0,sizebc-1 do
local op = p[xop]
@@ -129,46 +132,79 @@
end
p = p + 4
end
- return ffi.string(start, n)
+ local ndump = ffi.string(start, n)
+ -- Fixup hi-part of 0x4dp80 to LJ_KEYINDEX.
+ ndump = ndump:gsub("\x80\x80\xcd\xaa\x04", "\xff\xff\xf9\xff\x0f")
+ return { dump = ndump, startbc = startbc, sizebc = sizebc }
end
-local function find_defs(src)
+local function find_defs(src, mode)
local defs = {}
for name, code in string.gmatch(src, "LJLIB_LUA%(([^)]*)%)%s*/%*(.-)%*/") do
- local env = {}
local tcode, fixup = transform_lua(code)
- local func = assert(load(tcode, "", nil, env))()
- defs[name] = fixup_dump(string.dump(func, true), fixup)
+ local func = assert(load(tcode, "", mode))
+ defs[name] = fixup_dump(string.dump(func, mode), fixup)
defs[#defs+1] = name
end
return defs
end
-local function gen_header(defs)
+local function gen_header(defs32, defs64)
local t = {}
local function w(x) t[#t+1] = x end
w("/* This is a generated file. DO NOT EDIT! */\n\n")
w("static const int libbc_endian = ") w(isbe and 1 or 0) w(";\n\n")
- local s = ""
- for _,name in ipairs(defs) do
- s = s .. defs[name]
+ for j,defs in ipairs{defs64, defs32} do
+ local s, sb = "", ""
+ for i,name in ipairs(defs) do
+ local d = defs[name]
+ s = s .. d.dump
+ sb = sb .. string.char(i) .. ("\0"):rep(d.startbc - 1)
+ .. (isbe and "\0\0\0\255" or "\255\0\0\0"):rep(d.sizebc)
+ .. ("\0"):rep(#d.dump - d.startbc - d.sizebc*4)
+ end
+ if j == 1 then
+ w("static const uint8_t libbc_code[] = {\n#if LJ_FR2\n")
+ else
+ w("\n#else\n")
+ end
+ local n = 0
+ for i=1,#s do
+ local x = string.byte(s, i)
+ local xb = string.byte(sb, i)
+ if xb == 255 then
+ local name = BCN[x]
+ local m = #name + 4
+ if n + m > 78 then n = 0; w("\n") end
+ n = n + m
+ w("BC_"); w(name)
+ else
+ local m = x < 10 and 2 or (x < 100 and 3 or 4)
+ if xb == 0 then
+ if n + m > 78 then n = 0; w("\n") end
+ else
+ local name = defs[xb]:gsub("_", ".")
+ if n ~= 0 then w("\n") end
+ w("/* "); w(name); w(" */ ")
+ n = #name + 7
+ end
+ n = n + m
+ w(x)
+ end
+ w(",")
+ end
end
- w("static const uint8_t libbc_code[] = {\n")
- local n = 0
- for i=1,#s do
- local x = string.byte(s, i)
- w(x); w(",")
- n = n + (x < 10 and 2 or (x < 100 and 3 or 4))
- if n >= 75 then n = 0; w("\n") end
- end
- w("0\n};\n\n")
+ w("\n#endif\n0\n};\n\n")
w("static const struct { const char *name; int ofs; } libbc_map[] = {\n")
- local m = 0
- for _,name in ipairs(defs) do
- w('{"'); w(name); w('",'); w(m) w('},\n')
- m = m + #defs[name]
+ local m32, m64 = 0, 0
+ for i,name in ipairs(defs32) do
+ assert(name == defs64[i])
+ w('{"'); w(name); w('",'); w(m32) w('},\n')
+ m32 = m32 + #defs32[name].dump
+ m64 = m64 + #defs64[name].dump
+ assert(m32 == m64)
end
- w("{NULL,"); w(m); w("}\n};\n\n")
+ w("{NULL,"); w(m32); w("}\n};\n\n")
return table.concat(t)
end
@@ -191,7 +227,8 @@
local outfile = parse_arg(arg)
local src = read_files(arg)
-local defs = find_defs(src)
-local hdr = gen_header(defs)
+local defs32 = find_defs(src, "Wdts")
+local defs64 = find_defs(src, "Xdts")
+local hdr = gen_header(defs32, defs64)
write_file(outfile, hdr)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genminilua.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genminilua.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genminilua.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -2,7 +2,7 @@
-- Lua script to generate a customized, minified version of Lua.
-- The resulting 'minilua' is used for the build process of LuaJIT.
----------------------------------------------------------------------------
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
@@ -327,6 +327,12 @@
return gsub(src, "ZY([%w_]+)", "union %1")
end
+local function fix_bugs_and_warnings(src)
+ src = gsub(src, "(luaD_checkstack%(L,p%->maxstacksize)%)", "%1+p->numparams)")
+ src = gsub(src, "if%(sep==%-1%)(return'%[';)\nelse (luaX_lexerror%b();)", "if (sep!=-1)%2\n%1")
+ return gsub(src, "(default:{\nNode%*n=mainposition)", "/*fallthrough*/\n%1")
+end
+
local function func_gather(src)
local nodes, list = {}, {}
local pos, len = 1, #src
@@ -425,5 +431,6 @@
src = func_collect(src)
src = rename_tokens2(src)
src = restore_strings(src)
+src = fix_bugs_and_warnings(src)
src = merge_header(src, license)
io.write(src)
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genversion.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genversion.lua (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genversion.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,45 @@
+----------------------------------------------------------------------------
+-- Lua script to embed the rolling release version in luajit.h.
+----------------------------------------------------------------------------
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
+-- Released under the MIT license. See Copyright Notice in luajit.h
+----------------------------------------------------------------------------
+
+local arg = {...}
+local FILE_ROLLING_H = arg[1] or "luajit_rolling.h"
+local FILE_RELVER_TXT = arg[2] or "luajit_relver.txt"
+local FILE_LUAJIT_H = arg[3] or "luajit.h"
+
+local function file_read(file)
+ local fp = assert(io.open(file, "rb"), "run from the wrong directory")
+ local data = assert(fp:read("*a"))
+ fp:close()
+ return data
+end
+
+local function file_write_mod(file, data)
+ local fp = io.open(file, "rb")
+ if fp then
+ local odata = assert(fp:read("*a"))
+ fp:close()
+ if odata == data then return end
+ end
+ fp = assert(io.open(file, "wb"))
+ assert(fp:write(data))
+ assert(fp:close())
+end
+
+local text = file_read(FILE_ROLLING_H):gsub("#error.-\n", "")
+local relver = file_read(FILE_RELVER_TXT):match("(%d+)")
+
+if relver then
+ text = text:gsub("ROLLING", relver)
+else
+ io.stderr:write([[
+**** WARNING Cannot determine rolling release version from git log.
+**** WARNING The 'git' command must be available during the build.
+]])
+ file_write_mod(FILE_RELVER_TXT, "ROLLING\n") -- Fallback for install target.
+end
+
+file_write_mod(FILE_LUAJIT_H, text)
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/genversion.lua
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/host/minilua.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/host/minilua.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/host/minilua.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1134,7 +1134,7 @@
CallInfo*ci;
StkId st,base;
Proto*p=cl->p;
-luaD_checkstack(L,p->maxstacksize);
+luaD_checkstack(L,p->maxstacksize+p->numparams);
func=restorestack(L,funcr);
if(!p->is_vararg){
base=func+1;
@@ -1639,6 +1639,7 @@
if(luai_numeq(cast_num(k),nvalue(key)))
return luaH_getnum(t,k);
}
+/*fallthrough*/
default:{
Node*n=mainposition(t,key);
do{
@@ -2905,8 +2906,8 @@
read_long_string(ls,seminfo,sep);
return TK_STRING;
}
-else if(sep==-1)return'[';
-else luaX_lexerror(ls,"invalid long string delimiter",TK_STRING);
+else if (sep!=-1)luaX_lexerror(ls,"invalid long string delimiter",TK_STRING);
+return'[';
}
case'=':{
next(ls);
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/.gitignore
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/.gitignore (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/.gitignore 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1 @@
+vmdef.lua
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bc.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bc.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bc.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT bytecode listing module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
@@ -41,7 +41,6 @@
-- Cache some library functions and objects.
local jit = require("jit")
-assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
local jutil = require("jit.util")
local vmdef = require("jit.vmdef")
local bit = require("bit")
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bcsave.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bcsave.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/bcsave.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT module to save/list bytecode.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
@@ -11,12 +11,16 @@
------------------------------------------------------------------------------
local jit = require("jit")
-assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
+assert(jit.version_num == 20199, "LuaJIT core/library version mismatch")
local bit = require("bit")
-- Symbol name prefix for LuaJIT bytecode.
local LJBC_PREFIX = "luaJIT_BC_"
+local type, assert = type, assert
+local format = string.format
+local tremove, tconcat = table.remove, table.concat
+
------------------------------------------------------------------------------
local function usage()
@@ -25,15 +29,19 @@
-l Only list bytecode.
-s Strip debug info (default).
-g Keep debug info.
+ -W Generate 32 bit (non-GC64) bytecode.
+ -X Generate 64 bit (GC64) bytecode.
+ -d Generate bytecode in deterministic manner.
-n name Set module name (default: auto-detect from input name).
-t type Set output file type (default: auto-detect from output name).
-a arch Override architecture for object files (default: native).
-o os Override OS for object files (default: native).
+ -F name Override filename (default: input filename).
-e chunk Use chunk string as input.
-- Stop handling options.
- Use stdin as input and/or stdout as output.
-File types: c h obj o raw (default)
+File types: c cc h obj o raw (default)
]]
os.exit(1)
end
@@ -45,10 +53,23 @@
os.exit(1)
end
-local function readfile(input)
- if type(input) == "function" then return input end
- if input == "-" then input = nil end
- return check(loadfile(input))
+local function readfile(ctx, input)
+ if ctx.string then
+ return check(loadstring(input, nil, ctx.mode))
+ elseif ctx.filename then
+ local data
+ if input == "-" then
+ data = io.stdin:read("*a")
+ else
+ local fp = assert(io.open(input, "rb"))
+ data = assert(fp:read("*a"))
+ assert(fp:close())
+ end
+ return check(load(data, ctx.filename, ctx.mode))
+ else
+ if input == "-" then input = nil end
+ return check(loadfile(input, ctx.mode))
+ end
end
local function savefile(name, mode)
@@ -56,15 +77,30 @@
return check(io.open(name, mode))
end
+local function set_stdout_binary(ffi)
+ ffi.cdef[[int _setmode(int fd, int mode);]]
+ ffi.C._setmode(1, 0x8000)
+end
+
------------------------------------------------------------------------------
local map_type = {
- raw = "raw", c = "c", h = "h", o = "obj", obj = "obj",
+ raw = "raw", c = "c", cc = "c", h = "h", o = "obj", obj = "obj",
}
local map_arch = {
- x86 = true, x64 = true, arm = true, arm64 = true, arm64be = true,
- ppc = true, mips = true, mipsel = true,
+ x86 = { e = "le", b = 32, m = 3, p = 0x14c, },
+ x64 = { e = "le", b = 64, m = 62, p = 0x8664, },
+ arm = { e = "le", b = 32, m = 40, p = 0x1c0, },
+ arm64 = { e = "le", b = 64, m = 183, p = 0xaa64, },
+ arm64be = { e = "be", b = 64, m = 183, },
+ ppc = { e = "be", b = 32, m = 20, },
+ mips = { e = "be", b = 32, m = 8, f = 0x50001006, },
+ mipsel = { e = "le", b = 32, m = 8, f = 0x50001006, },
+ mips64 = { e = "be", b = 64, m = 8, f = 0x80000007, },
+ mips64el = { e = "le", b = 64, m = 8, f = 0x80000007, },
+ mips64r6 = { e = "be", b = 64, m = 8, f = 0xa0000407, },
+ mips64r6el = { e = "le", b = 64, m = 8, f = 0xa0000407, },
}
local map_os = {
@@ -73,33 +109,33 @@
}
local function checkarg(str, map, err)
- str = string.lower(str)
+ str = str:lower()
local s = check(map[str], "unknown ", err)
- return s == true and str or s
+ return type(s) == "string" and s or str
end
local function detecttype(str)
- local ext = string.match(string.lower(str), "%.(%a+)$")
+ local ext = str:lower():match("%.(%a+)$")
return map_type[ext] or "raw"
end
local function checkmodname(str)
- check(string.match(str, "^[%w_.%-]+$"), "bad module name")
- return string.gsub(str, "[%.%-]", "_")
+ check(str:match("^[%w_.%-]+$"), "bad module name")
+ return str:gsub("[%.%-]", "_")
end
local function detectmodname(str)
if type(str) == "string" then
- local tail = string.match(str, "[^/\\]+$")
+ local tail = str:match("[^/\\]+$")
if tail then str = tail end
- local head = string.match(str, "^(.*)%.[^.]*$")
+ local head = str:match("^(.*)%.[^.]*$")
if head then str = head end
- str = string.match(str, "^[%w_.%-]+")
+ str = str:match("^[%w_.%-]+")
else
str = nil
end
check(str, "cannot derive module name, use -n name")
- return string.gsub(str, "[%.%-]", "_")
+ return str:gsub("[%.%-]", "_")
end
------------------------------------------------------------------------------
@@ -111,6 +147,11 @@
end
local function bcsave_raw(output, s)
+ if output == "-" and jit.os == "Windows" then
+ local ok, ffi = pcall(require, "ffi")
+ check(ok, "FFI library required to write binary file to stdout")
+ set_stdout_binary(ffi)
+ end
local fp = savefile(output, "wb")
bcsave_tail(fp, output, s)
end
@@ -118,8 +159,8 @@
local function bcsave_c(ctx, output, s)
local fp = savefile(output, "w")
if ctx.type == "c" then
- fp:write(string.format([[
-#ifdef _cplusplus
+ fp:write(format([[
+#ifdef __cplusplus
extern "C"
#endif
#ifdef _WIN32
@@ -128,7 +169,7 @@
const unsigned char %s%s[] = {
]], LJBC_PREFIX, ctx.modname))
else
- fp:write(string.format([[
+ fp:write(format([[
#define %s%s_SIZE %d
static const unsigned char %s%s[] = {
]], LJBC_PREFIX, ctx.modname, #s, LJBC_PREFIX, ctx.modname))
@@ -138,13 +179,13 @@
local b = tostring(string.byte(s, i))
m = m + #b + 1
if m > 78 then
- fp:write(table.concat(t, ",", 1, n), ",\n")
+ fp:write(tconcat(t, ",", 1, n), ",\n")
n, m = 0, #b + 1
end
n = n + 1
t[n] = b
end
- bcsave_tail(fp, output, table.concat(t, ",", 1, n).."\n};\n")
+ bcsave_tail(fp, output, tconcat(t, ",", 1, n).."\n};\n")
end
local function bcsave_elfobj(ctx, output, s, ffi)
@@ -199,12 +240,8 @@
} ELF64obj;
]]
local symname = LJBC_PREFIX..ctx.modname
- local is64, isbe = false, false
- if ctx.arch == "x64" or ctx.arch == "arm64" or ctx.arch == "arm64be" then
- is64 = true
- elseif ctx.arch == "ppc" or ctx.arch == "mips" then
- isbe = true
- end
+ local ai = assert(map_arch[ctx.arch])
+ local is64, isbe = ai.b == 64, ai.e == "be"
-- Handle different host/target endianess.
local function f32(x) return x end
@@ -237,10 +274,8 @@
hdr.eendian = isbe and 2 or 1
hdr.eversion = 1
hdr.type = f16(1)
- hdr.machine = f16(({ x86=3, x64=62, arm=40, arm64=183, arm64be=183, ppc=20, mips=8, mipsel=8 })[ctx.arch])
- if ctx.arch == "mips" or ctx.arch == "mipsel" then
- hdr.flags = f32(0x50001006)
- end
+ hdr.machine = f16(ai.m)
+ hdr.flags = f32(ai.f or 0)
hdr.version = f32(1)
hdr.shofs = fofs(ffi.offsetof(o, "sect"))
hdr.ehsize = f16(ffi.sizeof(hdr))
@@ -275,7 +310,7 @@
o.sect[2].size = fofs(ofs)
o.sect[3].type = f32(3) -- .strtab
o.sect[3].ofs = fofs(sofs + ofs)
- o.sect[3].size = fofs(#symname+1)
+ o.sect[3].size = fofs(#symname+2)
ffi.copy(o.space+ofs+1, symname)
ofs = ofs + #symname + 2
o.sect[4].type = f32(1) -- .rodata
@@ -336,12 +371,8 @@
} PEobj;
]]
local symname = LJBC_PREFIX..ctx.modname
- local is64 = false
- if ctx.arch == "x86" then
- symname = "_"..symname
- elseif ctx.arch == "x64" then
- is64 = true
- end
+ local ai = assert(map_arch[ctx.arch])
+ local is64 = ai.b == 64
local symexport = " /EXPORT:"..symname..",DATA "
-- The file format is always little-endian. Swap if the host is big-endian.
@@ -355,7 +386,7 @@
-- Create PE object and fill in header.
local o = ffi.new("PEobj")
local hdr = o.hdr
- hdr.arch = f16(({ x86=0x14c, x64=0x8664, arm=0x1c0, ppc=0x1f2, mips=0x366, mipsel=0x366 })[ctx.arch])
+ hdr.arch = f16(assert(ai.p))
hdr.nsects = f16(2)
hdr.symtabofs = f32(ffi.offsetof(o, "sym0"))
hdr.nsyms = f32(6)
@@ -411,23 +442,11 @@
typedef struct {
uint32_t cmd, cmdsize;
char segname[16];
- uint32_t vmaddr, vmsize, fileoff, filesize;
- uint32_t maxprot, initprot, nsects, flags;
-} mach_segment_command;
-typedef struct {
- uint32_t cmd, cmdsize;
- char segname[16];
uint64_t vmaddr, vmsize, fileoff, filesize;
uint32_t maxprot, initprot, nsects, flags;
} mach_segment_command_64;
typedef struct {
char sectname[16], segname[16];
- uint32_t addr, size;
- uint32_t offset, align, reloff, nreloc, flags;
- uint32_t reserved1, reserved2;
-} mach_section;
-typedef struct {
- char sectname[16], segname[16];
uint64_t addr, size;
uint32_t offset, align, reloff, nreloc, flags;
uint32_t reserved1, reserved2, reserved3;
@@ -438,124 +457,61 @@
typedef struct {
int32_t strx;
uint8_t type, sect;
- int16_t desc;
- uint32_t value;
-} mach_nlist;
-typedef struct {
- uint32_t strx;
- uint8_t type, sect;
uint16_t desc;
uint64_t value;
} mach_nlist_64;
-typedef struct
-{
- uint32_t magic, nfat_arch;
-} mach_fat_header;
-typedef struct
-{
- uint32_t cputype, cpusubtype, offset, size, align;
-} mach_fat_arch;
typedef struct {
- struct {
- mach_header hdr;
- mach_segment_command seg;
- mach_section sec;
- mach_symtab_command sym;
- } arch[1];
- mach_nlist sym_entry;
- uint8_t space[4096];
-} mach_obj;
-typedef struct {
- struct {
- mach_header_64 hdr;
- mach_segment_command_64 seg;
- mach_section_64 sec;
- mach_symtab_command sym;
- } arch[1];
+ mach_header_64 hdr;
+ mach_segment_command_64 seg;
+ mach_section_64 sec;
+ mach_symtab_command sym;
mach_nlist_64 sym_entry;
uint8_t space[4096];
} mach_obj_64;
-typedef struct {
- mach_fat_header fat;
- mach_fat_arch fat_arch[2];
- struct {
- mach_header hdr;
- mach_segment_command seg;
- mach_section sec;
- mach_symtab_command sym;
- } arch[2];
- mach_nlist sym_entry;
- uint8_t space[4096];
-} mach_fat_obj;
]]
local symname = '_'..LJBC_PREFIX..ctx.modname
- local isfat, is64, align, mobj = false, false, 4, "mach_obj"
- if ctx.arch == "x64" then
- is64, align, mobj = true, 8, "mach_obj_64"
- elseif ctx.arch == "arm" then
- isfat, mobj = true, "mach_fat_obj"
- elseif ctx.arch == "arm64" then
- is64, align, isfat, mobj = true, 8, true, "mach_fat_obj"
- else
- check(ctx.arch == "x86", "unsupported architecture for OSX")
+ local cputype, cpusubtype = 0x01000007, 3
+ if ctx.arch ~= "x64" then
+ check(ctx.arch == "arm64", "unsupported architecture for OSX")
+ cputype, cpusubtype = 0x0100000c, 0
end
local function aligned(v, a) return bit.band(v+a-1, -a) end
- local be32 = bit.bswap -- Mach-O FAT is BE, supported archs are LE.
-- Create Mach-O object and fill in header.
- local o = ffi.new(mobj)
- local mach_size = aligned(ffi.offsetof(o, "space")+#symname+2, align)
- local cputype = ({ x86={7}, x64={0x01000007}, arm={7,12}, arm64={0x01000007,0x0100000c} })[ctx.arch]
- local cpusubtype = ({ x86={3}, x64={3}, arm={3,9}, arm64={3,0} })[ctx.arch]
- if isfat then
- o.fat.magic = be32(0xcafebabe)
- o.fat.nfat_arch = be32(#cpusubtype)
- end
+ local o = ffi.new("mach_obj_64")
+ local mach_size = aligned(ffi.offsetof(o, "space")+#symname+2, 8)
-- Fill in sections and symbols.
- for i=0,#cpusubtype-1 do
- local ofs = 0
- if isfat then
- local a = o.fat_arch[i]
- a.cputype = be32(cputype[i+1])
- a.cpusubtype = be32(cpusubtype[i+1])
- -- Subsequent slices overlap each other to share data.
- ofs = ffi.offsetof(o, "arch") + i*ffi.sizeof(o.arch[0])
- a.offset = be32(ofs)
- a.size = be32(mach_size-ofs+#s)
- end
- local a = o.arch[i]
- a.hdr.magic = is64 and 0xfeedfacf or 0xfeedface
- a.hdr.cputype = cputype[i+1]
- a.hdr.cpusubtype = cpusubtype[i+1]
- a.hdr.filetype = 1
- a.hdr.ncmds = 2
- a.hdr.sizeofcmds = ffi.sizeof(a.seg)+ffi.sizeof(a.sec)+ffi.sizeof(a.sym)
- a.seg.cmd = is64 and 0x19 or 0x1
- a.seg.cmdsize = ffi.sizeof(a.seg)+ffi.sizeof(a.sec)
- a.seg.vmsize = #s
- a.seg.fileoff = mach_size-ofs
- a.seg.filesize = #s
- a.seg.maxprot = 1
- a.seg.initprot = 1
- a.seg.nsects = 1
- ffi.copy(a.sec.sectname, "__data")
- ffi.copy(a.sec.segname, "__DATA")
- a.sec.size = #s
- a.sec.offset = mach_size-ofs
- a.sym.cmd = 2
- a.sym.cmdsize = ffi.sizeof(a.sym)
- a.sym.symoff = ffi.offsetof(o, "sym_entry")-ofs
- a.sym.nsyms = 1
- a.sym.stroff = ffi.offsetof(o, "sym_entry")+ffi.sizeof(o.sym_entry)-ofs
- a.sym.strsize = aligned(#symname+2, align)
- end
+ o.hdr.magic = 0xfeedfacf
+ o.hdr.cputype = cputype
+ o.hdr.cpusubtype = cpusubtype
+ o.hdr.filetype = 1
+ o.hdr.ncmds = 2
+ o.hdr.sizeofcmds = ffi.sizeof(o.seg)+ffi.sizeof(o.sec)+ffi.sizeof(o.sym)
+ o.seg.cmd = 0x19
+ o.seg.cmdsize = ffi.sizeof(o.seg)+ffi.sizeof(o.sec)
+ o.seg.vmsize = #s
+ o.seg.fileoff = mach_size
+ o.seg.filesize = #s
+ o.seg.maxprot = 1
+ o.seg.initprot = 1
+ o.seg.nsects = 1
+ ffi.copy(o.sec.sectname, "__data")
+ ffi.copy(o.sec.segname, "__DATA")
+ o.sec.size = #s
+ o.sec.offset = mach_size
+ o.sym.cmd = 2
+ o.sym.cmdsize = ffi.sizeof(o.sym)
+ o.sym.symoff = ffi.offsetof(o, "sym_entry")
+ o.sym.nsyms = 1
+ o.sym.stroff = ffi.offsetof(o, "sym_entry")+ffi.sizeof(o.sym_entry)
+ o.sym.strsize = aligned(#symname+2, 8)
o.sym_entry.type = 0xf
o.sym_entry.sect = 1
o.sym_entry.strx = 1
ffi.copy(o.space+1, symname)
- -- Write Macho-O object file.
+ -- Write Mach-O object file.
local fp = savefile(output, "wb")
fp:write(ffi.string(o, mach_size))
bcsave_tail(fp, output, s)
@@ -564,6 +520,9 @@
local function bcsave_obj(ctx, output, s)
local ok, ffi = pcall(require, "ffi")
check(ok, "FFI library required to write this file type")
+ if output == "-" and jit.os == "Windows" then
+ set_stdout_binary(ffi)
+ end
if ctx.os == "windows" then
return bcsave_peobj(ctx, output, s, ffi)
elseif ctx.os == "osx" then
@@ -575,14 +534,14 @@
------------------------------------------------------------------------------
-local function bclist(input, output)
- local f = readfile(input)
+local function bclist(ctx, input, output)
+ local f = readfile(ctx, input)
require("jit.bc").dump(f, savefile(output, "w"), true)
end
local function bcsave(ctx, input, output)
- local f = readfile(input)
- local s = string.dump(f, ctx.strip)
+ local f = readfile(ctx, input)
+ local s = string.dump(f, ctx.mode)
local t = ctx.type
if not t then
t = detecttype(output)
@@ -605,35 +564,43 @@
local n = 1
local list = false
local ctx = {
- strip = true, arch = jit.arch, os = string.lower(jit.os),
- type = false, modname = false,
+ mode = "bt", arch = jit.arch, os = jit.os:lower(),
+ type = false, modname = false, string = false,
}
+ local strip = "s"
+ local gc64 = ""
while n <= #arg do
local a = arg[n]
- if type(a) == "string" and string.sub(a, 1, 1) == "-" and a ~= "-" then
- table.remove(arg, n)
+ if type(a) == "string" and a:sub(1, 1) == "-" and a ~= "-" then
+ tremove(arg, n)
if a == "--" then break end
for m=2,#a do
- local opt = string.sub(a, m, m)
+ local opt = a:sub(m, m)
if opt == "l" then
list = true
elseif opt == "s" then
- ctx.strip = true
+ strip = "s"
elseif opt == "g" then
- ctx.strip = false
+ strip = ""
+ elseif opt == "W" or opt == "X" then
+ gc64 = opt
+ elseif opt == "d" then
+ ctx.mode = ctx.mode .. opt
else
if arg[n] == nil or m ~= #a then usage() end
if opt == "e" then
if n ~= 1 then usage() end
- arg[1] = check(loadstring(arg[1]))
+ ctx.string = true
elseif opt == "n" then
- ctx.modname = checkmodname(table.remove(arg, n))
+ ctx.modname = checkmodname(tremove(arg, n))
elseif opt == "t" then
- ctx.type = checkarg(table.remove(arg, n), map_type, "file type")
+ ctx.type = checkarg(tremove(arg, n), map_type, "file type")
elseif opt == "a" then
- ctx.arch = checkarg(table.remove(arg, n), map_arch, "architecture")
+ ctx.arch = checkarg(tremove(arg, n), map_arch, "architecture")
elseif opt == "o" then
- ctx.os = checkarg(table.remove(arg, n), map_os, "OS name")
+ ctx.os = checkarg(tremove(arg, n), map_os, "OS name")
+ elseif opt == "F" then
+ ctx.filename = "@"..tremove(arg, n)
else
usage()
end
@@ -643,9 +610,10 @@
n = n + 1
end
end
+ ctx.mode = ctx.mode .. strip .. gc64
if list then
if #arg == 0 or #arg > 2 then usage() end
- bclist(arg[1], arg[2] or "-")
+ bclist(ctx, arg[1], arg[2] or "-")
else
if #arg ~= 2 then usage() end
bcsave(ctx, arg[1], arg[2])
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT ARM disassembler module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This is a helper module used by the LuaJIT machine code dumper module.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT ARM64 disassembler module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
--
-- Contributed by Djordje Kovacevic and Stefan Pejic from RT-RK.com.
@@ -107,24 +107,20 @@
[0] = {
shift = 29, mask = 3,
[0] = {
- shift = 21, mask = 7,
- [0] = "andDNMSg", "bicDNMSg", "andDNMSg", "bicDNMSg",
- "andDNMSg", "bicDNMSg", "andDNMg", "bicDNMg"
+ shift = 21, mask = 1,
+ [0] = "andDNMSg", "bicDNMSg"
},
{
- shift = 21, mask = 7,
- [0] ="orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0MSg", "orn|mvnDN0MSg",
- "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0Mg", "orn|mvnDN0Mg"
+ shift = 21, mask = 1,
+ [0] = "orr|movDN0MSg", "orn|mvnDN0MSg"
},
{
- shift = 21, mask = 7,
- [0] = "eorDNMSg", "eonDNMSg", "eorDNMSg", "eonDNMSg",
- "eorDNMSg", "eonDNMSg", "eorDNMg", "eonDNMg"
+ shift = 21, mask = 1,
+ [0] = "eorDNMSg", "eonDNMSg"
},
{
- shift = 21, mask = 7,
- [0] = "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMSg", "bicsDNMSg",
- "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMg", "bicsDNMg"
+ shift = 21, mask = 1,
+ [0] = "ands|tstD0NMSg", "bicsDNMSg"
}
},
false -- unallocated
@@ -132,24 +128,20 @@
{
shift = 29, mask = 3,
[0] = {
- shift = 21, mask = 7,
- [0] = "andDNMSg", "bicDNMSg", "andDNMSg", "bicDNMSg",
- "andDNMSg", "bicDNMSg", "andDNMg", "bicDNMg"
+ shift = 21, mask = 1,
+ [0] = "andDNMSg", "bicDNMSg"
},
{
- shift = 21, mask = 7,
- [0] = "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0MSg", "orn|mvnDN0MSg",
- "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0Mg", "orn|mvnDN0Mg"
+ shift = 21, mask = 1,
+ [0] = "orr|movDN0MSg", "orn|mvnDN0MSg"
},
{
- shift = 21, mask = 7,
- [0] = "eorDNMSg", "eonDNMSg", "eorDNMSg", "eonDNMSg",
- "eorDNMSg", "eonDNMSg", "eorDNMg", "eonDNMg"
+ shift = 21, mask = 1,
+ [0] = "eorDNMSg", "eonDNMSg"
},
{
- shift = 21, mask = 7,
- [0] = "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMSg", "bicsDNMSg",
- "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMg", "bicsDNMg"
+ shift = 21, mask = 1,
+ [0] = "ands|tstD0NMSg", "bicsDNMSg"
}
}
}
@@ -666,6 +658,10 @@
}
}
}
+ },
+ { -- 010
+ shift = 0, mask = 0x81f8fc00,
+ [0x100e400] = "moviDdG"
}
}
@@ -735,7 +731,7 @@
"hi", "ls", "ge", "lt", "gt", "le", "al",
}
-local map_shift = { [0] = "lsl", "lsr", "asr", }
+local map_shift = { [0] = "lsl", "lsr", "asr", "ror"}
local map_extend = {
[0] = "uxtb", "uxth", "uxtw", "uxtx", "sxtb", "sxth", "sxtw", "sxtx",
@@ -840,6 +836,20 @@
return sign * frac * 2^exp
end
+local function decode_fpmovi(op)
+ local lo = rshift(op, 5)
+ local hi = rshift(op, 9)
+ lo = bor(band(lo, 1) * 0xff, band(lo, 2) * 0x7f80, band(lo, 4) * 0x3fc000,
+ band(lo, 8) * 0x1fe00000)
+ hi = bor(band(hi, 1) * 0xff, band(hi, 0x80) * 0x1fe,
+ band(hi, 0x100) * 0xff00, band(hi, 0x200) * 0x7f8000)
+ if hi ~= 0 then
+ return fmt_hex32(hi)..tohex(lo)
+ else
+ return fmt_hex32(lo)
+ end
+end
+
local function prefer_bfx(sf, uns, imms, immr)
if imms < immr or imms == 31 or imms == 63 then
return false
@@ -956,7 +966,7 @@
elseif p == "U" then
local rn = map_regs.x[band(rshift(op, 5), 31)]
local sz = band(rshift(op, 30), 3)
- local imm12 = lshift(arshift(lshift(op, 10), 20), sz)
+ local imm12 = lshift(rshift(lshift(op, 10), 20), sz)
if imm12 ~= 0 then
x = "["..rn..", #"..imm12.."]"
else
@@ -993,8 +1003,7 @@
x = x.."]"
end
elseif p == "P" then
- local opcv, sh = rshift(op, 26), 2
- if opcv >= 0x2a then sh = 4 elseif opcv >= 0x1b then sh = 3 end
+ local sh = 2 + rshift(op, 31 - band(rshift(op, 26), 1))
local imm7 = lshift(arshift(lshift(op, 10), 25), sh)
local rn = map_regs.x[band(rshift(op, 5), 31)]
local ind = band(rshift(op, 23), 3)
@@ -1089,7 +1098,7 @@
last = "#"..(sf+32 - immr)
operands[#operands] = last
x = x + 1
- elseif x >= immr then
+ else
name = a2
x = x - immr + 1
end
@@ -1140,6 +1149,8 @@
x = 0
elseif p == "F" then
x = parse_fpimm8(op)
+ elseif p == "G" then
+ x = "#0x"..decode_fpmovi(op)
elseif p == "g" or p == "f" or p == "x" or p == "w" or
p == "d" or p == "s" then
-- These are handled in D/N/M/A.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64be.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64be.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_arm64be.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT ARM64BE disassembler wrapper module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- ARM64 instructions are always little-endian. So just forward to the
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT MIPS disassembler module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT/X license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This is a helper module used by the LuaJIT machine code dumper module.
@@ -19,13 +19,34 @@
local lshift, rshift, arshift = bit.lshift, bit.rshift, bit.arshift
------------------------------------------------------------------------------
--- Primary and extended opcode maps
+-- Extended opcode maps common to all MIPS releases
------------------------------------------------------------------------------
-local map_movci = { shift = 16, mask = 1, [0] = "movfDSC", "movtDSC", }
local map_srl = { shift = 21, mask = 1, [0] = "srlDTA", "rotrDTA", }
local map_srlv = { shift = 6, mask = 1, [0] = "srlvDTS", "rotrvDTS", }
+local map_cop0 = {
+ shift = 25, mask = 1,
+ [0] = {
+ shift = 21, mask = 15,
+ [0] = "mfc0TDW", [4] = "mtc0TDW",
+ [10] = "rdpgprDT",
+ [11] = { shift = 5, mask = 1, [0] = "diT0", "eiT0", },
+ [14] = "wrpgprDT",
+ }, {
+ shift = 0, mask = 63,
+ [1] = "tlbr", [2] = "tlbwi", [6] = "tlbwr", [8] = "tlbp",
+ [24] = "eret", [31] = "deret",
+ [32] = "wait",
+ },
+}
+
+------------------------------------------------------------------------------
+-- Primary and extended opcode maps for MIPS R1-R5
+------------------------------------------------------------------------------
+
+local map_movci = { shift = 16, mask = 1, [0] = "movfDSC", "movtDSC", }
+
local map_special = {
shift = 0, mask = 63,
[0] = { shift = 0, mask = -1, [0] = "nop", _ = "sllDTA" },
@@ -87,22 +108,6 @@
false, false, false, "synciSO",
}
-local map_cop0 = {
- shift = 25, mask = 1,
- [0] = {
- shift = 21, mask = 15,
- [0] = "mfc0TDW", [4] = "mtc0TDW",
- [10] = "rdpgprDT",
- [11] = { shift = 5, mask = 1, [0] = "diT0", "eiT0", },
- [14] = "wrpgprDT",
- }, {
- shift = 0, mask = 63,
- [1] = "tlbr", [2] = "tlbwi", [6] = "tlbwr", [8] = "tlbp",
- [24] = "eret", [31] = "deret",
- [32] = "wait",
- },
-}
-
local map_cop1s = {
shift = 0, mask = 63,
[0] = "add.sFGH", "sub.sFGH", "mul.sFGH", "div.sFGH",
@@ -234,7 +239,209 @@
}
------------------------------------------------------------------------------
+-- Primary and extended opcode maps for MIPS R6
+------------------------------------------------------------------------------
+local map_mul_r6 = { shift = 6, mask = 3, [2] = "mulDST", [3] = "muhDST" }
+local map_mulu_r6 = { shift = 6, mask = 3, [2] = "muluDST", [3] = "muhuDST" }
+local map_div_r6 = { shift = 6, mask = 3, [2] = "divDST", [3] = "modDST" }
+local map_divu_r6 = { shift = 6, mask = 3, [2] = "divuDST", [3] = "moduDST" }
+local map_dmul_r6 = { shift = 6, mask = 3, [2] = "dmulDST", [3] = "dmuhDST" }
+local map_dmulu_r6 = { shift = 6, mask = 3, [2] = "dmuluDST", [3] = "dmuhuDST" }
+local map_ddiv_r6 = { shift = 6, mask = 3, [2] = "ddivDST", [3] = "dmodDST" }
+local map_ddivu_r6 = { shift = 6, mask = 3, [2] = "ddivuDST", [3] = "dmoduDST" }
+
+local map_special_r6 = {
+ shift = 0, mask = 63,
+ [0] = { shift = 0, mask = -1, [0] = "nop", _ = "sllDTA" },
+ false, map_srl, "sraDTA",
+ "sllvDTS", false, map_srlv, "sravDTS",
+ "jrS", "jalrD1S", false, false,
+ "syscallY", "breakY", false, "sync",
+ "clzDS", "cloDS", "dclzDS", "dcloDS",
+ "dsllvDST", "dlsaDSTA", "dsrlvDST", "dsravDST",
+ map_mul_r6, map_mulu_r6, map_div_r6, map_divu_r6,
+ map_dmul_r6, map_dmulu_r6, map_ddiv_r6, map_ddivu_r6,
+ "addDST", "addu|moveDST0", "subDST", "subu|neguDS0T",
+ "andDST", "or|moveDST0", "xorDST", "nor|notDST0",
+ false, false, "sltDST", "sltuDST",
+ "daddDST", "dadduDST", "dsubDST", "dsubuDST",
+ "tgeSTZ", "tgeuSTZ", "tltSTZ", "tltuSTZ",
+ "teqSTZ", "seleqzDST", "tneSTZ", "selnezDST",
+ "dsllDTA", false, "dsrlDTA", "dsraDTA",
+ "dsll32DTA", false, "dsrl32DTA", "dsra32DTA",
+}
+
+local map_bshfl_r6 = {
+ shift = 9, mask = 3,
+ [1] = "alignDSTa",
+ _ = {
+ shift = 6, mask = 31,
+ [0] = "bitswapDT",
+ [2] = "wsbhDT",
+ [16] = "sebDT",
+ [24] = "sehDT",
+ }
+}
+
+local map_dbshfl_r6 = {
+ shift = 9, mask = 3,
+ [1] = "dalignDSTa",
+ _ = {
+ shift = 6, mask = 31,
+ [0] = "dbitswapDT",
+ [2] = "dsbhDT",
+ [5] = "dshdDT",
+ }
+}
+
+local map_special3_r6 = {
+ shift = 0, mask = 63,
+ [0] = "extTSAK", [1] = "dextmTSAP", [3] = "dextTSAK",
+ [4] = "insTSAL", [6] = "dinsuTSEQ", [7] = "dinsTSAL",
+ [32] = map_bshfl_r6, [36] = map_dbshfl_r6, [59] = "rdhwrTD",
+}
+
+local map_regimm_r6 = {
+ shift = 16, mask = 31,
+ [0] = "bltzSB", [1] = "bgezSB",
+ [6] = "dahiSI", [30] = "datiSI",
+ [23] = "sigrieI", [31] = "synciSO",
+}
+
+local map_pcrel_r6 = {
+ shift = 19, mask = 3,
+ [0] = "addiupcS2", "lwpcS2", "lwupcS2", {
+ shift = 18, mask = 1,
+ [0] = "ldpcS3", { shift = 16, mask = 3, [2] = "auipcSI", [3] = "aluipcSI" }
+ }
+}
+
+local map_cop1s_r6 = {
+ shift = 0, mask = 63,
+ [0] = "add.sFGH", "sub.sFGH", "mul.sFGH", "div.sFGH",
+ "sqrt.sFG", "abs.sFG", "mov.sFG", "neg.sFG",
+ "round.l.sFG", "trunc.l.sFG", "ceil.l.sFG", "floor.l.sFG",
+ "round.w.sFG", "trunc.w.sFG", "ceil.w.sFG", "floor.w.sFG",
+ "sel.sFGH", false, false, false,
+ "seleqz.sFGH", "recip.sFG", "rsqrt.sFG", "selnez.sFGH",
+ "maddf.sFGH", "msubf.sFGH", "rint.sFG", "class.sFG",
+ "min.sFGH", "mina.sFGH", "max.sFGH", "maxa.sFGH",
+ false, "cvt.d.sFG", false, false,
+ "cvt.w.sFG", "cvt.l.sFG",
+}
+
+local map_cop1d_r6 = {
+ shift = 0, mask = 63,
+ [0] = "add.dFGH", "sub.dFGH", "mul.dFGH", "div.dFGH",
+ "sqrt.dFG", "abs.dFG", "mov.dFG", "neg.dFG",
+ "round.l.dFG", "trunc.l.dFG", "ceil.l.dFG", "floor.l.dFG",
+ "round.w.dFG", "trunc.w.dFG", "ceil.w.dFG", "floor.w.dFG",
+ "sel.dFGH", false, false, false,
+ "seleqz.dFGH", "recip.dFG", "rsqrt.dFG", "selnez.dFGH",
+ "maddf.dFGH", "msubf.dFGH", "rint.dFG", "class.dFG",
+ "min.dFGH", "mina.dFGH", "max.dFGH", "maxa.dFGH",
+ "cvt.s.dFG", false, false, false,
+ "cvt.w.dFG", "cvt.l.dFG",
+}
+
+local map_cop1w_r6 = {
+ shift = 0, mask = 63,
+ [0] = "cmp.af.sFGH", "cmp.un.sFGH", "cmp.eq.sFGH", "cmp.ueq.sFGH",
+ "cmp.lt.sFGH", "cmp.ult.sFGH", "cmp.le.sFGH", "cmp.ule.sFGH",
+ "cmp.saf.sFGH", "cmp.sun.sFGH", "cmp.seq.sFGH", "cmp.sueq.sFGH",
+ "cmp.slt.sFGH", "cmp.sult.sFGH", "cmp.sle.sFGH", "cmp.sule.sFGH",
+ false, "cmp.or.sFGH", "cmp.une.sFGH", "cmp.ne.sFGH",
+ false, false, false, false,
+ false, "cmp.sor.sFGH", "cmp.sune.sFGH", "cmp.sne.sFGH",
+ false, false, false, false,
+ "cvt.s.wFG", "cvt.d.wFG",
+}
+
+local map_cop1l_r6 = {
+ shift = 0, mask = 63,
+ [0] = "cmp.af.dFGH", "cmp.un.dFGH", "cmp.eq.dFGH", "cmp.ueq.dFGH",
+ "cmp.lt.dFGH", "cmp.ult.dFGH", "cmp.le.dFGH", "cmp.ule.dFGH",
+ "cmp.saf.dFGH", "cmp.sun.dFGH", "cmp.seq.dFGH", "cmp.sueq.dFGH",
+ "cmp.slt.dFGH", "cmp.sult.dFGH", "cmp.sle.dFGH", "cmp.sule.dFGH",
+ false, "cmp.or.dFGH", "cmp.une.dFGH", "cmp.ne.dFGH",
+ false, false, false, false,
+ false, "cmp.sor.dFGH", "cmp.sune.dFGH", "cmp.sne.dFGH",
+ false, false, false, false,
+ "cvt.s.lFG", "cvt.d.lFG",
+}
+
+local map_cop1_r6 = {
+ shift = 21, mask = 31,
+ [0] = "mfc1TG", "dmfc1TG", "cfc1TG", "mfhc1TG",
+ "mtc1TG", "dmtc1TG", "ctc1TG", "mthc1TG",
+ false, "bc1eqzHB", false, false,
+ false, "bc1nezHB", false, false,
+ map_cop1s_r6, map_cop1d_r6, false, false,
+ map_cop1w_r6, map_cop1l_r6,
+}
+
+local function maprs_popTS(rs, rt)
+ if rt == 0 then return 0 elseif rs == 0 then return 1
+ elseif rs == rt then return 2 else return 3 end
+end
+
+local map_pop06_r6 = {
+ maprs = maprs_popTS, [0] = "blezSB", "blezalcTB", "bgezalcTB", "bgeucSTB"
+}
+local map_pop07_r6 = {
+ maprs = maprs_popTS, [0] = "bgtzSB", "bgtzalcTB", "bltzalcTB", "bltucSTB"
+}
+local map_pop26_r6 = {
+ maprs = maprs_popTS, "blezcTB", "bgezcTB", "bgecSTB"
+}
+local map_pop27_r6 = {
+ maprs = maprs_popTS, "bgtzcTB", "bltzcTB", "bltcSTB"
+}
+
+local function maprs_popS(rs, rt)
+ if rs == 0 then return 0 else return 1 end
+end
+
+local map_pop66_r6 = {
+ maprs = maprs_popS, [0] = "jicTI", "beqzcSb"
+}
+local map_pop76_r6 = {
+ maprs = maprs_popS, [0] = "jialcTI", "bnezcSb"
+}
+
+local function maprs_popST(rs, rt)
+ if rs >= rt then return 0 elseif rs == 0 then return 1 else return 2 end
+end
+
+local map_pop10_r6 = {
+ maprs = maprs_popST, [0] = "bovcSTB", "beqzalcTB", "beqcSTB"
+}
+local map_pop30_r6 = {
+ maprs = maprs_popST, [0] = "bnvcSTB", "bnezalcTB", "bnecSTB"
+}
+
+local map_pri_r6 = {
+ [0] = map_special_r6, map_regimm_r6, "jJ", "jalJ",
+ "beq|beqz|bST00B", "bne|bnezST0B", map_pop06_r6, map_pop07_r6,
+ map_pop10_r6, "addiu|liTS0I", "sltiTSI", "sltiuTSI",
+ "andiTSU", "ori|liTS0U", "xoriTSU", "aui|luiTS0U",
+ map_cop0, map_cop1_r6, false, false,
+ false, false, map_pop26_r6, map_pop27_r6,
+ map_pop30_r6, "daddiuTSI", false, false,
+ false, "dauiTSI", false, map_special3_r6,
+ "lbTSO", "lhTSO", false, "lwTSO",
+ "lbuTSO", "lhuTSO", false, false,
+ "sbTSO", "shTSO", false, "swTSO",
+ false, false, false, false,
+ false, "lwc1HSO", "bc#", false,
+ false, "ldc1HSO", map_pop66_r6, "ldTSO",
+ false, "swc1HSO", "balc#", map_pcrel_r6,
+ false, "sdc1HSO", map_pop76_r6, "sdTSO",
+}
+
+------------------------------------------------------------------------------
+
local map_gpr = {
[0] = "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7",
"r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15",
@@ -287,10 +494,14 @@
ctx.op = op
ctx.rel = nil
- local opat = map_pri[rshift(op, 26)]
+ local opat = ctx.map_pri[rshift(op, 26)]
while type(opat) ~= "string" do
if not opat then return unknown(ctx) end
- opat = opat[band(rshift(op, opat.shift), opat.mask)] or opat._
+ if opat.maprs then
+ opat = opat[opat.maprs(band(rshift(op,21),31), band(rshift(op,16),31))]
+ else
+ opat = opat[band(rshift(op, opat.shift), opat.mask)] or opat._
+ end
end
local name, pat = match(opat, "^([a-z0-9_.]*)(.*)")
local altname, pat2 = match(pat, "|([a-z0-9_.|]*)(.*)")
@@ -314,6 +525,8 @@
x = "f"..band(rshift(op, 21), 31)
elseif p == "A" then
x = band(rshift(op, 6), 31)
+ elseif p == "a" then
+ x = band(rshift(op, 6), 7)
elseif p == "E" then
x = band(rshift(op, 6), 31) + 32
elseif p == "M" then
@@ -333,6 +546,10 @@
x = band(rshift(op, 11), 31) - last + 33
elseif p == "I" then
x = arshift(lshift(op, 16), 16)
+ elseif p == "2" then
+ x = arshift(lshift(op, 13), 11)
+ elseif p == "3" then
+ x = arshift(lshift(op, 14), 11)
elseif p == "U" then
x = band(op, 0xffff)
elseif p == "O" then
@@ -342,9 +559,17 @@
local index = map_gpr[band(rshift(op, 16), 31)]
operands[#operands] = format("%s(%s)", index, last)
elseif p == "B" then
- x = ctx.addr + ctx.pos + arshift(lshift(op, 16), 16)*4 + 4
+ x = ctx.addr + ctx.pos + arshift(lshift(op, 16), 14) + 4
ctx.rel = x
x = format("0x%08x", x)
+ elseif p == "b" then
+ x = ctx.addr + ctx.pos + arshift(lshift(op, 11), 9) + 4
+ ctx.rel = x
+ x = format("0x%08x", x)
+ elseif p == "#" then
+ x = ctx.addr + ctx.pos + arshift(lshift(op, 6), 4) + 4
+ ctx.rel = x
+ x = format("0x%08x", x)
elseif p == "J" then
local a = ctx.addr + ctx.pos
x = a - band(a, 0x0fffffff) + band(op, 0x03ffffff)*4
@@ -408,6 +633,7 @@
ctx.disass = disass_block
ctx.hexdump = 8
ctx.get = get_be
+ ctx.map_pri = map_pri
return ctx
end
@@ -417,6 +643,19 @@
return ctx
end
+local function create_r6(code, addr, out)
+ local ctx = create(code, addr, out)
+ ctx.map_pri = map_pri_r6
+ return ctx
+end
+
+local function create_r6_el(code, addr, out)
+ local ctx = create(code, addr, out)
+ ctx.get = get_le
+ ctx.map_pri = map_pri_r6
+ return ctx
+end
+
-- Simple API: disassemble code (a string) at address and output via out.
local function disass(code, addr, out)
create(code, addr, out):disass()
@@ -426,6 +665,14 @@
create_el(code, addr, out):disass()
end
+local function disass_r6(code, addr, out)
+ create_r6(code, addr, out):disass()
+end
+
+local function disass_r6_el(code, addr, out)
+ create_r6_el(code, addr, out):disass()
+end
+
-- Return register name for RID.
local function regname(r)
if r < 32 then return map_gpr[r] end
@@ -436,8 +683,12 @@
return {
create = create,
create_el = create_el,
+ create_r6 = create_r6,
+ create_r6_el = create_r6_el,
disass = disass,
disass_el = disass_el,
+ disass_r6 = disass_r6,
+ disass_r6_el = disass_r6_el,
regname = regname
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT MIPS64 disassembler wrapper module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This module just exports the big-endian functions from the
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64el.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64el.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64el.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT MIPS64EL disassembler wrapper module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This module just exports the little-endian functions from the
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6.lua (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,17 @@
+----------------------------------------------------------------------------
+-- LuaJIT MIPS64R6 disassembler wrapper module.
+--
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
+-- Released under the MIT license. See Copyright Notice in luajit.h
+----------------------------------------------------------------------------
+-- This module just exports the r6 big-endian functions from the
+-- MIPS disassembler module. All the interesting stuff is there.
+------------------------------------------------------------------------------
+
+local dis_mips = require((string.match(..., ".*%.") or "").."dis_mips")
+return {
+ create = dis_mips.create_r6,
+ disass = dis_mips.disass_r6,
+ regname = dis_mips.regname
+}
+
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6.lua
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6el.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6el.lua (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6el.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,17 @@
+----------------------------------------------------------------------------
+-- LuaJIT MIPS64R6EL disassembler wrapper module.
+--
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
+-- Released under the MIT license. See Copyright Notice in luajit.h
+----------------------------------------------------------------------------
+-- This module just exports the r6 little-endian functions from the
+-- MIPS disassembler module. All the interesting stuff is there.
+------------------------------------------------------------------------------
+
+local dis_mips = require((string.match(..., ".*%.") or "").."dis_mips")
+return {
+ create = dis_mips.create_r6_el,
+ disass = dis_mips.disass_r6_el,
+ regname = dis_mips.regname
+}
+
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mips64r6el.lua
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mipsel.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mipsel.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_mipsel.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT MIPSEL disassembler wrapper module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This module just exports the little-endian functions from the
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_ppc.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_ppc.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_ppc.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT PPC disassembler module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT/X license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This is a helper module used by the LuaJIT machine code dumper module.
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x64.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x64.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x64.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT x64 disassembler wrapper module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This module just exports the 64 bit functions from the combined
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x86.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x86.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dis_x86.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT x86/x64 disassembler module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
-- This is a helper module used by the LuaJIT machine code dumper module.
@@ -239,6 +239,24 @@
--8x
[0x8c] = "||pmaskmovXrvVSm",
[0x8e] = "||pmaskmovVSmXvr",
+--9x
+[0x96] = "||fmaddsub132pHXrvm",[0x97] = "||fmsubadd132pHXrvm",
+[0x98] = "||fmadd132pHXrvm",[0x99] = "||fmadd132sHXrvm",
+[0x9a] = "||fmsub132pHXrvm",[0x9b] = "||fmsub132sHXrvm",
+[0x9c] = "||fnmadd132pHXrvm",[0x9d] = "||fnmadd132sHXrvm",
+[0x9e] = "||fnmsub132pHXrvm",[0x9f] = "||fnmsub132sHXrvm",
+--Ax
+[0xa6] = "||fmaddsub213pHXrvm",[0xa7] = "||fmsubadd213pHXrvm",
+[0xa8] = "||fmadd213pHXrvm",[0xa9] = "||fmadd213sHXrvm",
+[0xaa] = "||fmsub213pHXrvm",[0xab] = "||fmsub213sHXrvm",
+[0xac] = "||fnmadd213pHXrvm",[0xad] = "||fnmadd213sHXrvm",
+[0xae] = "||fnmsub213pHXrvm",[0xaf] = "||fnmsub213sHXrvm",
+--Bx
+[0xb6] = "||fmaddsub231pHXrvm",[0xb7] = "||fmsubadd231pHXrvm",
+[0xb8] = "||fmadd231pHXrvm",[0xb9] = "||fmadd231sHXrvm",
+[0xba] = "||fmsub231pHXrvm",[0xbb] = "||fmsub231sHXrvm",
+[0xbc] = "||fnmadd231pHXrvm",[0xbd] = "||fnmadd231sHXrvm",
+[0xbe] = "||fnmsub231pHXrvm",[0xbf] = "||fnmsub231sHXrvm",
--Dx
[0xdc] = "||aesencXrvm", [0xdd] = "||aesenclastXrvm",
[0xde] = "||aesdecXrvm", [0xdf] = "||aesdeclastXrvm",
@@ -483,7 +501,7 @@
local operands, regs, sz, mode, sp, rm, sc, rx, sdisp
local code, pos, stop, vexl = ctx.code, ctx.pos, ctx.stop, ctx.vexl
- -- Chars used: 1DFGIMPQRSTUVWXYabcdfgijlmoprstuvwxyz
+ -- Chars used: 1DFGHIMPQRSTUVWXYabcdfgijlmoprstuvwxyz
for p in gmatch(pat, ".") do
local x = nil
if p == "V" or p == "U" then
@@ -506,6 +524,9 @@
sz = ctx.o16 and "X" or "M"; ctx.o16 = false
if sz == "X" and vexl then sz = "Y"; ctx.vexl = false end
regs = map_regs[sz]
+ elseif p == "H" then
+ name = name..(ctx.rexw and "d" or "s")
+ ctx.rexw = false
elseif p == "S" then
name = name..lower(sz)
elseif p == "s" then
@@ -735,6 +756,7 @@
V = putpat, U = putpat, T = putpat,
M = putpat, X = putpat, P = putpat,
F = putpat, G = putpat, Y = putpat,
+ H = putpat,
-- Collect prefixes.
[":"] = function(ctx, name, pat)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dump.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dump.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/dump.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT compiler dump module.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
@@ -55,7 +55,6 @@
-- Cache some library functions and objects.
local jit = require("jit")
-assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
local jutil = require("jit.util")
local vmdef = require("jit.vmdef")
local funcinfo, funcbc = jutil.funcinfo, jutil.funcbc
@@ -102,10 +101,12 @@
local function fillsymtab(tr, nexit)
local t = symtab
if nexitsym == 0 then
+ local maskaddr = jit.arch == "arm" and -2
local ircall = vmdef.ircall
for i=0,#ircall do
local addr = ircalladdr(i)
if addr ~= 0 then
+ if maskaddr then addr = band(addr, maskaddr) end
if addr < 0 then addr = addr + 2^32 end
t[addr] = ircall[i]
end
@@ -217,8 +218,10 @@
return s
end
-local function colorize_ansi(s, t)
- return format(colortype_ansi[t], s)
+local function colorize_ansi(s, t, extra)
+ local out = format(colortype_ansi[t], s)
+ if extra then out = "\027[3m"..out end
+ return out
end
local irtype_ansi = setmetatable({},
@@ -227,9 +230,10 @@
local html_escape = { ["<"] = "<", [">"] = ">", ["&"] = "&", }
-local function colorize_html(s, t)
+local function colorize_html(s, t, extra)
s = gsub(s, "[<>&]", html_escape)
- return format('<span class="irt_%s">%s</span>', irtype_text[t], s)
+ return format('<span class="irt_%s%s">%s</span>',
+ irtype_text[t], extra and " irt_extra" or "", s)
end
local irtype_html = setmetatable({},
@@ -254,6 +258,7 @@
span.irt_udt, span.irt_lud { color: #00c0c0; }
span.irt_num { color: #4040c0; }
span.irt_int, span.irt_i8, span.irt_u8, span.irt_i16, span.irt_u16 { color: #b040b0; }
+span.irt_extra { font-style: italic; }
</style>
]]
@@ -269,6 +274,7 @@
if band(mode, 8) ~= 0 then s = s.."C" end
if band(mode, 16) ~= 0 then s = s.."R" end
if band(mode, 32) ~= 0 then s = s.."I" end
+ if band(mode, 64) ~= 0 then s = s.."K" end
t[mode] = s
return s
end}),
@@ -277,8 +283,10 @@
local s = irtype[band(mode, 31)]
s = irtype[band(shr(mode, 5), 31)].."."..s
if band(mode, 0x800) ~= 0 then s = s.." sext" end
- local c = shr(mode, 14)
- if c == 2 then s = s.." index" elseif c == 3 then s = s.." check" end
+ local c = shr(mode, 12)
+ if c == 1 then s = s.." none"
+ elseif c == 2 then s = s.." index"
+ elseif c == 3 then s = s.." check" end
t[mode] = s
return s
end}),
@@ -285,7 +293,8 @@
["FLOAD "] = vmdef.irfield,
["FREF "] = vmdef.irfield,
["FPMATH"] = vmdef.irfpm,
- ["BUFHDR"] = { [0] = "RESET", "APPEND" },
+ ["TMPREF"] = { [0] = "", "IN", "OUT", "INOUT", "", "", "OUT2", "INOUT2" },
+ ["BUFHDR"] = { [0] = "RESET", "APPEND", "WRITE" },
["TOSTR "] = { [0] = "INT", "NUM", "CHAR" },
}
@@ -315,7 +324,9 @@
local tn = type(k)
local s
if tn == "number" then
- if band(sn or 0, 0x30000) ~= 0 then
+ if t < 12 then
+ s = k == 0 and "NULL" or format("[0x%08x]", k)
+ elseif band(sn or 0, 0x30000) ~= 0 then
s = band(sn, 0x20000) ~= 0 and "contpc" or "ftsz"
elseif k == 2^52+2^51 then
s = "bias"
@@ -343,7 +354,7 @@
else
s = tostring(k) -- For primitives.
end
- s = colorize(format("%-4s", s), t)
+ s = colorize(format("%-4s", s), t, band(sn or 0, 0x100000) ~= 0)
if slot then
s = format("%s @%d", s, slot)
end
@@ -363,7 +374,7 @@
out:write(colorize(format("%04d/%04d", ref, ref+1), 14))
else
local m, ot, op1, op2 = traceir(tr, ref)
- out:write(colorize(format("%04d", ref), band(ot, 31)))
+ out:write(colorize(format("%04d", ref), band(ot, 31), band(sn, 0x100000) ~= 0))
end
out:write(band(sn, 0x10000) == 0 and " " or "|") -- SNAP_FRAME
else
@@ -541,7 +552,12 @@
local function fmterr(err, info)
if type(err) == "number" then
if type(info) == "function" then info = fmtfunc(info) end
- err = format(vmdef.traceerr[err], info)
+ local fmt = vmdef.traceerr[err]
+ if fmt == "NYI: bytecode %s" then
+ local oidx = 6 * info
+ info = sub(vmdef.bcnames, oidx+1, oidx+6)
+ end
+ err = format(fmt, info)
end
return err
end
@@ -582,7 +598,7 @@
end
-- Dump recorded bytecode.
-local function dump_record(tr, func, pc, depth, callee)
+local function dump_record(tr, func, pc, depth)
if depth ~= recdepth then
recdepth = depth
recprefix = rep(" .", depth)
@@ -593,7 +609,6 @@
if dumpmode.H then line = gsub(line, "[<>&]", html_escape) end
else
line = "0000 "..recprefix.." FUNCC \n"
- callee = func
end
if pc <= 0 then
out:write(sub(line, 1, -2), " ; ", fmtfunc(func), "\n")
@@ -607,12 +622,15 @@
------------------------------------------------------------------------------
+local gpr64 = jit.arch:match("64")
+local fprmips32 = jit.arch == "mips" or jit.arch == "mipsel"
+
-- Dump taken trace exits.
local function dump_texit(tr, ex, ngpr, nfpr, ...)
out:write("---- TRACE ", tr, " exit ", ex, "\n")
if dumpmode.X then
local regs = {...}
- if jit.arch == "x64" then
+ if gpr64 then
for i=1,ngpr do
out:write(format(" %016x", regs[i]))
if i % 4 == 0 then out:write("\n") end
@@ -623,7 +641,7 @@
if i % 8 == 0 then out:write("\n") end
end
end
- if jit.arch == "mips" or jit.arch == "mipsel" then
+ if fprmips32 then
for i=1,nfpr,2 do
out:write(format(" %+17.14g", regs[ngpr+i]))
if i % 8 == 7 then out:write("\n") end
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/p.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/p.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/p.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT profiler.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
@@ -41,7 +41,6 @@
-- Cache some library functions and objects.
local jit = require("jit")
-assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
local profile = require("jit.profile")
local vmdef = require("jit.vmdef")
local math = math
@@ -228,9 +227,7 @@
local samples = prof_samples
if samples == 0 then
if prof_raw ~= true then out:write("[No samples collected]\n") end
- return
- end
- if prof_ann then
+ elseif prof_ann then
prof_annotate(prof_count1, samples)
else
prof_top(prof_count1, prof_count2, samples, "")
@@ -238,6 +235,7 @@
prof_count1 = nil
prof_count2 = nil
prof_ud = nil
+ if out ~= stdout then out:close() end
end
end
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/v.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/v.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/v.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- Verbose mode of the LuaJIT compiler.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
@@ -59,11 +59,10 @@
-- Cache some library functions and objects.
local jit = require("jit")
-assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
local jutil = require("jit.util")
local vmdef = require("jit.vmdef")
local funcinfo, traceinfo = jutil.funcinfo, jutil.traceinfo
-local type, format = type, string.format
+local type, sub, format = type, string.sub, string.format
local stdout, stderr = io.stdout, io.stderr
-- Active flag and output file handle.
@@ -90,7 +89,12 @@
local function fmterr(err, info)
if type(err) == "number" then
if type(info) == "function" then info = fmtfunc(info) end
- err = format(vmdef.traceerr[err], info)
+ local fmt = vmdef.traceerr[err]
+ if fmt == "NYI: bytecode %s" then
+ local oidx = 6 * info
+ info = sub(vmdef.bcnames, oidx+1, oidx+6)
+ end
+ err = format(fmt, info)
end
return err
end
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/zone.lua
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/zone.lua 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/jit/zone.lua 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
----------------------------------------------------------------------------
-- LuaJIT profiler zones.
--
--- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
+-- Copyright (C) 2005-2025 Mike Pall. All rights reserved.
-- Released under the MIT license. See Copyright Notice in luajit.h
----------------------------------------------------------------------------
--
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lauxlib.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lauxlib.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lauxlib.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -92,7 +92,6 @@
LUALIB_API void (luaL_setmetatable) (lua_State *L, const char *tname);
-
/*
** {======================================================
** File handles for IO library
@@ -117,6 +116,7 @@
+
/*
** ===============================================================
** some useful macros
@@ -180,7 +180,7 @@
LUALIB_API void (luaL_addvalue) (luaL_Buffer *B);
LUALIB_API void (luaL_pushresult) (luaL_Buffer *B);
-/* -- Luajittex needs this one because it's faster than make it Lua -- */
+/* -- Luajittex needs this one because it's faster than make it with Lua -- */
LUA_API int (RESERVED_lua_dump) (lua_State *L, lua_Writer writer, void *data, int strip);
/* Luajittex needs this one because it overloads loadfile */
LUALIB_API int (RESERVED_load_aux_JIT) (lua_State *L, int status, int envarg);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_aux.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_aux.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_aux.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Auxiliary library for the Lua/C API.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major parts taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -21,6 +21,7 @@
#include "lj_state.h"
#include "lj_trace.h"
#include "lj_lib.h"
+#include "lj_vmevent.h"
#if LJ_TARGET_POSIX
#include <sys/wait.h>
@@ -218,8 +219,15 @@
LUALIB_API void luaL_addlstring(luaL_Buffer *B, const char *s, size_t l)
{
- while (l--)
- luaL_addchar(B, *s++);
+ if (l <= bufffree(B)) {
+ memcpy(B->p, s, l);
+ B->p += l;
+ } else {
+ emptybuffer(B);
+ lua_pushlstring(B->L, s, l);
+ B->lvl++;
+ adjuststack(B);
+ }
}
LUALIB_API void luaL_addstring(luaL_Buffer *B, const char *s)
@@ -311,6 +319,18 @@
return 0;
}
+#ifndef LUAJIT_DISABLE_VMEVENT
+static int error_finalizer(lua_State *L)
+{
+ const char *s = lua_tostring(L, -1);
+ fputs("ERROR in finalizer: ", stderr);
+ fputs(s ? s : "?", stderr);
+ fputc('\n', stderr);
+ fflush(stderr);
+ return 0;
+}
+#endif
+
#ifdef LUAJIT_USE_SYSMALLOC
#if LJ_64 && !LJ_GC64 && !defined(LUAJIT_USE_VALGRIND)
@@ -332,25 +352,39 @@
LUALIB_API lua_State *luaL_newstate(void)
{
lua_State *L = lua_newstate(mem_alloc, NULL);
- if (L) G(L)->panic = panic;
+ if (L) {
+ G(L)->panic = panic;
+#ifndef LUAJIT_DISABLE_VMEVENT
+ luaL_findtable(L, LUA_REGISTRYINDEX, LJ_VMEVENTS_REGKEY, LJ_VMEVENTS_HSIZE);
+ lua_pushcfunction(L, error_finalizer);
+ lua_rawseti(L, -2, VMEVENT_HASH(LJ_VMEVENT_ERRFIN));
+ G(L)->vmevmask = VMEVENT_MASK(LJ_VMEVENT_ERRFIN);
+ L->top--;
+#endif
+ }
return L;
}
#else
-#include "lj_alloc.h"
-
LUALIB_API lua_State *luaL_newstate(void)
{
lua_State *L;
- void *ud = lj_alloc_create();
- if (ud == NULL) return NULL;
#if LJ_64 && !LJ_GC64
- L = lj_state_newstate(lj_alloc_f, ud);
+ L = lj_state_newstate(LJ_ALLOCF_INTERNAL, NULL);
#else
- L = lua_newstate(lj_alloc_f, ud);
+ L = lua_newstate(LJ_ALLOCF_INTERNAL, NULL);
#endif
- if (L) G(L)->panic = panic;
+ if (L) {
+ G(L)->panic = panic;
+#ifndef LUAJIT_DISABLE_VMEVENT
+ luaL_findtable(L, LUA_REGISTRYINDEX, LJ_VMEVENTS_REGKEY, LJ_VMEVENTS_HSIZE);
+ lua_pushcfunction(L, error_finalizer);
+ lua_rawseti(L, -2, VMEVENT_HASH(LJ_VMEVENT_ERRFIN));
+ G(L)->vmevmask = VMEVENT_MASK(LJ_VMEVENT_ERRFIN);
+ L->top--;
+#endif
+ }
return L;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_base.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_base.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_base.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Base and coroutine library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2011 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -19,6 +19,7 @@
#include "lj_gc.h"
#include "lj_err.h"
#include "lj_debug.h"
+#include "lj_buf.h"
#include "lj_str.h"
#include "lj_tab.h"
#include "lj_meta.h"
@@ -42,13 +43,13 @@
LJLIB_ASM(assert) LJLIB_REC(.)
{
- GCstr *s;
lj_lib_checkany(L, 1);
- s = lj_lib_optstr(L, 2);
- if (s)
- lj_err_callermsg(L, strdata(s));
+ if (L->top == L->base+1)
+ lj_err_caller(L, LJ_ERR_ASSERT);
+ else if (tvisstr(L->base+1) || tvisnumber(L->base+1))
+ lj_err_callermsg(L, strdata(lj_lib_checkstr(L, 2)));
else
- lj_err_caller(L, LJ_ERR_ASSERT);
+ lj_err_run(L);
return FFH_UNREACHABLE;
}
@@ -75,9 +76,10 @@
/* This solves a circular dependency problem -- change FF_next_N as needed. */
LJ_STATIC_ASSERT((int)FF_next == FF_next_N);
-LJLIB_ASM(next)
+LJLIB_ASM(next) LJLIB_REC(.)
{
lj_lib_checktab(L, 1);
+ lj_err_msg(L, LJ_ERR_NEXTIDX);
return FFH_UNREACHABLE;
}
@@ -144,6 +146,8 @@
cTValue *o = L->base;
if (!(o < L->top && tvisfunc(o))) {
int level = lj_lib_optint(L, 1, 1);
+ if (level < 0)
+ lj_err_arg(L, 1, LJ_ERR_INVLVL);
o = lj_debug_frame(L, level, &level);
if (o == NULL)
lj_err_arg(L, 1, LJ_ERR_INVLVL);
@@ -166,6 +170,8 @@
setgcref(L->env, obj2gco(t));
return 0;
}
+ if (level < 0)
+ lj_err_arg(L, 1, LJ_ERR_INVLVL);
o = lj_debug_frame(L, level, &level);
if (o == NULL)
lj_err_arg(L, 1, LJ_ERR_INVLVL);
@@ -224,9 +230,11 @@
int32_t n, i = lj_lib_optint(L, 2, 1);
int32_t e = (L->base+3-1 < L->top && !tvisnil(L->base+3-1)) ?
lj_lib_checkint(L, 3) : (int32_t)lj_tab_len(t);
+ uint32_t nu;
if (i > e) return 0;
- n = e - i + 1;
- if (n <= 0 || !lua_checkstack(L, n))
+ nu = (uint32_t)e - (uint32_t)i;
+ n = (int32_t)(nu+1);
+ if (nu >= LUAI_MAXCSTACK || !lua_checkstack(L, n))
lj_err_caller(L, LJ_ERR_UNPACK);
do {
cTValue *tv = lj_tab_getint(t, i);
@@ -287,18 +295,27 @@
} else {
const char *p = strdata(lj_lib_checkstr(L, 1));
char *ep;
+ unsigned int neg = 0;
unsigned long ul;
if (base < 2 || base > 36)
lj_err_arg(L, 2, LJ_ERR_BASERNG);
- ul = strtoul(p, &ep, base);
- if (p != ep) {
- while (lj_char_isspace((unsigned char)(*ep))) ep++;
- if (*ep == '\0') {
- if (LJ_DUALNUM && LJ_LIKELY(ul < 0x80000000u))
- setintV(L->base-1-LJ_FR2, (int32_t)ul);
- else
- setnumV(L->base-1-LJ_FR2, (lua_Number)ul);
- return FFH_RES(1);
+ while (lj_char_isspace((unsigned char)(*p))) p++;
+ if (*p == '-') { p++; neg = 1; } else if (*p == '+') { p++; }
+ if (lj_char_isalnum((unsigned char)(*p))) {
+ ul = strtoul(p, &ep, base);
+ if (p != ep) {
+ while (lj_char_isspace((unsigned char)(*ep))) ep++;
+ if (*ep == '\0') {
+ if (LJ_DUALNUM && LJ_LIKELY(ul < 0x80000000u+neg)) {
+ if (neg) ul = ~ul+1u;
+ setintV(L->base-1-LJ_FR2, (int32_t)ul);
+ } else {
+ lua_Number n = (lua_Number)ul;
+ if (neg) n = -n;
+ setnumV(L->base-1-LJ_FR2, n);
+ }
+ return FFH_RES(1);
+ }
}
}
}
@@ -347,7 +364,11 @@
static int load_aux(lua_State *L, int status, int envarg)
{
if (status == LUA_OK) {
- if (tvistab(L->base+envarg-1)) {
+ /*
+ ** Set environment table for top-level function.
+ ** Don't do this for non-native bytecode, which returns a prototype.
+ */
+ if (tvistab(L->base+envarg-1) && tvisfunc(L->top-1)) {
GCfunc *fn = funcV(L->top-1);
GCtab *t = tabV(L->base+envarg-1);
setgcref(fn->c.env, obj2gco(t));
@@ -395,10 +416,22 @@
GCstr *name = lj_lib_optstr(L, 2);
GCstr *mode = lj_lib_optstr(L, 3);
int status;
- if (L->base < L->top && (tvisstr(L->base) || tvisnumber(L->base))) {
- GCstr *s = lj_lib_checkstr(L, 1);
+ if (L->base < L->top &&
+ (tvisstr(L->base) || tvisnumber(L->base) || tvisbuf(L->base))) {
+ const char *s;
+ MSize len;
+ if (tvisbuf(L->base)) {
+ SBufExt *sbx = bufV(L->base);
+ s = sbx->r;
+ len = sbufxlen(sbx);
+ if (!name) name = &G(L)->strempty; /* Buffers are not NUL-terminated. */
+ } else {
+ GCstr *str = lj_lib_checkstr(L, 1);
+ s = strdata(str);
+ len = str->len;
+ }
lua_settop(L, 4); /* Ensure env arg exists. */
- status = luaL_loadbufferx(L, strdata(s), s->len, strdata(name ? name : s),
+ status = luaL_loadbufferx(L, s, len, name ? strdata(name) : s,
mode ? strdata(mode) : NULL);
} else {
lj_lib_checkfunc(L, 1);
@@ -493,7 +526,8 @@
lua_gettable(L, LUA_GLOBALSINDEX);
tv = L->top-1;
}
- shortcut = (tvisfunc(tv) && funcV(tv)->c.ffid == FF_tostring);
+ shortcut = (tvisfunc(tv) && funcV(tv)->c.ffid == FF_tostring) &&
+ !gcrefu(basemt_it(G(L), LJ_TNUMX));
for (i = 0; i < nargs; i++) {
cTValue *o = &L->base[i];
const char *str;
@@ -590,7 +624,10 @@
setstrV(L, L->base-LJ_FR2, lj_err_str(L, em));
return FFH_RES(2);
}
- lj_state_growstack(co, (MSize)(L->top - L->base));
+ if (lj_state_cpgrowstack(co, (MSize)(L->top - L->base)) != LUA_OK) {
+ cTValue *msg = --co->top;
+ lj_err_callermsg(L, strVdata(msg));
+ }
return FFH_RETRY;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_bit.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_bit.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_bit.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bit manipulation library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lib_bit_c
@@ -98,7 +98,7 @@
x = lj_carith_shift64(x, sh, curr_func(L)->c.ffid - (int)FF_bit_lshift);
return bit_result64(L, id, x);
}
- if (id2) setintV(L->base+1, sh);
+ setintV(L->base+1, sh);
return FFH_RETRY;
#else
lj_lib_checknumber(L, 1);
@@ -155,7 +155,8 @@
#endif
SBuf *sb = lj_buf_tmp_(L);
SFormat sf = (STRFMT_UINT|STRFMT_T_HEX);
- if (n < 0) { n = -n; sf |= STRFMT_F_UPPER; }
+ if (n < 0) { n = (int32_t)(~(uint32_t)n+1u); sf |= STRFMT_F_UPPER; }
+ if ((uint32_t)n > 254) n = 254;
sf |= ((SFormat)((n+1)&255) << STRFMT_SH_PREC);
#if LJ_HASFFI
if (n < 16) b &= ((uint64_t)1 << 4*n)-1;
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_buffer.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_buffer.c (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_buffer.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,360 @@
+/*
+** Buffer library.
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
+*/
+
+#define lib_buffer_c
+#define LUA_LIB
+
+#include "lua.h"
+#include "lauxlib.h"
+#include "lualib.h"
+
+#include "lj_obj.h"
+
+#if LJ_HASBUFFER
+#include "lj_gc.h"
+#include "lj_err.h"
+#include "lj_buf.h"
+#include "lj_str.h"
+#include "lj_tab.h"
+#include "lj_udata.h"
+#include "lj_meta.h"
+#if LJ_HASFFI
+#include "lj_ctype.h"
+#include "lj_cdata.h"
+#include "lj_cconv.h"
+#endif
+#include "lj_strfmt.h"
+#include "lj_serialize.h"
+#include "lj_lib.h"
+
+/* -- Helper functions ---------------------------------------------------- */
+
+/* Check that the first argument is a string buffer. */
+static SBufExt *buffer_tobuf(lua_State *L)
+{
+ if (!(L->base < L->top && tvisbuf(L->base)))
+ lj_err_argtype(L, 1, "buffer");
+ return bufV(L->base);
+}
+
+/* Ditto, but for writers. */
+static LJ_AINLINE SBufExt *buffer_tobufw(lua_State *L)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ setsbufXL_(sbx, L);
+ return sbx;
+}
+
+#define buffer_toudata(sbx) ((GCudata *)(sbx)-1)
+
+/* -- Buffer methods ------------------------------------------------------ */
+
+#define LJLIB_MODULE_buffer_method
+
+LJLIB_CF(buffer_method_free)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ lj_bufx_free(L, sbx);
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_reset) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ lj_bufx_reset(sbx);
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_skip) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ MSize n = (MSize)lj_lib_checkintrange(L, 2, 0, LJ_MAX_BUF);
+ MSize len = sbufxlen(sbx);
+ if (n < len) {
+ sbx->r += n;
+ } else if (sbufiscow(sbx)) {
+ sbx->r = sbx->w;
+ } else {
+ sbx->r = sbx->w = sbx->b;
+ }
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_set) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ GCobj *ref;
+ const char *p;
+ MSize len;
+#if LJ_HASFFI
+ if (tviscdata(L->base+1)) {
+ CTState *cts = ctype_cts(L);
+ lj_cconv_ct_tv(cts, ctype_get(cts, CTID_P_CVOID), (uint8_t *)&p,
+ L->base+1, CCF_ARG(2));
+ len = (MSize)lj_lib_checkintrange(L, 3, 0, LJ_MAX_BUF);
+ } else
+#endif
+ {
+ GCstr *str = lj_lib_checkstrx(L, 2);
+ p = strdata(str);
+ len = str->len;
+ }
+ lj_bufx_free(L, sbx);
+ lj_bufx_set_cow(L, sbx, p, len);
+ ref = gcV(L->base+1);
+ setgcref(sbx->cowref, ref);
+ lj_gc_objbarrier(L, buffer_toudata(sbx), ref);
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_put) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ ptrdiff_t arg, narg = L->top - L->base;
+ for (arg = 1; arg < narg; arg++) {
+ cTValue *o = &L->base[arg], *mo = NULL;
+ retry:
+ if (tvisstr(o)) {
+ lj_buf_putstr((SBuf *)sbx, strV(o));
+ } else if (tvisint(o)) {
+ lj_strfmt_putint((SBuf *)sbx, intV(o));
+ } else if (tvisnum(o)) {
+ lj_strfmt_putfnum((SBuf *)sbx, STRFMT_G14, numV(o));
+ } else if (tvisbuf(o)) {
+ SBufExt *sbx2 = bufV(o);
+ if (sbx2 == sbx) lj_err_arg(L, (int)(arg+1), LJ_ERR_BUFFER_SELF);
+ lj_buf_putmem((SBuf *)sbx, sbx2->r, sbufxlen(sbx2));
+ } else if (!mo && !tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
+ /* Call __tostring metamethod inline. */
+ copyTV(L, L->top++, mo);
+ copyTV(L, L->top++, o);
+ lua_call(L, 1, 1);
+ o = &L->base[arg]; /* The stack may have been reallocated. */
+ copyTV(L, &L->base[arg], L->top-1);
+ L->top = L->base + narg;
+ goto retry; /* Retry with the result. */
+ } else {
+ lj_err_argtype(L, (int)(arg+1), "string/number/__tostring");
+ }
+ /* Probably not useful to inline other __tostring MMs, e.g. FFI numbers. */
+ }
+ L->top = L->base+1; /* Chain buffer object. */
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_method_putf) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ lj_strfmt_putarg(L, (SBuf *)sbx, 2, 2);
+ L->top = L->base+1; /* Chain buffer object. */
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_method_get) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ ptrdiff_t arg, narg = L->top - L->base;
+ if (narg == 1) {
+ narg++;
+ setnilV(L->top++); /* get() is the same as get(nil). */
+ }
+ for (arg = 1; arg < narg; arg++) {
+ TValue *o = &L->base[arg];
+ MSize n = tvisnil(o) ? LJ_MAX_BUF :
+ (MSize) lj_lib_checkintrange(L, (int)(arg+1), 0, LJ_MAX_BUF);
+ MSize len = sbufxlen(sbx);
+ if (n > len) n = len;
+ setstrV(L, o, lj_str_new(L, sbx->r, n));
+ sbx->r += n;
+ }
+ if (sbx->r == sbx->w && !sbufiscow(sbx)) sbx->r = sbx->w = sbx->b;
+ lj_gc_check(L);
+ return (int)(narg-1);
+}
+
+#if LJ_HASFFI
+LJLIB_CF(buffer_method_putcdata) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ const char *p;
+ MSize len;
+ if (tviscdata(L->base+1)) {
+ CTState *cts = ctype_cts(L);
+ lj_cconv_ct_tv(cts, ctype_get(cts, CTID_P_CVOID), (uint8_t *)&p,
+ L->base+1, CCF_ARG(2));
+ } else {
+ lj_err_argtype(L, 2, "cdata");
+ }
+ len = (MSize)lj_lib_checkintrange(L, 3, 0, LJ_MAX_BUF);
+ lj_buf_putmem((SBuf *)sbx, p, len);
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_reserve) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ MSize sz = (MSize)lj_lib_checkintrange(L, 2, 0, LJ_MAX_BUF);
+ GCcdata *cd;
+ lj_buf_more((SBuf *)sbx, sz);
+ ctype_loadffi(L);
+ cd = lj_cdata_new_(L, CTID_P_UINT8, CTSIZE_PTR);
+ *(void **)cdataptr(cd) = sbx->w;
+ setcdataV(L, L->top++, cd);
+ setintV(L->top++, sbufleft(sbx));
+ return 2;
+}
+
+LJLIB_CF(buffer_method_commit) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ MSize len = (MSize)lj_lib_checkintrange(L, 2, 0, LJ_MAX_BUF);
+ if (len > sbufleft(sbx)) lj_err_arg(L, 2, LJ_ERR_NUMRNG);
+ sbx->w += len;
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_ref) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ GCcdata *cd;
+ ctype_loadffi(L);
+ cd = lj_cdata_new_(L, CTID_P_UINT8, CTSIZE_PTR);
+ *(void **)cdataptr(cd) = sbx->r;
+ setcdataV(L, L->top++, cd);
+ setintV(L->top++, sbufxlen(sbx));
+ return 2;
+}
+#endif
+
+LJLIB_CF(buffer_method_encode) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ cTValue *o = lj_lib_checkany(L, 2);
+ lj_serialize_put(sbx, o);
+ lj_gc_check(L);
+ L->top = L->base+1; /* Chain buffer object. */
+ return 1;
+}
+
+LJLIB_CF(buffer_method_decode) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobufw(L);
+ setnilV(L->top++);
+ sbx->r = lj_serialize_get(sbx, L->top-1);
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_method___gc)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ lj_bufx_free(L, sbx);
+ return 0;
+}
+
+LJLIB_CF(buffer_method___tostring) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ setstrV(L, L->top-1, lj_str_new(L, sbx->r, sbufxlen(sbx)));
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_method___len) LJLIB_REC(.)
+{
+ SBufExt *sbx = buffer_tobuf(L);
+ setintV(L->top-1, (int32_t)sbufxlen(sbx));
+ return 1;
+}
+
+LJLIB_PUSH("buffer") LJLIB_SET(__metatable)
+LJLIB_PUSH(top-1) LJLIB_SET(__index)
+
+/* -- Buffer library functions -------------------------------------------- */
+
+#define LJLIB_MODULE_buffer
+
+LJLIB_PUSH(top-2) LJLIB_SET(!) /* Set environment. */
+
+LJLIB_CF(buffer_new)
+{
+ MSize sz = 0;
+ int targ = 1;
+ GCtab *env, *dict_str = NULL, *dict_mt = NULL;
+ GCudata *ud;
+ SBufExt *sbx;
+ if (L->base < L->top && !tvistab(L->base)) {
+ targ = 2;
+ if (!tvisnil(L->base))
+ sz = (MSize)lj_lib_checkintrange(L, 1, 0, LJ_MAX_BUF);
+ }
+ if (L->base+targ-1 < L->top) {
+ GCtab *options = lj_lib_checktab(L, targ);
+ cTValue *opt_dict, *opt_mt;
+ opt_dict = lj_tab_getstr(options, lj_str_newlit(L, "dict"));
+ if (opt_dict && tvistab(opt_dict)) {
+ dict_str = tabV(opt_dict);
+ lj_serialize_dict_prep_str(L, dict_str);
+ }
+ opt_mt = lj_tab_getstr(options, lj_str_newlit(L, "metatable"));
+ if (opt_mt && tvistab(opt_mt)) {
+ dict_mt = tabV(opt_mt);
+ lj_serialize_dict_prep_mt(L, dict_mt);
+ }
+ }
+ env = tabref(curr_func(L)->c.env);
+ ud = lj_udata_new(L, sizeof(SBufExt), env);
+ ud->udtype = UDTYPE_BUFFER;
+ /* NOBARRIER: The GCudata is new (marked white). */
+ setgcref(ud->metatable, obj2gco(env));
+ setudataV(L, L->top++, ud);
+ sbx = (SBufExt *)uddata(ud);
+ lj_bufx_init(L, sbx);
+ setgcref(sbx->dict_str, obj2gco(dict_str));
+ setgcref(sbx->dict_mt, obj2gco(dict_mt));
+ if (sz > 0) lj_buf_need2((SBuf *)sbx, sz);
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_encode) LJLIB_REC(.)
+{
+ cTValue *o = lj_lib_checkany(L, 1);
+ setstrV(L, L->top++, lj_serialize_encode(L, o));
+ lj_gc_check(L);
+ return 1;
+}
+
+LJLIB_CF(buffer_decode) LJLIB_REC(.)
+{
+ GCstr *str = lj_lib_checkstrx(L, 1);
+ setnilV(L->top++);
+ lj_serialize_decode(L, L->top-1, str);
+ lj_gc_check(L);
+ return 1;
+}
+
+/* ------------------------------------------------------------------------ */
+
+#include "lj_libdef.h"
+
+int luaopen_string_buffer(lua_State *L)
+{
+ LJ_LIB_REG(L, NULL, buffer_method);
+ lua_getfield(L, -1, "__tostring");
+ lua_setfield(L, -2, "tostring");
+ LJ_LIB_REG(L, NULL, buffer);
+ return 1;
+}
+
+#endif
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_buffer.c
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_debug.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_debug.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_debug.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Debug library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -231,8 +231,8 @@
int32_t n = lj_lib_checkint(L, 2) - 1;
if ((uint32_t)n >= fn->l.nupvalues)
lj_err_arg(L, 2, LJ_ERR_IDXRNG);
- setlightudV(L->top-1, isluafunc(fn) ? (void *)gcref(fn->l.uvptr[n]) :
- (void *)&fn->c.upvalue[n]);
+ lua_pushlightuserdata(L, isluafunc(fn) ? (void *)gcref(fn->l.uvptr[n]) :
+ (void *)&fn->c.upvalue[n]);
return 1;
}
@@ -283,13 +283,13 @@
/* ------------------------------------------------------------------------ */
-#define KEY_HOOK ((void *)0x3004)
+#define KEY_HOOK (U64x(80000000,00000000)|'h')
static void hookf(lua_State *L, lua_Debug *ar)
{
static const char *const hooknames[] =
{"call", "return", "line", "count", "tail return"};
- lua_pushlightuserdata(L, KEY_HOOK);
+ (L->top++)->u64 = KEY_HOOK;
lua_rawget(L, LUA_REGISTRYINDEX);
if (lua_isfunction(L, -1)) {
lua_pushstring(L, hooknames[(int)ar->event]);
@@ -334,7 +334,7 @@
count = luaL_optint(L, arg+3, 0);
func = hookf; mask = makemask(smask, count);
}
- lua_pushlightuserdata(L, KEY_HOOK);
+ (L->top++)->u64 = KEY_HOOK;
lua_pushvalue(L, arg+1);
lua_rawset(L, LUA_REGISTRYINDEX);
lua_sethook(L, func, mask, count);
@@ -349,7 +349,7 @@
if (hook != NULL && hook != hookf) { /* external hook? */
lua_pushliteral(L, "external hook");
} else {
- lua_pushlightuserdata(L, KEY_HOOK);
+ (L->top++)->u64 = KEY_HOOK;
lua_rawget(L, LUA_REGISTRYINDEX); /* get hook */
}
lua_pushstring(L, unmakemask(mask, buff));
@@ -369,7 +369,8 @@
return 0;
if (luaL_loadbuffer(L, buffer, strlen(buffer), "=(debug command)") ||
lua_pcall(L, 0, 0, 0)) {
- fputs(lua_tostring(L, -1), stderr);
+ const char *s = lua_tostring(L, -1);
+ fputs(s ? s : "(error object is not a string)", stderr);
fputs("\n", stderr);
}
lua_settop(L, 0); /* remove eventual returns */
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_ffi.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_ffi.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_ffi.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lib_ffi_c
@@ -305,7 +305,7 @@
p = *(void **)p;
} else if (ctype_isenum(ct->info)) {
msg = "cdata<%s>: %d";
- p = (void *)(uintptr_t)*(uint32_t **)p;
+ p = (void *)(uintptr_t)*(uint32_t *)p;
} else {
if (ctype_isptr(ct->info)) {
p = cdata_getptr(p, ct->size);
@@ -513,7 +513,7 @@
/* Handle ctype __gc metamethod. Use the fast lookup here. */
cTValue *tv = lj_tab_getinth(cts->miscmap, -(int32_t)id);
if (tv && tvistab(tv) && (tv = lj_meta_fast(L, tabV(tv), MM_gc))) {
- GCtab *t = cts->finalizer;
+ GCtab *t = tabref(G(L)->gcroot[GCROOT_FFI_FIN]);
if (gcref(t->metatable)) {
/* Add to finalizer table, if still enabled. */
copyTV(L, lj_tab_set(L, t, o-1), tv);
@@ -573,6 +573,7 @@
setintV(lj_tab_setstr(L, t, lj_str_newlit(L, "sib")), (int32_t)ct->sib);
if (gcref(ct->name)) {
GCstr *s = gco2str(gcref(ct->name));
+ if (isdead(G(L), obj2gco(s))) flipwhite(obj2gco(s));
setstrV(L, lj_tab_setstr(L, t, lj_str_newlit(L, "name")), s);
}
lj_gc_check(L);
@@ -638,7 +639,7 @@
CTState *cts = ctype_cts(L);
CTypeID id = ffi_checkctype(L, cts, NULL);
CTSize sz = 0;
- CTInfo info = lj_ctype_info(cts, id, &sz);
+ CTInfo info = lj_ctype_info_raw(cts, id, &sz);
setintV(L->top-1, 1 << ctype_align(info));
return 1;
}
@@ -720,49 +721,52 @@
return 0;
}
-#define H_(le, be) LJ_ENDIAN_SELECT(0x##le, 0x##be)
-
/* Test ABI string. */
LJLIB_CF(ffi_abi) LJLIB_REC(.)
{
GCstr *s = lj_lib_checkstr(L, 1);
- int b = 0;
- switch (s->hash) {
+ int b = lj_cparse_case(s,
#if LJ_64
- case H_(849858eb,ad35fd06): b = 1; break; /* 64bit */
+ "\00564bit"
#else
- case H_(662d3c79,d0e22477): b = 1; break; /* 32bit */
+ "\00532bit"
#endif
#if LJ_ARCH_HASFPU
- case H_(e33ee463,e33ee463): b = 1; break; /* fpu */
+ "\003fpu"
#endif
#if LJ_ABI_SOFTFP
- case H_(61211a23,c2e8c81c): b = 1; break; /* softfp */
+ "\006softfp"
#else
- case H_(539417a8,8ce0812f): b = 1; break; /* hardfp */
+ "\006hardfp"
#endif
#if LJ_ABI_EABI
- case H_(2182df8f,f2ed1152): b = 1; break; /* eabi */
+ "\004eabi"
#endif
#if LJ_ABI_WIN
- case H_(4ab624a8,4ab624a8): b = 1; break; /* win */
+ "\003win"
#endif
- case H_(3af93066,1f001464): b = 1; break; /* le/be */
+#if LJ_ABI_PAUTH
+ "\005pauth"
+#endif
+#if LJ_TARGET_UWP
+ "\003uwp"
+#endif
+#if LJ_LE
+ "\002le"
+#else
+ "\002be"
+#endif
#if LJ_GC64
- case H_(9e89d2c9,13c83c92): b = 1; break; /* gc64 */
+ "\004gc64"
#endif
- default:
- break;
- }
+ ) >= 0;
setboolV(L->top-1, b);
setboolV(&G(L)->tmptv2, b); /* Remember for trace recorder. */
return 1;
}
-#undef H_
+LJLIB_PUSH(top-7) LJLIB_SET(!) /* Store reference to miscmap table. */
-LJLIB_PUSH(top-8) LJLIB_SET(!) /* Store reference to miscmap table. */
-
LJLIB_CF(ffi_metatype)
{
CTState *cts = ctype_cts(L);
@@ -769,13 +773,13 @@
CTypeID id = ffi_checkctype(L, cts, NULL);
GCtab *mt = lj_lib_checktab(L, 2);
GCtab *t = cts->miscmap;
- CType *ct = ctype_get(cts, id); /* Only allow raw types. */
+ CType *ct = ctype_raw(cts, id);
TValue *tv;
GCcdata *cd;
if (!(ctype_isstruct(ct->info) || ctype_iscomplex(ct->info) ||
ctype_isvector(ct->info)))
lj_err_arg(L, 1, LJ_ERR_FFI_INVTYPE);
- tv = lj_tab_setinth(L, t, -(int32_t)id);
+ tv = lj_tab_setinth(L, t, -(int32_t)ctype_typeid(cts, ct));
if (!tvisnil(tv))
lj_err_caller(L, LJ_ERR_PROTMT);
settabV(L, tv, mt);
@@ -787,8 +791,6 @@
return 1;
}
-LJLIB_PUSH(top-7) LJLIB_SET(!) /* Store reference to finalizer table. */
-
LJLIB_CF(ffi_gc) LJLIB_REC(.)
{
GCcdata *cd = ffi_checkcdata(L, 1);
@@ -821,19 +823,6 @@
/* ------------------------------------------------------------------------ */
-/* Create special weak-keyed finalizer table. */
-static GCtab *ffi_finalizer(lua_State *L)
-{
- /* NOBARRIER: The table is new (marked white). */
- GCtab *t = lj_tab_new(L, 0, 1);
- settabV(L, L->top++, t);
- setgcref(t->metatable, obj2gco(t));
- setstrV(L, lj_tab_setstr(L, t, lj_str_newlit(L, "__mode")),
- lj_str_newlit(L, "k"));
- t->nomm = (uint8_t)(~(1u<<MM_mode));
- return t;
-}
-
/* Register FFI module as loaded. */
static void ffi_register_module(lua_State *L)
{
@@ -849,7 +838,6 @@
{
CTState *cts = lj_ctype_init(L);
settabV(L, L->top++, (cts->miscmap = lj_tab_new(L, 0, 1)));
- cts->finalizer = ffi_finalizer(L);
LJ_LIB_REG(L, NULL, ffi_meta);
/* NOBARRIER: basemt is a GC root. */
setgcref(basemt_it(G(L), LJ_TCDATA), obj2gco(tabV(L->top-1)));
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_init.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_init.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_init.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Library initialization.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major parts taken verbatim from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_io.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_io.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_io.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** I/O library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2011 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -60,12 +60,12 @@
return iof;
}
-static FILE *io_stdfile(lua_State *L, ptrdiff_t id)
+static IOFileUD *io_stdfile(lua_State *L, ptrdiff_t id)
{
IOFileUD *iof = IOSTDF_IOF(L, id);
if (iof->fp == NULL)
lj_err_caller(L, LJ_ERR_IOSTDCL);
- return iof->fp;
+ return iof;
}
static IOFileUD *io_file_new(lua_State *L)
@@ -99,11 +99,8 @@
int stat = -1;
#if LJ_TARGET_POSIX
stat = pclose(iof->fp);
-#elif LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE
+#elif LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE && !LJ_TARGET_UWP
stat = _pclose(iof->fp);
-#else
- lua_assert(0);
- return 0;
#endif
#if LJ_52
iof->fp = NULL;
@@ -112,7 +109,8 @@
ok = (stat != -1);
#endif
} else {
- lua_assert((iof->type & IOFILE_TYPE_MASK) == IOFILE_TYPE_STDF);
+ lj_assertL((iof->type & IOFILE_TYPE_MASK) == IOFILE_TYPE_STDF,
+ "close of unknown FILE* type");
setnilV(L->top++);
lua_pushliteral(L, "cannot close standard file");
return 2;
@@ -180,7 +178,7 @@
MSize n = (MSize)fread(buf, 1, m, fp);
setstrV(L, L->top++, lj_str_new(L, buf, (size_t)n));
lj_gc_check(L);
- return (n > 0 || m == 0);
+ return n > 0;
} else {
int c = getc(fp);
ungetc(c, fp);
@@ -189,8 +187,9 @@
}
}
-static int io_file_read(lua_State *L, FILE *fp, int start)
+static int io_file_read(lua_State *L, IOFileUD *iof, int start)
{
+ FILE *fp = iof->fp;
int ok, n, nargs = (int)(L->top - L->base) - start;
clearerr(fp);
if (nargs == 0) {
@@ -226,8 +225,9 @@
return n - start;
}
-static int io_file_write(lua_State *L, FILE *fp, int start)
+static int io_file_write(lua_State *L, IOFileUD *iof, int start)
{
+ FILE *fp = iof->fp;
cTValue *tv;
int status = 1;
for (tv = L->base+start; tv < L->top; tv++) {
@@ -255,13 +255,11 @@
lj_err_caller(L, LJ_ERR_IOCLFL);
L->top = L->base;
if (n) { /* Copy upvalues with options to stack. */
- if (n > LUAI_MAXCSTACK)
- lj_err_caller(L, LJ_ERR_STKOV);
lj_state_checkstack(L, (MSize)n);
memcpy(L->top, &fn->c.upvalue[1], n*sizeof(TValue));
L->top += n;
}
- n = io_file_read(L, iof->fp, 0);
+ n = io_file_read(L, iof, 0);
if (ferror(iof->fp))
lj_err_callermsg(L, strVdata(L->top-2));
if (tvisnil(L->base) && (iof->type & IOFILE_FLAG_CLOSE)) {
@@ -286,19 +284,25 @@
LJLIB_CF(io_method_close)
{
- IOFileUD *iof = L->base < L->top ? io_tofile(L) :
- IOSTDF_IOF(L, GCROOT_IO_OUTPUT);
+ IOFileUD *iof;
+ if (L->base < L->top) {
+ iof = io_tofile(L);
+ } else {
+ iof = IOSTDF_IOF(L, GCROOT_IO_OUTPUT);
+ if (iof->fp == NULL)
+ lj_err_caller(L, LJ_ERR_IOCLFL);
+ }
return io_file_close(L, iof);
}
LJLIB_CF(io_method_read)
{
- return io_file_read(L, io_tofile(L)->fp, 1);
+ return io_file_read(L, io_tofile(L), 1);
}
LJLIB_CF(io_method_write) LJLIB_REC(io_write 0)
{
- return io_file_write(L, io_tofile(L)->fp, 1);
+ return io_file_write(L, io_tofile(L), 1);
}
LJLIB_CF(io_method_flush) LJLIB_REC(io_flush 0)
@@ -306,6 +310,14 @@
return luaL_fileresult(L, fflush(io_tofile(L)->fp) == 0, NULL);
}
+#if LJ_32 && defined(__ANDROID__) && __ANDROID_API__ < 24
+/* The Android NDK is such an unmatched marvel of engineering. */
+extern int fseeko32(FILE *, long int, int) __asm__("fseeko");
+extern long int ftello32(FILE *) __asm__("ftello");
+#define fseeko(fp, pos, whence) (fseeko32((fp), (pos), (whence)))
+#define ftello(fp) (ftello32((fp)))
+#endif
+
LJLIB_CF(io_method_seek)
{
FILE *fp = io_tofile(L)->fp;
@@ -406,7 +418,7 @@
LJLIB_CF(io_popen)
{
-#if LJ_TARGET_POSIX || (LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE)
+#if LJ_TARGET_POSIX || (LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE && !LJ_TARGET_UWP)
const char *fname = strdata(lj_lib_checkstr(L, 1));
GCstr *s = lj_lib_optstr(L, 2);
const char *mode = s ? strdata(s) : "r";
@@ -427,7 +439,7 @@
LJLIB_CF(io_tmpfile)
{
IOFileUD *iof = io_file_new(L);
-#if LJ_TARGET_PS3 || LJ_TARGET_PS4 || LJ_TARGET_PSVITA
+#if LJ_TARGET_PS3 || LJ_TARGET_PS4 || LJ_TARGET_PS5 || LJ_TARGET_PSVITA || LJ_TARGET_NX
iof->fp = NULL; errno = ENOSYS;
#else
iof->fp = tmpfile();
@@ -452,7 +464,7 @@
LJLIB_CF(io_flush) LJLIB_REC(io_flush GCROOT_IO_OUTPUT)
{
- return luaL_fileresult(L, fflush(io_stdfile(L, GCROOT_IO_OUTPUT)) == 0, NULL);
+ return luaL_fileresult(L, fflush(io_stdfile(L, GCROOT_IO_OUTPUT)->fp) == 0, NULL);
}
static int io_std_getset(lua_State *L, ptrdiff_t id, const char *mode)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_jit.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_jit.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_jit.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** JIT library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lib_jit_c
@@ -104,8 +104,8 @@
jit_State *J = L2J(L);
L->top = L->base;
setboolV(L->top++, (J->flags & JIT_F_ON) ? 1 : 0);
- flagbits_to_strings(L, J->flags, JIT_F_CPU_FIRST, JIT_F_CPUSTRING);
- flagbits_to_strings(L, J->flags, JIT_F_OPT_FIRST, JIT_F_OPTSTRING);
+ flagbits_to_strings(L, J->flags, JIT_F_CPU, JIT_F_CPUSTRING);
+ flagbits_to_strings(L, J->flags, JIT_F_OPT, JIT_F_OPTSTRING);
return (int)(L->top - L->base);
#else
setboolV(L->top++, 0);
@@ -113,6 +113,13 @@
#endif
}
+LJLIB_CF(jit_security)
+{
+ int idx = lj_lib_checkopt(L, 1, -1, LJ_SECURITY_MODESTRING);
+ setintV(L->top++, ((LJ_SECURITY_MODE >> (2*idx)) & 3));
+ return 1;
+}
+
LJLIB_CF(jit_attach)
{
#ifdef LUAJIT_DISABLE_VMEVENT
@@ -154,24 +161,6 @@
/* -- Reflection API for Lua functions ------------------------------------ */
-/* Return prototype of first argument (Lua function or prototype object) */
-static GCproto *check_Lproto(lua_State *L, int nolua)
-{
- TValue *o = L->base;
- if (L->top > o) {
- if (tvisproto(o)) {
- return protoV(o);
- } else if (tvisfunc(o)) {
- if (isluafunc(funcV(o)))
- return funcproto(funcV(o));
- else if (nolua)
- return NULL;
- }
- }
- lj_err_argt(L, 1, LUA_TFUNCTION);
- return NULL; /* unreachable */
-}
-
static void setintfield(lua_State *L, GCtab *t, const char *name, int32_t val)
{
setintV(lj_tab_setstr(L, t, lj_str_newz(L, name)), val);
@@ -180,7 +169,7 @@
/* local info = jit.util.funcinfo(func [,pc]) */
LJLIB_CF(jit_util_funcinfo)
{
- GCproto *pt = check_Lproto(L, 1);
+ GCproto *pt = lj_lib_checkLproto(L, 1, 1);
if (pt) {
BCPos pc = (BCPos)lj_lib_optint(L, 2, 0);
GCtab *t;
@@ -222,12 +211,12 @@
/* local ins, m = jit.util.funcbc(func, pc) */
LJLIB_CF(jit_util_funcbc)
{
- GCproto *pt = check_Lproto(L, 0);
+ GCproto *pt = lj_lib_checkLproto(L, 1, 0);
BCPos pc = (BCPos)lj_lib_checkint(L, 2);
if (pc < pt->sizebc) {
BCIns ins = proto_bc(pt)[pc];
BCOp op = bc_op(ins);
- lua_assert(op < BC__MAX);
+ lj_assertL(op < BC__MAX, "bad bytecode op %d", op);
setintV(L->top, ins);
setintV(L->top+1, lj_bc_mode[op]);
L->top += 2;
@@ -239,7 +228,7 @@
/* local k = jit.util.funck(func, idx) */
LJLIB_CF(jit_util_funck)
{
- GCproto *pt = check_Lproto(L, 0);
+ GCproto *pt = lj_lib_checkLproto(L, 1, 0);
ptrdiff_t idx = (ptrdiff_t)lj_lib_checkint(L, 2);
if (idx >= 0) {
if (idx < (ptrdiff_t)pt->sizekn) {
@@ -259,7 +248,7 @@
/* local name = jit.util.funcuvname(func, idx) */
LJLIB_CF(jit_util_funcuvname)
{
- GCproto *pt = check_Lproto(L, 0);
+ GCproto *pt = lj_lib_checkLproto(L, 1, 0);
uint32_t idx = (uint32_t)lj_lib_checkint(L, 2);
if (idx < pt->sizeuv) {
setstrV(L, L->top-1, lj_str_newz(L, lj_debug_uvname(pt, idx)));
@@ -339,11 +328,7 @@
ir = &T->ir[ir->op1];
}
#if LJ_HASFFI
- if (ir->o == IR_KINT64 && !ctype_ctsG(G(L))) {
- ptrdiff_t oldtop = savestack(L, L->top);
- luaopen_ffi(L); /* Load FFI library on-demand. */
- L->top = restorestack(L, oldtop);
- }
+ if (ir->o == IR_KINT64) ctype_loadffi(L);
#endif
lj_ir_kvalue(L, L->top-2, ir);
setintV(L->top-1, (int32_t)irt_type(ir->t));
@@ -419,7 +404,8 @@
{
uint32_t idx = (uint32_t)lj_lib_checkint(L, 1);
if (idx < IRCALL__MAX) {
- setintptrV(L->top-1, (intptr_t)(void *)lj_ir_callinfo[idx].func);
+ ASMFunction func = lj_ir_callinfo[idx].func;
+ setintptrV(L->top-1, (intptr_t)(void *)lj_ptr_strip(func));
return 1;
}
return 0;
@@ -471,7 +457,7 @@
str += str[2] == '-' ? 3 : 2;
set = 0;
}
- for (opt = JIT_F_OPT_FIRST; ; opt <<= 1) {
+ for (opt = JIT_F_OPT; ; opt <<= 1) {
size_t len = *(const uint8_t *)lst;
if (len == 0)
break;
@@ -491,7 +477,7 @@
int i;
for (i = 0; i < JIT_P__MAX; i++) {
size_t len = *(const uint8_t *)lst;
- lua_assert(len != 0);
+ lj_assertJ(len != 0, "bad JIT_P_STRING");
if (strncmp(str, lst+1, len) == 0 && str[len] == '=') {
int32_t n = 0;
const char *p = &str[len+1];
@@ -540,8 +526,8 @@
/* Not loaded by default, use: local profile = require("jit.profile") */
-static const char KEY_PROFILE_THREAD = 't';
-static const char KEY_PROFILE_FUNC = 'f';
+#define KEY_PROFILE_THREAD (U64x(80000000,00000000)|'t')
+#define KEY_PROFILE_FUNC (U64x(80000000,00000000)|'f')
static void jit_profile_callback(lua_State *L2, lua_State *L, int samples,
int vmstate)
@@ -548,7 +534,7 @@
{
TValue key;
cTValue *tv;
- setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
+ key.u64 = KEY_PROFILE_FUNC;
tv = lj_tab_get(L, tabV(registry(L)), &key);
if (tvisfunc(tv)) {
char vmst = (char)vmstate;
@@ -575,9 +561,9 @@
lua_State *L2 = lua_newthread(L); /* Thread that runs profiler callback. */
TValue key;
/* Anchor thread and function in registry. */
- setlightudV(&key, (void *)&KEY_PROFILE_THREAD);
+ key.u64 = KEY_PROFILE_THREAD;
setthreadV(L, lj_tab_set(L, registry, &key), L2);
- setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
+ key.u64 = KEY_PROFILE_FUNC;
setfuncV(L, lj_tab_set(L, registry, &key), func);
lj_gc_anybarriert(L, registry);
luaJIT_profile_start(L, mode ? strdata(mode) : "",
@@ -592,9 +578,9 @@
TValue key;
luaJIT_profile_stop(L);
registry = tabV(registry(L));
- setlightudV(&key, (void *)&KEY_PROFILE_THREAD);
+ key.u64 = KEY_PROFILE_THREAD;
setnilV(lj_tab_set(L, registry, &key));
- setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
+ key.u64 = KEY_PROFILE_FUNC;
setnilV(lj_tab_set(L, registry, &key));
lj_gc_anybarriert(L, registry);
return 0;
@@ -640,49 +626,32 @@
#undef JIT_PARAMINIT
0
};
-#endif
#if LJ_TARGET_ARM && LJ_TARGET_LINUX
#include <sys/utsname.h>
#endif
-/* Arch-dependent CPU detection. */
-static uint32_t jit_cpudetect(lua_State *L)
+/* Arch-dependent CPU feature detection. */
+static uint32_t jit_cpudetect(void)
{
uint32_t flags = 0;
#if LJ_TARGET_X86ORX64
+
uint32_t vendor[4];
uint32_t features[4];
if (lj_vm_cpuid(0, vendor) && lj_vm_cpuid(1, features)) {
-#if !LJ_HASJIT
-#define JIT_F_SSE2 2
-#endif
- flags |= ((features[3] >> 26)&1) * JIT_F_SSE2;
-#if LJ_HASJIT
flags |= ((features[2] >> 0)&1) * JIT_F_SSE3;
flags |= ((features[2] >> 19)&1) * JIT_F_SSE4_1;
- if (vendor[2] == 0x6c65746e) { /* Intel. */
- if ((features[0] & 0x0fff0ff0) == 0x000106c0) /* Atom. */
- flags |= JIT_F_LEA_AGU;
- } else if (vendor[2] == 0x444d4163) { /* AMD. */
- uint32_t fam = (features[0] & 0x0ff00f00);
- if (fam >= 0x00000f00) /* K8, K10. */
- flags |= JIT_F_PREFER_IMUL;
- }
if (vendor[0] >= 7) {
uint32_t xfeatures[4];
lj_vm_cpuid(7, xfeatures);
flags |= ((xfeatures[1] >> 8)&1) * JIT_F_BMI2;
}
-#endif
}
- /* Check for required instruction set support on x86 (unnecessary on x64). */
-#if LJ_TARGET_X86
- if (!(flags & JIT_F_SSE2))
- luaL_error(L, "CPU with SSE2 required");
-#endif
+ /* Don't bother checking for SSE2 -- the VM will crash before getting here. */
+
#elif LJ_TARGET_ARM
-#if LJ_HASJIT
+
int ver = LJ_ARCH_VERSION; /* Compile-time ARM CPU detection. */
#if LJ_TARGET_LINUX
if (ver < 70) { /* Runtime ARM CPU detection. */
@@ -689,10 +658,9 @@
struct utsname ut;
uname(&ut);
if (strncmp(ut.machine, "armv", 4) == 0) {
- if (ut.machine[4] >= '7')
- ver = 70;
- else if (ut.machine[4] == '6')
- ver = 60;
+ if (ut.machine[4] >= '8') ver = 80;
+ else if (ut.machine[4] == '7') ver = 70;
+ else if (ut.machine[4] == '6') ver = 60;
}
}
#endif
@@ -700,11 +668,13 @@
ver >= 61 ? JIT_F_ARMV6T2_ :
ver >= 60 ? JIT_F_ARMV6_ : 0;
flags |= LJ_ARCH_HASFPU == 0 ? 0 : ver >= 70 ? JIT_F_VFPV3 : JIT_F_VFPV2;
-#endif
+
#elif LJ_TARGET_ARM64
+
/* No optional CPU features to detect (for now). */
+
#elif LJ_TARGET_PPC
-#if LJ_HASJIT
+
#if LJ_ARCH_SQRT
flags |= JIT_F_SQRT;
#endif
@@ -711,9 +681,9 @@
#if LJ_ARCH_ROUND
flags |= JIT_F_ROUND;
#endif
-#endif
+
#elif LJ_TARGET_MIPS
-#if LJ_HASJIT
+
/* Compile-time MIPS CPU detection. */
#if LJ_ARCH_VERSION >= 20
flags |= JIT_F_MIPSXXR2;
@@ -731,11 +701,10 @@
if (x) flags |= JIT_F_MIPSXXR2; /* Either 0x80000000 (R2) or 0 (R1). */
}
#endif
-#endif
+
#else
#error "Missing CPU detection for this architecture"
#endif
- UNUSED(L);
return flags;
}
@@ -742,23 +711,21 @@
/* Initialize JIT compiler. */
static void jit_init(lua_State *L)
{
- uint32_t flags = jit_cpudetect(L);
-#if LJ_HASJIT
jit_State *J = L2J(L);
- J->flags = flags | JIT_F_ON | JIT_F_OPT_DEFAULT;
+ J->flags = jit_cpudetect() | JIT_F_ON | JIT_F_OPT_DEFAULT;
memcpy(J->param, jit_param_default, sizeof(J->param));
lj_dispatch_update(G(L));
-#else
- UNUSED(flags);
+}
#endif
-}
LUALIB_API int luaopen_jit(lua_State *L)
{
+#if LJ_HASJIT
jit_init(L);
+#endif
lua_pushliteral(L, LJ_OS_NAME);
lua_pushliteral(L, LJ_ARCH_NAME);
- lua_pushinteger(L, LUAJIT_VERSION_NUM);
+ lua_pushinteger(L, LUAJIT_VERSION_NUM); /* Deprecated. */
lua_pushliteral(L, LUAJIT_VERSION);
LJ_LIB_REG(L, LUA_JITLIBNAME, jit);
#if LJ_HASPROFILE
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_math.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_math.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_math.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Math library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include <math.h>
@@ -13,8 +13,10 @@
#include "lualib.h"
#include "lj_obj.h"
+#include "lj_err.h"
#include "lj_lib.h"
#include "lj_vm.h"
+#include "lj_prng.h"
/* ------------------------------------------------------------------------ */
@@ -33,19 +35,19 @@
lj_lib_checknum(L, 1);
return FFH_RETRY;
}
-LJLIB_ASM_(math_log10) LJLIB_REC(math_unary IRFPM_LOG10)
-LJLIB_ASM_(math_exp) LJLIB_REC(math_unary IRFPM_EXP)
-LJLIB_ASM_(math_sin) LJLIB_REC(math_unary IRFPM_SIN)
-LJLIB_ASM_(math_cos) LJLIB_REC(math_unary IRFPM_COS)
-LJLIB_ASM_(math_tan) LJLIB_REC(math_unary IRFPM_TAN)
-LJLIB_ASM_(math_asin) LJLIB_REC(math_atrig FF_math_asin)
-LJLIB_ASM_(math_acos) LJLIB_REC(math_atrig FF_math_acos)
-LJLIB_ASM_(math_atan) LJLIB_REC(math_atrig FF_math_atan)
-LJLIB_ASM_(math_sinh) LJLIB_REC(math_htrig IRCALL_sinh)
-LJLIB_ASM_(math_cosh) LJLIB_REC(math_htrig IRCALL_cosh)
-LJLIB_ASM_(math_tanh) LJLIB_REC(math_htrig IRCALL_tanh)
+LJLIB_ASM_(math_log10) LJLIB_REC(math_call IRCALL_log10)
+LJLIB_ASM_(math_exp) LJLIB_REC(math_call IRCALL_exp)
+LJLIB_ASM_(math_sin) LJLIB_REC(math_call IRCALL_sin)
+LJLIB_ASM_(math_cos) LJLIB_REC(math_call IRCALL_cos)
+LJLIB_ASM_(math_tan) LJLIB_REC(math_call IRCALL_tan)
+LJLIB_ASM_(math_asin) LJLIB_REC(math_call IRCALL_asin)
+LJLIB_ASM_(math_acos) LJLIB_REC(math_call IRCALL_acos)
+LJLIB_ASM_(math_atan) LJLIB_REC(math_call IRCALL_atan)
+LJLIB_ASM_(math_sinh) LJLIB_REC(math_call IRCALL_sinh)
+LJLIB_ASM_(math_cosh) LJLIB_REC(math_call IRCALL_cosh)
+LJLIB_ASM_(math_tanh) LJLIB_REC(math_call IRCALL_tanh)
LJLIB_ASM_(math_frexp)
-LJLIB_ASM_(math_modf) LJLIB_REC(.)
+LJLIB_ASM_(math_modf)
LJLIB_ASM(math_log) LJLIB_REC(math_log)
{
@@ -105,35 +107,12 @@
** Full-period ME-CF generator with L=64, J=4, k=223, N1=49.
*/
-/* PRNG state. */
-struct RandomState {
- uint64_t gen[4]; /* State of the 4 LFSR generators. */
- int valid; /* State is valid. */
-};
-
/* Union needed for bit-pattern conversion between uint64_t and double. */
typedef union { uint64_t u64; double d; } U64double;
-/* Update generator i and compute a running xor of all states. */
-#define TW223_GEN(i, k, q, s) \
- z = rs->gen[i]; \
- z = (((z<<q)^z) >> (k-s)) ^ ((z&((uint64_t)(int64_t)-1 << (64-k)))<<s); \
- r ^= z; rs->gen[i] = z;
-
-/* PRNG step function. Returns a double in the range 1.0 <= d < 2.0. */
-LJ_NOINLINE uint64_t LJ_FASTCALL lj_math_random_step(RandomState *rs)
+/* PRNG seeding function. */
+static void random_seed(PRNGState *rs, double d)
{
- uint64_t z, r = 0;
- TW223_GEN(0, 63, 31, 18)
- TW223_GEN(1, 58, 19, 28)
- TW223_GEN(2, 55, 24, 7)
- TW223_GEN(3, 47, 21, 8)
- return (r & U64x(000fffff,ffffffff)) | U64x(3ff00000,00000000);
-}
-
-/* PRNG initialization function. */
-static void random_init(RandomState *rs, double d)
-{
uint32_t r = 0x11090601; /* 64-k[i] as four 8 bit constants. */
int i;
for (i = 0; i < 4; i++) {
@@ -141,24 +120,22 @@
uint32_t m = 1u << (r&255);
r >>= 8;
u.d = d = d * 3.14159265358979323846 + 2.7182818284590452354;
- if (u.u64 < m) u.u64 += m; /* Ensure k[i] MSB of gen[i] are non-zero. */
- rs->gen[i] = u.u64;
+ if (u.u64 < m) u.u64 += m; /* Ensure k[i] MSB of u[i] are non-zero. */
+ rs->u[i] = u.u64;
}
- rs->valid = 1;
for (i = 0; i < 10; i++)
- lj_math_random_step(rs);
+ (void)lj_prng_u64(rs);
}
/* PRNG extract function. */
-LJLIB_PUSH(top-2) /* Upvalue holds userdata with RandomState. */
+LJLIB_PUSH(top-2) /* Upvalue holds userdata with PRNGState. */
LJLIB_CF(math_random) LJLIB_REC(.)
{
int n = (int)(L->top - L->base);
- RandomState *rs = (RandomState *)(uddata(udataV(lj_lib_upvalue(L, 1))));
+ PRNGState *rs = (PRNGState *)(uddata(udataV(lj_lib_upvalue(L, 1))));
U64double u;
double d;
- if (LJ_UNLIKELY(!rs->valid)) random_init(rs, 0.0);
- u.u64 = lj_math_random_step(rs);
+ u.u64 = lj_prng_u64d(rs);
d = u.d - 1.0;
if (n > 0) {
#if LJ_DUALNUM
@@ -203,11 +180,14 @@
}
/* PRNG seed function. */
-LJLIB_PUSH(top-2) /* Upvalue holds userdata with RandomState. */
+LJLIB_PUSH(top-2) /* Upvalue holds userdata with PRNGState. */
LJLIB_CF(math_randomseed)
{
- RandomState *rs = (RandomState *)(uddata(udataV(lj_lib_upvalue(L, 1))));
- random_init(rs, lj_lib_checknum(L, 1));
+ PRNGState *rs = (PRNGState *)(uddata(udataV(lj_lib_upvalue(L, 1))));
+ if (L->base != L->top)
+ random_seed(rs, lj_lib_checknum(L, 1));
+ else if (!lj_prng_seed_secure(rs))
+ lj_err_caller(L, LJ_ERR_PRNGSD);
return 0;
}
@@ -217,9 +197,8 @@
LUALIB_API int luaopen_math(lua_State *L)
{
- RandomState *rs;
- rs = (RandomState *)lua_newuserdata(L, sizeof(RandomState));
- rs->valid = 0; /* Use lazy initialization to save some time on startup. */
+ PRNGState *rs = (PRNGState *)lua_newuserdata(L, sizeof(PRNGState));
+ lj_prng_seed_fixed(rs);
LJ_LIB_REG(L, LUA_MATHLIBNAME, math);
return 1;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_os.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_os.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_os.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** OS library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -76,7 +76,7 @@
LJLIB_CF(os_tmpname)
{
-#if LJ_TARGET_PS3 || LJ_TARGET_PS4 || LJ_TARGET_PSVITA
+#if LJ_TARGET_PS3 || LJ_TARGET_PS4 || LJ_TARGET_PS5 || LJ_TARGET_PSVITA || LJ_TARGET_NX
lj_err_caller(L, LJ_ERR_OSUNIQF);
return 0;
#else
@@ -205,12 +205,12 @@
setboolfield(L, "isdst", stm->tm_isdst);
} else if (*s) {
SBuf *sb = &G(L)->tmpbuf;
- MSize sz = 0;
+ MSize sz = 0, retry = 4;
const char *q;
for (q = s; *q; q++)
sz += (*q == '%') ? 30 : 1; /* Overflow doesn't matter. */
setsbufL(sb, L);
- for (;;) {
+ while (retry--) { /* Limit growth for invalid format or empty result. */
char *buf = lj_buf_need(sb, sz);
size_t len = strftime(buf, sbufsz(sb), s, stm);
if (len) {
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_package.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_package.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_package.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Package library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2012 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -57,7 +57,7 @@
static const char *ll_bcsym(void *lib, const char *sym)
{
-#if defined(RTLD_DEFAULT)
+#if defined(RTLD_DEFAULT) && !defined(NO_RTLD_DEFAULT)
if (lib == NULL) lib = RTLD_DEFAULT;
#elif LJ_TARGET_OSX || LJ_TARGET_BSD
if (lib == NULL) lib = (void *)(intptr_t)-2;
@@ -76,6 +76,20 @@
BOOL WINAPI GetModuleHandleExA(DWORD, LPCSTR, HMODULE*);
#endif
+#if LJ_TARGET_UWP
+void *LJ_WIN_LOADLIBA(const char *path)
+{
+ DWORD err = GetLastError();
+ wchar_t wpath[256];
+ HANDLE lib = NULL;
+ if (MultiByteToWideChar(CP_ACP, 0, path, -1, wpath, 256) > 0) {
+ lib = LoadPackagedLibrary(wpath, 0);
+ }
+ SetLastError(err);
+ return lib;
+}
+#endif
+
#undef setprogdir
static void setprogdir(lua_State *L)
@@ -119,7 +133,7 @@
static void *ll_load(lua_State *L, const char *path, int gl)
{
- HINSTANCE lib = LoadLibraryExA(path, NULL, 0);
+ HINSTANCE lib = LJ_WIN_LOADLIBA(path);
if (lib == NULL) pusherror(L);
UNUSED(gl);
return lib;
@@ -132,11 +146,18 @@
return f;
}
+#if LJ_TARGET_UWP
+EXTERN_C IMAGE_DOS_HEADER __ImageBase;
+#endif
+
static const char *ll_bcsym(void *lib, const char *sym)
{
if (lib) {
return (const char *)GetProcAddress((HINSTANCE)lib, sym);
} else {
+#if LJ_TARGET_UWP
+ return (const char *)GetProcAddress((HINSTANCE)&__ImageBase, sym);
+#else
HINSTANCE h = GetModuleHandleA(NULL);
const char *p = (const char *)GetProcAddress(h, sym);
if (p == NULL && GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS|GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT,
@@ -143,6 +164,7 @@
(const char *)ll_bcsym, &h))
p = (const char *)GetProcAddress(h, sym);
return p;
+#endif
}
}
@@ -215,7 +237,12 @@
static int ll_loadfunc(lua_State *L, const char *path, const char *name, int r)
{
- void **reg = ll_register(L, path);
+ void **reg;
+ if (strlen(path) >= 4096) {
+ lua_pushliteral(L, "path too long");
+ return PACKAGE_ERR_LIB;
+ }
+ reg = ll_register(L, path);
if (*reg == NULL) *reg = ll_load(L, path, (*name == '*'));
if (*reg == NULL) {
return PACKAGE_ERR_LIB; /* Unable to load library. */
@@ -233,7 +260,7 @@
const char *bcdata = ll_bcsym(*reg, mksymname(L, name, SYMPREFIX_BC));
lua_pop(L, 1);
if (bcdata) {
- if (luaL_loadbuffer(L, bcdata, LJ_MAX_BUF, name) != 0)
+ if (luaL_loadbuffer(L, bcdata, ~(size_t)0, name) != 0)
return PACKAGE_ERR_LOAD;
return 0;
}
@@ -351,16 +378,6 @@
return 1; /* library loaded successfully */
}
-static int lj_cf_package_loader_c(lua_State *L)
-{
- const char *name = luaL_checkstring(L, 1);
- const char *filename = findfile(L, name, "cpath");
- if (filename == NULL) return 1; /* library not found in this path */
- if (ll_loadfunc(L, filename, name, 0) != 0)
- loaderror(L, filename);
- return 1; /* library loaded successfully */
-}
-
#define LUA_POF "luaopen_"
#define LUA_OFSEP "_"
#define POF LUA_POF
@@ -383,6 +400,17 @@
return 1; /* library loaded successfully */
}
+
+static int lj_cf_package_loader_c(lua_State *L)
+{
+ const char *name = luaL_checkstring(L, 1);
+ const char *filename = findfile(L, name, "cpath");
+ if (filename == NULL) return 1; /* library not found in this path */
+ if (ll_loadfunc(L, filename, name, 0) != 0)
+ loaderror(L, filename);
+ return 1; /* library loaded successfully */
+}
+
static int lj_cf_package_loader_croot(lua_State *L)
{
const char *filename;
@@ -426,7 +454,7 @@
if (lua_isnil(L, -1)) { /* Not found? */
const char *bcname = mksymname(L, name, SYMPREFIX_BC);
const char *bcdata = ll_bcsym(NULL, bcname);
- if (bcdata == NULL || luaL_loadbuffer(L, bcdata, LJ_MAX_BUF, name) != 0)
+ if (bcdata == NULL || luaL_loadbuffer(L, bcdata, ~(size_t)0, name) != 0)
lua_pushfstring(L, "\n\tno field package.preload['%s']", name);
}
return 1;
@@ -434,7 +462,7 @@
/* ------------------------------------------------------------------------ */
-#define sentinel ((void *)0x4004)
+#define KEY_SENTINEL (U64x(80000000,00000000)|'s')
static int lj_cf_package_require(lua_State *L)
{
@@ -444,7 +472,7 @@
lua_getfield(L, LUA_REGISTRYINDEX, "_LOADED");
lua_getfield(L, 2, name);
if (lua_toboolean(L, -1)) { /* is it there? */
- if (lua_touserdata(L, -1) == sentinel) /* check loops */
+ if ((L->top-1)->u64 == KEY_SENTINEL) /* check loops */
luaL_error(L, "loop or previous error loading module " LUA_QS, name);
return 1; /* package is already loaded */
}
@@ -467,7 +495,7 @@
else
lua_pop(L, 1);
}
- lua_pushlightuserdata(L, sentinel);
+ (L->top++)->u64 = KEY_SENTINEL;
lua_setfield(L, 2, name); /* _LOADED[name] = sentinel */
lua_pushstring(L, name); /* pass name as argument to module */
lua_call(L, 1, 1); /* run loaded module */
@@ -474,7 +502,7 @@
if (!lua_isnil(L, -1)) /* non-nil return? */
lua_setfield(L, 2, name); /* _LOADED[name] = returned value */
lua_getfield(L, 2, name);
- if (lua_touserdata(L, -1) == sentinel) { /* module did not set a value? */
+ if ((L->top-1)->u64 == KEY_SENTINEL) { /* module did not set a value? */
lua_pushboolean(L, 1); /* use true as result */
lua_pushvalue(L, -1); /* extra copy to be returned */
lua_setfield(L, 2, name); /* _LOADED[name] = true */
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_string.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_string.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_string.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** String library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -122,11 +122,25 @@
LJLIB_CF(string_dump)
{
- GCfunc *fn = lj_lib_checkfunc(L, 1);
- int strip = L->base+1 < L->top && tvistruecond(L->base+1);
- SBuf *sb = lj_buf_tmp_(L); /* Assumes lj_bcwrite() doesn't use tmpbuf. */
+ GCproto *pt = lj_lib_checkLproto(L, 1, 1);
+ uint32_t flags = 0;
+ SBuf *sb;
+ TValue *o = L->base+1;
+ if (o < L->top) {
+ if (tvisstr(o)) {
+ const char *mode = strVdata(o);
+ char c;
+ while ((c = *mode++)) {
+ if (c == 's') flags |= BCDUMP_F_STRIP;
+ if (c == 'd') flags |= BCDUMP_F_DETERMINISTIC;
+ }
+ } else if (tvistruecond(o)) {
+ flags |= BCDUMP_F_STRIP;
+ }
+ }
+ sb = lj_buf_tmp_(L); /* Assumes lj_bcwrite() doesn't use tmpbuf. */
L->top = L->base+1;
- if (!isluafunc(fn) || lj_bcwrite(L, funcproto(fn), writer_buf, sb, strip))
+ if (!pt || lj_bcwrite(L, pt, writer_buf, sb, flags))
lj_err_caller(L, LJ_ERR_STRDUMP);
setstrV(L, L->top-1, lj_buf_str(L, sb));
lj_gc_check(L);
@@ -136,7 +150,7 @@
/* ------------------------------------------------------------------------ */
/* macro to `unsign' a character */
-#define uchar(c) ((unsigned char)(c))
+#define uchar(c) ((unsigned char)(c))
#define CAP_UNFINISHED (-1)
#define CAP_POSITION (-2)
@@ -640,89 +654,14 @@
/* ------------------------------------------------------------------------ */
-/* Emulate tostring() inline. */
-static GCstr *string_fmt_tostring(lua_State *L, int arg, int retry)
-{
- TValue *o = L->base+arg-1;
- cTValue *mo;
- lua_assert(o < L->top); /* Caller already checks for existence. */
- if (LJ_LIKELY(tvisstr(o)))
- return strV(o);
- if (retry != 2 && !tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
- copyTV(L, L->top++, mo);
- copyTV(L, L->top++, o);
- lua_call(L, 1, 1);
- copyTV(L, L->base+arg-1, --L->top);
- return NULL; /* Buffer may be overwritten, retry. */
- }
- return lj_strfmt_obj(L, o);
-}
-
LJLIB_CF(string_format) LJLIB_REC(.)
{
- int arg, top = (int)(L->top - L->base);
- GCstr *fmt;
+ int retry = 0;
SBuf *sb;
- FormatState fs;
- SFormat sf;
- int retry = 0;
-again:
- arg = 1;
- sb = lj_buf_tmp_(L);
- fmt = lj_lib_checkstr(L, arg);
- lj_strfmt_init(&fs, strdata(fmt), fmt->len);
- while ((sf = lj_strfmt_parse(&fs)) != STRFMT_EOF) {
- if (sf == STRFMT_LIT) {
- lj_buf_putmem(sb, fs.str, fs.len);
- } else if (sf == STRFMT_ERR) {
- lj_err_callerv(L, LJ_ERR_STRFMT, strdata(lj_str_new(L, fs.str, fs.len)));
- } else {
- if (++arg > top)
- luaL_argerror(L, arg, lj_obj_typename[0]);
- switch (STRFMT_TYPE(sf)) {
- case STRFMT_INT:
- if (tvisint(L->base+arg-1)) {
- int32_t k = intV(L->base+arg-1);
- if (sf == STRFMT_INT)
- lj_strfmt_putint(sb, k); /* Shortcut for plain %d. */
- else
- lj_strfmt_putfxint(sb, sf, k);
- } else {
- lj_strfmt_putfnum_int(sb, sf, lj_lib_checknum(L, arg));
- }
- break;
- case STRFMT_UINT:
- if (tvisint(L->base+arg-1))
- lj_strfmt_putfxint(sb, sf, intV(L->base+arg-1));
- else
- lj_strfmt_putfnum_uint(sb, sf, lj_lib_checknum(L, arg));
- break;
- case STRFMT_NUM:
- lj_strfmt_putfnum(sb, sf, lj_lib_checknum(L, arg));
- break;
- case STRFMT_STR: {
- GCstr *str = string_fmt_tostring(L, arg, retry);
- if (str == NULL)
- retry = 1;
- else if ((sf & STRFMT_T_QUOTED))
- lj_strfmt_putquoted(sb, str); /* No formatting. */
- else
- lj_strfmt_putfstr(sb, sf, str);
- break;
- }
- case STRFMT_CHAR:
- lj_strfmt_putfchar(sb, sf, lj_lib_checkint(L, arg));
- break;
- case STRFMT_PTR: /* No formatting. */
- lj_strfmt_putptr(sb, lj_obj_ptr(L->base+arg-1));
- break;
- default:
- lua_assert(0);
- break;
- }
- }
- }
- if (retry++ == 1) goto again;
+ do {
+ sb = lj_buf_tmp_(L);
+ retry = lj_strfmt_putarg(L, sb, 1, -retry);
+ } while (retry > 0);
setstrV(L, L->top-1, lj_buf_str(L, sb));
lj_gc_check(L);
return 1;
@@ -743,6 +682,9 @@
setgcref(basemt_it(g, LJ_TSTR), obj2gco(mt));
settabV(L, lj_tab_setstr(L, mt, mmname_str(g, MM_index)), tabV(L->top-1));
mt->nomm = (uint8_t)(~(1u<<MM_index));
+#if LJ_HASBUFFER
+ lj_lib_prereg(L, LUA_STRLIBNAME ".buffer", luaopen_string_buffer, tabV(L->top-1));
+#endif
return 1;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_table.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_table.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lib_table.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Table library.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -159,7 +159,7 @@
SBuf *sb = lj_buf_tmp_(L);
SBuf *sbx = lj_buf_puttab(sb, t, sep, i, e);
if (LJ_UNLIKELY(!sbx)) { /* Error: bad element type. */
- int32_t idx = (int32_t)(intptr_t)sbufP(sb);
+ int32_t idx = (int32_t)(intptr_t)sb->w;
cTValue *o = lj_tab_getint(t, idx);
lj_err_callerv(L, LJ_ERR_TABCAT,
lj_obj_itypename[o ? itypemap(o) : ~LJ_TNIL], idx);
Deleted: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj.supp
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj.supp 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj.supp 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,41 +0,0 @@
-# Valgrind suppression file for LuaJIT 2.0.
-{
- Optimized string compare
- Memcheck:Addr4
- fun:lj_str_cmp
-}
-{
- Optimized string compare
- Memcheck:Addr1
- fun:lj_str_cmp
-}
-{
- Optimized string compare
- Memcheck:Addr4
- fun:lj_str_new
-}
-{
- Optimized string compare
- Memcheck:Addr1
- fun:lj_str_new
-}
-{
- Optimized string compare
- Memcheck:Cond
- fun:lj_str_new
-}
-{
- Optimized string compare
- Memcheck:Addr4
- fun:str_fastcmp
-}
-{
- Optimized string compare
- Memcheck:Addr1
- fun:str_fastcmp
-}
-{
- Optimized string compare
- Memcheck:Cond
- fun:str_fastcmp
-}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -6,7 +6,7 @@
**
** This is a version (aka dlmalloc) of malloc/free/realloc written by
** Doug Lea and released to the public domain, as explained at
-** http://creativecommons.org/licenses/publicdomain.
+** https://creativecommons.org/licenses/publicdomain.
**
** * Version pre-2.8.4 Wed Mar 29 19:46:29 2006 (dl at gee)
**
@@ -16,8 +16,8 @@
** If you want to use dlmalloc in another project, you should get
** the original from: ftp://gee.cs.oswego.edu/pub/misc/
** For thread-safe derivatives, take a look at:
-** - ptmalloc: http://www.malloc.de/
-** - nedmalloc: http://www.nedprod.com/programs/portable/nedmalloc/
+** - ptmalloc: https://www.malloc.de/
+** - nedmalloc: https://www.nedprod.com/programs/portable/nedmalloc/
*/
#define lj_alloc_c
@@ -31,6 +31,7 @@
#include "lj_def.h"
#include "lj_arch.h"
#include "lj_alloc.h"
+#include "lj_prng.h"
#ifndef LUAJIT_USE_SYSMALLOC
@@ -123,7 +124,7 @@
#if LJ_ALLOC_NTAVM
/* Undocumented, but hey, that's what we all love so much about Windows. */
-typedef long (*PNTAVM)(HANDLE handle, void **addr, ULONG zbits,
+typedef long (*PNTAVM)(HANDLE handle, void **addr, ULONG_PTR zbits,
size_t *size, ULONG alloctype, ULONG prot);
static PNTAVM ntavm;
@@ -140,7 +141,7 @@
#define INIT_MMAP() init_mmap()
/* Win64 32 bit MMAP via NtAllocateVirtualMemory. */
-static void *CALL_MMAP(size_t size)
+static void *mmap_plain(size_t size)
{
DWORD olderr = GetLastError();
void *ptr = NULL;
@@ -151,7 +152,7 @@
}
/* For direct MMAP, use MEM_TOP_DOWN to minimize interference */
-static void *DIRECT_MMAP(size_t size)
+static void *direct_mmap(size_t size)
{
DWORD olderr = GetLastError();
void *ptr = NULL;
@@ -164,20 +165,20 @@
#else
/* Win32 MMAP via VirtualAlloc */
-static void *CALL_MMAP(size_t size)
+static void *mmap_plain(size_t size)
{
DWORD olderr = GetLastError();
- void *ptr = VirtualAlloc(0, size, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
+ void *ptr = LJ_WIN_VALLOC(0, size, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
SetLastError(olderr);
return ptr ? ptr : MFAIL;
}
/* For direct MMAP, use MEM_TOP_DOWN to minimize interference */
-static void *DIRECT_MMAP(size_t size)
+static void *direct_mmap(size_t size)
{
DWORD olderr = GetLastError();
- void *ptr = VirtualAlloc(0, size, MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN,
- PAGE_READWRITE);
+ void *ptr = LJ_WIN_VALLOC(0, size, MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN,
+ PAGE_READWRITE);
SetLastError(olderr);
return ptr ? ptr : MFAIL;
}
@@ -184,6 +185,9 @@
#endif
+#define CALL_MMAP(prng, size) mmap_plain(size)
+#define DIRECT_MMAP(prng, size) direct_mmap(size)
+
/* This function supports releasing coalesed segments */
static int CALL_MUNMAP(void *ptr, size_t size)
{
@@ -226,36 +230,17 @@
#define LJ_ALLOC_MMAP_PROBE_LOWER ((uintptr_t)0x4000)
-/* No point in a giant ifdef mess. Just try to open /dev/urandom.
-** It doesn't really matter if this fails, since we get some ASLR bits from
-** every unsuitable allocation, too. And we prefer linear allocation, anyway.
-*/
-#include <fcntl.h>
-#include <unistd.h>
-
-static uintptr_t mmap_probe_seed(void)
+static void *mmap_probe(PRNGState *rs, size_t size)
{
- uintptr_t val;
- int fd = open("/dev/urandom", O_RDONLY);
- if (fd != -1) {
- int ok = ((size_t)read(fd, &val, sizeof(val)) == sizeof(val));
- (void)close(fd);
- if (ok) return val;
- }
- return 1; /* Punt. */
-}
-
-static void *mmap_probe(size_t size)
-{
/* Hint for next allocation. Doesn't need to be thread-safe. */
static uintptr_t hint_addr = 0;
- static uintptr_t hint_prng = 0;
int olderr = errno;
int retry;
for (retry = 0; retry < LJ_ALLOC_MMAP_PROBE_MAX; retry++) {
void *p = mmap((void *)hint_addr, size, MMAP_PROT, MMAP_FLAGS_PROBE, -1, 0);
uintptr_t addr = (uintptr_t)p;
- if ((addr >> LJ_ALLOC_MBITS) == 0 && addr >= LJ_ALLOC_MMAP_PROBE_LOWER) {
+ if ((addr >> LJ_ALLOC_MBITS) == 0 && addr >= LJ_ALLOC_MMAP_PROBE_LOWER &&
+ ((addr + size) >> LJ_ALLOC_MBITS) == 0) {
/* We got a suitable address. Bump the hint address. */
hint_addr = addr + size;
errno = olderr;
@@ -280,15 +265,8 @@
}
}
/* Finally, try pseudo-random probing. */
- if (LJ_UNLIKELY(hint_prng == 0)) {
- hint_prng = mmap_probe_seed();
- }
- /* The unsuitable address we got has some ASLR PRNG bits. */
- hint_addr ^= addr & ~((uintptr_t)(LJ_PAGESIZE-1));
- do { /* The PRNG itself is very weak, but see above. */
- hint_prng = hint_prng * 1103515245 + 12345;
- hint_addr ^= hint_prng * (uintptr_t)LJ_PAGESIZE;
- hint_addr &= (((uintptr_t)1 << LJ_ALLOC_MBITS)-1);
+ do {
+ hint_addr = lj_prng_u64(rs) & (((uintptr_t)1<<LJ_ALLOC_MBITS)-LJ_PAGESIZE);
} while (hint_addr < LJ_ALLOC_MMAP_PROBE_LOWER);
}
errno = olderr;
@@ -299,18 +277,22 @@
#if LJ_ALLOC_MMAP32
-#if defined(__sun__)
+#if LJ_TARGET_SOLARIS
#define LJ_ALLOC_MMAP32_START ((uintptr_t)0x1000)
#else
#define LJ_ALLOC_MMAP32_START ((uintptr_t)0)
#endif
+#if LJ_ALLOC_MMAP_PROBE
+static void *mmap_map32(PRNGState *rs, size_t size)
+#else
static void *mmap_map32(size_t size)
+#endif
{
#if LJ_ALLOC_MMAP_PROBE
static int fallback = 0;
if (fallback)
- return mmap_probe(size);
+ return mmap_probe(rs, size);
#endif
{
int olderr = errno;
@@ -320,7 +302,7 @@
#if LJ_ALLOC_MMAP_PROBE
if (ptr == MFAIL) {
fallback = 1;
- return mmap_probe(size);
+ return mmap_probe(rs, size);
}
#endif
return ptr;
@@ -330,11 +312,15 @@
#endif
#if LJ_ALLOC_MMAP32
-#define CALL_MMAP(size) mmap_map32(size)
+#if LJ_ALLOC_MMAP_PROBE
+#define CALL_MMAP(prng, size) mmap_map32(prng, size)
+#else
+#define CALL_MMAP(prng, size) mmap_map32(size)
+#endif
#elif LJ_ALLOC_MMAP_PROBE
-#define CALL_MMAP(size) mmap_probe(size)
+#define CALL_MMAP(prng, size) mmap_probe(prng, size)
#else
-static void *CALL_MMAP(size_t size)
+static void *mmap_plain(size_t size)
{
int olderr = errno;
void *ptr = mmap(NULL, size, MMAP_PROT, MMAP_FLAGS, -1, 0);
@@ -341,9 +327,10 @@
errno = olderr;
return ptr;
}
+#define CALL_MMAP(prng, size) mmap_plain(size)
#endif
-#if ((defined(__FreeBSD__) && __FreeBSD__ < 10) || defined(__FreeBSD_kernel__)) && !LJ_TARGET_PS4
+#if LJ_64 && !LJ_GC64 && ((defined(__FreeBSD__) && __FreeBSD__ < 10) || defined(__FreeBSD_kernel__)) && !LJ_TARGET_PS4 && !LJ_TARGET_PS5
#include <sys/resource.h>
@@ -378,7 +365,7 @@
#define CALL_MREMAP(addr, osz, nsz, mv) CALL_MREMAP_((addr), (osz), (nsz), (mv))
#define CALL_MREMAP_NOMOVE 0
#define CALL_MREMAP_MAYMOVE 1
-#if LJ_64 && !LJ_GC64
+#if LJ_64 && (!LJ_GC64 || LJ_TARGET_ARM64)
#define CALL_MREMAP_MV CALL_MREMAP_NOMOVE
#else
#define CALL_MREMAP_MV CALL_MREMAP_MAYMOVE
@@ -393,7 +380,7 @@
#endif
#ifndef DIRECT_MMAP
-#define DIRECT_MMAP(s) CALL_MMAP(s)
+#define DIRECT_MMAP(prng, s) CALL_MMAP(prng, s)
#endif
#ifndef CALL_MREMAP
@@ -552,6 +539,7 @@
mchunkptr smallbins[(NSMALLBINS+1)*2];
tbinptr treebins[NTREEBINS];
msegment seg;
+ PRNGState *prng;
};
typedef struct malloc_state *mstate;
@@ -609,7 +597,7 @@
noncontiguous segments are added.
*/
#define TOP_FOOT_SIZE\
- (align_offset(chunk2mem(0))+pad_request(sizeof(struct malloc_segment))+MIN_CHUNK_SIZE)
+ (align_offset(TWO_SIZE_T_SIZES)+pad_request(sizeof(struct malloc_segment))+MIN_CHUNK_SIZE)
/* ---------------------------- Indexing Bins ---------------------------- */
@@ -834,11 +822,11 @@
/* ----------------------- Direct-mmapping chunks ----------------------- */
-static void *direct_alloc(size_t nb)
+static void *direct_alloc(mstate m, size_t nb)
{
size_t mmsize = mmap_align(nb + SIX_SIZE_T_SIZES + CHUNK_ALIGN_MASK);
if (LJ_LIKELY(mmsize > nb)) { /* Check for wrap around 0 */
- char *mm = (char *)(DIRECT_MMAP(mmsize));
+ char *mm = (char *)(DIRECT_MMAP(m->prng, mmsize));
if (mm != CMFAIL) {
size_t offset = align_offset(chunk2mem(mm));
size_t psize = mmsize - offset - DIRECT_FOOT_PAD;
@@ -850,6 +838,7 @@
return chunk2mem(p);
}
}
+ UNUSED(m);
return NULL;
}
@@ -998,7 +987,7 @@
/* Directly map large chunks */
if (LJ_UNLIKELY(nb >= DEFAULT_MMAP_THRESHOLD)) {
- void *mem = direct_alloc(nb);
+ void *mem = direct_alloc(m, nb);
if (mem != 0)
return mem;
}
@@ -1007,7 +996,7 @@
size_t req = nb + TOP_FOOT_SIZE + SIZE_T_ONE;
size_t rsize = granularity_align(req);
if (LJ_LIKELY(rsize > nb)) { /* Fail if wraps around zero */
- char *mp = (char *)(CALL_MMAP(rsize));
+ char *mp = (char *)(CALL_MMAP(m->prng, rsize));
if (mp != CMFAIL) {
tbase = mp;
tsize = rsize;
@@ -1068,7 +1057,7 @@
mchunkptr p = align_as_chunk(base);
size_t psize = chunksize(p);
/* Can unmap if first chunk holds entire segment and not pinned */
- if (!cinuse(p) && (char *)p + psize >= base + size - TOP_FOOT_SIZE) {
+ if (!cinuse(p) && (char *)p + psize == (char *)mem2chunk(sp)) {
tchunkptr tp = (tchunkptr)p;
if (p == m->dv) {
m->dv = 0;
@@ -1234,12 +1223,13 @@
/* ----------------------------------------------------------------------- */
-void *lj_alloc_create(void)
+void *lj_alloc_create(PRNGState *rs)
{
size_t tsize = DEFAULT_GRANULARITY;
char *tbase;
INIT_MMAP();
- tbase = (char *)(CALL_MMAP(tsize));
+ UNUSED(rs);
+ tbase = (char *)(CALL_MMAP(rs, tsize));
if (tbase != CMFAIL) {
size_t msize = pad_request(sizeof(struct malloc_state));
mchunkptr mn;
@@ -1258,6 +1248,12 @@
return NULL;
}
+void lj_alloc_setprng(void *msp, PRNGState *rs)
+{
+ mstate ms = (mstate)msp;
+ ms->prng = rs;
+}
+
void lj_alloc_destroy(void *msp)
{
mstate ms = (mstate)msp;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_alloc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -9,7 +9,8 @@
#include "lj_def.h"
#ifndef LUAJIT_USE_SYSMALLOC
-LJ_FUNC void *lj_alloc_create(void);
+LJ_FUNC void *lj_alloc_create(PRNGState *rs);
+LJ_FUNC void lj_alloc_setprng(void *msp, PRNGState *rs);
LJ_FUNC void lj_alloc_destroy(void *msp);
LJ_FUNC void *lj_alloc_f(void *msp, void *ptr, size_t osize, size_t nsize);
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_api.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_api.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_api.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Public Lua/C API.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -28,8 +28,8 @@
/* -- Common helper functions --------------------------------------------- */
-#define api_checknelems(L, n) api_check(L, (n) <= (L->top - L->base))
-#define api_checkvalidindex(L, i) api_check(L, (i) != niltv(L))
+#define lj_checkapi_slot(idx) \
+ lj_checkapi((idx) <= (L->top - L->base), "stack slot %d out of range", (idx))
static TValue *index2adr(lua_State *L, int idx)
{
@@ -37,7 +37,8 @@
TValue *o = L->base + (idx - 1);
return o < L->top ? o : niltv(L);
} else if (idx > LUA_REGISTRYINDEX) {
- api_check(L, idx != 0 && -idx <= L->top - L->base);
+ lj_checkapi(idx != 0 && -idx <= L->top - L->base,
+ "bad stack slot %d", idx);
return L->top + idx;
} else if (idx == LUA_GLOBALSINDEX) {
TValue *o = &G(L)->tmptv;
@@ -47,7 +48,8 @@
return registry(L);
} else {
GCfunc *fn = curr_func(L);
- api_check(L, fn->c.gct == ~LJ_TFUNC && !isluafunc(fn));
+ lj_checkapi(fn->c.gct == ~LJ_TFUNC && !isluafunc(fn),
+ "calling frame is not a C function");
if (idx == LUA_ENVIRONINDEX) {
TValue *o = &G(L)->tmptv;
settabV(L, o, tabref(fn->c.env));
@@ -59,13 +61,27 @@
}
}
-static TValue *stkindex2adr(lua_State *L, int idx)
+static LJ_AINLINE TValue *index2adr_check(lua_State *L, int idx)
{
+ TValue *o = index2adr(L, idx);
+ lj_checkapi(o != niltv(L), "invalid stack slot %d", idx);
+ return o;
+}
+
+static TValue *index2adr_stack(lua_State *L, int idx)
+{
if (idx > 0) {
TValue *o = L->base + (idx - 1);
+ if (o < L->top) {
+ return o;
+ } else {
+ lj_checkapi(0, "invalid stack slot %d", idx);
+ return niltv(L);
+ }
return o < L->top ? o : niltv(L);
} else {
- api_check(L, idx != 0 && -idx <= L->top - L->base);
+ lj_checkapi(idx != 0 && -idx <= L->top - L->base,
+ "invalid stack slot %d", idx);
return L->top + idx;
}
}
@@ -88,7 +104,12 @@
if (size > LUAI_MAXCSTACK || (L->top - L->base + size) > LUAI_MAXCSTACK) {
return 0; /* Stack overflow. */
} else if (size > 0) {
- lj_state_checkstack(L, (MSize)size);
+ int avail = (int)(mref(L->maxstack, TValue) - L->top);
+ if (size > avail &&
+ lj_state_cpgrowstack(L, (MSize)(size - avail)) != LUA_OK) {
+ L->top--;
+ return 0; /* Out of memory. */
+ }
}
return 1;
}
@@ -99,17 +120,17 @@
lj_err_callerv(L, LJ_ERR_STKOVM, msg);
}
-LUA_API void lua_xmove(lua_State *from, lua_State *to, int n)
+LUA_API void lua_xmove(lua_State *L, lua_State *to, int n)
{
TValue *f, *t;
- if (from == to) return;
- api_checknelems(from, n);
- api_check(from, G(from) == G(to));
+ if (L == to) return;
+ lj_checkapi_slot(n);
+ lj_checkapi(G(L) == G(to), "move across global states");
lj_state_checkstack(to, (MSize)n);
- f = from->top;
+ f = L->top;
t = to->top = to->top + n;
while (--n >= 0) copyTV(to, --t, --f);
- from->top = f;
+ L->top = f;
}
LUA_API const lua_Number *lua_version(lua_State *L)
@@ -129,7 +150,7 @@
LUA_API void lua_settop(lua_State *L, int idx)
{
if (idx >= 0) {
- api_check(L, idx <= tvref(L->maxstack) - L->base);
+ lj_checkapi(idx <= tvref(L->maxstack) - L->base, "bad stack slot %d", idx);
if (L->base + idx > L->top) {
if (L->base + idx >= tvref(L->maxstack))
lj_state_growstack(L, (MSize)idx - (MSize)(L->top - L->base));
@@ -138,7 +159,7 @@
L->top = L->base + idx;
}
} else {
- api_check(L, -(idx+1) <= (L->top - L->base));
+ lj_checkapi(-(idx+1) <= (L->top - L->base), "bad stack slot %d", idx);
L->top += idx+1; /* Shrinks top (idx < 0). */
}
}
@@ -145,8 +166,7 @@
LUA_API void lua_remove(lua_State *L, int idx)
{
- TValue *p = stkindex2adr(L, idx);
- api_checkvalidindex(L, p);
+ TValue *p = index2adr_stack(L, idx);
while (++p < L->top) copyTV(L, p-1, p);
L->top--;
}
@@ -153,8 +173,7 @@
LUA_API void lua_insert(lua_State *L, int idx)
{
- TValue *q, *p = stkindex2adr(L, idx);
- api_checkvalidindex(L, p);
+ TValue *q, *p = index2adr_stack(L, idx);
for (q = L->top; q > p; q--) copyTV(L, q, q-1);
copyTV(L, p, L->top);
}
@@ -162,7 +181,7 @@
static void copy_slot(lua_State *L, TValue *f, int idx)
{
if (idx == LUA_GLOBALSINDEX) {
- api_check(L, tvistab(f));
+ lj_checkapi(tvistab(f), "stack slot %d is not a table", idx);
/* NOBARRIER: A thread (i.e. L) is never black. */
setgcref(L->env, obj2gco(tabV(f)));
} else if (idx == LUA_ENVIRONINDEX) {
@@ -169,12 +188,11 @@
GCfunc *fn = curr_func(L);
if (fn->c.gct != ~LJ_TFUNC)
lj_err_msg(L, LJ_ERR_NOENV);
- api_check(L, tvistab(f));
+ lj_checkapi(tvistab(f), "stack slot %d is not a table", idx);
setgcref(fn->c.env, obj2gco(tabV(f)));
lj_gc_barrier(L, fn, f);
} else {
- TValue *o = index2adr(L, idx);
- api_checkvalidindex(L, o);
+ TValue *o = index2adr_check(L, idx);
copyTV(L, o, f);
if (idx < LUA_GLOBALSINDEX) /* Need a barrier for upvalues. */
lj_gc_barrier(L, curr_func(L), f);
@@ -183,7 +201,7 @@
LUA_API void lua_replace(lua_State *L, int idx)
{
- api_checknelems(L, 1);
+ lj_checkapi_slot(1);
copy_slot(L, L->top - 1, idx);
L->top--;
}
@@ -219,7 +237,7 @@
#else
int tt = (int)(((t < 8 ? 0x98042110u : 0x75a06u) >> 4*(t&7)) & 15u);
#endif
- lua_assert(tt != LUA_TNIL || tvisnil(o));
+ lj_assertL(tt != LUA_TNIL || tvisnil(o), "bad tag conversion");
return tt;
}
}
@@ -595,7 +613,7 @@
if (tvisudata(o))
return uddata(udataV(o));
else if (tvislightud(o))
- return lightudV(o);
+ return lightudV(G(L), o);
else
return NULL;
}
@@ -608,7 +626,7 @@
LUA_API const void *lua_topointer(lua_State *L, int idx)
{
- return lj_obj_ptr(index2adr(L, idx));
+ return lj_obj_ptr(G(L), index2adr(L, idx));
}
/* -- Stack setters (object creation) ------------------------------------- */
@@ -677,7 +695,7 @@
{
GCfunc *fn;
lj_gc_check(L);
- api_checknelems(L, n);
+ lj_checkapi_slot(n);
fn = lj_func_newC(L, (MSize)n, getcurrenv(L));
fn->c.f = f;
L->top -= n;
@@ -684,7 +702,7 @@
while (n--)
copyTV(L, &fn->c.upvalue[n], L->top+n);
setfuncV(L, L->top, fn);
- lua_assert(iswhite(obj2gco(fn)));
+ lj_assertL(iswhite(obj2gco(fn)), "new GC object is not white");
incr_top(L);
}
@@ -696,7 +714,10 @@
LUA_API void lua_pushlightuserdata(lua_State *L, void *p)
{
- setlightudV(L->top, checklightudptr(L, p));
+#if LJ_64
+ p = lj_lightud_intern(L, p);
+#endif
+ setrawlightudV(L->top, p);
incr_top(L);
}
@@ -754,7 +775,7 @@
LUA_API void lua_concat(lua_State *L, int n)
{
- api_checknelems(L, n);
+ lj_checkapi_slot(n);
if (n >= 2) {
n--;
do {
@@ -763,7 +784,7 @@
L->top -= n;
break;
}
- n -= (int)(L->top - top);
+ n -= (int)(L->top - (top - 2*LJ_FR2));
L->top = top+2;
lj_vm_call(L, top, 1+1);
L->top -= 1+LJ_FR2;
@@ -780,9 +801,8 @@
LUA_API void lua_gettable(lua_State *L, int idx)
{
- cTValue *v, *t = index2adr(L, idx);
- api_checkvalidindex(L, t);
- v = lj_meta_tget(L, t, L->top-1);
+ cTValue *t = index2adr_check(L, idx);
+ cTValue *v = lj_meta_tget(L, t, L->top-1);
if (v == NULL) {
L->top += 2;
lj_vm_call(L, L->top-2, 1+1);
@@ -794,9 +814,8 @@
LUA_API void lua_getfield(lua_State *L, int idx, const char *k)
{
- cTValue *v, *t = index2adr(L, idx);
+ cTValue *v, *t = index2adr_check(L, idx);
TValue key;
- api_checkvalidindex(L, t);
setstrV(L, &key, lj_str_newz(L, k));
v = lj_meta_tget(L, t, &key);
if (v == NULL) {
@@ -812,7 +831,7 @@
LUA_API void lua_rawget(lua_State *L, int idx)
{
cTValue *t = index2adr(L, idx);
- api_check(L, tvistab(t));
+ lj_checkapi(tvistab(t), "stack slot %d is not a table", idx);
copyTV(L, L->top-1, lj_tab_get(L, tabV(t), L->top-1));
}
@@ -819,7 +838,7 @@
LUA_API void lua_rawgeti(lua_State *L, int idx, int n)
{
cTValue *v, *t = index2adr(L, idx);
- api_check(L, tvistab(t));
+ lj_checkapi(tvistab(t), "stack slot %d is not a table", idx);
v = lj_tab_getint(tabV(t), n);
if (v) {
copyTV(L, L->top, v);
@@ -861,8 +880,7 @@
LUA_API void lua_getfenv(lua_State *L, int idx)
{
- cTValue *o = index2adr(L, idx);
- api_checkvalidindex(L, o);
+ cTValue *o = index2adr_check(L, idx);
if (tvisfunc(o)) {
settabV(L, L->top, tabref(funcV(o)->c.env));
} else if (tvisudata(o)) {
@@ -879,12 +897,14 @@
{
cTValue *t = index2adr(L, idx);
int more;
- api_check(L, tvistab(t));
- more = lj_tab_next(L, tabV(t), L->top-1);
- if (more) {
+ lj_checkapi(tvistab(t), "stack slot %d is not a table", idx);
+ more = lj_tab_next(tabV(t), L->top-1, L->top-1);
+ if (more > 0) {
incr_top(L); /* Return new key and value slot. */
- } else { /* End of traversal. */
+ } else if (!more) { /* End of traversal. */
L->top--; /* Remove key slot. */
+ } else {
+ lj_err_msg(L, LJ_ERR_NEXTIDX);
}
return more;
}
@@ -892,7 +912,8 @@
LUA_API const char *lua_getupvalue(lua_State *L, int idx, int n)
{
TValue *val;
- const char *name = lj_debug_uvnamev(index2adr(L, idx), (uint32_t)(n-1), &val);
+ GCobj *o;
+ const char *name = lj_debug_uvnamev(index2adr(L, idx), (uint32_t)(n-1), &val, &o);
if (name) {
copyTV(L, L->top, val);
incr_top(L);
@@ -904,7 +925,7 @@
{
GCfunc *fn = funcV(index2adr(L, idx));
n--;
- api_check(L, (uint32_t)n < fn->l.nupvalues);
+ lj_checkapi((uint32_t)n < fn->l.nupvalues, "bad upvalue %d", n);
return isluafunc(fn) ? (void *)gcref(fn->l.uvptr[n]) :
(void *)&fn->c.upvalue[n];
}
@@ -914,8 +935,10 @@
GCfunc *fn1 = funcV(index2adr(L, idx1));
GCfunc *fn2 = funcV(index2adr(L, idx2));
n1--; n2--;
- api_check(L, isluafunc(fn1) && (uint32_t)n1 < fn1->l.nupvalues);
- api_check(L, isluafunc(fn2) && (uint32_t)n2 < fn2->l.nupvalues);
+ lj_checkapi(isluafunc(fn1), "stack slot %d is not a Lua function", idx1);
+ lj_checkapi(isluafunc(fn2), "stack slot %d is not a Lua function", idx2);
+ lj_checkapi((uint32_t)n1 < fn1->l.nupvalues, "bad upvalue %d", n1+1);
+ lj_checkapi((uint32_t)n2 < fn2->l.nupvalues, "bad upvalue %d", n2+1);
setgcrefr(fn1->l.uvptr[n1], fn2->l.uvptr[n2]);
lj_gc_objbarrier(L, fn1, gcref(fn1->l.uvptr[n1]));
}
@@ -944,9 +967,8 @@
LUA_API void lua_settable(lua_State *L, int idx)
{
TValue *o;
- cTValue *t = index2adr(L, idx);
- api_checknelems(L, 2);
- api_checkvalidindex(L, t);
+ cTValue *t = index2adr_check(L, idx);
+ lj_checkapi_slot(2);
o = lj_meta_tset(L, t, L->top-2);
if (o) {
/* NOBARRIER: lj_meta_tset ensures the table is not black. */
@@ -965,9 +987,8 @@
{
TValue *o;
TValue key;
- cTValue *t = index2adr(L, idx);
- api_checknelems(L, 1);
- api_checkvalidindex(L, t);
+ cTValue *t = index2adr_check(L, idx);
+ lj_checkapi_slot(1);
setstrV(L, &key, lj_str_newz(L, k));
o = lj_meta_tset(L, t, &key);
if (o) {
@@ -986,7 +1007,7 @@
{
GCtab *t = tabV(index2adr(L, idx));
TValue *dst, *key;
- api_checknelems(L, 2);
+ lj_checkapi_slot(2);
key = L->top-2;
dst = lj_tab_set(L, t, key);
copyTV(L, dst, key+1);
@@ -998,7 +1019,7 @@
{
GCtab *t = tabV(index2adr(L, idx));
TValue *dst, *src;
- api_checknelems(L, 1);
+ lj_checkapi_slot(1);
dst = lj_tab_setint(L, t, n);
src = L->top-1;
copyTV(L, dst, src);
@@ -1010,13 +1031,12 @@
{
global_State *g;
GCtab *mt;
- cTValue *o = index2adr(L, idx);
- api_checknelems(L, 1);
- api_checkvalidindex(L, o);
+ cTValue *o = index2adr_check(L, idx);
+ lj_checkapi_slot(1);
if (tvisnil(L->top-1)) {
mt = NULL;
} else {
- api_check(L, tvistab(L->top-1));
+ lj_checkapi(tvistab(L->top-1), "top stack slot is not a table");
mt = tabV(L->top-1);
}
g = G(L);
@@ -1032,6 +1052,7 @@
/* Flush cache, since traces specialize to basemt. But not during __gc. */
if (lj_trace_flushall(L))
lj_err_caller(L, LJ_ERR_NOGCMM);
+ o = index2adr(L, idx); /* Stack may have been reallocated. */
if (tvisbool(o)) {
/* NOBARRIER: basemt is a GC root. */
setgcref(basemt_it(g, LJ_TTRUE), obj2gco(mt));
@@ -1053,11 +1074,10 @@
LUA_API int lua_setfenv(lua_State *L, int idx)
{
- cTValue *o = index2adr(L, idx);
+ cTValue *o = index2adr_check(L, idx);
GCtab *t;
- api_checknelems(L, 1);
- api_checkvalidindex(L, o);
- api_check(L, tvistab(L->top-1));
+ lj_checkapi_slot(1);
+ lj_checkapi(tvistab(L->top-1), "top stack slot is not a table");
t = tabV(L->top-1);
if (tvisfunc(o)) {
setgcref(funcV(o)->c.env, obj2gco(t));
@@ -1078,13 +1098,14 @@
{
cTValue *f = index2adr(L, idx);
TValue *val;
+ GCobj *o;
const char *name;
- api_checknelems(L, 1);
- name = lj_debug_uvnamev(f, (uint32_t)(n-1), &val);
+ lj_checkapi_slot(1);
+ name = lj_debug_uvnamev(f, (uint32_t)(n-1), &val, &o);
if (name) {
L->top--;
copyTV(L, val, L->top);
- lj_gc_barrier(L, funcV(f), L->top);
+ lj_gc_barrier(L, o, L->top);
}
return name;
}
@@ -1106,8 +1127,9 @@
LUA_API void lua_call(lua_State *L, int nargs, int nresults)
{
- api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
- api_checknelems(L, nargs+1);
+ lj_checkapi(L->status == LUA_OK || L->status == LUA_ERRERR,
+ "thread called in wrong state %d", L->status);
+ lj_checkapi_slot(nargs+1);
lj_vm_call(L, api_call_base(L, nargs), nresults+1);
}
@@ -1117,13 +1139,13 @@
uint8_t oldh = hook_save(g);
ptrdiff_t ef;
int status;
- api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
- api_checknelems(L, nargs+1);
+ lj_checkapi(L->status == LUA_OK || L->status == LUA_ERRERR,
+ "thread called in wrong state %d", L->status);
+ lj_checkapi_slot(nargs+1);
if (errfunc == 0) {
ef = 0;
} else {
- cTValue *o = stkindex2adr(L, errfunc);
- api_checkvalidindex(L, o);
+ cTValue *o = index2adr_stack(L, errfunc);
ef = savestack(L, o);
}
status = lj_vm_pcall(L, api_call_base(L, nargs), nresults+1, ef);
@@ -1138,7 +1160,10 @@
fn->c.f = func;
setfuncV(L, top++, fn);
if (LJ_FR2) setnilV(top++);
- setlightudV(top++, checklightudptr(L, ud));
+#if LJ_64
+ ud = lj_lightud_intern(L, ud);
+#endif
+ setrawlightudV(top++, ud);
cframe_nres(L->cframe) = 1+0; /* Zero results. */
L->top = top;
return top-1; /* Now call the newly allocated C function. */
@@ -1149,7 +1174,8 @@
global_State *g = G(L);
uint8_t oldh = hook_save(g);
int status;
- api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
+ lj_checkapi(L->status == LUA_OK || L->status == LUA_ERRERR,
+ "thread called in wrong state %d", L->status);
status = lj_vm_cpcall(L, func, ud, cpcall);
if (status) hook_restore(g, oldh);
return status;
@@ -1198,11 +1224,12 @@
setcont(top, lj_cont_hook);
if (LJ_FR2) top++;
setframe_pc(top, cframe_pc(cf)-1);
+ top++;
+ setframe_gc(top, obj2gco(L), LJ_TTHREAD);
if (LJ_FR2) top++;
- setframe_gc(top, obj2gco(L), LJ_TTHREAD);
setframe_ftsz(top, ((char *)(top+1)-(char *)L->base)+FRAME_CONT);
L->top = L->base = top+1;
-#if LJ_TARGET_X64
+#if ((defined(__GNUC__) || defined(__clang__)) && (LJ_TARGET_X64 || defined(LUAJIT_UNWIND_EXTERNAL)) && !LJ_NO_UNWIND) || LJ_TARGET_WINDOWS
lj_err_throw(L, LUA_YIELD);
#else
L->cframe = NULL;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_arch.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_arch.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_arch.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Target architecture selection.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_ARCH_H
@@ -8,6 +8,8 @@
#include "lua.h"
+/* -- Target definitions -------------------------------------------------- */
+
/* Target endianess. */
#define LUAJIT_LE 0
#define LUAJIT_BE 1
@@ -38,6 +40,14 @@
#define LUAJIT_OS_BSD 4
#define LUAJIT_OS_POSIX 5
+/* Number mode. */
+#define LJ_NUMMODE_SINGLE 0 /* Single-number mode only. */
+#define LJ_NUMMODE_SINGLE_DUAL 1 /* Default to single-number mode. */
+#define LJ_NUMMODE_DUAL 2 /* Dual-number mode only. */
+#define LJ_NUMMODE_DUAL_SINGLE 3 /* Default to dual-number mode. */
+
+/* -- Target detection ---------------------------------------------------- */
+
/* Select native target if no target defined. */
#ifndef LUAJIT_TARGET
@@ -47,7 +57,7 @@
#define LUAJIT_TARGET LUAJIT_ARCH_X64
#elif defined(__arm__) || defined(__arm) || defined(__ARM__) || defined(__ARM)
#define LUAJIT_TARGET LUAJIT_ARCH_ARM
-#elif defined(__aarch64__)
+#elif defined(__aarch64__) || defined(_M_ARM64)
#define LUAJIT_TARGET LUAJIT_ARCH_ARM64
#elif defined(__ppc__) || defined(__ppc) || defined(__PPC__) || defined(__PPC) || defined(__powerpc__) || defined(__powerpc) || defined(__POWERPC__) || defined(__POWERPC) || defined(_M_PPC)
#define LUAJIT_TARGET LUAJIT_ARCH_PPC
@@ -56,7 +66,7 @@
#elif defined(__mips__) || defined(__mips) || defined(__MIPS__) || defined(__MIPS)
#define LUAJIT_TARGET LUAJIT_ARCH_MIPS32
#else
-#error "No support for this architecture (yet)"
+#error "Architecture not supported (in this version), see: https://luajit.org/status.html#architectures"
#endif
#endif
@@ -69,16 +79,23 @@
#elif defined(__linux__)
#define LUAJIT_OS LUAJIT_OS_LINUX
#elif defined(__MACH__) && defined(__APPLE__)
+#include "TargetConditionals.h"
#define LUAJIT_OS LUAJIT_OS_OSX
#elif (defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || \
defined(__NetBSD__) || defined(__OpenBSD__) || \
- defined(__DragonFly__)) && !defined(__ORBIS__)
+ defined(__DragonFly__)) && !defined(__ORBIS__) && !defined(__PROSPERO__)
#define LUAJIT_OS LUAJIT_OS_BSD
-#elif (defined(__sun__) && defined(__svr4__)) || defined(__HAIKU__)
+#elif (defined(__sun__) && defined(__svr4__))
+#define LJ_TARGET_SOLARIS 1
#define LUAJIT_OS LUAJIT_OS_POSIX
+#elif defined(__HAIKU__)
+#define LUAJIT_OS LUAJIT_OS_POSIX
#elif defined(__CYGWIN__)
#define LJ_TARGET_CYGWIN 1
#define LUAJIT_OS LUAJIT_OS_POSIX
+#elif defined(__QNX__)
+#define LJ_TARGET_QNX 1
+#define LUAJIT_OS LUAJIT_OS_POSIX
#else
#define LUAJIT_OS LUAJIT_OS_OTHER
#endif
@@ -103,10 +120,16 @@
#define LJ_TARGET_WINDOWS (LUAJIT_OS == LUAJIT_OS_WINDOWS)
#define LJ_TARGET_LINUX (LUAJIT_OS == LUAJIT_OS_LINUX)
#define LJ_TARGET_OSX (LUAJIT_OS == LUAJIT_OS_OSX)
-#define LJ_TARGET_IOS (LJ_TARGET_OSX && (LUAJIT_TARGET == LUAJIT_ARCH_ARM || LUAJIT_TARGET == LUAJIT_ARCH_ARM64))
+#define LJ_TARGET_BSD (LUAJIT_OS == LUAJIT_OS_BSD)
#define LJ_TARGET_POSIX (LUAJIT_OS > LUAJIT_OS_WINDOWS)
#define LJ_TARGET_DLOPEN LJ_TARGET_POSIX
+#if defined(TARGET_OS_IPHONE) && TARGET_OS_IPHONE
+#define LJ_TARGET_IOS 1
+#else
+#define LJ_TARGET_IOS 0
+#endif
+
#ifdef __CELLOS_LV2__
#define LJ_TARGET_PS3 1
#define LJ_TARGET_CONSOLE 1
@@ -119,6 +142,13 @@
#define NULL ((void*)0)
#endif
+#ifdef __PROSPERO__
+#define LJ_TARGET_PS5 1
+#define LJ_TARGET_CONSOLE 1
+#undef NULL
+#define NULL ((void*)0)
+#endif
+
#ifdef __psp2__
#define LJ_TARGET_PSVITA 1
#define LJ_TARGET_CONSOLE 1
@@ -135,11 +165,22 @@
#define LJ_TARGET_GC64 1
#endif
-#define LJ_NUMMODE_SINGLE 0 /* Single-number mode only. */
-#define LJ_NUMMODE_SINGLE_DUAL 1 /* Default to single-number mode. */
-#define LJ_NUMMODE_DUAL 2 /* Dual-number mode only. */
-#define LJ_NUMMODE_DUAL_SINGLE 3 /* Default to dual-number mode. */
+#ifdef __NX__
+#define LJ_TARGET_NX 1
+#define LJ_TARGET_CONSOLE 1
+#undef NULL
+#define NULL ((void*)0)
+#endif
+#ifdef _UWP
+#define LJ_TARGET_UWP 1
+#if LUAJIT_TARGET == LUAJIT_ARCH_X64
+#define LJ_TARGET_GC64 1
+#endif
+#endif
+
+/* -- Arch-specific settings ---------------------------------------------- */
+
/* Set target architecture properties. */
#if LUAJIT_TARGET == LUAJIT_ARCH_X86
@@ -146,14 +187,10 @@
#define LJ_ARCH_NAME "x86"
#define LJ_ARCH_BITS 32
#define LJ_ARCH_ENDIAN LUAJIT_LE
-#if LJ_TARGET_WINDOWS || LJ_TARGET_CYGWIN
-#define LJ_ABI_WIN 1
-#else
-#define LJ_ABI_WIN 0
-#endif
#define LJ_TARGET_X86 1
#define LJ_TARGET_X86ORX64 1
#define LJ_TARGET_EHRETREG 0
+#define LJ_TARGET_EHRAREG 8
#define LJ_TARGET_MASKSHIFT 1
#define LJ_TARGET_MASKROT 1
#define LJ_TARGET_UNALIGNED 1
@@ -164,21 +201,19 @@
#define LJ_ARCH_NAME "x64"
#define LJ_ARCH_BITS 64
#define LJ_ARCH_ENDIAN LUAJIT_LE
-#if LJ_TARGET_WINDOWS || LJ_TARGET_CYGWIN
-#define LJ_ABI_WIN 1
-#else
-#define LJ_ABI_WIN 0
-#endif
#define LJ_TARGET_X64 1
#define LJ_TARGET_X86ORX64 1
#define LJ_TARGET_EHRETREG 0
+#define LJ_TARGET_EHRAREG 16
#define LJ_TARGET_JUMPRANGE 31 /* +-2^31 = +-2GB */
#define LJ_TARGET_MASKSHIFT 1
#define LJ_TARGET_MASKROT 1
#define LJ_TARGET_UNALIGNED 1
#define LJ_ARCH_NUMMODE LJ_NUMMODE_SINGLE_DUAL
-#ifdef LUAJIT_ENABLE_GC64
+#ifndef LUAJIT_DISABLE_GC64
#define LJ_TARGET_GC64 1
+#elif LJ_TARGET_OSX
+#error "macOS requires GC64 -- don't disable it"
#endif
#elif LUAJIT_TARGET == LUAJIT_ARCH_ARM
@@ -195,6 +230,7 @@
#define LJ_ABI_EABI 1
#define LJ_TARGET_ARM 1
#define LJ_TARGET_EHRETREG 0
+#define LJ_TARGET_EHRAREG 14
#define LJ_TARGET_JUMPRANGE 25 /* +-2^25 = +-32MB */
#define LJ_TARGET_MASKSHIFT 0
#define LJ_TARGET_MASKROT 1
@@ -201,13 +237,13 @@
#define LJ_TARGET_UNIFYROT 2 /* Want only IR_BROR. */
#define LJ_ARCH_NUMMODE LJ_NUMMODE_DUAL
-#if __ARM_ARCH____ARM_ARCH_8__ || __ARM_ARCH_8A__
+#if __ARM_ARCH >= 8 || __ARM_ARCH_8__ || __ARM_ARCH_8A__
#define LJ_ARCH_VERSION 80
-#elif __ARM_ARCH_7__ || __ARM_ARCH_7A__ || __ARM_ARCH_7R__ || __ARM_ARCH_7S__ || __ARM_ARCH_7VE__
+#elif __ARM_ARCH == 7 || __ARM_ARCH_7__ || __ARM_ARCH_7A__ || __ARM_ARCH_7R__ || __ARM_ARCH_7S__ || __ARM_ARCH_7VE__
#define LJ_ARCH_VERSION 70
#elif __ARM_ARCH_6T2__
#define LJ_ARCH_VERSION 61
-#elif __ARM_ARCH_6__ || __ARM_ARCH_6J__ || __ARM_ARCH_6K__ || __ARM_ARCH_6Z__ || __ARM_ARCH_6ZK__
+#elif __ARM_ARCH == 6 || __ARM_ARCH_6__ || __ARM_ARCH_6J__ || __ARM_ARCH_6K__ || __ARM_ARCH_6Z__ || __ARM_ARCH_6ZK__
#define LJ_ARCH_VERSION 60
#else
#define LJ_ARCH_VERSION 50
@@ -223,8 +259,12 @@
#define LJ_ARCH_NAME "arm64"
#define LJ_ARCH_ENDIAN LUAJIT_LE
#endif
+#if !defined(LJ_ABI_PAUTH) && defined(__arm64e__)
+#define LJ_ABI_PAUTH 1
+#endif
#define LJ_TARGET_ARM64 1
#define LJ_TARGET_EHRETREG 0
+#define LJ_TARGET_EHRAREG 30
#define LJ_TARGET_JUMPRANGE 27 /* +-2^27 = +-128MB */
#define LJ_TARGET_MASKSHIFT 1
#define LJ_TARGET_MASKROT 1
@@ -254,23 +294,44 @@
#else
#define LJ_ARCH_BITS 32
#define LJ_ARCH_NAME "ppc"
+
+#if !defined(LJ_ARCH_HASFPU)
+#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
+#define LJ_ARCH_HASFPU 0
+#else
+#define LJ_ARCH_HASFPU 1
#endif
+#endif
+#if !defined(LJ_ABI_SOFTFP)
+#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
+#define LJ_ABI_SOFTFP 1
+#else
+#define LJ_ABI_SOFTFP 0
+#endif
+#endif
+#endif
+
+#if LJ_ABI_SOFTFP
+#define LJ_ARCH_NUMMODE LJ_NUMMODE_DUAL
+#else
+#define LJ_ARCH_NUMMODE LJ_NUMMODE_DUAL_SINGLE
+#endif
+
#define LJ_TARGET_PPC 1
#define LJ_TARGET_EHRETREG 3
+#define LJ_TARGET_EHRAREG 65
#define LJ_TARGET_JUMPRANGE 25 /* +-2^25 = +-32MB */
#define LJ_TARGET_MASKSHIFT 0
#define LJ_TARGET_MASKROT 1
#define LJ_TARGET_UNIFYROT 1 /* Want only IR_BROL. */
-#define LJ_ARCH_NUMMODE LJ_NUMMODE_DUAL_SINGLE
#if LJ_TARGET_CONSOLE
#define LJ_ARCH_PPC32ON64 1
#define LJ_ARCH_NOFFI 1
#elif LJ_ARCH_BITS == 64
-#define LJ_ARCH_PPC64 1
-#define LJ_TARGET_GC64 1
-#define LJ_ARCH_NOJIT 1 /* NYI */
+#error "No support for PPC64"
+#undef LJ_TARGET_PPC
#endif
#if _ARCH_PWR7
@@ -302,18 +363,38 @@
#elif LUAJIT_TARGET == LUAJIT_ARCH_MIPS32 || LUAJIT_TARGET == LUAJIT_ARCH_MIPS64
#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL)
+#if __mips_isa_rev >= 6
+#define LJ_TARGET_MIPSR6 1
+#define LJ_TARGET_UNALIGNED 1
+#endif
#if LUAJIT_TARGET == LUAJIT_ARCH_MIPS32
+#if LJ_TARGET_MIPSR6
+#define LJ_ARCH_NAME "mips32r6el"
+#else
#define LJ_ARCH_NAME "mipsel"
+#endif
#else
+#if LJ_TARGET_MIPSR6
+#define LJ_ARCH_NAME "mips64r6el"
+#else
#define LJ_ARCH_NAME "mips64el"
#endif
+#endif
#define LJ_ARCH_ENDIAN LUAJIT_LE
#else
#if LUAJIT_TARGET == LUAJIT_ARCH_MIPS32
+#if LJ_TARGET_MIPSR6
+#define LJ_ARCH_NAME "mips32r6"
+#else
#define LJ_ARCH_NAME "mips"
+#endif
#else
+#if LJ_TARGET_MIPSR6
+#define LJ_ARCH_NAME "mips64r6"
+#else
#define LJ_ARCH_NAME "mips64"
#endif
+#endif
#define LJ_ARCH_ENDIAN LUAJIT_BE
#endif
@@ -337,9 +418,6 @@
#define LJ_ARCH_BITS 32
#define LJ_TARGET_MIPS32 1
#else
-#if LJ_ABI_SOFTFP || !LJ_ARCH_HASFPU
-#define LJ_ARCH_NOJIT 1 /* NYI */
-#endif
#define LJ_ARCH_BITS 64
#define LJ_TARGET_MIPS64 1
#define LJ_TARGET_GC64 1
@@ -346,6 +424,7 @@
#endif
#define LJ_TARGET_MIPS 1
#define LJ_TARGET_EHRETREG 4
+#define LJ_TARGET_EHRAREG 31
#define LJ_TARGET_JUMPRANGE 27 /* 2*2^27 = 256MB-aligned region */
#define LJ_TARGET_MASKSHIFT 1
#define LJ_TARGET_MASKROT 1
@@ -352,7 +431,9 @@
#define LJ_TARGET_UNIFYROT 2 /* Want only IR_BROR. */
#define LJ_ARCH_NUMMODE LJ_NUMMODE_DUAL
-#if _MIPS_ARCH_MIPS32R2 || _MIPS_ARCH_MIPS64R2
+#if LJ_TARGET_MIPSR6
+#define LJ_ARCH_VERSION 60
+#elif _MIPS_ARCH_MIPS32R2 || _MIPS_ARCH_MIPS64R2
#define LJ_ARCH_VERSION 20
#else
#define LJ_ARCH_VERSION 10
@@ -362,9 +443,7 @@
#error "No target architecture defined"
#endif
-#ifndef LJ_PAGESIZE
-#define LJ_PAGESIZE 4096
-#endif
+/* -- Checks for requirements --------------------------------------------- */
/* Check for minimum required compiler versions. */
#if defined(__GNUC__)
@@ -376,7 +455,7 @@
#if __GNUC__ < 4
#error "Need at least GCC 4.0 or newer"
#endif
-#elif LJ_TARGET_ARM || LJ_TARGET_PPC
+#elif LJ_TARGET_ARM || LJ_TARGET_PPC
#if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 2)
#error "Need at least GCC 4.2 or newer"
#endif
@@ -391,11 +470,17 @@
#endif
#endif
#elif !LJ_TARGET_PS3
+#if __clang__
+#if ((__clang_major__ < 3) || ((__clang_major__ == 3) && __clang_minor__ < 5))
+#error "Need at least Clang 3.5 or newer"
+#endif
+#else
#if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 3)
#error "Need at least GCC 4.3 or newer"
#endif
#endif
#endif
+#endif
/* Check target-specific constraints. */
#ifndef _BUILDVM_H
@@ -406,41 +491,51 @@
#elif LJ_TARGET_ARM
#if defined(__ARMEB__)
#error "No support for big-endian ARM"
+#undef LJ_TARGET_ARM
#endif
#if __ARM_ARCH_6M__ || __ARM_ARCH_7M__ || __ARM_ARCH_7EM__
#error "No support for Cortex-M CPUs"
+#undef LJ_TARGET_ARM
#endif
#if !(__ARM_EABI__ || LJ_TARGET_IOS)
#error "Only ARM EABI or iOS 3.0+ ABI is supported"
+#undef LJ_TARGET_ARM
#endif
#elif LJ_TARGET_ARM64
#if defined(_ILP32)
#error "No support for ILP32 model on ARM64"
+#undef LJ_TARGET_ARM64
#endif
#elif LJ_TARGET_PPC
-#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
-#error "No support for PowerPC CPUs without double-precision FPU"
-#endif
-#if !LJ_ARCH_PPC64 && LJ_ARCH_ENDIAN == LUAJIT_LE
+#if defined(_LITTLE_ENDIAN) && (!defined(_BYTE_ORDER) || (_BYTE_ORDER == _LITTLE_ENDIAN))
#error "No support for little-endian PPC32"
+#undef LJ_TARGET_PPC
#endif
-#if LJ_ARCH_PPC64
-#error "No support for PowerPC 64 bit mode (yet)"
+#if defined(__NO_FPRS__) && !defined(_SOFT_FLOAT)
+#error "No support for PPC/e500, use LuaJIT 2.0"
+#undef LJ_TARGET_PPC
#endif
-#ifdef __NO_FPRS__
-#error "No support for PPC/e500 anymore (use LuaJIT 2.0)"
-#endif
#elif LJ_TARGET_MIPS32
#if !((defined(_MIPS_SIM_ABI32) && _MIPS_SIM == _MIPS_SIM_ABI32) || (defined(_ABIO32) && _MIPS_SIM == _ABIO32))
#error "Only o32 ABI supported for MIPS32"
+#undef LJ_TARGET_MIPS
#endif
+#if LJ_TARGET_MIPSR6
+/* Not that useful, since most available r6 CPUs are 64 bit. */
+#error "No support for MIPS32R6"
+#undef LJ_TARGET_MIPS
+#endif
#elif LJ_TARGET_MIPS64
#if !((defined(_MIPS_SIM_ABI64) && _MIPS_SIM == _MIPS_SIM_ABI64) || (defined(_ABI64) && _MIPS_SIM == _ABI64))
+/* MIPS32ON64 aka n32 ABI support might be desirable, but difficult. */
#error "Only n64 ABI supported for MIPS64"
+#undef LJ_TARGET_MIPS
#endif
#endif
#endif
+/* -- Derived defines ----------------------------------------------------- */
+
/* Enable or disable the dual-number mode for the VM. */
#if (LJ_ARCH_NUMMODE == LJ_NUMMODE_SINGLE && LUAJIT_NUMMODE == 2) || \
(LJ_ARCH_NUMMODE == LJ_NUMMODE_DUAL && LUAJIT_NUMMODE == 1)
@@ -490,6 +585,13 @@
#define LJ_HASFFI 1
#endif
+/* Disable or enable the string buffer extension. */
+#if defined(LUAJIT_DISABLE_BUFFER)
+#define LJ_HASBUFFER 0
+#else
+#define LJ_HASBUFFER 1
+#endif
+
#if defined(LUAJIT_DISABLE_PROFILE)
#define LJ_HASPROFILE 0
#elif LJ_TARGET_POSIX
@@ -512,7 +614,12 @@
#define LJ_ABI_SOFTFP 0
#endif
#define LJ_SOFTFP (!LJ_ARCH_HASFPU)
+#define LJ_SOFTFP32 (LJ_SOFTFP && LJ_32)
+#ifndef LJ_ABI_PAUTH
+#define LJ_ABI_PAUTH 0
+#endif
+
#if LJ_ARCH_ENDIAN == LUAJIT_BE
#define LJ_LE 0
#define LJ_BE 1
@@ -537,26 +644,52 @@
#define LJ_TARGET_UNALIGNED 0
#endif
+#ifndef LJ_PAGESIZE
+#define LJ_PAGESIZE 4096
+#endif
+
/* Various workarounds for embedded operating systems or weak C runtimes. */
#if defined(__ANDROID__) || defined(__symbian__) || LJ_TARGET_XBOX360 || LJ_TARGET_WINDOWS
#define LUAJIT_NO_LOG2
#endif
-#if defined(__symbian__) || LJ_TARGET_WINDOWS
-#define LUAJIT_NO_EXP2
-#endif
#if LJ_TARGET_CONSOLE || (LJ_TARGET_IOS && __IPHONE_OS_VERSION_MIN_REQUIRED >= __IPHONE_8_0)
#define LJ_NO_SYSTEM 1
#endif
-#if !defined(LUAJIT_NO_UNWIND) && __GNU_COMPACT_EH__
-/* NYI: no support for compact unwind specification, yet. */
-#define LUAJIT_NO_UNWIND 1
+#if LJ_TARGET_WINDOWS || LJ_TARGET_CYGWIN
+#define LJ_ABI_WIN 1
+#else
+#define LJ_ABI_WIN 0
#endif
-#if defined(LUAJIT_NO_UNWIND) || defined(__symbian__) || LJ_TARGET_IOS || LJ_TARGET_PS3 || LJ_TARGET_PS4
+#if LJ_TARGET_WINDOWS
+#if LJ_TARGET_UWP
+#define LJ_WIN_VALLOC VirtualAllocFromApp
+#define LJ_WIN_VPROTECT VirtualProtectFromApp
+extern void *LJ_WIN_LOADLIBA(const char *path);
+#else
+#define LJ_WIN_VALLOC VirtualAlloc
+#define LJ_WIN_VPROTECT VirtualProtect
+#define LJ_WIN_LOADLIBA(path) LoadLibraryExA((path), NULL, 0)
+#endif
+#endif
+
+#if defined(LUAJIT_NO_UNWIND) || __GNU_COMPACT_EH__ || defined(__symbian__) || LJ_TARGET_IOS || LJ_TARGET_PS3 || LJ_TARGET_PS4 || LJ_TARGET_PS5
#define LJ_NO_UNWIND 1
#endif
+#if !LJ_NO_UNWIND && !defined(LUAJIT_UNWIND_INTERNAL) && (LJ_ABI_WIN || (defined(LUAJIT_UNWIND_EXTERNAL) && (defined(__GNUC__) || defined(__clang__))))
+#define LJ_UNWIND_EXT 1
+#else
+#define LJ_UNWIND_EXT 0
+#endif
+
+#if LJ_UNWIND_EXT && LJ_HASJIT && !LJ_TARGET_ARM && !(LJ_ABI_WIN && LJ_TARGET_X86)
+#define LJ_UNWIND_JIT 1
+#else
+#define LJ_UNWIND_JIT 0
+#endif
+
/* Compatibility with Lua 5.1 vs. 5.2. */
#ifdef LUAJIT_ENABLE_LUA52COMPAT
#define LJ_52 1
@@ -564,4 +697,46 @@
#define LJ_52 0
#endif
+/* -- VM security --------------------------------------------------------- */
+
+/* Don't make any changes here. Instead build with:
+** make "XCFLAGS=-DLUAJIT_SECURITY_flag=value"
+**
+** Important note to distro maintainers: DO NOT change the defaults for a
+** regular distro build -- neither upwards, nor downwards!
+** These build-time configurable security flags are intended for embedders
+** who may have specific needs wrt. security vs. performance.
+*/
+
+/* Security defaults. */
+#ifndef LUAJIT_SECURITY_PRNG
+/* PRNG init: 0 = fixed/insecure, 1 = secure from OS. */
+#define LUAJIT_SECURITY_PRNG 1
#endif
+
+#ifndef LUAJIT_SECURITY_STRHASH
+/* String hash: 0 = sparse only, 1 = sparse + dense. */
+#define LUAJIT_SECURITY_STRHASH 1
+#endif
+
+#ifndef LUAJIT_SECURITY_STRID
+/* String IDs: 0 = linear, 1 = reseed < 255, 2 = reseed < 15, 3 = random. */
+#define LUAJIT_SECURITY_STRID 1
+#endif
+
+#ifndef LUAJIT_SECURITY_MCODE
+/* Machine code page protection: 0 = insecure RWX, 1 = secure RW^X. */
+#define LUAJIT_SECURITY_MCODE 1
+#endif
+
+#define LJ_SECURITY_MODE \
+ ( 0u \
+ | ((LUAJIT_SECURITY_PRNG & 3) << 0) \
+ | ((LUAJIT_SECURITY_STRHASH & 3) << 2) \
+ | ((LUAJIT_SECURITY_STRID & 3) << 4) \
+ | ((LUAJIT_SECURITY_MCODE & 3) << 6) \
+ )
+#define LJ_SECURITY_MODESTRING \
+ "\004prng\007strhash\005strid\005mcode"
+
+#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_asm_c
@@ -11,6 +11,7 @@
#if LJ_HASJIT
#include "lj_gc.h"
+#include "lj_buf.h"
#include "lj_str.h"
#include "lj_tab.h"
#include "lj_frame.h"
@@ -22,7 +23,6 @@
#include "lj_ircall.h"
#include "lj_iropt.h"
#include "lj_mcode.h"
-#include "lj_iropt.h"
#include "lj_trace.h"
#include "lj_snap.h"
#include "lj_asm.h"
@@ -29,6 +29,7 @@
#include "lj_dispatch.h"
#include "lj_vm.h"
#include "lj_target.h"
+#include "lj_prng.h"
#ifdef LUA_USE_ASSERT
#include <stdio.h>
@@ -72,6 +73,8 @@
IRRef snaprename; /* Rename highwater mark for snapshot check. */
SnapNo snapno; /* Current snapshot number. */
SnapNo loopsnapno; /* Loop snapshot number. */
+ int snapalloc; /* Current snapshot needs allocation. */
+ BloomFilter snapfilt1, snapfilt2; /* Filled with snapshot refs. */
IRRef fuseref; /* Fusion limit (loopref, 0 or FUSE_DISABLED). */
IRRef sectref; /* Section base reference (loopref or 0). */
@@ -85,11 +88,18 @@
MCode *mcbot; /* Bottom of reserved MCode. */
MCode *mctop; /* Top of generated MCode. */
+ MCode *mctoporig; /* Original top of generated MCode. */
MCode *mcloop; /* Pointer to loop MCode (or NULL). */
MCode *invmcp; /* Points to invertible loop branch (or NULL). */
MCode *flagmcp; /* Pending opportunity to merge flag setting ins. */
MCode *realign; /* Realign loop if not NULL. */
+#ifdef LUAJIT_RANDOM_RA
+ /* Randomize register allocation. OK for fuzz testing, not for production. */
+ uint64_t prngbits;
+ PRNGState prngstate;
+#endif
+
#ifdef RID_NUM_KREF
intptr_t krefk[RID_NUM_KREF];
#endif
@@ -97,6 +107,12 @@
uint16_t parentmap[LJ_MAX_JSLOTS]; /* Parent instruction to RegSP map. */
} ASMState;
+#ifdef LUA_USE_ASSERT
+#define lj_assertA(c, ...) lj_assertG_(J2G(as->J), (c), __VA_ARGS__)
+#else
+#define lj_assertA(c, ...) ((void)as)
+#endif
+
#define IR(ref) (&as->ir[(ref)])
#define ASMREF_TMP1 REF_TRUE /* Temp. register. */
@@ -128,9 +144,8 @@
#ifdef LUA_USE_ASSERT
if (as->mcp + MCLIM_REDZONE < as->mcp_prev) {
IRIns *ir = IR(as->curins+1);
- fprintf(stderr, "RED ZONE OVERFLOW: %p IR %04d %02d %04d %04d\n", as->mcp,
- as->curins+1-REF_BIAS, ir->o, ir->op1-REF_BIAS, ir->op2-REF_BIAS);
- lua_assert(0);
+ lj_assertA(0, "red zone overflow: %p IR %04d %02d %04d %04d\n", as->mcp,
+ as->curins+1-REF_BIAS, ir->o, ir->op1-REF_BIAS, ir->op2-REF_BIAS);
}
#endif
if (LJ_UNLIKELY(as->mcp < as->mclim)) asm_mclimit(as);
@@ -165,6 +180,41 @@
0
};
+#ifdef LUAJIT_RANDOM_RA
+/* Return a fixed number of random bits from the local PRNG state. */
+static uint32_t ra_random_bits(ASMState *as, uint32_t nbits) {
+ uint64_t b = as->prngbits;
+ uint32_t res = (1u << nbits) - 1u;
+ if (b <= res) b = lj_prng_u64(&as->prngstate) | (1ull << 63);
+ res &= (uint32_t)b;
+ as->prngbits = b >> nbits;
+ return res;
+}
+
+/* Pick a random register from a register set. */
+static Reg rset_pickrandom(ASMState *as, RegSet rs)
+{
+ Reg r = rset_pickbot_(rs);
+ rs >>= r;
+ if (rs > 1) { /* More than one bit set? */
+ while (1) {
+ /* We need to sample max. the GPR or FPR half of the set. */
+ uint32_t d = ra_random_bits(as, RSET_BITS-1);
+ if ((rs >> d) & 1) {
+ r += d;
+ break;
+ }
+ }
+ }
+ return r;
+}
+#define rset_picktop(rs) rset_pickrandom(as, rs)
+#define rset_pickbot(rs) rset_pickrandom(as, rs)
+#else
+#define rset_picktop(rs) rset_picktop_(rs)
+#define rset_pickbot(rs) rset_pickbot_(rs)
+#endif
+
/* -- Target-specific instruction emitter --------------------------------- */
#if LJ_TARGET_X86ORX64
@@ -244,7 +294,7 @@
*p++ = *q >= 'A' && *q <= 'Z' ? *q + 0x20 : *q;
} else {
*p++ = '?';
- lua_assert(0);
+ lj_assertA(0, "bad register %d for debug format \"%s\"", r, fmt);
}
} else if (e[1] == 'f' || e[1] == 'i') {
IRRef ref;
@@ -262,7 +312,7 @@
} else if (e[1] == 'x') {
p += sprintf(p, "%08x", va_arg(argp, int32_t));
} else {
- lua_assert(0);
+ lj_assertA(0, "bad debug format code");
}
fmt = e+2;
}
@@ -321,7 +371,7 @@
Reg r;
if (ra_iskref(ref)) {
r = ra_krefreg(ref);
- lua_assert(!rset_test(as->freeset, r));
+ lj_assertA(!rset_test(as->freeset, r), "rematk of free reg %d", r);
ra_free(as, r);
ra_modified(as, r);
#if LJ_64
@@ -333,12 +383,14 @@
}
ir = IR(ref);
r = ir->r;
- lua_assert(ra_hasreg(r) && !ra_hasspill(ir->s));
+ lj_assertA(ra_hasreg(r), "rematk of K%03d has no reg", REF_BIAS - ref);
+ lj_assertA(!ra_hasspill(ir->s),
+ "rematk of K%03d has spill slot [%x]", REF_BIAS - ref, ir->s);
ra_free(as, r);
ra_modified(as, r);
ir->r = RID_INIT; /* Do not keep any hint. */
RA_DBGX((as, "remat $i $r", ir, r));
-#if !LJ_SOFTFP
+#if !LJ_SOFTFP32
if (ir->o == IR_KNUM) {
emit_loadk64(as, r, ir);
} else
@@ -347,7 +399,8 @@
ra_sethint(ir->r, RID_BASE); /* Restore BASE register hint. */
emit_getgl(as, r, jit_base);
} else if (emit_canremat(ASMREF_L) && ir->o == IR_KPRI) {
- lua_assert(irt_isnil(ir->t)); /* REF_NIL stores ASMREF_L register. */
+ /* REF_NIL stores ASMREF_L register. */
+ lj_assertA(irt_isnil(ir->t), "rematk of bad ASMREF_L");
emit_getgl(as, r, cur_L);
#if LJ_64
} else if (ir->o == IR_KINT64) {
@@ -360,8 +413,9 @@
#endif
#endif
} else {
- lua_assert(ir->o == IR_KINT || ir->o == IR_KGC ||
- ir->o == IR_KPTR || ir->o == IR_KKPTR || ir->o == IR_KNULL);
+ lj_assertA(ir->o == IR_KINT || ir->o == IR_KGC ||
+ ir->o == IR_KPTR || ir->o == IR_KKPTR || ir->o == IR_KNULL,
+ "rematk of bad IR op %d", ir->o);
emit_loadi(as, r, ir->i);
}
return r;
@@ -371,7 +425,8 @@
static int32_t ra_spill(ASMState *as, IRIns *ir)
{
int32_t slot = ir->s;
- lua_assert(ir >= as->ir + REF_TRUE);
+ lj_assertA(ir >= as->ir + REF_TRUE,
+ "spill of K%03d", REF_BIAS - (int)(ir - as->ir));
if (!ra_hasspill(slot)) {
if (irt_is64(ir->t)) {
slot = as->evenspill;
@@ -396,7 +451,9 @@
{
IRIns *ir = IR(ref);
Reg r = ir->r;
- lua_assert(ra_hasreg(r) && !ra_hasspill(ir->s));
+ lj_assertA(ra_hasreg(r), "release of TMP%d has no reg", ref-ASMREF_TMP1+1);
+ lj_assertA(!ra_hasspill(ir->s),
+ "release of TMP%d has spill slot [%x]", ref-ASMREF_TMP1+1, ir->s);
ra_free(as, r);
ra_modified(as, r);
ir->r = RID_INIT;
@@ -412,7 +469,7 @@
IRIns *ir = IR(ref);
int32_t ofs = ra_spill(as, ir); /* Force a spill slot. */
Reg r = ir->r;
- lua_assert(ra_hasreg(r));
+ lj_assertA(ra_hasreg(r), "restore of IR %04d has no reg", ref - REF_BIAS);
ra_sethint(ir->r, r); /* Keep hint. */
ra_free(as, r);
if (!rset_test(as->weakset, r)) { /* Only restore non-weak references. */
@@ -441,7 +498,7 @@
{
IRRef ref;
RegCost cost = ~(RegCost)0;
- lua_assert(allow != RSET_EMPTY);
+ lj_assertA(allow != RSET_EMPTY, "evict from empty set");
if (RID_NUM_FPR == 0 || allow < RID2RSET(RID_MAX_GPR)) {
GPRDEF(MINCOST)
} else {
@@ -448,7 +505,8 @@
FPRDEF(MINCOST)
}
ref = regcost_ref(cost);
- lua_assert(ra_iskref(ref) || (ref >= as->T->nk && ref < as->T->nins));
+ lj_assertA(ra_iskref(ref) || (ref >= as->T->nk && ref < as->T->nins),
+ "evict of out-of-range IR %04d", ref - REF_BIAS);
/* Preferably pick any weak ref instead of a non-weak, non-const ref. */
if (!irref_isk(ref) && (as->weakset & allow)) {
IRIns *ir = IR(ref);
@@ -548,7 +606,11 @@
IRIns *ir = IR(ref);
if ((ir->o == IR_KINT64 && k == (int64_t)ir_kint64(ir)->u64) ||
#if LJ_GC64
+#if LJ_TARGET_ARM64
+ (ir->o == IR_KINT && (uint64_t)k == (uint32_t)ir->i) ||
+#else
(ir->o == IR_KINT && k == ir->i) ||
+#endif
(ir->o == IR_KGC && k == (intptr_t)ir_kgc(ir)) ||
((ir->o == IR_KPTR || ir->o == IR_KKPTR) &&
k == (intptr_t)ir_kptr(ir))
@@ -606,7 +668,8 @@
IRIns *ir = IR(ref);
RegSet pick = as->freeset & allow;
Reg r;
- lua_assert(ra_noreg(ir->r));
+ lj_assertA(ra_noreg(ir->r),
+ "IR %04d already has reg %d", ref - REF_BIAS, ir->r);
if (pick) {
/* First check register hint from propagation or PHI. */
if (ra_hashint(ir->r)) {
@@ -670,8 +733,10 @@
IRIns *ir = IR(ref);
ir->r = (uint8_t)up;
as->cost[down] = 0;
- lua_assert((down < RID_MAX_GPR) == (up < RID_MAX_GPR));
- lua_assert(!rset_test(as->freeset, down) && rset_test(as->freeset, up));
+ lj_assertA((down < RID_MAX_GPR) == (up < RID_MAX_GPR),
+ "rename between GPR/FPR %d and %d", down, up);
+ lj_assertA(!rset_test(as->freeset, down), "rename from free reg %d", down);
+ lj_assertA(rset_test(as->freeset, up), "rename to non-free reg %d", up);
ra_free(as, down); /* 'down' is free ... */
ra_modified(as, down);
rset_clear(as->freeset, up); /* ... and 'up' is now allocated. */
@@ -679,7 +744,14 @@
RA_DBGX((as, "rename $f $r $r", regcost_ref(as->cost[up]), down, up));
emit_movrr(as, ir, down, up); /* Backwards codegen needs inverse move. */
if (!ra_hasspill(IR(ref)->s)) { /* Add the rename to the IR. */
- ra_addrename(as, down, ref, as->snapno);
+ /*
+ ** The rename is effective at the subsequent (already emitted) exit
+ ** branch. This is for the current snapshot (as->snapno). Except if we
+ ** haven't yet allocated any refs for the snapshot (as->snapalloc == 1),
+ ** then it belongs to the next snapshot.
+ ** See also the discussion at asm_snap_checkrename().
+ */
+ ra_addrename(as, down, ref, as->snapno + as->snapalloc);
}
}
@@ -712,7 +784,7 @@
{
Reg dest = ra_dest(as, ir, RID2RSET(r));
if (dest != r) {
- lua_assert(rset_test(as->freeset, r));
+ lj_assertA(rset_test(as->freeset, r), "dest reg %d is not free", r);
ra_modified(as, r);
emit_movrr(as, ir, dest, r);
}
@@ -745,8 +817,9 @@
#endif
#endif
} else if (ir->o != IR_KPRI) {
- lua_assert(ir->o == IR_KINT || ir->o == IR_KGC ||
- ir->o == IR_KPTR || ir->o == IR_KKPTR || ir->o == IR_KNULL);
+ lj_assertA(ir->o == IR_KINT || ir->o == IR_KGC ||
+ ir->o == IR_KPTR || ir->o == IR_KKPTR || ir->o == IR_KNULL,
+ "K%03d has bad IR op %d", REF_BIAS - lref, ir->o);
emit_loadi(as, dest, ir->i);
return;
}
@@ -791,11 +864,11 @@
}
#endif
-#if !LJ_64
/* Force a RID_RETLO/RID_RETHI destination register pair (marked as free). */
static void ra_destpair(ASMState *as, IRIns *ir)
{
Reg destlo = ir->r, desthi = (ir+1)->r;
+ IRIns *irx = (LJ_64 && !irt_is64(ir->t)) ? ir+1 : ir;
/* First spill unrelated refs blocking the destination registers. */
if (!rset_test(as->freeset, RID_RETLO) &&
destlo != RID_RETLO && desthi != RID_RETLO)
@@ -819,29 +892,29 @@
/* Check for conflicts and shuffle the registers as needed. */
if (destlo == RID_RETHI) {
if (desthi == RID_RETLO) {
-#if LJ_TARGET_X86
+#if LJ_TARGET_X86ORX64
*--as->mcp = XI_XCHGa + RID_RETHI;
+ if (LJ_64 && irt_is64(irx->t)) *--as->mcp = 0x48;
#else
- emit_movrr(as, ir, RID_RETHI, RID_TMP);
- emit_movrr(as, ir, RID_RETLO, RID_RETHI);
- emit_movrr(as, ir, RID_TMP, RID_RETLO);
+ emit_movrr(as, irx, RID_RETHI, RID_TMP);
+ emit_movrr(as, irx, RID_RETLO, RID_RETHI);
+ emit_movrr(as, irx, RID_TMP, RID_RETLO);
#endif
} else {
- emit_movrr(as, ir, RID_RETHI, RID_RETLO);
- if (desthi != RID_RETHI) emit_movrr(as, ir, desthi, RID_RETHI);
+ emit_movrr(as, irx, RID_RETHI, RID_RETLO);
+ if (desthi != RID_RETHI) emit_movrr(as, irx, desthi, RID_RETHI);
}
} else if (desthi == RID_RETLO) {
- emit_movrr(as, ir, RID_RETLO, RID_RETHI);
- if (destlo != RID_RETLO) emit_movrr(as, ir, destlo, RID_RETLO);
+ emit_movrr(as, irx, RID_RETLO, RID_RETHI);
+ if (destlo != RID_RETLO) emit_movrr(as, irx, destlo, RID_RETLO);
} else {
- if (desthi != RID_RETHI) emit_movrr(as, ir, desthi, RID_RETHI);
- if (destlo != RID_RETLO) emit_movrr(as, ir, destlo, RID_RETLO);
+ if (desthi != RID_RETHI) emit_movrr(as, irx, desthi, RID_RETHI);
+ if (destlo != RID_RETLO) emit_movrr(as, irx, destlo, RID_RETLO);
}
/* Restore spill slots (if any). */
if (ra_hasspill((ir+1)->s)) ra_save(as, ir+1, RID_RETHI);
if (ra_hasspill(ir->s)) ra_save(as, ir, RID_RETLO);
}
-#endif
/* -- Snapshot handling --------- ----------------------------------------- */
@@ -876,8 +949,11 @@
static void asm_snap_alloc1(ASMState *as, IRRef ref)
{
IRIns *ir = IR(ref);
- if (!irref_isk(ref) && (!(ra_used(ir) || ir->r == RID_SUNK))) {
- if (ir->r == RID_SINK) {
+ if (!irref_isk(ref)) {
+ bloomset(as->snapfilt1, ref);
+ bloomset(as->snapfilt2, hashrot(ref, ref + HASH_BIAS));
+ if (ra_used(ir)) return;
+ if (ir->r == RID_SINK || ir->r == RID_SUNK) {
ir->r = RID_SUNK;
#if LJ_HASFFI
if (ir->o == IR_CNEWI) { /* Allocate CNEWI value. */
@@ -888,11 +964,14 @@
#endif
{ /* Allocate stored values for TNEW, TDUP and CNEW. */
IRIns *irs;
- lua_assert(ir->o == IR_TNEW || ir->o == IR_TDUP || ir->o == IR_CNEW);
+ lj_assertA(ir->o == IR_TNEW || ir->o == IR_TDUP || ir->o == IR_CNEW,
+ "sink of IR %04d has bad op %d", ref - REF_BIAS, ir->o);
for (irs = IR(as->snapref-1); irs > ir; irs--)
if (irs->r == RID_SINK && asm_sunk_store(as, ir, irs)) {
- lua_assert(irs->o == IR_ASTORE || irs->o == IR_HSTORE ||
- irs->o == IR_FSTORE || irs->o == IR_XSTORE);
+ lj_assertA(irs->o == IR_ASTORE || irs->o == IR_HSTORE ||
+ irs->o == IR_FSTORE || irs->o == IR_XSTORE,
+ "sunk store IR %04d has bad op %d",
+ (int)(irs - as->ir) - REF_BIAS, irs->o);
asm_snap_alloc1(as, irs->op2);
if (LJ_32 && (irs+1)->o == IR_HIOP)
asm_snap_alloc1(as, (irs+1)->op2);
@@ -928,11 +1007,12 @@
}
/* Allocate refs escaping to a snapshot. */
-static void asm_snap_alloc(ASMState *as)
+static void asm_snap_alloc(ASMState *as, int snapno)
{
- SnapShot *snap = &as->T->snap[as->snapno];
+ SnapShot *snap = &as->T->snap[snapno];
SnapEntry *map = &as->T->snapmap[snap->mapofs];
MSize n, nent = snap->nent;
+ as->snapfilt1 = as->snapfilt2 = 0;
for (n = 0; n < nent; n++) {
SnapEntry sn = map[n];
IRRef ref = snap_ref(sn);
@@ -939,7 +1019,9 @@
if (!irref_isk(ref)) {
asm_snap_alloc1(as, ref);
if (LJ_SOFTFP && (sn & SNAP_SOFTFPNUM)) {
- lua_assert(irt_type(IR(ref+1)->t) == IRT_SOFTFP);
+ lj_assertA(irt_type(IR(ref+1)->t) == IRT_SOFTFP,
+ "snap %d[%d] points to bad SOFTFP IR %04d",
+ snapno, n, ref - REF_BIAS);
asm_snap_alloc1(as, ref+1);
}
}
@@ -955,35 +1037,26 @@
*/
static int asm_snap_checkrename(ASMState *as, IRRef ren)
{
- SnapShot *snap = &as->T->snap[as->snapno];
- SnapEntry *map = &as->T->snapmap[snap->mapofs];
- MSize n, nent = snap->nent;
- for (n = 0; n < nent; n++) {
- SnapEntry sn = map[n];
- IRRef ref = snap_ref(sn);
- if (ref == ren || (LJ_SOFTFP && (sn & SNAP_SOFTFPNUM) && ++ref == ren)) {
- IRIns *ir = IR(ref);
- ra_spill(as, ir); /* Register renamed, so force a spill slot. */
- RA_DBGX((as, "snaprensp $f $s", ref, ir->s));
- return 1; /* Found. */
- }
+ if (bloomtest(as->snapfilt1, ren) &&
+ bloomtest(as->snapfilt2, hashrot(ren, ren + HASH_BIAS))) {
+ IRIns *ir = IR(ren);
+ ra_spill(as, ir); /* Register renamed, so force a spill slot. */
+ RA_DBGX((as, "snaprensp $f $s", ren, ir->s));
+ return 1; /* Found. */
}
return 0; /* Not found. */
}
-/* Prepare snapshot for next guard instruction. */
+/* Prepare snapshot for next guard or throwing instruction. */
static void asm_snap_prep(ASMState *as)
{
- if (as->curins < as->snapref) {
- do {
- if (as->snapno == 0) return; /* Called by sunk stores before snap #0. */
- as->snapno--;
- as->snapref = as->T->snap[as->snapno].ref;
- } while (as->curins < as->snapref);
- asm_snap_alloc(as);
+ if (as->snapalloc) {
+ /* Alloc on first invocation for each snapshot. */
+ as->snapalloc = 0;
+ asm_snap_alloc(as, as->snapno);
as->snaprename = as->T->nins;
} else {
- /* Process any renames above the highwater mark. */
+ /* Check any renames above the highwater mark. */
for (; as->snaprename < as->T->nins; as->snaprename++) {
IRIns *ir = &as->T->ir[as->snaprename];
if (asm_snap_checkrename(as, ir->op1))
@@ -992,6 +1065,35 @@
}
}
+/* Move to previous snapshot when we cross the current snapshot ref. */
+static void asm_snap_prev(ASMState *as)
+{
+ if (as->curins < as->snapref) {
+ uintptr_t ofs = (uintptr_t)(as->mctoporig - as->mcp);
+ if (ofs >= 0x10000) lj_trace_err(as->J, LJ_TRERR_MCODEOV);
+ do {
+ if (as->snapno == 0) return;
+ as->snapno--;
+ as->snapref = as->T->snap[as->snapno].ref;
+ as->T->snap[as->snapno].mcofs = (uint16_t)ofs; /* Remember mcode ofs. */
+ } while (as->curins < as->snapref); /* May have no ins inbetween. */
+ as->snapalloc = 1;
+ }
+}
+
+/* Fixup snapshot mcode offsetst. */
+static void asm_snap_fixup_mcofs(ASMState *as)
+{
+ uint32_t sz = (uint32_t)(as->mctoporig - as->mcp);
+ SnapShot *snap = as->T->snap;
+ SnapNo i;
+ for (i = as->T->nsnap-1; i > 0; i--) {
+ /* Compute offset from mcode start and store in correct snapshot. */
+ snap[i].mcofs = (uint16_t)(sz - snap[i-1].mcofs);
+ }
+ snap[0].mcofs = 0;
+}
+
/* -- Miscellaneous helpers ----------------------------------------------- */
/* Calculate stack adjustment. */
@@ -1003,21 +1105,26 @@
}
/* Must match with hash*() in lj_tab.c. */
-static uint32_t ir_khash(IRIns *ir)
+static uint32_t ir_khash(ASMState *as, IRIns *ir)
{
uint32_t lo, hi;
+ UNUSED(as);
if (irt_isstr(ir->t)) {
- return ir_kstr(ir)->hash;
+ return ir_kstr(ir)->sid;
} else if (irt_isnum(ir->t)) {
lo = ir_knum(ir)->u32.lo;
hi = ir_knum(ir)->u32.hi << 1;
} else if (irt_ispri(ir->t)) {
- lua_assert(!irt_isnil(ir->t));
+ lj_assertA(!irt_isnil(ir->t), "hash of nil key");
return irt_type(ir->t)-IRT_FALSE;
} else {
- lua_assert(irt_isgcv(ir->t));
+ lj_assertA(irt_isgcv(ir->t), "hash of bad IR type %d", irt_type(ir->t));
lo = u32ptr(ir_kgc(ir));
+#if LJ_GC64
+ hi = (uint32_t)(u64ptr(ir_kgc(ir)) >> 32) | (irt_toitype(ir->t) << 15);
+#else
hi = lo + HASH_BIAS;
+#endif
}
return hashrot(lo, hi);
}
@@ -1031,6 +1138,7 @@
{
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_str_new];
IRRef args[3];
+ asm_snap_prep(as);
args[0] = ASMREF_L; /* lua_State *L */
args[1] = ir->op1; /* const char *str */
args[2] = ir->op2; /* size_t len */
@@ -1043,6 +1151,7 @@
{
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_tab_new1];
IRRef args[2];
+ asm_snap_prep(as);
args[0] = ASMREF_L; /* lua_State *L */
args[1] = ASMREF_TMP1; /* uint32_t ahsize */
as->gcsteps++;
@@ -1055,6 +1164,7 @@
{
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_tab_dup];
IRRef args[2];
+ asm_snap_prep(as);
args[0] = ASMREF_L; /* lua_State *L */
args[1] = ir->op1; /* const GCtab *kt */
as->gcsteps++;
@@ -1080,17 +1190,29 @@
/* -- Buffer operations --------------------------------------------------- */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref);
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode);
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb);
+#endif
static void asm_bufhdr(ASMState *as, IRIns *ir)
{
Reg sb = ra_dest(as, ir, RSET_GPR);
- if ((ir->op2 & IRBUFHDR_APPEND)) {
+ switch (ir->op2) {
+ case IRBUFHDR_RESET: {
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irbp;
+ irbp.ot = IRT(0, IRT_PTR); /* Buffer data pointer type. */
+ emit_storeofs(as, &irbp, tmp, sb, offsetof(SBuf, w));
+ emit_loadofs(as, &irbp, tmp, sb, offsetof(SBuf, b));
+ break;
+ }
+ case IRBUFHDR_APPEND: {
/* Rematerialize const buffer pointer instead of likely spill. */
IRIns *irp = IR(ir->op1);
if (!(ra_hasreg(irp->r) || irp == ir-1 ||
(irp == ir-2 && !ra_used(ir-1)))) {
- while (!(irp->o == IR_BUFHDR && !(irp->op2 & IRBUFHDR_APPEND)))
+ while (!(irp->o == IR_BUFHDR && irp->op2 == IRBUFHDR_RESET))
irp = IR(irp->op1);
if (irref_isk(irp->op1)) {
ra_weak(as, ra_allocref(as, ir->op1, RSET_GPR));
@@ -1097,11 +1219,14 @@
ir = irp;
}
}
- } else {
- Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
- /* Passing ir isn't strictly correct, but it's an IRT_PGC, too. */
- emit_storeofs(as, ir, tmp, sb, offsetof(SBuf, p));
- emit_loadofs(as, ir, tmp, sb, offsetof(SBuf, b));
+ break;
+ }
+#if LJ_HASBUFFER
+ case IRBUFHDR_WRITE:
+ asm_bufhdr_write(as, sb);
+ break;
+#endif
+ default: lj_assertA(0, "bad BUFHDR op2 %d", ir->op2); break;
}
#if LJ_TARGET_X86ORX64
ra_left(as, sb, ir->op1);
@@ -1115,15 +1240,16 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_buf_putstr];
IRRef args[3];
IRIns *irs;
- int kchar = -1;
+ int kchar = -129;
args[0] = ir->op1; /* SBuf * */
args[1] = ir->op2; /* GCstr * */
irs = IR(ir->op2);
- lua_assert(irt_isstr(irs->t));
+ lj_assertA(irt_isstr(irs->t),
+ "BUFPUT of non-string IR %04d", ir->op2 - REF_BIAS);
if (irs->o == IR_KGC) {
GCstr *s = ir_kstr(irs);
if (s->len == 1) { /* Optimize put of single-char string constant. */
- kchar = strdata(s)[0];
+ kchar = (int8_t)strdata(s)[0]; /* Signed! */
args[1] = ASMREF_TMP1; /* int, truncated to char */
ci = &lj_ir_callinfo[IRCALL_lj_buf_putchar];
}
@@ -1133,7 +1259,8 @@
args[1] = ASMREF_TMP1; /* TValue * */
ci = &lj_ir_callinfo[IRCALL_lj_strfmt_putnum];
} else {
- lua_assert(irt_isinteger(IR(irs->op1)->t));
+ lj_assertA(irt_isinteger(IR(irs->op1)->t),
+ "TOSTR of non-numeric IR %04d", irs->op1);
args[1] = irs->op1; /* int */
if (irs->op2 == IRTOSTR_INT)
ci = &lj_ir_callinfo[IRCALL_lj_strfmt_putint];
@@ -1150,8 +1277,8 @@
asm_gencall(as, ci, args);
if (args[1] == ASMREF_TMP1) {
Reg tmp = ra_releasetmp(as, ASMREF_TMP1);
- if (kchar == -1)
- asm_tvptr(as, tmp, irs->op1);
+ if (kchar == -129)
+ asm_tvptr(as, tmp, irs->op1, IRTMPREF_IN1);
else
ra_allockreg(as, kchar, tmp);
}
@@ -1173,6 +1300,7 @@
{
const CCallInfo *ci;
IRRef args[2];
+ asm_snap_prep(as);
args[0] = ASMREF_L;
as->gcsteps++;
if (ir->op2 == IRTOSTR_NUM) {
@@ -1188,7 +1316,7 @@
asm_setupresult(as, ir, ci); /* GCstr * */
asm_gencall(as, ci, args);
if (ir->op2 == IRTOSTR_NUM)
- asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op1);
+ asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op1, IRTMPREF_IN1);
}
#if LJ_32 && LJ_HASFFI && !LJ_SOFTFP && !LJ_TARGET_X86
@@ -1198,7 +1326,8 @@
IRType dt = (((ir-1)->op2 & IRCONV_DSTMASK) >> IRCONV_DSH);
IRCallID id;
IRRef args[2];
- lua_assert((ir-1)->o == IR_CONV && ir->o == IR_HIOP);
+ lj_assertA((ir-1)->o == IR_CONV && ir->o == IR_HIOP,
+ "not a CONV/HIOP pair at IR %04d", (int)(ir - as->ir) - REF_BIAS);
args[LJ_BE] = (ir-1)->op1;
args[LJ_LE] = ir->op1;
if (st == IRT_NUM || st == IRT_FLOAT) {
@@ -1228,14 +1357,21 @@
IRRef args[3];
if (ir->r == RID_SINK)
return;
+ asm_snap_prep(as);
args[0] = ASMREF_L; /* lua_State *L */
args[1] = ir->op1; /* GCtab *t */
args[2] = ASMREF_TMP1; /* cTValue *key */
asm_setupresult(as, ir, ci); /* TValue * */
asm_gencall(as, ci, args);
- asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op2);
+ asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op2, IRTMPREF_IN1);
}
+static void asm_tmpref(ASMState *as, IRIns *ir)
+{
+ Reg r = ra_dest(as, ir, RSET_GPR);
+ asm_tvptr(as, r, ir->op1, ir->op2);
+}
+
static void asm_lref(ASMState *as, IRIns *ir)
{
Reg r = ra_dest(as, ir, RSET_GPR);
@@ -1253,15 +1389,16 @@
const CCallInfo *ci, IRRef *args)
{
uint32_t n = CCI_XNARGS(ci);
- lua_assert(n <= CCI_NARGS_MAX*2); /* Account for split args. */
+ /* Account for split args. */
+ lj_assertA(n <= CCI_NARGS_MAX*2, "too many args %d to collect", n);
if ((ci->flags & CCI_L)) { *args++ = ASMREF_L; n--; }
while (n-- > 1) {
ir = IR(ir->op1);
- lua_assert(ir->o == IR_CARG);
+ lj_assertA(ir->o == IR_CARG, "malformed CALL arg tree");
args[n] = ir->op2 == REF_NIL ? 0 : ir->op2;
}
args[0] = ir->op1 == REF_NIL ? 0 : ir->op1;
- lua_assert(IR(ir->op1)->o != IR_CARG);
+ lj_assertA(IR(ir->op1)->o != IR_CARG, "malformed CALL arg tree");
}
/* Reconstruct CCallInfo flags for CALLX*. */
@@ -1305,32 +1442,6 @@
asm_gencall(as, ci, args);
}
-#if !LJ_SOFTFP
-static void asm_fppow(ASMState *as, IRIns *ir, IRRef lref, IRRef rref)
-{
- const CCallInfo *ci = &lj_ir_callinfo[IRCALL_pow];
- IRRef args[2];
- args[0] = lref;
- args[1] = rref;
- asm_setupresult(as, ir, ci);
- asm_gencall(as, ci, args);
-}
-
-static int asm_fpjoin_pow(ASMState *as, IRIns *ir)
-{
- IRIns *irp = IR(ir->op1);
- if (irp == ir-1 && irp->o == IR_MUL && !ra_used(irp)) {
- IRIns *irpp = IR(irp->op1);
- if (irpp == ir-2 && irpp->o == IR_FPMATH &&
- irpp->op2 == IRFPM_LOG2 && !ra_used(irpp)) {
- asm_fppow(as, ir, irpp->op1, irp->op2);
- return 1;
- }
- }
- return 0;
-}
-#endif
-
/* -- PHI and loop handling ----------------------------------------------- */
/* Break a PHI cycle by renaming to a free register (evict if needed). */
@@ -1601,6 +1712,64 @@
#error "Missing assembler for target CPU"
#endif
+/* -- Common instruction helpers ------------------------------------------ */
+
+#if !LJ_SOFTFP32
+#if !LJ_TARGET_X86ORX64
+#define asm_ldexp(as, ir) asm_callid(as, ir, IRCALL_ldexp)
+#endif
+
+static void asm_pow(ASMState *as, IRIns *ir)
+{
+#if LJ_64 && LJ_HASFFI
+ if (!irt_isnum(ir->t))
+ asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_powi64 :
+ IRCALL_lj_carith_powu64);
+ else
+#endif
+ asm_callid(as, ir, IRCALL_pow);
+}
+
+static void asm_div(ASMState *as, IRIns *ir)
+{
+#if LJ_64 && LJ_HASFFI
+ if (!irt_isnum(ir->t))
+ asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_divi64 :
+ IRCALL_lj_carith_divu64);
+ else
+#endif
+ asm_fpdiv(as, ir);
+}
+#endif
+
+static void asm_mod(ASMState *as, IRIns *ir)
+{
+#if LJ_64 && LJ_HASFFI
+ if (!irt_isint(ir->t))
+ asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_modi64 :
+ IRCALL_lj_carith_modu64);
+ else
+#endif
+ asm_callid(as, ir, IRCALL_lj_vm_modi);
+}
+
+static void asm_fuseequal(ASMState *as, IRIns *ir)
+{
+ /* Fuse HREF + EQ/NE. */
+ if ((ir-1)->o == IR_HREF && ir->op1 == as->curins-1) {
+ as->curins--;
+ asm_href(as, ir-1, (IROp)ir->o);
+ } else {
+ asm_equal(as, ir);
+ }
+}
+
+static void asm_alen(ASMState *as, IRIns *ir)
+{
+ asm_callid(as, ir, ir->op2 == REF_NIL ? IRCALL_lj_tab_len :
+ IRCALL_lj_tab_len_hint);
+}
+
/* -- Instruction dispatch ------------------------------------------------ */
/* Assemble a single instruction. */
@@ -1609,7 +1778,10 @@
switch ((IROp)ir->o) {
/* Miscellaneous ops. */
case IR_LOOP: asm_loop(as); break;
- case IR_NOP: case IR_XBAR: lua_assert(!ra_used(ir)); break;
+ case IR_NOP: case IR_XBAR:
+ lj_assertA(!ra_used(ir),
+ "IR %04d not unused", (int)(ir - as->ir) - REF_BIAS);
+ break;
case IR_USE:
ra_alloc1(as, ir->op1, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR); break;
case IR_PHI: asm_phi(as, ir); break;
@@ -1623,14 +1795,7 @@
case IR_ABC:
asm_comp(as, ir);
break;
- case IR_EQ: case IR_NE:
- if ((ir-1)->o == IR_HREF && ir->op1 == as->curins-1) {
- as->curins--;
- asm_href(as, ir-1, (IROp)ir->o);
- } else {
- asm_equal(as, ir);
- }
- break;
+ case IR_EQ: case IR_NE: asm_fuseequal(as, ir); break;
case IR_RETF: asm_retf(as, ir); break;
@@ -1652,16 +1817,17 @@
case IR_MUL: asm_mul(as, ir); break;
case IR_MOD: asm_mod(as, ir); break;
case IR_NEG: asm_neg(as, ir); break;
-#if LJ_SOFTFP
+#if LJ_SOFTFP32
case IR_DIV: case IR_POW: case IR_ABS:
- case IR_ATAN2: case IR_LDEXP: case IR_FPMATH: case IR_TOBIT:
- lua_assert(0); /* Unused for LJ_SOFTFP. */
+ case IR_LDEXP: case IR_FPMATH: case IR_TOBIT:
+ /* Unused for LJ_SOFTFP32. */
+ lj_assertA(0, "IR %04d with unused op %d",
+ (int)(ir - as->ir) - REF_BIAS, ir->o);
break;
#else
case IR_DIV: asm_div(as, ir); break;
case IR_POW: asm_pow(as, ir); break;
case IR_ABS: asm_abs(as, ir); break;
- case IR_ATAN2: asm_atan2(as, ir); break;
case IR_LDEXP: asm_ldexp(as, ir); break;
case IR_FPMATH: asm_fpmath(as, ir); break;
case IR_TOBIT: asm_tobit(as, ir); break;
@@ -1681,6 +1847,7 @@
case IR_NEWREF: asm_newref(as, ir); break;
case IR_UREFO: case IR_UREFC: asm_uref(as, ir); break;
case IR_FREF: asm_fref(as, ir); break;
+ case IR_TMPREF: asm_tmpref(as, ir); break;
case IR_STRREF: asm_strref(as, ir); break;
case IR_LREF: asm_lref(as, ir); break;
@@ -1691,6 +1858,7 @@
case IR_FLOAD: asm_fload(as, ir); break;
case IR_XLOAD: asm_xload(as, ir); break;
case IR_SLOAD: asm_sload(as, ir); break;
+ case IR_ALEN: asm_alen(as, ir); break;
case IR_ASTORE: case IR_HSTORE: case IR_USTORE: asm_ahustore(as, ir); break;
case IR_FSTORE: asm_fstore(as, ir); break;
@@ -1700,7 +1868,14 @@
case IR_SNEW: case IR_XSNEW: asm_snew(as, ir); break;
case IR_TNEW: asm_tnew(as, ir); break;
case IR_TDUP: asm_tdup(as, ir); break;
- case IR_CNEW: case IR_CNEWI: asm_cnew(as, ir); break;
+ case IR_CNEW: case IR_CNEWI:
+#if LJ_HASFFI
+ asm_cnew(as, ir);
+#else
+ lj_assertA(0, "IR %04d with unused op %d",
+ (int)(ir - as->ir) - REF_BIAS, ir->o);
+#endif
+ break;
/* Buffer operations. */
case IR_BUFHDR: asm_bufhdr(as, ir); break;
@@ -1759,6 +1934,8 @@
IRRef1 sloadins[RID_MAX];
RegSet allow = RSET_ALL; /* Inverse of all coalesced registers. */
RegSet live = RSET_EMPTY; /* Live parent registers. */
+ RegSet pallow = RSET_GPR; /* Registers needed by the parent stack check. */
+ Reg pbase;
IRIns *irp = &as->parent->ir[REF_BASE]; /* Parent base. */
int32_t spadj, spdelta;
int pass2 = 0;
@@ -1767,17 +1944,22 @@
if (as->snapno && as->topslot > as->parent->topslot) {
/* Force snap #0 alloc to prevent register overwrite in stack check. */
- as->snapno = 0;
- asm_snap_alloc(as);
+ asm_snap_alloc(as, 0);
}
- allow = asm_head_side_base(as, irp, allow);
+ pbase = asm_head_side_base(as, irp);
+ if (pbase != RID_NONE) {
+ rset_clear(allow, pbase);
+ rset_clear(pallow, pbase);
+ }
/* Scan all parent SLOADs and collect register dependencies. */
for (i = as->stopins; i > REF_BASE; i--) {
IRIns *ir = IR(i);
RegSP rs;
- lua_assert((ir->o == IR_SLOAD && (ir->op2 & IRSLOAD_PARENT)) ||
- (LJ_SOFTFP && ir->o == IR_HIOP) || ir->o == IR_PVAL);
+ lj_assertA((ir->o == IR_SLOAD && (ir->op2 & IRSLOAD_PARENT)) ||
+ (LJ_SOFTFP && ir->o == IR_HIOP) || ir->o == IR_PVAL,
+ "IR %04d has bad parent op %d",
+ (int)(ir - as->ir) - REF_BIAS, ir->o);
rs = as->parentmap[i - REF_FIRST];
if (ra_hasreg(ir->r)) {
rset_clear(allow, ir->r);
@@ -1798,6 +1980,7 @@
sloadins[rs] = (IRRef1)i;
rset_set(live, rs); /* Block live parent register. */
}
+ if (!ra_hasspill(regsp_spill(rs))) rset_clear(pallow, regsp_reg(rs));
}
/* Calculate stack frame adjustment. */
@@ -1914,7 +2097,7 @@
ExitNo exitno = as->J->exitno;
#endif
as->T->topslot = (uint8_t)as->topslot; /* Remember for child traces. */
- asm_stack_check(as, as->topslot, irp, allow & RSET_GPR, exitno);
+ asm_stack_check(as, as->topslot, irp, pallow, exitno);
}
}
@@ -2005,6 +2188,9 @@
#endif
ra_setup(as);
+#if LJ_TARGET_ARM64
+ ra_setkref(as, RID_GL, (intptr_t)J2G(as->J));
+#endif
/* Clear reg/sp for constants. */
for (ir = IR(T->nk), lastir = IR(REF_BASE); ir < lastir; ir++) {
@@ -2011,6 +2197,7 @@
ir->prev = REGSP_INIT;
if (irt_is64(ir->t) && ir->o != IR_KNULL) {
#if LJ_GC64
+ /* The false-positive of irt_is64() for ASMREF_L (REF_NIL) is OK here. */
ir->i = 0; /* Will become non-zero only for RIP-relative addresses. */
#else
/* Make life easier for backends by putting address of constant in i. */
@@ -2026,6 +2213,7 @@
as->snaprename = nins;
as->snapref = nins;
as->snapno = T->nsnap;
+ as->snapalloc = 0;
as->stopins = REF_BASE;
as->orignins = nins;
@@ -2035,7 +2223,7 @@
ir = IR(REF_FIRST);
if (as->parent) {
uint16_t *p;
- lastir = lj_snap_regspmap(as->parent, as->J->exitno, ir);
+ lastir = lj_snap_regspmap(as->J, as->parent, as->J->exitno, ir);
if (lastir - ir > LJ_MAX_JSLOTS)
lj_trace_err(as->J, LJ_TRERR_NYICOAL);
as->stopins = (IRRef)((lastir-1) - as->ir);
@@ -2074,6 +2262,10 @@
ir->prev = (uint16_t)REGSP_HINT((rload & 15));
rload = lj_ror(rload, 4);
continue;
+ case IR_TMPREF:
+ if ((ir->op2 & IRTMPREF_OUT2) && as->evenspill < 4)
+ as->evenspill = 4; /* TMPREF OUT2 needs two TValues on the stack. */
+ break;
#endif
case IR_CALLXS: {
CCallInfo ci;
@@ -2083,7 +2275,17 @@
as->modset |= RSET_SCRATCH;
continue;
}
- case IR_CALLN: case IR_CALLA: case IR_CALLL: case IR_CALLS: {
+ case IR_CALLL:
+ /* lj_vm_next needs two TValues on the stack. */
+#if LJ_TARGET_X64 && LJ_ABI_WIN
+ if (ir->op2 == IRCALL_lj_vm_next && as->evenspill < SPS_FIRST + 4)
+ as->evenspill = SPS_FIRST + 4;
+#else
+ if (SPS_FIRST < 4 && ir->op2 == IRCALL_lj_vm_next && as->evenspill < 4)
+ as->evenspill = 4;
+#endif
+ /* fallthrough */
+ case IR_CALLN: case IR_CALLA: case IR_CALLS: {
const CCallInfo *ci = &lj_ir_callinfo[ir->op2];
ir->prev = asm_setup_call_slots(as, ir, ci);
if (inloop)
@@ -2091,7 +2293,6 @@
(RSET_SCRATCH & ~RSET_FPR) : RSET_SCRATCH;
continue;
}
-#if LJ_SOFTFP || (LJ_32 && LJ_HASFFI)
case IR_HIOP:
switch ((ir-1)->o) {
#if LJ_SOFTFP && LJ_TARGET_ARM
@@ -2102,15 +2303,15 @@
}
break;
#endif
-#if !LJ_SOFTFP && LJ_NEED_FP64
+#if !LJ_SOFTFP && LJ_NEED_FP64 && LJ_32 && LJ_HASFFI
case IR_CONV:
if (irt_isfp((ir-1)->t)) {
ir->prev = REGSP_HINT(RID_FPRET);
continue;
}
- /* fallthrough */
#endif
- case IR_CALLN: case IR_CALLXS:
+ /* fallthrough */
+ case IR_CALLN: case IR_CALLL: case IR_CALLS: case IR_CALLXS:
#if LJ_SOFTFP
case IR_MIN: case IR_MAX:
#endif
@@ -2121,12 +2322,11 @@
break;
}
break;
-#endif
#if LJ_SOFTFP
case IR_MIN: case IR_MAX:
if ((ir+1)->o != IR_HIOP) break;
- /* fallthrough */
#endif
+ /* fallthrough */
/* C calls evict all scratch regs and return results in RID_RET. */
case IR_SNEW: case IR_XSNEW: case IR_NEWREF: case IR_BUFPUT:
if (REGARG_NUMGPR < 3 && as->evenspill < 3)
@@ -2137,9 +2337,12 @@
if (ir->op2 != REF_NIL && as->evenspill < 4)
as->evenspill = 4; /* lj_cdata_newv needs 4 args. */
}
+ /* fallthrough */
#else
+ /* fallthrough */
case IR_CNEW:
#endif
+ /* fallthrough */
case IR_TNEW: case IR_TDUP: case IR_CNEWI: case IR_TOSTR:
case IR_BUFSTR:
ir->prev = REGSP_HINT(RID_RET);
@@ -2151,20 +2354,20 @@
as->modset = RSET_SCRATCH;
break;
#if !LJ_SOFTFP
- case IR_ATAN2:
-#if LJ_TARGET_X86
- if (as->evenspill < 4) /* Leave room to call atan2(). */
- as->evenspill = 4;
-#endif
#if !LJ_TARGET_X86ORX64
case IR_LDEXP:
#endif
#endif
+ /* fallthrough */
case IR_POW:
if (!LJ_SOFTFP && irt_isnum(ir->t)) {
if (inloop)
as->modset |= RSET_SCRATCH;
#if LJ_TARGET_X86
+ if (irt_isnum(IR(ir->op2)->t)) {
+ if (as->evenspill < 4) /* Leave room to call pow(). */
+ as->evenspill = 4;
+ }
break;
#else
ir->prev = REGSP_HINT(RID_FPRET);
@@ -2171,9 +2374,9 @@
continue;
#endif
}
- /* fallthrough for integer POW */
+ /* fallthrough */ /* for integer POW */
case IR_DIV: case IR_MOD:
- if (!irt_isnum(ir->t)) {
+ if ((LJ_64 && LJ_SOFTFP) || !irt_isnum(ir->t)) {
ir->prev = REGSP_HINT(RID_RET);
if (inloop)
as->modset |= (RSET_SCRATCH & RSET_GPR);
@@ -2180,6 +2383,16 @@
continue;
}
break;
+#if LJ_64 && LJ_SOFTFP
+ case IR_ADD: case IR_SUB: case IR_MUL:
+ if (irt_isnum(ir->t)) {
+ ir->prev = REGSP_HINT(RID_RET);
+ if (inloop)
+ as->modset |= (RSET_SCRATCH & RSET_GPR);
+ continue;
+ }
+ break;
+#endif
case IR_FPMATH:
#if LJ_TARGET_X86ORX64
if (ir->op2 <= IRFPM_TRUNC) {
@@ -2190,9 +2403,6 @@
continue;
}
break;
- } else if (ir->op2 == IRFPM_EXP2 && !LJ_64) {
- if (as->evenspill < 4) /* Leave room to call pow(). */
- as->evenspill = 4;
}
#endif
if (inloop)
@@ -2208,6 +2418,7 @@
case IR_BSHL: case IR_BSHR: case IR_BSAR:
if ((as->flags & JIT_F_BMI2)) /* Except if BMI2 is available. */
break;
+ /* fallthrough */
case IR_BROL: case IR_BROR:
if (!irref_isk(ir->op2) && !ra_hashint(IR(ir->op2)->r)) {
IR(ir->op2)->r = REGSP_HINT(RID_ECX);
@@ -2252,7 +2463,6 @@
{
ASMState as_;
ASMState *as = &as_;
- MCode *origtop;
/* Remove nops/renames left over from ASM restart due to LJ_TRERR_MCODELM. */
{
@@ -2267,7 +2477,7 @@
/* Ensure an initialized instruction beyond the last one for HIOP checks. */
/* This also allows one RENAME to be added without reallocating curfinal. */
as->orignins = lj_ir_nextins(J);
- J->cur.ir[as->orignins].o = IR_NOP;
+ lj_ir_nop(&J->cur.ir[as->orignins]);
/* Setup initial state. Copy some fields to reduce indirections. */
as->J = J;
@@ -2278,9 +2488,12 @@
as->realign = NULL;
as->loopinv = 0;
as->parent = J->parent ? traceref(J, J->parent) : NULL;
+#ifdef LUAJIT_RANDOM_RA
+ (void)lj_prng_u64(&J2G(J)->prng); /* Ensure PRNG step between traces. */
+#endif
/* Reserve MCode memory. */
- as->mctop = origtop = lj_mcode_reserve(J, &as->mcbot);
+ as->mctop = as->mctoporig = lj_mcode_reserve(J, &as->mcbot);
as->mcp = as->mctop;
as->mclim = as->mcbot + MCLIM_REDZONE;
asm_setup_target(as);
@@ -2319,6 +2532,10 @@
#endif
as->ir = J->curfinal->ir; /* Use the copied IR. */
as->curins = J->cur.nins = as->orignins;
+#ifdef LUAJIT_RANDOM_RA
+ as->prngstate = J2G(J)->prng; /* Must (re)start from identical state. */
+ as->prngbits = 0;
+#endif
RA_DBG_START();
RA_DBGX((as, "===== STOP ====="));
@@ -2338,7 +2555,11 @@
/* Assemble a trace in linear backwards order. */
for (as->curins--; as->curins > as->stopins; as->curins--) {
IRIns *ir = IR(as->curins);
- lua_assert(!(LJ_32 && irt_isint64(ir->t))); /* Handled by SPLIT. */
+ /* 64 bit types handled by SPLIT for 32 bit archs. */
+ lj_assertA(!(LJ_32 && irt_isint64(ir->t)),
+ "IR %04d has unsplit 64 bit type",
+ (int)(ir - as->ir) - REF_BIAS);
+ asm_snap_prev(as);
if (!ra_used(ir) && !ir_sideeff(ir) && (as->flags & JIT_F_OPT_DCE))
continue; /* Dead-code elimination can be soooo easy. */
if (irt_isguard(ir->t))
@@ -2368,10 +2589,13 @@
asm_phi_fixup(as);
if (J->curfinal->nins >= T->nins) { /* IR didn't grow? */
- lua_assert(J->curfinal->nk == T->nk);
+ lj_assertA(J->curfinal->nk == T->nk, "unexpected IR constant growth");
memcpy(J->curfinal->ir + as->orignins, T->ir + as->orignins,
(T->nins - as->orignins) * sizeof(IRIns)); /* Copy RENAMEs. */
T->nins = J->curfinal->nins;
+ /* Fill mcofs of any unprocessed snapshots. */
+ as->curins = REF_FIRST;
+ asm_snap_prev(as);
break; /* Done. */
}
@@ -2390,13 +2614,16 @@
/* Set trace entry point before fixing up tail to allow link to self. */
T->mcode = as->mcp;
T->mcloop = as->mcloop ? (MSize)((char *)as->mcloop - (char *)as->mcp) : 0;
- if (!as->loopref)
+ if (as->loopref)
+ asm_loop_tail_fixup(as);
+ else
asm_tail_fixup(as, T->link); /* Note: this may change as->mctop! */
T->szmcode = (MSize)((char *)as->mctop - (char *)as->mcp);
+ asm_snap_fixup_mcofs(as);
#if LJ_TARGET_MCODE_FIXUP
asm_mcode_fixup(T->mcode, T->szmcode);
#endif
- lj_mcode_sync(T->mcode, origtop);
+ lj_mcode_sync(T->mcode, as->mctoporig);
}
#undef IR
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_ASM_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** ARM IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Register allocator extensions --------------------------------------- */
@@ -41,7 +41,7 @@
}
}
}
- lua_assert(rset_test(RSET_GPREVEN, r));
+ lj_assertA(rset_test(RSET_GPREVEN, r), "odd reg %d", r);
ra_modified(as, r);
ra_modified(as, r+1);
RA_DBGX((as, "scratchpair $r $r", r, r+1));
@@ -185,6 +185,9 @@
*ofsp = (ofs & 255); /* Mask out less bits to allow LDRD. */
return ra_allock(as, (ofs & ~255), allow);
}
+ } else if (ir->o == IR_TMPREF) {
+ *ofsp = 0;
+ return RID_SP;
}
}
*ofsp = 0;
@@ -269,7 +272,7 @@
return;
}
} else if (ir->o == IR_STRREF && !(!LJ_SOFTFP && (ai & 0x08000000))) {
- lua_assert(ofs == 0);
+ lj_assertA(ofs == 0, "bad usage");
ofs = (int32_t)sizeof(GCstr);
if (irref_isk(ir->op2)) {
ofs += IR(ir->op2)->i;
@@ -310,7 +313,11 @@
}
#if !LJ_SOFTFP
-/* Fuse to multiply-add/sub instruction. */
+/*
+** Fuse to multiply-add/sub instruction.
+** VMLA rounds twice (UMA, not FMA) -- no need to check for JIT_F_OPT_FMA.
+** VFMA needs VFPv4, which is uncommon on the remaining ARM32 targets.
+*/
static int asm_fusemadd(ASMState *as, IRIns *ir, ARMIns ai, ARMIns air)
{
IRRef lref = ir->op1, rref = ir->op2;
@@ -389,9 +396,11 @@
as->freeset |= (of & RSET_RANGE(REGARG_FIRSTGPR, REGARG_LASTGPR+1));
if (irt_isnum(ir->t)) gpr = (gpr+1) & ~1u;
if (gpr <= REGARG_LASTGPR) {
- lua_assert(rset_test(as->freeset, gpr)); /* Must have been evicted. */
+ lj_assertA(rset_test(as->freeset, gpr),
+ "reg %d not free", gpr); /* Must have been evicted. */
if (irt_isnum(ir->t)) {
- lua_assert(rset_test(as->freeset, gpr+1)); /* Ditto. */
+ lj_assertA(rset_test(as->freeset, gpr+1),
+ "reg %d not free", gpr+1); /* Ditto. */
emit_dnm(as, ARMI_VMOV_RR_D, gpr, gpr+1, (src & 15));
gpr += 2;
} else {
@@ -408,7 +417,8 @@
#endif
{
if (gpr <= REGARG_LASTGPR) {
- lua_assert(rset_test(as->freeset, gpr)); /* Must have been evicted. */
+ lj_assertA(rset_test(as->freeset, gpr),
+ "reg %d not free", gpr); /* Must have been evicted. */
if (ref) ra_leftov(as, gpr, ref);
gpr++;
} else {
@@ -433,7 +443,7 @@
rset_clear(drop, (ir+1)->r); /* Dest reg handled below. */
ra_evictset(as, drop); /* Evictions must be performed first. */
if (ra_used(ir)) {
- lua_assert(!irt_ispri(ir->t));
+ lj_assertA(!irt_ispri(ir->t), "PRI dest");
if (!LJ_SOFTFP && irt_isfp(ir->t)) {
if (LJ_ABI_SOFTFP || (ci->flags & (CCI_CASTU64|CCI_VARARG))) {
Reg dest = (ra_dest(as, ir, RSET_FPR) & 15);
@@ -495,6 +505,30 @@
emit_lso(as, ARMI_LDR, RID_TMP, base, -4);
}
+/* -- Buffer operations --------------------------------------------------- */
+
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb)
+{
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irgc;
+ int32_t addr = i32ptr((void *)&J2G(as->J)->cur_L);
+ irgc.ot = IRT(0, IRT_PGC); /* GC type. */
+ emit_storeofs(as, &irgc, RID_TMP, sb, offsetof(SBuf, L));
+ if ((as->flags & JIT_F_ARMV6T2)) {
+ emit_dnm(as, ARMI_BFI, RID_TMP, lj_fls(SBUF_MASK_FLAG), tmp);
+ } else {
+ emit_dnm(as, ARMI_ORR, RID_TMP, RID_TMP, tmp);
+ emit_dn(as, ARMI_AND|ARMI_K12|SBUF_MASK_FLAG, tmp, tmp);
+ }
+ emit_lso(as, ARMI_LDR, RID_TMP,
+ ra_allock(as, (addr & ~4095),
+ rset_exclude(rset_exclude(RSET_GPR, sb), tmp)),
+ (addr & 4095));
+ emit_loadofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+}
+#endif
+
/* -- Type conversions ---------------------------------------------------- */
#if !LJ_SOFTFP
@@ -530,13 +564,17 @@
#endif
IRRef lref = ir->op1;
/* 64 bit integer conversions are handled by SPLIT. */
- lua_assert(!irt_isint64(ir->t) && !(st == IRT_I64 || st == IRT_U64));
+ lj_assertA(!irt_isint64(ir->t) && !(st == IRT_I64 || st == IRT_U64),
+ "IR %04d has unsplit 64 bit type",
+ (int)(ir - as->ir) - REF_BIAS);
#if LJ_SOFTFP
/* FP conversions are handled by SPLIT. */
- lua_assert(!irt_isfp(ir->t) && !(st == IRT_NUM || st == IRT_FLOAT));
+ lj_assertA(!irt_isfp(ir->t) && !(st == IRT_NUM || st == IRT_FLOAT),
+ "IR %04d has FP type",
+ (int)(ir - as->ir) - REF_BIAS);
/* Can't check for same types: SPLIT uses CONV int.int + BXOR for sfp NEG. */
#else
- lua_assert(irt_type(ir->t) != st);
+ lj_assertA(irt_type(ir->t) != st, "inconsistent types for CONV");
if (irt_isfp(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
if (stfp) { /* FP to FP conversion. */
@@ -553,7 +591,8 @@
} else if (stfp) { /* FP to integer conversion. */
if (irt_isguard(ir->t)) {
/* Checked conversions are only supported from number to int. */
- lua_assert(irt_isint(ir->t) && st == IRT_NUM);
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
} else {
Reg left = ra_alloc1(as, lref, RSET_FPR);
@@ -572,7 +611,7 @@
Reg dest = ra_dest(as, ir, RSET_GPR);
if (st >= IRT_I8 && st <= IRT_U16) { /* Extend to 32 bit integer. */
Reg left = ra_alloc1(as, lref, RSET_GPR);
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t), "bad type for CONV EXT");
if ((as->flags & JIT_F_ARMV6)) {
ARMIns ai = st == IRT_I8 ? ARMI_SXTB :
st == IRT_U8 ? ARMI_UXTB :
@@ -658,35 +697,55 @@
/* -- Memory references --------------------------------------------------- */
/* Get pointer to TValue. */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode)
{
- IRIns *ir = IR(ref);
- if (irt_isnum(ir->t)) {
- if (irref_isk(ref)) {
- /* Use the number constant itself as a TValue. */
- ra_allockreg(as, i32ptr(ir_knum(ir)), dest);
- } else {
+ if ((mode & IRTMPREF_IN1)) {
+ IRIns *ir = IR(ref);
+ if (irt_isnum(ir->t)) {
+ if ((mode & IRTMPREF_OUT1)) {
#if LJ_SOFTFP
- lua_assert(0);
+ lj_assertA(irref_isk(ref), "unsplit FP op");
+ emit_dm(as, ARMI_MOV, dest, RID_SP);
+ emit_lso(as, ARMI_STR,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.lo, RSET_GPR),
+ RID_SP, 0);
+ emit_lso(as, ARMI_STR,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.hi, RSET_GPR),
+ RID_SP, 4);
#else
- /* Otherwise force a spill and use the spill slot. */
- emit_opk(as, ARMI_ADD, dest, RID_SP, ra_spill(as, ir), RSET_GPR);
+ Reg src = ra_alloc1(as, ref, RSET_FPR);
+ emit_dm(as, ARMI_MOV, dest, RID_SP);
+ emit_vlso(as, ARMI_VSTR_D, src, RID_SP, 0);
#endif
+ } else if (irref_isk(ref)) {
+ /* Use the number constant itself as a TValue. */
+ ra_allockreg(as, i32ptr(ir_knum(ir)), dest);
+ } else {
+#if LJ_SOFTFP
+ lj_assertA(0, "unsplit FP op");
+#else
+ /* Otherwise force a spill and use the spill slot. */
+ emit_opk(as, ARMI_ADD, dest, RID_SP, ra_spill(as, ir), RSET_GPR);
+#endif
+ }
+ } else {
+ /* Otherwise use [sp] and [sp+4] to hold the TValue.
+ ** This assumes the following call has max. 4 args.
+ */
+ Reg type;
+ emit_dm(as, ARMI_MOV, dest, RID_SP);
+ if (!irt_ispri(ir->t)) {
+ Reg src = ra_alloc1(as, ref, RSET_GPR);
+ emit_lso(as, ARMI_STR, src, RID_SP, 0);
+ }
+ if (LJ_SOFTFP && (ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t))
+ type = ra_alloc1(as, ref+1, RSET_GPR);
+ else
+ type = ra_allock(as, irt_toitype(ir->t), RSET_GPR);
+ emit_lso(as, ARMI_STR, type, RID_SP, 4);
}
} else {
- /* Otherwise use [sp] and [sp+4] to hold the TValue. */
- RegSet allow = rset_exclude(RSET_GPR, dest);
- Reg type;
emit_dm(as, ARMI_MOV, dest, RID_SP);
- if (!irt_ispri(ir->t)) {
- Reg src = ra_alloc1(as, ref, allow);
- emit_lso(as, ARMI_STR, src, RID_SP, 0);
- }
- if (LJ_SOFTFP && (ir+1)->o == IR_HIOP)
- type = ra_alloc1(as, ref+1, allow);
- else
- type = ra_allock(as, irt_toitype(ir->t), allow);
- emit_lso(as, ARMI_STR, type, RID_SP, 4);
}
}
@@ -811,16 +870,16 @@
*l_loop = ARMF_CC(ARMI_B, CC_NE) | ((as->mcp-l_loop-2) & 0x00ffffffu);
/* Load main position relative to tab->node into dest. */
- khash = irref_isk(refkey) ? ir_khash(irkey) : 1;
+ khash = irref_isk(refkey) ? ir_khash(as, irkey) : 1;
if (khash == 0) {
emit_lso(as, ARMI_LDR, dest, tab, (int32_t)offsetof(GCtab, node));
} else {
emit_dnm(as, ARMI_ADD|ARMF_SH(ARMSH_LSL, 3), dest, dest, tmp);
emit_dnm(as, ARMI_ADD|ARMF_SH(ARMSH_LSL, 1), tmp, tmp, tmp);
- if (irt_isstr(kt)) { /* Fetch of str->hash is cheaper than ra_allock. */
+ if (irt_isstr(kt)) { /* Fetch of str->sid is cheaper than ra_allock. */
emit_dnm(as, ARMI_AND, tmp, tmp+1, RID_TMP);
emit_lso(as, ARMI_LDR, dest, tab, (int32_t)offsetof(GCtab, node));
- emit_lso(as, ARMI_LDR, tmp+1, key, (int32_t)offsetof(GCstr, hash));
+ emit_lso(as, ARMI_LDR, tmp+1, key, (int32_t)offsetof(GCstr, sid));
emit_lso(as, ARMI_LDR, RID_TMP, tab, (int32_t)offsetof(GCtab, hmask));
} else if (irref_isk(refkey)) {
emit_opk(as, ARMI_AND, tmp, RID_TMP, (int32_t)khash,
@@ -867,7 +926,7 @@
Reg node = ra_alloc1(as, ir->op1, RSET_GPR);
Reg key = RID_NONE, type = RID_TMP, idx = node;
RegSet allow = rset_exclude(RSET_GPR, node);
- lua_assert(ofs % sizeof(Node) == 0);
+ lj_assertA(ofs % sizeof(Node) == 0, "unaligned HREFK slot");
if (ofs > 4095) {
idx = dest;
rset_clear(allow, dest);
@@ -910,24 +969,32 @@
static void asm_uref(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (irref_isk(ir->op1)) {
+ int guarded = (irt_t(ir->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC);
+ if (irref_isk(ir->op1) && !guarded) {
GCfunc *fn = ir_kfunc(IR(ir->op1));
MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
emit_lsptr(as, ARMI_LDR, dest, v);
} else {
- Reg uv = ra_scratch(as, RSET_GPR);
- Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
- if (ir->o == IR_UREFC) {
- asm_guardcc(as, CC_NE);
+ if (guarded) {
+ asm_guardcc(as, ir->o == IR_UREFC ? CC_NE : CC_EQ);
emit_n(as, ARMI_CMP|ARMI_K12|1, RID_TMP);
- emit_opk(as, ARMI_ADD, dest, uv,
+ }
+ if (ir->o == IR_UREFC)
+ emit_opk(as, ARMI_ADD, dest, dest,
(int32_t)offsetof(GCupval, tv), RSET_GPR);
- emit_lso(as, ARMI_LDRB, RID_TMP, uv, (int32_t)offsetof(GCupval, closed));
+ else
+ emit_lso(as, ARMI_LDR, dest, dest, (int32_t)offsetof(GCupval, v));
+ if (guarded)
+ emit_lso(as, ARMI_LDRB, RID_TMP, dest,
+ (int32_t)offsetof(GCupval, closed));
+ if (irref_isk(ir->op1)) {
+ GCfunc *fn = ir_kfunc(IR(ir->op1));
+ int32_t k = (int32_t)gcrefu(fn->l.uvptr[(ir->op2 >> 8)]);
+ emit_loadi(as, dest, k);
} else {
- emit_lso(as, ARMI_LDR, dest, uv, (int32_t)offsetof(GCupval, v));
+ emit_lso(as, ARMI_LDR, dest, ra_alloc1(as, ir->op1, RSET_GPR),
+ (int32_t)offsetof(GCfuncL, uvptr) + 4*(int32_t)(ir->op2 >> 8));
}
- emit_lso(as, ARMI_LDR, uv, func,
- (int32_t)offsetof(GCfuncL, uvptr) + 4*(int32_t)(ir->op2 >> 8));
}
}
@@ -934,7 +1001,7 @@
static void asm_fref(ASMState *as, IRIns *ir)
{
UNUSED(as); UNUSED(ir);
- lua_assert(!ra_used(ir));
+ lj_assertA(!ra_used(ir), "unfused FREF");
}
static void asm_strref(ASMState *as, IRIns *ir)
@@ -971,26 +1038,28 @@
/* -- Loads and stores ---------------------------------------------------- */
-static ARMIns asm_fxloadins(IRIns *ir)
+static ARMIns asm_fxloadins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: return ARMI_LDRSB;
case IRT_U8: return ARMI_LDRB;
case IRT_I16: return ARMI_LDRSH;
case IRT_U16: return ARMI_LDRH;
- case IRT_NUM: lua_assert(!LJ_SOFTFP); return ARMI_VLDR_D;
- case IRT_FLOAT: if (!LJ_SOFTFP) return ARMI_VLDR_S;
+ case IRT_NUM: lj_assertA(!LJ_SOFTFP, "unsplit FP op"); return ARMI_VLDR_D;
+ case IRT_FLOAT: if (!LJ_SOFTFP) return ARMI_VLDR_S; /* fallthrough */
default: return ARMI_LDR;
}
}
-static ARMIns asm_fxstoreins(IRIns *ir)
+static ARMIns asm_fxstoreins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: case IRT_U8: return ARMI_STRB;
case IRT_I16: case IRT_U16: return ARMI_STRH;
- case IRT_NUM: lua_assert(!LJ_SOFTFP); return ARMI_VSTR_D;
- case IRT_FLOAT: if (!LJ_SOFTFP) return ARMI_VSTR_S;
+ case IRT_NUM: lj_assertA(!LJ_SOFTFP, "unsplit FP op"); return ARMI_VSTR_D;
+ case IRT_FLOAT: if (!LJ_SOFTFP) return ARMI_VSTR_S; /* fallthrough */
default: return ARMI_STR;
}
}
@@ -997,13 +1066,15 @@
static void asm_fload(ASMState *as, IRIns *ir)
{
- if (ir->op1 == REF_NIL) {
- lua_assert(!ra_used(ir)); /* We can end up here if DCE is turned off. */
+ Reg dest = ra_dest(as, ir, RSET_GPR);
+ ARMIns ai = asm_fxloadins(as, ir);
+ Reg idx;
+ int32_t ofs;
+ if (ir->op1 == REF_NIL) { /* FLOAD from GG_State with offset. */
+ idx = ra_allock(as, (int32_t)(ir->op2<<2) + (int32_t)J2GG(as->J), RSET_GPR);
+ ofs = 0;
} else {
- Reg dest = ra_dest(as, ir, RSET_GPR);
- Reg idx = ra_alloc1(as, ir->op1, RSET_GPR);
- ARMIns ai = asm_fxloadins(ir);
- int32_t ofs;
+ idx = ra_alloc1(as, ir->op1, RSET_GPR);
if (ir->op2 == IRFL_TAB_ARRAY) {
ofs = asm_fuseabase(as, ir->op1);
if (ofs) { /* Turn the t->array load into an add for colocated arrays. */
@@ -1012,11 +1083,11 @@
}
}
ofs = field_ofs[ir->op2];
- if ((ai & 0x04000000))
- emit_lso(as, ai, dest, idx, ofs);
- else
- emit_lsox(as, ai, dest, idx, ofs);
}
+ if ((ai & 0x04000000))
+ emit_lso(as, ai, dest, idx, ofs);
+ else
+ emit_lsox(as, ai, dest, idx, ofs);
}
static void asm_fstore(ASMState *as, IRIns *ir)
@@ -1026,7 +1097,7 @@
IRIns *irf = IR(ir->op1);
Reg idx = ra_alloc1(as, irf->op1, rset_exclude(RSET_GPR, src));
int32_t ofs = field_ofs[irf->op2];
- ARMIns ai = asm_fxstoreins(ir);
+ ARMIns ai = asm_fxstoreins(as, ir);
if ((ai & 0x04000000))
emit_lso(as, ai, src, idx, ofs);
else
@@ -1038,8 +1109,8 @@
{
Reg dest = ra_dest(as, ir,
(!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
- lua_assert(!(ir->op2 & IRXLOAD_UNALIGNED));
- asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR, 0);
+ lj_assertA(!(ir->op2 & IRXLOAD_UNALIGNED), "unaligned XLOAD");
+ asm_fusexref(as, asm_fxloadins(as, ir), dest, ir->op1, RSET_GPR, 0);
}
static void asm_xstore_(ASMState *as, IRIns *ir, int32_t ofs)
@@ -1047,7 +1118,7 @@
if (ir->r != RID_SINK) {
Reg src = ra_alloc1(as, ir->op2,
(!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
- asm_fusexref(as, asm_fxstoreins(ir), src, ir->op1,
+ asm_fusexref(as, asm_fxstoreins(as, ir), src, ir->op1,
rset_exclude(RSET_GPR, src), ofs);
}
}
@@ -1066,13 +1137,15 @@
rset_clear(allow, type);
}
if (ra_used(ir)) {
- lua_assert((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
- irt_isint(ir->t) || irt_isaddr(ir->t));
+ lj_assertA((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
+ irt_isint(ir->t) || irt_isaddr(ir->t),
+ "bad load type %d", irt_type(ir->t));
dest = ra_dest(as, ir, (!LJ_SOFTFP && t == IRT_NUM) ? RSET_FPR : allow);
rset_clear(allow, dest);
}
idx = asm_fuseahuref(as, ir->op1, &ofs, allow,
(!LJ_SOFTFP && t == IRT_NUM) ? 1024 : 4096);
+ if (ir->o == IR_VLOAD) ofs += 8 * ir->op2;
if (!hiop || type == RID_NONE) {
rset_clear(allow, idx);
if (ofs < 256 && ra_hasreg(dest) && (dest & 1) == 0 &&
@@ -1133,10 +1206,13 @@
IRType t = hiop ? IRT_NUM : irt_type(ir->t);
Reg dest = RID_NONE, type = RID_NONE, base;
RegSet allow = RSET_GPR;
- lua_assert(!(ir->op2 & IRSLOAD_PARENT)); /* Handled by asm_head_side(). */
- lua_assert(irt_isguard(ir->t) || !(ir->op2 & IRSLOAD_TYPECHECK));
+ lj_assertA(!(ir->op2 & IRSLOAD_PARENT),
+ "bad parent SLOAD"); /* Handled by asm_head_side(). */
+ lj_assertA(irt_isguard(ir->t) || !(ir->op2 & IRSLOAD_TYPECHECK),
+ "inconsistent SLOAD variant");
#if LJ_SOFTFP
- lua_assert(!(ir->op2 & IRSLOAD_CONVERT)); /* Handled by LJ_SOFTFP SPLIT. */
+ lj_assertA(!(ir->op2 & IRSLOAD_CONVERT),
+ "unsplit SLOAD convert"); /* Handled by LJ_SOFTFP SPLIT. */
if (hiop && ra_used(ir+1)) {
type = ra_dest(as, ir+1, allow);
rset_clear(allow, type);
@@ -1152,8 +1228,9 @@
Reg tmp = RID_NONE;
if ((ir->op2 & IRSLOAD_CONVERT))
tmp = ra_scratch(as, t == IRT_INT ? RSET_FPR : RSET_GPR);
- lua_assert((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
- irt_isint(ir->t) || irt_isaddr(ir->t));
+ lj_assertA((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
+ irt_isint(ir->t) || irt_isaddr(ir->t),
+ "bad SLOAD type %d", irt_type(ir->t));
dest = ra_dest(as, ir, (!LJ_SOFTFP && t == IRT_NUM) ? RSET_FPR : allow);
rset_clear(allow, dest);
base = ra_alloc1(as, REF_BASE, allow);
@@ -1185,7 +1262,12 @@
}
}
asm_guardcc(as, t == IRT_NUM ? CC_HS : CC_NE);
- emit_n(as, ARMI_CMN|ARMI_K12|-irt_toitype_(t), type);
+ if ((ir->op2 & IRSLOAD_KEYINDEX)) {
+ emit_n(as, ARMI_CMN|ARMI_K12|1, type);
+ emit_dn(as, ARMI_EOR^emit_isk12(ARMI_EOR, ~LJ_KEYINDEX), type, type);
+ } else {
+ emit_n(as, ARMI_CMN|ARMI_K12|-irt_toitype_(t), type);
+ }
}
if (ra_hasreg(dest)) {
#if !LJ_SOFTFP
@@ -1218,7 +1300,8 @@
IRRef args[4];
RegSet allow = (RSET_GPR & ~RSET_SCRATCH);
RegSet drop = RSET_SCRATCH;
- lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
+ lj_assertA(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL),
+ "bad CNEW/CNEWI operands");
as->gcsteps++;
if (ra_hasreg(ir->r))
@@ -1230,10 +1313,10 @@
/* Initialize immutable cdata object. */
if (ir->o == IR_CNEWI) {
int32_t ofs = sizeof(GCcdata);
- lua_assert(sz == 4 || sz == 8);
+ lj_assertA(sz == 4 || sz == 8, "bad CNEWI size %d", sz);
if (sz == 8) {
ofs += 4; ir++;
- lua_assert(ir->o == IR_HIOP);
+ lj_assertA(ir->o == IR_HIOP, "expected HIOP for CNEWI");
}
for (;;) {
Reg r = ra_alloc1(as, ir->op2, allow);
@@ -1268,8 +1351,6 @@
ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
ra_releasetmp(as, ASMREF_TMP1));
}
-#else
-#define asm_cnew(as, ir) ((void)0)
#endif
/* -- Write barriers ------------------------------------------------------ */
@@ -1301,7 +1382,7 @@
MCLabel l_end;
Reg obj, val, tmp;
/* No need for other object barriers (yet). */
- lua_assert(IR(ir->op1)->o == IR_UREFC);
+ lj_assertA(IR(ir->op1)->o == IR_UREFC, "bad OBAR type");
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
args[0] = ASMREF_TMP1; /* global_State *g */
@@ -1364,8 +1445,6 @@
static void asm_fpmath(ASMState *as, IRIns *ir)
{
- if (ir->op2 == IRFPM_EXP2 && asm_fpjoin_pow(as, ir))
- return;
if (ir->op2 <= IRFPM_TRUNC)
asm_callround(as, ir, ir->op2);
else if (ir->op2 == IRFPM_SQRT)
@@ -1412,16 +1491,31 @@
emit_dn(as, ai^m, dest, left);
}
-static void asm_intop_s(ASMState *as, IRIns *ir, ARMIns ai)
+/* Try to drop cmp r, #0. */
+static ARMIns asm_drop_cmp0(ASMState *as, ARMIns ai)
{
- if (as->flagmcp == as->mcp) { /* Drop cmp r, #0. */
+ if (as->flagmcp == as->mcp) {
+ uint32_t cc = (as->mcp[1] >> 28);
as->flagmcp = NULL;
- as->mcp++;
- ai |= ARMI_S;
+ if (cc <= CC_NE) {
+ as->mcp++;
+ ai |= ARMI_S;
+ } else if (cc == CC_GE) {
+ *++as->mcp ^= ((CC_GE^CC_PL) << 28);
+ ai |= ARMI_S;
+ } else if (cc == CC_LT) {
+ *++as->mcp ^= ((CC_LT^CC_MI) << 28);
+ ai |= ARMI_S;
+ } /* else: other conds don't work in general. */
}
- asm_intop(as, ir, ai);
+ return ai;
}
+static void asm_intop_s(ASMState *as, IRIns *ir, ARMIns ai)
+{
+ asm_intop(as, ir, asm_drop_cmp0(as, ai));
+}
+
static void asm_intneg(ASMState *as, IRIns *ir, ARMIns ai)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -1492,15 +1586,10 @@
#define asm_mulov(as, ir) asm_mul(as, ir)
#if !LJ_SOFTFP
-#define asm_div(as, ir) asm_fparith(as, ir, ARMI_VDIV_D)
-#define asm_pow(as, ir) asm_callid(as, ir, IRCALL_lj_vm_powi)
+#define asm_fpdiv(as, ir) asm_fparith(as, ir, ARMI_VDIV_D)
#define asm_abs(as, ir) asm_fpunary(as, ir, ARMI_VABS_D)
-#define asm_atan2(as, ir) asm_callid(as, ir, IRCALL_atan2)
-#define asm_ldexp(as, ir) asm_callid(as, ir, IRCALL_ldexp)
#endif
-#define asm_mod(as, ir) asm_callid(as, ir, IRCALL_lj_vm_modi)
-
static void asm_neg(ASMState *as, IRIns *ir)
{
#if !LJ_SOFTFP
@@ -1514,20 +1603,7 @@
static void asm_bitop(ASMState *as, IRIns *ir, ARMIns ai)
{
- if (as->flagmcp == as->mcp) { /* Try to drop cmp r, #0. */
- uint32_t cc = (as->mcp[1] >> 28);
- as->flagmcp = NULL;
- if (cc <= CC_NE) {
- as->mcp++;
- ai |= ARMI_S;
- } else if (cc == CC_GE) {
- *++as->mcp ^= ((CC_GE^CC_PL) << 28);
- ai |= ARMI_S;
- } else if (cc == CC_LT) {
- *++as->mcp ^= ((CC_LT^CC_MI) << 28);
- ai |= ARMI_S;
- } /* else: other conds don't work with bit ops. */
- }
+ ai = asm_drop_cmp0(as, ai);
if (ir->op2 == 0) {
Reg dest = ra_dest(as, ir, RSET_GPR);
uint32_t m = asm_fuseopm(as, ai, ir->op1, RSET_GPR);
@@ -1582,7 +1658,7 @@
#define asm_bshr(as, ir) asm_bitshift(as, ir, ARMSH_LSR)
#define asm_bsar(as, ir) asm_bitshift(as, ir, ARMSH_ASR)
#define asm_bror(as, ir) asm_bitshift(as, ir, ARMSH_ROR)
-#define asm_brol(as, ir) lua_assert(0)
+#define asm_brol(as, ir) lj_assertA(0, "unexpected BROL")
static void asm_intmin_max(ASMState *as, IRIns *ir, int cc)
{
@@ -1657,8 +1733,8 @@
asm_intmin_max(as, ir, cc);
}
-#define asm_min(as, ir) asm_min_max(as, ir, CC_GT, CC_HI)
-#define asm_max(as, ir) asm_min_max(as, ir, CC_LT, CC_LO)
+#define asm_min(as, ir) asm_min_max(as, ir, CC_GT, CC_PL)
+#define asm_max(as, ir) asm_min_max(as, ir, CC_LT, CC_LE)
/* -- Comparisons --------------------------------------------------------- */
@@ -1733,7 +1809,8 @@
Reg left;
uint32_t m;
int cmpprev0 = 0;
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t),
+ "bad comparison data type %d", irt_type(ir->t));
if (asm_swapops(as, lref, rref)) {
Reg tmp = lref; lref = rref; rref = tmp;
if (cc >= CC_GE) cc ^= 7; /* LT <-> GT, LE <-> GE */
@@ -1825,15 +1902,15 @@
}
#endif
-/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
+/* -- Split register ops -------------------------------------------------- */
-/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
+/* Hiword op of a split 32/32 bit op. Previous op is the loword op. */
static void asm_hiop(ASMState *as, IRIns *ir)
{
-#if LJ_HASFFI || LJ_SOFTFP
/* HIOP is marked as a store because it needs its own DCE logic. */
int uselo = ra_used(ir-1), usehi = ra_used(ir); /* Loword/hiword used? */
if (LJ_UNLIKELY(!(as->flags & JIT_F_OPT_DCE))) uselo = usehi = 1;
+#if LJ_HASFFI || LJ_SOFTFP
if ((ir-1)->o <= IR_NE) { /* 64 bit integer or FP comparisons. ORDER IR. */
as->curins--; /* Always skip the loword comparison. */
#if LJ_SOFTFP
@@ -1850,7 +1927,7 @@
} else if ((ir-1)->o == IR_MIN || (ir-1)->o == IR_MAX) {
as->curins--; /* Always skip the loword min/max. */
if (uselo || usehi)
- asm_sfpmin_max(as, ir-1, (ir-1)->o == IR_MIN ? CC_HI : CC_LO);
+ asm_sfpmin_max(as, ir-1, (ir-1)->o == IR_MIN ? CC_PL : CC_LE);
return;
#elif LJ_HASFFI
} else if ((ir-1)->o == IR_CONV) {
@@ -1864,6 +1941,7 @@
asm_xstore_(as, ir, 4);
return;
}
+#endif
if (!usehi) return; /* Skip unused hiword op for all remaining ops. */
switch ((ir-1)->o) {
#if LJ_HASFFI
@@ -1882,6 +1960,9 @@
asm_intneg(as, ir, ARMI_RSC);
asm_intneg(as, ir-1, ARMI_RSB|ARMI_S);
break;
+ case IR_CNEWI:
+ /* Nothing to do here. Handled by lo op itself. */
+ break;
#endif
#if LJ_SOFTFP
case IR_SLOAD: case IR_ALOAD: case IR_HLOAD: case IR_ULOAD: case IR_VLOAD:
@@ -1889,24 +1970,16 @@
if (!uselo)
ra_allocref(as, ir->op1, RSET_GPR); /* Mark lo op as used. */
break;
+ case IR_ASTORE: case IR_HSTORE: case IR_USTORE: case IR_TOSTR: case IR_TMPREF:
+ /* Nothing to do here. Handled by lo op itself. */
+ break;
#endif
- case IR_CALLN:
- case IR_CALLS:
- case IR_CALLXS:
+ case IR_CALLN: case IR_CALLL: case IR_CALLS: case IR_CALLXS:
if (!uselo)
ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
break;
-#if LJ_SOFTFP
- case IR_ASTORE: case IR_HSTORE: case IR_USTORE: case IR_TOSTR:
-#endif
- case IR_CNEWI:
- /* Nothing to do here. Handled by lo op itself. */
- break;
- default: lua_assert(0); break;
+ default: lj_assertA(0, "bad HIOP for op %d", (ir-1)->o); break;
}
-#else
- UNUSED(as); UNUSED(ir); lua_assert(0);
-#endif
}
/* -- Profiling ----------------------------------------------------------- */
@@ -1925,24 +1998,27 @@
static void asm_stack_check(ASMState *as, BCReg topslot,
IRIns *irp, RegSet allow, ExitNo exitno)
{
+ int savereg = 0;
Reg pbase;
uint32_t k;
if (irp) {
if (!ra_hasspill(irp->s)) {
pbase = irp->r;
- lua_assert(ra_hasreg(pbase));
+ lj_assertA(ra_hasreg(pbase), "base reg lost");
} else if (allow) {
pbase = rset_pickbot(allow);
} else {
pbase = RID_RET;
- emit_lso(as, ARMI_LDR, RID_RET, RID_SP, 0); /* Restore temp. register. */
+ savereg = 1;
}
} else {
pbase = RID_BASE;
}
emit_branch(as, ARMF_CC(ARMI_BL, CC_LS), exitstub_addr(as->J, exitno));
+ if (savereg)
+ emit_lso(as, ARMI_LDR, RID_RET, RID_SP, 0); /* Restore temp. register. */
k = emit_isk12(0, (int32_t)(8*topslot));
- lua_assert(k);
+ lj_assertA(k, "slot offset %d does not fit in K12", 8*topslot);
emit_n(as, ARMI_CMP^k, RID_TMP);
emit_dnm(as, ARMI_SUB, RID_TMP, RID_TMP, pbase);
emit_lso(as, ARMI_LDR, RID_TMP, RID_TMP,
@@ -1952,7 +2028,7 @@
if (ra_hasspill(irp->s))
emit_lso(as, ARMI_LDR, pbase, RID_SP, sps_scale(irp->s));
emit_lso(as, ARMI_LDR, RID_TMP, RID_TMP, (i & 4095));
- if (ra_hasspill(irp->s) && !allow)
+ if (savereg)
emit_lso(as, ARMI_STR, RID_RET, RID_SP, 0); /* Save temp. register. */
emit_loadi(as, RID_TMP, (i & ~4095));
} else {
@@ -1966,11 +2042,12 @@
SnapEntry *map = &as->T->snapmap[snap->mapofs];
SnapEntry *flinks = &as->T->snapmap[snap_nextofs(as->T, snap)-1];
MSize n, nent = snap->nent;
+ int32_t bias = 0;
/* Store the value of all modified slots to the Lua stack. */
for (n = 0; n < nent; n++) {
SnapEntry sn = map[n];
BCReg s = snap_slot(sn);
- int32_t ofs = 8*((int32_t)s-1);
+ int32_t ofs = 8*((int32_t)s-1) - bias;
IRRef ref = snap_ref(sn);
IRIns *ir = IR(ref);
if ((sn & SNAP_NORESTORE))
@@ -1979,7 +2056,8 @@
#if LJ_SOFTFP
RegSet odd = rset_exclude(RSET_GPRODD, RID_BASE);
Reg tmp;
- lua_assert(irref_isk(ref)); /* LJ_SOFTFP: must be a number constant. */
+ /* LJ_SOFTFP: must be a number constant. */
+ lj_assertA(irref_isk(ref), "unsplit FP op");
tmp = ra_allock(as, (int32_t)ir_knum(ir)->u32.lo,
rset_exclude(RSET_GPREVEN, RID_BASE));
emit_lso(as, ARMI_STR, tmp, RID_BASE, ofs);
@@ -1988,12 +2066,19 @@
emit_lso(as, ARMI_STR, tmp, RID_BASE, ofs+4);
#else
Reg src = ra_alloc1(as, ref, RSET_FPR);
+ if (LJ_UNLIKELY(ofs < -1020 || ofs > 1020)) {
+ int32_t adj = ofs & 0xffffff00; /* K12-friendly. */
+ bias += adj;
+ ofs -= adj;
+ emit_addptr(as, RID_BASE, -adj);
+ }
emit_vlso(as, ARMI_VSTR_D, src, RID_BASE, ofs);
#endif
} else {
RegSet odd = rset_exclude(RSET_GPRODD, RID_BASE);
Reg type;
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t),
+ "restore of IR type %d", irt_type(ir->t));
if (!irt_ispri(ir->t)) {
Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPREVEN, RID_BASE));
emit_lso(as, ARMI_STR, src, RID_BASE, ofs);
@@ -2006,6 +2091,8 @@
} else if ((sn & SNAP_SOFTFPNUM)) {
type = ra_alloc1(as, ref+1, rset_exclude(RSET_GPRODD, RID_BASE));
#endif
+ } else if ((sn & SNAP_KEYINDEX)) {
+ type = ra_allock(as, (int32_t)LJ_KEYINDEX, odd);
} else {
type = ra_allock(as, (int32_t)irt_toitype(ir->t), odd);
}
@@ -2013,11 +2100,15 @@
}
checkmclim(as);
}
- lua_assert(map + nent == flinks);
+ emit_addptr(as, RID_BASE, bias);
+ lj_assertA(map + nent == flinks, "inconsistent frames in snapshot");
}
/* -- GC handling --------------------------------------------------------- */
+/* Marker to prevent patching the GC check exit. */
+#define ARM_NOPATCH_GC_CHECK (ARMI_BIC|ARMI_K12)
+
/* Check GC threshold and do one or more GC steps. */
static void asm_gc_check(ASMState *as)
{
@@ -2029,6 +2120,7 @@
l_end = emit_label(as);
/* Exit trace if in GCSatomic or GCSfinalize. Avoids syncing GC objects. */
asm_guardcc(as, CC_NE); /* Assumes asm_snap_prep() already done. */
+ *--as->mcp = ARM_NOPATCH_GC_CHECK;
emit_n(as, ARMI_CMP|ARMI_K12|0, RID_RET);
args[0] = ASMREF_TMP1; /* global_State *g */
args[1] = ASMREF_TMP2; /* MSize steps */
@@ -2063,6 +2155,12 @@
}
}
+/* Fixup the tail of the loop. */
+static void asm_loop_tail_fixup(ASMState *as)
+{
+ UNUSED(as); /* Nothing to do. */
+}
+
/* -- Head of trace ------------------------------------------------------- */
/* Reload L register from g->cur_L. */
@@ -2088,7 +2186,7 @@
}
/* Coalesce BASE register for a side trace. */
-static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
+static Reg asm_head_side_base(ASMState *as, IRIns *irp)
{
IRIns *ir;
asm_head_lreg(as);
@@ -2096,16 +2194,15 @@
if (ra_hasreg(ir->r) && (rset_test(as->modset, ir->r) || irt_ismarked(ir->t)))
ra_spill(as, ir);
if (ra_hasspill(irp->s)) {
- rset_clear(allow, ra_dest(as, ir, allow));
+ return ra_dest(as, ir, RSET_GPR);
} else {
Reg r = irp->r;
- lua_assert(ra_hasreg(r));
- rset_clear(allow, r);
+ lj_assertA(ra_hasreg(r), "base reg lost");
if (r != ir->r && !rset_test(as->freeset, r))
ra_restore(as, regcost_ref(as->cost[r]));
ra_destreg(as, ir, r);
+ return r;
}
- return allow;
}
/* -- Tail of trace ------------------------------------------------------- */
@@ -2121,7 +2218,7 @@
} else {
/* Patch stack adjustment. */
uint32_t k = emit_isk12(ARMI_ADD, spadj);
- lua_assert(k);
+ lj_assertA(k, "stack adjustment %d does not fit in K12", spadj);
p[-2] = (ARMI_ADD^k) | ARMF_D(RID_SP) | ARMF_N(RID_SP);
}
/* Patch exit branch. */
@@ -2174,7 +2271,7 @@
}
if (nslots > as->evenspill) /* Leave room for args in stack slots. */
as->evenspill = nslots;
- return REGSP_HINT(RID_RET);
+ return REGSP_HINT(irt_isfp(ir->t) ? RID_FPRET : RID_RET);
}
static void asm_setup_target(ASMState *as)
@@ -2197,13 +2294,14 @@
/* Look for bl_cc exitstub, replace with b_cc target. */
uint32_t ins = *p;
if ((ins & 0x0f000000u) == 0x0b000000u && ins < 0xf0000000u &&
- ((ins ^ (px-p)) & 0x00ffffffu) == 0) {
+ ((ins ^ (px-p)) & 0x00ffffffu) == 0 &&
+ p[-1] != ARM_NOPATCH_GC_CHECK) {
*p = (ins & 0xfe000000u) | (((target-p)-2) & 0x00ffffffu);
cend = p+1;
if (!cstart) cstart = p;
}
}
- lua_assert(cstart != NULL);
+ lj_assertJ(cstart != NULL, "exit stub %d not found", exitno);
lj_mcode_sync(cstart, cend);
lj_mcode_patch(J, mcarea, 1);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm64.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm64.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_arm64.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** ARM64 IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Contributed by Djordje Kovacevic and Stefan Pejic from RT-RK.com.
** Sponsored by Cisco Systems, Inc.
@@ -56,11 +56,11 @@
asm_mclimit(as);
/* 1: str lr,[sp]; bl ->vm_exit_handler; movz w0,traceno; bl <1; bl <1; ... */
for (i = nexits-1; (int32_t)i >= 0; i--)
- *--mxp = A64I_LE(A64I_BL|((-3-i)&0x03ffffffu));
- *--mxp = A64I_LE(A64I_MOVZw|A64F_U16(as->T->traceno));
+ *--mxp = A64I_LE(A64I_BL | A64F_S26(-3-i));
+ *--mxp = A64I_LE(A64I_MOVZw | A64F_U16(as->T->traceno));
mxp--;
- *mxp = A64I_LE(A64I_BL|(((MCode *)(void *)lj_vm_exit_handler-mxp)&0x03ffffffu));
- *--mxp = A64I_LE(A64I_STRx|A64F_D(RID_LR)|A64F_N(RID_SP));
+ *mxp = A64I_LE(A64I_BL | A64F_S26(((MCode *)(void *)lj_vm_exit_handler-mxp)));
+ *--mxp = A64I_LE(A64I_STRx | A64F_D(RID_LR) | A64F_N(RID_SP));
as->mctop = mxp;
}
@@ -77,7 +77,7 @@
MCode *p = as->mcp;
if (LJ_UNLIKELY(p == as->invmcp)) {
as->loopinv = 1;
- *p = A64I_B | ((target-p) & 0x03ffffffu);
+ *p = A64I_B | A64F_S26(target-p);
emit_cond_branch(as, cc^1, p-1);
return;
}
@@ -84,18 +84,23 @@
emit_cond_branch(as, cc, target);
}
-/* Emit test and branch instruction to exit for guard. */
-static void asm_guardtnb(ASMState *as, A64Ins ai, Reg r, uint32_t bit)
+/* Emit test and branch instruction to exit for guard, if in range. */
+static int asm_guardtnb(ASMState *as, A64Ins ai, Reg r, uint32_t bit)
{
MCode *target = asm_exitstub_addr(as, as->snapno);
MCode *p = as->mcp;
+ ptrdiff_t delta = target - p;
if (LJ_UNLIKELY(p == as->invmcp)) {
+ if (as->orignins > 1023) return 0; /* Delta might end up too large. */
as->loopinv = 1;
- *p = A64I_B | ((target-p) & 0x03ffffffu);
- emit_tnb(as, ai^0x01000000u, r, bit, p-1);
- return;
+ *p = A64I_B | A64F_S26(delta);
+ ai ^= 0x01000000u;
+ target = p-1;
+ } else if (LJ_UNLIKELY(delta >= 0x1fff)) {
+ return 0;
}
emit_tnb(as, ai, r, bit, target);
+ return 1;
}
/* Emit compare and branch instruction to exit for guard. */
@@ -105,7 +110,7 @@
MCode *p = as->mcp;
if (LJ_UNLIKELY(p == as->invmcp)) {
as->loopinv = 1;
- *p = A64I_B | ((target-p) & 0x03ffffffu);
+ *p = A64I_B | A64F_S26(target-p);
emit_cnb(as, ai^0x01000000u, r, p-1);
return;
}
@@ -198,6 +203,9 @@
return RID_GL;
}
}
+ } else if (ir->o == IR_TMPREF) {
+ *ofsp = (int32_t)glofs(as, &J2G(as->J)->tmptv);
+ return RID_GL;
}
}
*ofsp = 0;
@@ -208,16 +216,14 @@
static uint32_t asm_fuseopm(ASMState *as, A64Ins ai, IRRef ref, RegSet allow)
{
IRIns *ir = IR(ref);
+ int logical = (ai & 0x1f000000) == 0x0a000000;
if (ra_hasreg(ir->r)) {
ra_noweak(as, ir->r);
return A64F_M(ir->r);
} else if (irref_isk(ref)) {
- uint32_t m;
- int64_t k = get_k64val(ir);
- if ((ai & 0x1f000000) == 0x0a000000)
- m = emit_isk13(k, irt_is64(ir->t));
- else
- m = emit_isk12(k);
+ int64_t k = get_k64val(as, ref);
+ uint32_t m = logical ? emit_isk13(k, irt_is64(ir->t)) :
+ emit_isk12(irt_is64(ir->t) ? k : (int32_t)k);
if (m)
return m;
} else if (mayfuse(as, ref)) {
@@ -229,7 +235,7 @@
(IR(ir->op2)->i & (irt_is64(ir->t) ? 63 : 31));
IRIns *irl = IR(ir->op1);
if (sh == A64SH_LSL &&
- irl->o == IR_CONV &&
+ irl->o == IR_CONV && !logical &&
irl->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT) &&
shift <= 4 &&
canfuse(as, irl)) {
@@ -239,7 +245,11 @@
Reg m = ra_alloc1(as, ir->op1, allow);
return A64F_M(m) | A64F_SH(sh, shift);
}
- } else if (ir->o == IR_CONV &&
+ } else if (ir->o == IR_BROR && logical && irref_isk(ir->op2)) {
+ Reg m = ra_alloc1(as, ir->op1, allow);
+ int shift = (IR(ir->op2)->i & (irt_is64(ir->t) ? 63 : 31));
+ return A64F_M(m) | A64F_SH(A64SH_ROR, shift);
+ } else if (ir->o == IR_CONV && !logical &&
ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)) {
Reg m = ra_alloc1(as, ir->op1, allow);
return A64F_M(m) | A64F_EX(A64EX_SXTW);
@@ -295,8 +305,10 @@
} else if (asm_isk32(as, ir->op1, &ofs)) {
ref = ir->op2;
} else {
- Reg rn = ra_alloc1(as, ir->op1, allow);
- IRIns *irr = IR(ir->op2);
+ Reg refk = irref_isk(ir->op1) ? ir->op1 : ir->op2;
+ Reg refv = irref_isk(ir->op1) ? ir->op2 : ir->op1;
+ Reg rn = ra_alloc1(as, refv, allow);
+ IRIns *irr = IR(refk);
uint32_t m;
if (irr+1 == ir && !ra_used(irr) &&
irr->o == IR_ADD && irref_isk(irr->op2)) {
@@ -307,7 +319,7 @@
goto skipopm;
}
}
- m = asm_fuseopm(as, 0, ir->op2, rset_exclude(allow, rn));
+ m = asm_fuseopm(as, 0, refk, rset_exclude(allow, rn));
ofs = sizeof(GCstr);
skipopm:
emit_lso(as, ai, rd, rd, ofs);
@@ -332,7 +344,8 @@
{
IRRef lref = ir->op1, rref = ir->op2;
IRIns *irm;
- if (lref != rref &&
+ if ((as->flags & JIT_F_OPT_FMA) &&
+ lref != rref &&
((mayfuse(as, lref) && (irm = IR(lref), irm->o == IR_MUL) &&
ra_noreg(irm->r)) ||
(mayfuse(as, rref) && (irm = IR(rref), irm->o == IR_MUL) &&
@@ -352,9 +365,9 @@
static int asm_fuseandshift(ASMState *as, IRIns *ir)
{
IRIns *irl = IR(ir->op1);
- lua_assert(ir->o == IR_BAND);
+ lj_assertA(ir->o == IR_BAND, "bad usage");
if (canfuse(as, irl) && irref_isk(ir->op2)) {
- uint64_t mask = get_k64val(IR(ir->op2));
+ uint64_t mask = get_k64val(as, ir->op2);
if (irref_isk(irl->op2) && (irl->o == IR_BSHR || irl->o == IR_BSHL)) {
int32_t shmask = irt_is64(irl->t) ? 63 : 31;
int32_t shift = (IR(irl->op2)->i & shmask);
@@ -382,7 +395,7 @@
static int asm_fuseorshift(ASMState *as, IRIns *ir)
{
IRIns *irl = IR(ir->op1), *irr = IR(ir->op2);
- lua_assert(ir->o == IR_BOR);
+ lj_assertA(ir->o == IR_BOR, "bad usage");
if (canfuse(as, irl) && canfuse(as, irr) &&
((irl->o == IR_BSHR && irr->o == IR_BSHL) ||
(irl->o == IR_BSHL && irr->o == IR_BSHR))) {
@@ -413,13 +426,18 @@
static void asm_gencall(ASMState *as, const CCallInfo *ci, IRRef *args)
{
uint32_t n, nargs = CCI_XNARGS(ci);
- int32_t ofs = 0;
+ int32_t spofs = 0, spalign = LJ_HASFFI && LJ_TARGET_OSX ? 0 : 7;
Reg gpr, fpr = REGARG_FIRSTFPR;
- if ((void *)ci->func)
- emit_call(as, (void *)ci->func);
+ if (ci->func)
+ emit_call(as, ci->func);
for (gpr = REGARG_FIRSTGPR; gpr <= REGARG_LASTGPR; gpr++)
as->cost[gpr] = REGCOST(~0u, ASMREF_L);
gpr = REGARG_FIRSTGPR;
+#if LJ_HASFFI && LJ_ABI_WIN
+ if ((ci->flags & CCI_VARARG)) {
+ fpr = REGARG_LASTFPR+1;
+ }
+#endif
for (n = 0; n < nargs; n++) { /* Setup args. */
IRRef ref = args[n];
IRIns *ir = IR(ref);
@@ -426,25 +444,55 @@
if (ref) {
if (irt_isfp(ir->t)) {
if (fpr <= REGARG_LASTFPR) {
- lua_assert(rset_test(as->freeset, fpr)); /* Must have been evicted. */
+ lj_assertA(rset_test(as->freeset, fpr),
+ "reg %d not free", fpr); /* Must have been evicted. */
ra_leftov(as, fpr, ref);
fpr++;
+#if LJ_HASFFI && LJ_ABI_WIN
+ } else if ((ci->flags & CCI_VARARG) && (gpr <= REGARG_LASTGPR)) {
+ Reg rf = ra_alloc1(as, ref, RSET_FPR);
+ emit_dn(as, A64I_FMOV_R_D, gpr++, rf & 31);
+#endif
} else {
Reg r = ra_alloc1(as, ref, RSET_FPR);
- emit_spstore(as, ir, r, ofs + ((LJ_BE && !irt_isnum(ir->t)) ? 4 : 0));
- ofs += 8;
+ int32_t al = spalign;
+#if LJ_HASFFI && LJ_TARGET_OSX
+ al |= irt_isnum(ir->t) ? 7 : 3;
+#endif
+ spofs = (spofs + al) & ~al;
+ if (LJ_BE && al >= 7 && !irt_isnum(ir->t)) spofs += 4, al -= 4;
+ emit_spstore(as, ir, r, spofs);
+ spofs += al + 1;
}
} else {
if (gpr <= REGARG_LASTGPR) {
- lua_assert(rset_test(as->freeset, gpr)); /* Must have been evicted. */
+ lj_assertA(rset_test(as->freeset, gpr),
+ "reg %d not free", gpr); /* Must have been evicted. */
ra_leftov(as, gpr, ref);
gpr++;
} else {
Reg r = ra_alloc1(as, ref, RSET_GPR);
- emit_spstore(as, ir, r, ofs + ((LJ_BE && !irt_is64(ir->t)) ? 4 : 0));
- ofs += 8;
+ int32_t al = spalign;
+#if LJ_HASFFI && LJ_TARGET_OSX
+ al |= irt_size(ir->t) - 1;
+#endif
+ spofs = (spofs + al) & ~al;
+ if (al >= 3) {
+ if (LJ_BE && al >= 7 && !irt_is64(ir->t)) spofs += 4, al -= 4;
+ emit_spstore(as, ir, r, spofs);
+ } else {
+ lj_assertA(al == 0 || al == 1, "size %d unexpected", al + 1);
+ emit_lso(as, al ? A64I_STRH : A64I_STRB, r, RID_SP, spofs);
+ }
+ spofs += al + 1;
}
}
+#if LJ_HASFFI && LJ_TARGET_OSX
+ } else { /* Marker for start of varargs. */
+ gpr = REGARG_LASTGPR+1;
+ fpr = REGARG_LASTFPR+1;
+ spalign = 7;
+#endif
}
}
}
@@ -453,11 +501,14 @@
static void asm_setupresult(ASMState *as, IRIns *ir, const CCallInfo *ci)
{
RegSet drop = RSET_SCRATCH;
+ int hiop = ((ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t));
if (ra_hasreg(ir->r))
rset_clear(drop, ir->r); /* Dest reg handled below. */
+ if (hiop && ra_hasreg((ir+1)->r))
+ rset_clear(drop, (ir+1)->r); /* Dest reg handled below. */
ra_evictset(as, drop); /* Evictions must be performed first. */
if (ra_used(ir)) {
- lua_assert(!irt_ispri(ir->t));
+ lj_assertA(!irt_ispri(ir->t), "PRI dest");
if (irt_isfp(ir->t)) {
if (ci->flags & CCI_CASTU64) {
Reg dest = ra_dest(as, ir, RSET_FPR) & 31;
@@ -466,6 +517,8 @@
} else {
ra_destreg(as, ir, RID_FPRET);
}
+ } else if (hiop) {
+ ra_destpair(as, ir);
} else {
ra_destreg(as, ir, RID_RET);
}
@@ -488,7 +541,7 @@
ci.func = (ASMFunction)(ir_k64(irf)->u64);
} else { /* Need a non-argument register for indirect calls. */
Reg freg = ra_alloc1(as, func, RSET_RANGE(RID_X8, RID_MAX_GPR)-RSET_FIXED);
- emit_n(as, A64I_BLR, freg);
+ emit_n(as, A64I_BLR_AUTH, freg);
ci.func = (ASMFunction)(void *)0;
}
asm_gencall(as, &ci, args);
@@ -505,8 +558,6 @@
as->topslot -= (BCReg)delta;
if ((int32_t)as->topslot < 0) as->topslot = 0;
irt_setmark(IR(REF_BASE)->t); /* Children must not coalesce with BASE reg. */
- /* Need to force a spill on REF_BASE now to update the stack slot. */
- emit_lso(as, A64I_STRx, base, RID_SP, ra_spill(as, IR(REF_BASE)));
emit_setgl(as, base, jit_base);
emit_addptr(as, base, -8*delta);
asm_guardcc(as, CC_NE);
@@ -515,6 +566,21 @@
emit_lso(as, A64I_LDRx, RID_TMP, base, -8);
}
+/* -- Buffer operations --------------------------------------------------- */
+
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb)
+{
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irgc;
+ irgc.ot = IRT(0, IRT_PGC); /* GC type. */
+ emit_storeofs(as, &irgc, RID_TMP, sb, offsetof(SBuf, L));
+ emit_dn(as, A64I_BFMx | A64F_IMMS(lj_fls(SBUF_MASK_FLAG)) | A64F_IMMR(0), RID_TMP, tmp);
+ emit_getgl(as, RID_TMP, cur_L);
+ emit_loadofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+}
+#endif
+
/* -- Type conversions ---------------------------------------------------- */
static void asm_tointg(ASMState *as, IRIns *ir, Reg left)
@@ -544,7 +610,7 @@
int st64 = (st == IRT_I64 || st == IRT_U64 || st == IRT_P64);
int stfp = (st == IRT_NUM || st == IRT_FLOAT);
IRRef lref = ir->op1;
- lua_assert(irt_type(ir->t) != st);
+ lj_assertA(irt_type(ir->t) != st, "inconsistent types for CONV");
if (irt_isfp(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
if (stfp) { /* FP to FP conversion. */
@@ -564,7 +630,8 @@
} else if (stfp) { /* FP to integer conversion. */
if (irt_isguard(ir->t)) {
/* Checked conversions are only supported from number to int. */
- lua_assert(irt_isint(ir->t) && st == IRT_NUM);
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
} else {
Reg left = ra_alloc1(as, lref, RSET_FPR);
@@ -584,7 +651,7 @@
A64Ins ai = st == IRT_I8 ? A64I_SXTBw :
st == IRT_U8 ? A64I_UXTBw :
st == IRT_I16 ? A64I_SXTHw : A64I_UXTHw;
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t), "bad type for CONV EXT");
emit_dn(as, ai, dest, left);
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -597,7 +664,7 @@
emit_dn(as, A64I_SXTW, dest, left);
}
} else {
- if (st64) {
+ if (st64 && !(ir->op2 & IRCONV_NONE)) {
/* This is either a 32 bit reg/reg mov which zeroes the hiword
** or a load of the loword from a 64 bit address.
*/
@@ -614,14 +681,12 @@
{
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_strscan_num];
IRRef args[2];
- Reg dest = 0, tmp;
- int destused = ra_used(ir);
+ Reg tmp;
int32_t ofs = 0;
ra_evictset(as, RSET_SCRATCH);
- if (destused) {
+ if (ra_used(ir)) {
if (ra_hasspill(ir->s)) {
ofs = sps_scale(ir->s);
- destused = 0;
if (ra_hasreg(ir->r)) {
ra_free(as, ir->r);
ra_modified(as, ir->r);
@@ -628,11 +693,10 @@
emit_spload(as, ir, ir->r, ofs);
}
} else {
- dest = ra_dest(as, ir, RSET_FPR);
+ Reg dest = ra_dest(as, ir, RSET_FPR);
+ emit_lso(as, A64I_LDRd, (dest & 31), RID_SP, 0);
}
}
- if (destused)
- emit_lso(as, A64I_LDRd, (dest & 31), RID_SP, 0);
asm_guardcnb(as, A64I_CBZ, RID_RET);
args[0] = ir->op1; /* GCstr *str */
args[1] = ASMREF_TMP1; /* TValue *n */
@@ -648,7 +712,8 @@
{
RegSet allow = rset_exclude(RSET_GPR, base);
IRIns *ir = IR(ref);
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t),
+ "store of IR type %d", irt_type(ir->t));
if (irref_isk(ref)) {
TValue k;
lj_ir_kvalue(as->J->L, &k, ir);
@@ -669,22 +734,23 @@
}
/* Get pointer to TValue. */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode)
{
- IRIns *ir = IR(ref);
- if (irt_isnum(ir->t)) {
- if (irref_isk(ref)) {
- /* Use the number constant itself as a TValue. */
- ra_allockreg(as, i64ptr(ir_knum(ir)), dest);
+ if ((mode & IRTMPREF_IN1)) {
+ IRIns *ir = IR(ref);
+ if (irt_isnum(ir->t)) {
+ if (irref_isk(ref) && !(mode & IRTMPREF_OUT1)) {
+ /* Use the number constant itself as a TValue. */
+ ra_allockreg(as, i64ptr(ir_knum(ir)), dest);
+ return;
+ }
+ emit_lso(as, A64I_STRd, (ra_alloc1(as, ref, RSET_FPR) & 31), dest, 0);
} else {
- /* Otherwise force a spill and use the spill slot. */
- emit_opk(as, A64I_ADDx, dest, RID_SP, ra_spill(as, ir), RSET_GPR);
+ asm_tvstore64(as, dest, 0, ref);
}
- } else {
- /* Otherwise use g->tmptv to hold the TValue. */
- asm_tvstore64(as, dest, 0, ref);
- ra_allockreg(as, i64ptr(&J2G(as->J)->tmptv), dest);
}
+ /* g->tmptv holds the TValue(s). */
+ emit_dn(as, A64I_ADDx^emit_isk12(glofs(as, &J2G(as->J)->tmptv)), dest, RID_GL);
}
static void asm_aref(ASMState *as, IRIns *ir)
@@ -721,49 +787,49 @@
int destused = ra_used(ir);
Reg dest = ra_dest(as, ir, allow);
Reg tab = ra_alloc1(as, ir->op1, rset_clear(allow, dest));
- Reg key = 0, tmp = RID_TMP;
+ Reg tmp = RID_TMP, type = RID_NONE, key = RID_NONE, tkey;
IRRef refkey = ir->op2;
IRIns *irkey = IR(refkey);
- int isk = irref_isk(ir->op2);
+ int isk = irref_isk(refkey);
IRType1 kt = irkey->t;
uint32_t k = 0;
uint32_t khash;
- MCLabel l_end, l_loop, l_next;
+ MCLabel l_end, l_loop;
rset_clear(allow, tab);
- if (!isk) {
- key = ra_alloc1(as, ir->op2, irt_isnum(kt) ? RSET_FPR : allow);
- rset_clear(allow, key);
- if (!irt_isstr(kt)) {
- tmp = ra_scratch(as, allow);
- rset_clear(allow, tmp);
+ /* Allocate register for tkey outside of the loop. */
+ if (isk) {
+ int64_t kk;
+ if (irt_isaddr(kt)) {
+ kk = ((int64_t)irt_toitype(kt) << 47) | irkey[1].tv.u64;
+ } else if (irt_isnum(kt)) {
+ kk = (int64_t)ir_knum(irkey)->u64;
+ /* Assumes -0.0 is already canonicalized to +0.0. */
+ } else {
+ lj_assertA(irt_ispri(kt) && !irt_isnil(kt), "bad HREF key type");
+ kk = ~((int64_t)~irt_toitype(kt) << 47);
}
- } else if (irt_isnum(kt)) {
- int64_t val = (int64_t)ir_knum(irkey)->u64;
- if (!(k = emit_isk12(val))) {
- key = ra_allock(as, val, allow);
- rset_clear(allow, key);
- }
- } else if (!irt_ispri(kt)) {
- if (!(k = emit_isk12(irkey->i))) {
- key = ra_alloc1(as, refkey, allow);
- rset_clear(allow, key);
- }
+ k = emit_isk12(kk);
+ tkey = k ? 0 : ra_allock(as, kk, allow);
+ } else {
+ tkey = ra_scratch(as, allow);
}
/* Key not found in chain: jump to exit (if merged) or load niltv. */
l_end = emit_label(as);
as->invmcp = NULL;
- if (merge == IR_NE)
+ if (merge == IR_NE) {
asm_guardcc(as, CC_AL);
- else if (destused)
- emit_loada(as, dest, niltvg(J2G(as->J)));
+ } else if (destused) {
+ uint32_t k12 = emit_isk12(offsetof(global_State, nilnode.val));
+ lj_assertA(k12 != 0, "Cannot k12 encode niltv(L)");
+ emit_dn(as, A64I_ADDx^k12, dest, RID_GL);
+ }
/* Follow hash chain until the end. */
l_loop = --as->mcp;
- emit_n(as, A64I_CMPx^A64I_K12^0, dest);
- emit_lso(as, A64I_LDRx, dest, dest, offsetof(Node, next));
- l_next = emit_label(as);
+ if (destused)
+ emit_lso(as, A64I_LDRx, dest, dest, offsetof(Node, next));
/* Type and value comparison. */
if (merge == IR_EQ)
@@ -770,56 +836,28 @@
asm_guardcc(as, CC_EQ);
else
emit_cond_branch(as, CC_EQ, l_end);
+ emit_nm(as, A64I_CMPx^k, tmp, tkey);
+ if (!destused)
+ emit_lso(as, A64I_LDRx, dest, dest, offsetof(Node, next));
+ emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key));
+ *l_loop = A64I_X | A64I_CBNZ | A64F_S19(as->mcp - l_loop) | dest;
- if (irt_isnum(kt)) {
- if (isk) {
- /* Assumes -0.0 is already canonicalized to +0.0. */
- if (k)
- emit_n(as, A64I_CMPx^k, tmp);
- else
- emit_nm(as, A64I_CMPx, key, tmp);
- emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.u64));
+ /* Construct tkey as canonicalized or tagged key. */
+ if (!isk) {
+ if (irt_isnum(kt)) {
+ key = ra_alloc1(as, refkey, RSET_FPR);
+ emit_dnm(as, A64I_CSELx | A64F_CC(CC_EQ), tkey, RID_ZERO, tkey);
+ /* A64I_FMOV_R_D from key to tkey done below. */
} else {
- Reg tisnum = ra_allock(as, LJ_TISNUM << 15, allow);
- Reg ftmp = ra_scratch(as, rset_exclude(RSET_FPR, key));
- rset_clear(allow, tisnum);
- emit_nm(as, A64I_FCMPd, key, ftmp);
- emit_dn(as, A64I_FMOV_D_R, (ftmp & 31), (tmp & 31));
- emit_cond_branch(as, CC_LO, l_next);
- emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32), tisnum, tmp);
- emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.n));
+ lj_assertA(irt_isaddr(kt), "bad HREF key type");
+ key = ra_alloc1(as, refkey, allow);
+ type = ra_allock(as, irt_toitype(kt) << 15, rset_clear(allow, key));
+ emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 32), tkey, key, type);
}
- } else if (irt_isaddr(kt)) {
- Reg scr;
- if (isk) {
- int64_t kk = ((int64_t)irt_toitype(irkey->t) << 47) | irkey[1].tv.u64;
- scr = ra_allock(as, kk, allow);
- emit_nm(as, A64I_CMPx, scr, tmp);
- emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.u64));
- } else {
- scr = ra_scratch(as, allow);
- emit_nm(as, A64I_CMPx, tmp, scr);
- emit_lso(as, A64I_LDRx, scr, dest, offsetof(Node, key.u64));
- }
- rset_clear(allow, scr);
- } else {
- Reg type, scr;
- lua_assert(irt_ispri(kt) && !irt_isnil(kt));
- type = ra_allock(as, ~((int64_t)~irt_toitype(ir->t) << 47), allow);
- scr = ra_scratch(as, rset_clear(allow, type));
- rset_clear(allow, scr);
- emit_nm(as, A64I_CMPw, scr, type);
- emit_lso(as, A64I_LDRx, scr, dest, offsetof(Node, key));
}
- *l_loop = A64I_BCC | A64F_S19(as->mcp - l_loop) | CC_NE;
- if (!isk && irt_isaddr(kt)) {
- Reg type = ra_allock(as, (int32_t)irt_toitype(kt), allow);
- emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 47), tmp, key, type);
- rset_clear(allow, type);
- }
/* Load main position relative to tab->node into dest. */
- khash = isk ? ir_khash(irkey) : 1;
+ khash = isk ? ir_khash(as, irkey) : 1;
if (khash == 0) {
emit_lso(as, A64I_LDRx, dest, tab, offsetof(GCtab, node));
} else {
@@ -831,9 +869,8 @@
emit_dnm(as, A64I_ANDw, dest, dest, tmphash);
emit_lso(as, A64I_LDRw, dest, tab, offsetof(GCtab, hmask));
} else if (irt_isstr(kt)) {
- /* Fetch of str->hash is cheaper than ra_allock. */
emit_dnm(as, A64I_ANDw, dest, dest, tmp);
- emit_lso(as, A64I_LDRw, tmp, key, offsetof(GCstr, hash));
+ emit_lso(as, A64I_LDRw, tmp, key, offsetof(GCstr, sid));
emit_lso(as, A64I_LDRw, dest, tab, offsetof(GCtab, hmask));
} else { /* Must match with hash*() in lj_tab.c. */
emit_dnm(as, A64I_ANDw, dest, dest, tmp);
@@ -840,23 +877,18 @@
emit_lso(as, A64I_LDRw, tmp, tab, offsetof(GCtab, hmask));
emit_dnm(as, A64I_SUBw, dest, dest, tmp);
emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT3)), tmp, tmp, tmp);
- emit_dnm(as, A64I_EORw, dest, dest, tmp);
- emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT2)), dest, dest, dest);
+ emit_dnm(as, A64I_EORw | A64F_SH(A64SH_ROR, 32-HASH_ROT2), dest, tmp, dest);
emit_dnm(as, A64I_SUBw, tmp, tmp, dest);
emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT1)), dest, dest, dest);
- emit_dnm(as, A64I_EORw, tmp, tmp, dest);
if (irt_isnum(kt)) {
+ emit_dnm(as, A64I_EORw, tmp, tkey, dest);
emit_dnm(as, A64I_ADDw, dest, dest, dest);
- emit_dn(as, A64I_LSRx | A64F_IMMR(32)|A64F_IMMS(32), dest, dest);
- emit_dm(as, A64I_MOVw, tmp, dest);
- emit_dn(as, A64I_FMOV_R_D, dest, (key & 31));
+ emit_dn(as, A64I_LSRx | A64F_IMMR(32)|A64F_IMMS(32), dest, tkey);
+ emit_nm(as, A64I_FCMPZd, (key & 31), 0);
+ emit_dn(as, A64I_FMOV_R_D, tkey, (key & 31));
} else {
- checkmclim(as);
- emit_dm(as, A64I_MOVw, tmp, key);
- emit_dnm(as, A64I_EORw, dest, dest,
- ra_allock(as, irt_toitype(kt) << 15, allow));
- emit_dn(as, A64I_LSRx | A64F_IMMR(32)|A64F_IMMS(32), dest, dest);
- emit_dm(as, A64I_MOVx, dest, key);
+ emit_dnm(as, A64I_EORw, tmp, key, dest);
+ emit_dnm(as, A64I_EORx | A64F_SH(A64SH_LSR, 32), dest, type, key);
}
}
}
@@ -868,15 +900,13 @@
IRIns *irkey = IR(kslot->op1);
int32_t ofs = (int32_t)(kslot->op2 * sizeof(Node));
int32_t kofs = ofs + (int32_t)offsetof(Node, key);
- int bigofs = !emit_checkofs(A64I_LDRx, ofs);
- RegSet allow = RSET_GPR;
+ int bigofs = !emit_checkofs(A64I_LDRx, kofs);
Reg dest = (ra_used(ir) || bigofs) ? ra_dest(as, ir, RSET_GPR) : RID_NONE;
- Reg node = ra_alloc1(as, ir->op1, allow);
- Reg key = ra_scratch(as, rset_clear(allow, node));
+ Reg node = ra_alloc1(as, ir->op1, RSET_GPR);
Reg idx = node;
+ RegSet allow = rset_exclude(RSET_GPR, node);
uint64_t k;
- lua_assert(ofs % sizeof(Node) == 0);
- rset_clear(allow, key);
+ lj_assertA(ofs % sizeof(Node) == 0, "unaligned HREFK slot");
if (bigofs) {
idx = dest;
rset_clear(allow, dest);
@@ -892,33 +922,39 @@
} else {
k = ((uint64_t)irt_toitype(irkey->t) << 47) | (uint64_t)ir_kgc(irkey);
}
- emit_nm(as, A64I_CMPx, key, ra_allock(as, k, allow));
- emit_lso(as, A64I_LDRx, key, idx, kofs);
+ emit_nm(as, A64I_CMPx, RID_TMP, ra_allock(as, k, allow));
+ emit_lso(as, A64I_LDRx, RID_TMP, idx, kofs);
if (bigofs)
- emit_opk(as, A64I_ADDx, dest, node, ofs, RSET_GPR);
+ emit_opk(as, A64I_ADDx, dest, node, ofs, rset_exclude(RSET_GPR, node));
}
static void asm_uref(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (irref_isk(ir->op1)) {
+ int guarded = (irt_t(ir->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC);
+ if (irref_isk(ir->op1) && !guarded) {
GCfunc *fn = ir_kfunc(IR(ir->op1));
MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
emit_lsptr(as, A64I_LDRx, dest, v);
} else {
- Reg uv = ra_scratch(as, RSET_GPR);
- Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
- if (ir->o == IR_UREFC) {
- asm_guardcc(as, CC_NE);
- emit_n(as, (A64I_CMPx^A64I_K12) | A64F_U12(1), RID_TMP);
- emit_opk(as, A64I_ADDx, dest, uv,
+ if (guarded)
+ asm_guardcnb(as, ir->o == IR_UREFC ? A64I_CBZ : A64I_CBNZ, RID_TMP);
+ if (ir->o == IR_UREFC)
+ emit_opk(as, A64I_ADDx, dest, dest,
(int32_t)offsetof(GCupval, tv), RSET_GPR);
- emit_lso(as, A64I_LDRB, RID_TMP, uv, (int32_t)offsetof(GCupval, closed));
+ else
+ emit_lso(as, A64I_LDRx, dest, dest, (int32_t)offsetof(GCupval, v));
+ if (guarded)
+ emit_lso(as, A64I_LDRB, RID_TMP, dest,
+ (int32_t)offsetof(GCupval, closed));
+ if (irref_isk(ir->op1)) {
+ GCfunc *fn = ir_kfunc(IR(ir->op1));
+ uint64_t k = gcrefu(fn->l.uvptr[(ir->op2 >> 8)]);
+ emit_loadu64(as, dest, k);
} else {
- emit_lso(as, A64I_LDRx, dest, uv, (int32_t)offsetof(GCupval, v));
+ emit_lso(as, A64I_LDRx, dest, ra_alloc1(as, ir->op1, RSET_GPR),
+ (int32_t)offsetof(GCfuncL, uvptr) + 8*(int32_t)(ir->op2 >> 8));
}
- emit_lso(as, A64I_LDRx, uv, func,
- (int32_t)offsetof(GCfuncL, uvptr) + 8*(int32_t)(ir->op2 >> 8));
}
}
@@ -925,7 +961,7 @@
static void asm_fref(ASMState *as, IRIns *ir)
{
UNUSED(as); UNUSED(ir);
- lua_assert(!ra_used(ir));
+ lj_assertA(!ra_used(ir), "unfused FREF");
}
static void asm_strref(ASMState *as, IRIns *ir)
@@ -977,7 +1013,7 @@
Reg idx;
A64Ins ai = asm_fxloadins(ir);
int32_t ofs;
- if (ir->op1 == REF_NIL) {
+ if (ir->op1 == REF_NIL) { /* FLOAD from GG_State with offset. */
idx = RID_GL;
ofs = (ir->op2 << 2) - GG_OFS(g);
} else {
@@ -1008,7 +1044,7 @@
static void asm_xload(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR);
- lua_assert(!(ir->op2 & IRXLOAD_UNALIGNED));
+ lj_assertA(!(ir->op2 & IRXLOAD_UNALIGNED), "unaligned XLOAD");
asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR);
}
@@ -1023,11 +1059,12 @@
static void asm_ahuvload(ASMState *as, IRIns *ir)
{
- Reg idx, tmp, type;
+ Reg idx, tmp;
int32_t ofs = 0;
RegSet gpr = RSET_GPR, allow = irt_isnum(ir->t) ? RSET_FPR : RSET_GPR;
- lua_assert(irt_isnum(ir->t) || irt_ispri(ir->t) || irt_isaddr(ir->t) ||
- irt_isint(ir->t));
+ lj_assertA(irt_isnum(ir->t) || irt_ispri(ir->t) || irt_isaddr(ir->t) ||
+ irt_isint(ir->t),
+ "bad load type %d", irt_type(ir->t));
if (ra_used(ir)) {
Reg dest = ra_dest(as, ir, allow);
tmp = irt_isnum(ir->t) ? ra_scratch(as, rset_clear(gpr, dest)) : dest;
@@ -1041,22 +1078,25 @@
} else {
tmp = ra_scratch(as, gpr);
}
- type = ra_scratch(as, rset_clear(gpr, tmp));
- idx = asm_fuseahuref(as, ir->op1, &ofs, rset_clear(gpr, type), A64I_LDRx);
+ idx = asm_fuseahuref(as, ir->op1, &ofs, rset_clear(gpr, tmp), A64I_LDRx);
+ rset_clear(gpr, idx);
+ if (ofs & FUSE_REG) rset_clear(gpr, ofs & 31);
+ if (ir->o == IR_VLOAD) ofs += 8 * ir->op2;
/* Always do the type check, even if the load result is unused. */
asm_guardcc(as, irt_isnum(ir->t) ? CC_LS : CC_NE);
if (irt_type(ir->t) >= IRT_NUM) {
- lua_assert(irt_isinteger(ir->t) || irt_isnum(ir->t));
+ lj_assertA(irt_isinteger(ir->t) || irt_isnum(ir->t),
+ "bad load type %d", irt_type(ir->t));
emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
- ra_allock(as, LJ_TISNUM << 15, rset_exclude(gpr, idx)), tmp);
+ ra_allock(as, LJ_TISNUM << 15, gpr), tmp);
} else if (irt_isaddr(ir->t)) {
- emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(ir->t)), type);
- emit_dn(as, A64I_ASRx | A64F_IMMR(47), type, tmp);
+ emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(ir->t)), RID_TMP);
+ emit_dn(as, A64I_ASRx | A64F_IMMR(47), RID_TMP, tmp);
} else if (irt_isnil(ir->t)) {
emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(1), tmp);
} else {
emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
- ra_allock(as, (irt_toitype(ir->t) << 15) | 0x7fff, allow), tmp);
+ ra_allock(as, (irt_toitype(ir->t) << 15) | 0x7fff, gpr), tmp);
}
if (ofs & FUSE_REG)
emit_dnm(as, (A64I_LDRx^A64I_LS_R)|A64I_LS_UXTWx|A64I_LS_SH, tmp, idx, (ofs & 31));
@@ -1111,8 +1151,10 @@
IRType1 t = ir->t;
Reg dest = RID_NONE, base;
RegSet allow = RSET_GPR;
- lua_assert(!(ir->op2 & IRSLOAD_PARENT)); /* Handled by asm_head_side(). */
- lua_assert(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK));
+ lj_assertA(!(ir->op2 & IRSLOAD_PARENT),
+ "bad parent SLOAD"); /* Handled by asm_head_side(). */
+ lj_assertA(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK),
+ "inconsistent SLOAD variant");
if ((ir->op2 & IRSLOAD_CONVERT) && irt_isguard(t) && irt_isint(t)) {
dest = ra_scratch(as, RSET_FPR);
asm_tointg(as, ir, dest);
@@ -1121,7 +1163,8 @@
Reg tmp = RID_NONE;
if ((ir->op2 & IRSLOAD_CONVERT))
tmp = ra_scratch(as, irt_isint(t) ? RSET_FPR : RSET_GPR);
- lua_assert((irt_isnum(t)) || irt_isint(t) || irt_isaddr(t));
+ lj_assertA((irt_isnum(t)) || irt_isint(t) || irt_isaddr(t),
+ "bad SLOAD type %d", irt_type(t));
dest = ra_dest(as, ir, irt_isnum(t) ? RSET_FPR : allow);
base = ra_alloc1(as, REF_BASE, rset_clear(allow, dest));
if (irt_isaddr(t)) {
@@ -1156,14 +1199,15 @@
tmp = ra_scratch(as, allow);
rset_clear(allow, tmp);
}
- if (irt_isnum(t) && !(ir->op2 & IRSLOAD_CONVERT))
+ if (ra_hasreg(dest) && tmp != dest)
emit_dn(as, A64I_FMOV_D_R, (dest & 31), tmp);
/* Need type check, even if the load result is unused. */
asm_guardcc(as, irt_isnum(t) ? CC_LS : CC_NE);
if (irt_type(t) >= IRT_NUM) {
- lua_assert(irt_isinteger(t) || irt_isnum(t));
+ lj_assertA(irt_isinteger(t) || irt_isnum(t),
+ "bad SLOAD type %d", irt_type(t));
emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
- ra_allock(as, LJ_TISNUM << 15, allow), tmp);
+ ra_allock(as, (ir->op2 & IRSLOAD_KEYINDEX) ? LJ_KEYINDEX : (LJ_TISNUM << 15), allow), tmp);
} else if (irt_isnil(t)) {
emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(1), tmp);
} else if (irt_ispri(t)) {
@@ -1170,9 +1214,8 @@
emit_nm(as, A64I_CMPx,
ra_allock(as, ~((int64_t)~irt_toitype(t) << 47) , allow), tmp);
} else {
- Reg type = ra_scratch(as, allow);
- emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(t)), type);
- emit_dn(as, A64I_ASRx | A64F_IMMR(47), type, tmp);
+ emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(t)), RID_TMP);
+ emit_dn(as, A64I_ASRx | A64F_IMMR(47), RID_TMP, tmp);
}
emit_lso(as, A64I_LDRx, tmp, base, ofs);
return;
@@ -1196,7 +1239,8 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
IRRef args[4];
RegSet allow = (RSET_GPR & ~RSET_SCRATCH);
- lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
+ lj_assertA(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL),
+ "bad CNEW/CNEWI operands");
as->gcsteps++;
asm_setupresult(as, ir, ci); /* GCcdata * */
@@ -1204,7 +1248,7 @@
if (ir->o == IR_CNEWI) {
int32_t ofs = sizeof(GCcdata);
Reg r = ra_alloc1(as, ir->op2, allow);
- lua_assert(sz == 4 || sz == 8);
+ lj_assertA(sz == 4 || sz == 8, "bad CNEWI size %d", sz);
emit_lso(as, sz == 8 ? A64I_STRx : A64I_STRw, r, RID_RET, ofs);
} else if (ir->op2 != REF_NIL) { /* Create VLA/VLS/aligned cdata. */
ci = &lj_ir_callinfo[IRCALL_lj_cdata_newv];
@@ -1231,8 +1275,6 @@
ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
ra_releasetmp(as, ASMREF_TMP1));
}
-#else
-#define asm_cnew(as, ir) ((void)0)
#endif
/* -- Write barriers ------------------------------------------------------ */
@@ -1241,17 +1283,14 @@
{
Reg tab = ra_alloc1(as, ir->op1, RSET_GPR);
Reg link = ra_scratch(as, rset_exclude(RSET_GPR, tab));
- Reg gr = ra_allock(as, i64ptr(J2G(as->J)),
- rset_exclude(rset_exclude(RSET_GPR, tab), link));
Reg mark = RID_TMP;
MCLabel l_end = emit_label(as);
+ emit_lso(as, A64I_STRB, mark, tab, (int32_t)offsetof(GCtab, marked));
+ /* Keep STRx in the middle to avoid LDP/STP fusion with surrounding code. */
emit_lso(as, A64I_STRx, link, tab, (int32_t)offsetof(GCtab, gclist));
- emit_lso(as, A64I_STRB, mark, tab, (int32_t)offsetof(GCtab, marked));
- emit_lso(as, A64I_STRx, tab, gr,
- (int32_t)offsetof(global_State, gc.grayagain));
+ emit_setgl(as, tab, gc.grayagain);
emit_dn(as, A64I_ANDw^emit_isk13(~LJ_GC_BLACK, 0), mark, mark);
- emit_lso(as, A64I_LDRx, link, gr,
- (int32_t)offsetof(global_State, gc.grayagain));
+ emit_getgl(as, link, gc.grayagain);
emit_cond_branch(as, CC_EQ, l_end);
emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_BLACK, 0), mark);
emit_lso(as, A64I_LDRB, mark, tab, (int32_t)offsetof(GCtab, marked));
@@ -1262,25 +1301,23 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_gc_barrieruv];
IRRef args[2];
MCLabel l_end;
- RegSet allow = RSET_GPR;
Reg obj, val, tmp;
/* No need for other object barriers (yet). */
- lua_assert(IR(ir->op1)->o == IR_UREFC);
+ lj_assertA(IR(ir->op1)->o == IR_UREFC, "bad OBAR type");
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
args[0] = ASMREF_TMP1; /* global_State *g */
args[1] = ir->op1; /* TValue *tv */
asm_gencall(as, ci, args);
- ra_allockreg(as, i64ptr(J2G(as->J)), ra_releasetmp(as, ASMREF_TMP1) );
+ emit_dm(as, A64I_MOVx, ra_releasetmp(as, ASMREF_TMP1), RID_GL);
obj = IR(ir->op1)->r;
- tmp = ra_scratch(as, rset_exclude(allow, obj));
+ tmp = ra_scratch(as, rset_exclude(RSET_GPR, obj));
+ emit_tnb(as, A64I_TBZ, tmp, lj_ffs(LJ_GC_BLACK), l_end);
emit_cond_branch(as, CC_EQ, l_end);
- emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_BLACK, 0), tmp);
- emit_cond_branch(as, CC_EQ, l_end);
emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_WHITES, 0), RID_TMP);
val = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, obj));
emit_lso(as, A64I_LDRB, tmp, obj,
- (int32_t)offsetof(GCupval, marked)-(int32_t)offsetof(GCupval, tv));
+ (int32_t)offsetof(GCupval, marked)-(int32_t)offsetof(GCupval, tv));
emit_lso(as, A64I_LDRB, RID_TMP, val, (int32_t)offsetof(GChead, marked));
}
@@ -1309,8 +1346,6 @@
} else if (fpm <= IRFPM_TRUNC) {
asm_fpunary(as, ir, fpm == IRFPM_FLOOR ? A64I_FRINTMd :
fpm == IRFPM_CEIL ? A64I_FRINTPd : A64I_FRINTZd);
- } else if (fpm == IRFPM_EXP2 && asm_fpjoin_pow(as, ir)) {
- return;
} else {
asm_callid(as, ir, IRCALL_lj_vm_floor + fpm);
}
@@ -1324,12 +1359,12 @@
if (irref_isk(lref))
return 1; /* But swap constants to the right. */
ir = IR(rref);
- if ((ir->o >= IR_BSHL && ir->o <= IR_BSAR) ||
+ if ((ir->o >= IR_BSHL && ir->o <= IR_BROR) ||
(ir->o == IR_ADD && ir->op1 == ir->op2) ||
(ir->o == IR_CONV && ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)))
return 0; /* Don't swap fusable operands to the left. */
ir = IR(lref);
- if ((ir->o >= IR_BSHL && ir->o <= IR_BSAR) ||
+ if ((ir->o >= IR_BSHL && ir->o <= IR_BROR) ||
(ir->o == IR_ADD && ir->op1 == ir->op2) ||
(ir->o == IR_CONV && ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)))
return 1; /* But swap fusable operands to the right. */
@@ -1375,13 +1410,12 @@
static void asm_intmul(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- Reg left = ra_alloc1(as, ir->op1, rset_exclude(RSET_GPR, dest));
+ Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
Reg right = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, left));
if (irt_isguard(ir->t)) { /* IR_MULOV */
asm_guardcc(as, CC_NE);
emit_dm(as, A64I_MOVw, dest, dest); /* Zero-extend. */
- emit_nm(as, A64I_CMPw | A64F_SH(A64SH_ASR, 31), RID_TMP, dest);
- emit_dn(as, A64I_ASRx | A64F_IMMR(32), RID_TMP, dest);
+ emit_nm(as, A64I_CMPx | A64F_EX(A64EX_SXTW), dest, dest);
emit_dnm(as, A64I_SMULL, dest, right, left);
} else {
emit_dnm(as, irt_is64(ir->t) ? A64I_MULx : A64I_MULw, dest, left, right);
@@ -1417,47 +1451,13 @@
asm_intmul(as, ir);
}
-static void asm_div(ASMState *as, IRIns *ir)
-{
-#if LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_divi64 :
- IRCALL_lj_carith_divu64);
- else
-#endif
- asm_fparith(as, ir, A64I_FDIVd);
-}
-
-static void asm_pow(ASMState *as, IRIns *ir)
-{
-#if LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_powi64 :
- IRCALL_lj_carith_powu64);
- else
-#endif
- asm_callid(as, ir, IRCALL_lj_vm_powi);
-}
-
#define asm_addov(as, ir) asm_add(as, ir)
#define asm_subov(as, ir) asm_sub(as, ir)
#define asm_mulov(as, ir) asm_mul(as, ir)
+#define asm_fpdiv(as, ir) asm_fparith(as, ir, A64I_FDIVd)
#define asm_abs(as, ir) asm_fpunary(as, ir, A64I_FABS)
-#define asm_atan2(as, ir) asm_callid(as, ir, IRCALL_atan2)
-#define asm_ldexp(as, ir) asm_callid(as, ir, IRCALL_ldexp)
-static void asm_mod(ASMState *as, IRIns *ir)
-{
-#if LJ_HASFFI
- if (!irt_isint(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_modi64 :
- IRCALL_lj_carith_modu64);
- else
-#endif
- asm_callid(as, ir, IRCALL_lj_vm_modi);
-}
-
static void asm_neg(ASMState *as, IRIns *ir)
{
if (irt_isnum(ir->t)) {
@@ -1571,7 +1571,7 @@
#define asm_bshr(as, ir) asm_bitshift(as, ir, A64I_UBFMw, A64SH_LSR)
#define asm_bsar(as, ir) asm_bitshift(as, ir, A64I_SBFMw, A64SH_ASR)
#define asm_bror(as, ir) asm_bitshift(as, ir, A64I_EXTRw, A64SH_ROR)
-#define asm_brol(as, ir) lua_assert(0)
+#define asm_brol(as, ir) lj_assertA(0, "unexpected BROL")
static void asm_intmin_max(ASMState *as, IRIns *ir, A64CC cc)
{
@@ -1587,7 +1587,7 @@
Reg dest = (ra_dest(as, ir, RSET_FPR) & 31);
Reg right, left = ra_alloc2(as, ir, RSET_FPR);
right = ((left >> 8) & 31); left &= 31;
- emit_dnm(as, A64I_FCSELd | A64F_CC(fcc), dest, left, right);
+ emit_dnm(as, A64I_FCSELd | A64F_CC(fcc), dest, right, left);
emit_nm(as, A64I_FCMPd, left, right);
}
@@ -1599,8 +1599,8 @@
asm_intmin_max(as, ir, cc);
}
-#define asm_max(as, ir) asm_min_max(as, ir, CC_GT, CC_HI)
-#define asm_min(as, ir) asm_min_max(as, ir, CC_LT, CC_LO)
+#define asm_min(as, ir) asm_min_max(as, ir, CC_LT, CC_PL)
+#define asm_max(as, ir) asm_min_max(as, ir, CC_GT, CC_LE)
/* -- Comparisons --------------------------------------------------------- */
@@ -1652,8 +1652,9 @@
Reg left;
uint32_t m;
int cmpprev0 = 0;
- lua_assert(irt_is64(ir->t) || irt_isint(ir->t) ||
- irt_isu32(ir->t) || irt_isaddr(ir->t) || irt_isu8(ir->t));
+ lj_assertA(irt_is64(ir->t) || irt_isint(ir->t) ||
+ irt_isu32(ir->t) || irt_isaddr(ir->t) || irt_isu8(ir->t),
+ "bad comparison data type %d", irt_type(ir->t));
if (asm_swapops(as, lref, rref)) {
IRRef tmp = lref; lref = rref; rref = tmp;
if (cc >= CC_GE) cc ^= 7; /* LT <-> GT, LE <-> GE */
@@ -1660,7 +1661,7 @@
else if (cc > CC_NE) cc ^= 11; /* LO <-> HI, LS <-> HS */
}
oldcc = cc;
- if (irref_isk(rref) && get_k64val(IR(rref)) == 0) {
+ if (irref_isk(rref) && get_k64val(as, rref) == 0) {
IRIns *irl = IR(lref);
if (cc == CC_GE) cc = CC_PL;
else if (cc == CC_LT) cc = CC_MI;
@@ -1674,16 +1675,15 @@
if (asm_swapops(as, blref, brref)) {
Reg tmp = blref; blref = brref; brref = tmp;
}
+ bleft = ra_alloc1(as, blref, RSET_GPR);
if (irref_isk(brref)) {
- uint64_t k = get_k64val(IR(brref));
- if (k && !(k & (k-1)) && (cc == CC_EQ || cc == CC_NE)) {
- asm_guardtnb(as, cc == CC_EQ ? A64I_TBZ : A64I_TBNZ,
- ra_alloc1(as, blref, RSET_GPR), emit_ctz64(k));
+ uint64_t k = get_k64val(as, brref);
+ if (k && !(k & (k-1)) && (cc == CC_EQ || cc == CC_NE) &&
+ asm_guardtnb(as, cc == CC_EQ ? A64I_TBZ : A64I_TBNZ, bleft,
+ emit_ctz64(k)))
return;
- }
m2 = emit_isk13(k, irt_is64(irl->t));
}
- bleft = ra_alloc1(as, blref, RSET_GPR);
ai = (irt_is64(irl->t) ? A64I_TSTx : A64I_TSTw);
if (!m2)
m2 = asm_fuseopm(as, ai, brref, rset_exclude(RSET_GPR, bleft));
@@ -1719,12 +1719,25 @@
#define asm_equal(as, ir) asm_comp(as, ir)
-/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
+/* -- Split register ops -------------------------------------------------- */
-/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
+/* Hiword op of a split 64/64 bit op. Previous op is the loword op. */
static void asm_hiop(ASMState *as, IRIns *ir)
{
- UNUSED(as); UNUSED(ir); lua_assert(0); /* Unused on 64 bit. */
+ /* HIOP is marked as a store because it needs its own DCE logic. */
+ int uselo = ra_used(ir-1), usehi = ra_used(ir); /* Loword/hiword used? */
+ if (LJ_UNLIKELY(!(as->flags & JIT_F_OPT_DCE))) uselo = usehi = 1;
+ if (!usehi) return; /* Skip unused hiword op for all remaining ops. */
+ switch ((ir-1)->o) {
+ case IR_CALLN:
+ case IR_CALLL:
+ case IR_CALLS:
+ case IR_CALLXS:
+ if (!uselo)
+ ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
+ break;
+ default: lj_assertA(0, "bad HIOP for op %d", (ir-1)->o); break;
+ }
}
/* -- Profiling ----------------------------------------------------------- */
@@ -1732,7 +1745,7 @@
static void asm_prof(ASMState *as, IRIns *ir)
{
uint32_t k = emit_isk13(HOOK_PROFILE, 0);
- lua_assert(k != 0);
+ lj_assertA(k != 0, "HOOK_PROFILE does not fit in K13");
UNUSED(ir);
asm_guardcc(as, CC_NE);
emit_n(as, A64I_TSTw^k, RID_TMP);
@@ -1745,37 +1758,28 @@
static void asm_stack_check(ASMState *as, BCReg topslot,
IRIns *irp, RegSet allow, ExitNo exitno)
{
- Reg pbase;
uint32_t k;
+ Reg pbase = RID_BASE;
if (irp) {
- if (!ra_hasspill(irp->s)) {
- pbase = irp->r;
- lua_assert(ra_hasreg(pbase));
- } else if (allow) {
- pbase = rset_pickbot(allow);
- } else {
- pbase = RID_RET;
- emit_lso(as, A64I_LDRx, RID_RET, RID_SP, 0); /* Restore temp register. */
- }
- } else {
- pbase = RID_BASE;
+ pbase = irp->r;
+ if (!ra_hasreg(pbase))
+ pbase = allow ? (0x40 | rset_pickbot(allow)) : (0xC0 | RID_RET);
}
emit_cond_branch(as, CC_LS, asm_exitstub_addr(as, exitno));
+ if (pbase & 0x80) /* Restore temp. register. */
+ emit_lso(as, A64I_LDRx, (pbase & 31), RID_SP, 0);
k = emit_isk12((8*topslot));
- lua_assert(k);
+ lj_assertA(k, "slot offset %d does not fit in K12", 8*topslot);
emit_n(as, A64I_CMPx^k, RID_TMP);
- emit_dnm(as, A64I_SUBx, RID_TMP, RID_TMP, pbase);
+ emit_dnm(as, A64I_SUBx, RID_TMP, RID_TMP, (pbase & 31));
emit_lso(as, A64I_LDRx, RID_TMP, RID_TMP,
(int32_t)offsetof(lua_State, maxstack));
- if (irp) { /* Must not spill arbitrary registers in head of side trace. */
- if (ra_hasspill(irp->s))
- emit_lso(as, A64I_LDRx, pbase, RID_SP, sps_scale(irp->s));
- emit_lso(as, A64I_LDRx, RID_TMP, RID_GL, glofs(as, &J2G(as->J)->cur_L));
- if (ra_hasspill(irp->s) && !allow)
- emit_lso(as, A64I_STRx, RID_RET, RID_SP, 0); /* Save temp register. */
- } else {
- emit_getgl(as, RID_TMP, cur_L);
+ if (pbase & 0x40) {
+ emit_getgl(as, (pbase & 31), jit_base);
+ if (pbase & 0x80) /* Save temp register. */
+ emit_lso(as, A64I_STRx, (pbase & 31), RID_SP, 0);
}
+ emit_getgl(as, RID_TMP, cur_L);
}
/* Restore Lua stack from on-trace state. */
@@ -1795,7 +1799,14 @@
IRIns *ir = IR(ref);
if ((sn & SNAP_NORESTORE))
continue;
- if (irt_isnum(ir->t)) {
+ if ((sn & SNAP_KEYINDEX)) {
+ RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
+ Reg r = irref_isk(ref) ? ra_allock(as, ir->i, allow) :
+ ra_alloc1(as, ref, allow);
+ rset_clear(allow, r);
+ emit_lso(as, A64I_STRw, r, RID_BASE, ofs);
+ emit_lso(as, A64I_STRw, ra_allock(as, LJ_KEYINDEX, allow), RID_BASE, ofs+4);
+ } else if (irt_isnum(ir->t)) {
Reg src = ra_alloc1(as, ref, RSET_FPR);
emit_lso(as, A64I_STRd, (src & 31), RID_BASE, ofs);
} else {
@@ -1803,11 +1814,15 @@
}
checkmclim(as);
}
- lua_assert(map + nent == flinks);
+ lj_assertA(map + nent == flinks, "inconsistent frames in snapshot");
}
/* -- GC handling --------------------------------------------------------- */
+/* Marker to prevent patching the GC check exit. */
+#define ARM64_NOPATCH_GC_CHECK \
+ (A64I_ORRx|A64F_D(RID_ZERO)|A64F_M(RID_ZERO)|A64F_N(RID_ZERO))
+
/* Check GC threshold and do one or more GC steps. */
static void asm_gc_check(ASMState *as)
{
@@ -1814,25 +1829,23 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_gc_step_jit];
IRRef args[2];
MCLabel l_end;
- Reg tmp1, tmp2;
+ Reg tmp2;
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
/* Exit trace if in GCSatomic or GCSfinalize. Avoids syncing GC objects. */
asm_guardcnb(as, A64I_CBNZ, RID_RET); /* Assumes asm_snap_prep() is done. */
+ *--as->mcp = ARM64_NOPATCH_GC_CHECK;
args[0] = ASMREF_TMP1; /* global_State *g */
args[1] = ASMREF_TMP2; /* MSize steps */
asm_gencall(as, ci, args);
- tmp1 = ra_releasetmp(as, ASMREF_TMP1);
+ emit_dm(as, A64I_MOVx, ra_releasetmp(as, ASMREF_TMP1), RID_GL);
tmp2 = ra_releasetmp(as, ASMREF_TMP2);
emit_loadi(as, tmp2, as->gcsteps);
/* Jump around GC step if GC total < GC threshold. */
emit_cond_branch(as, CC_LS, l_end);
emit_nm(as, A64I_CMPx, RID_TMP, tmp2);
- emit_lso(as, A64I_LDRx, tmp2, tmp1,
- (int32_t)offsetof(global_State, gc.threshold));
- emit_lso(as, A64I_LDRx, RID_TMP, tmp1,
- (int32_t)offsetof(global_State, gc.total));
- ra_allockreg(as, i64ptr(J2G(as->J)), tmp1);
+ emit_getgl(as, tmp2, gc.threshold);
+ emit_getgl(as, RID_TMP, gc.total);
as->gcsteps = 0;
checkmclim(as);
}
@@ -1851,53 +1864,52 @@
p[-2] |= ((uint32_t)delta & mask) << 5;
} else {
ptrdiff_t delta = target - (p - 1);
- p[-1] = A64I_B | ((uint32_t)(delta) & 0x03ffffffu);
+ p[-1] = A64I_B | A64F_S26(delta);
}
}
-/* -- Head of trace ------------------------------------------------------- */
-
-/* Reload L register from g->cur_L. */
-static void asm_head_lreg(ASMState *as)
+/* Fixup the tail of the loop. */
+static void asm_loop_tail_fixup(ASMState *as)
{
- IRIns *ir = IR(ASMREF_L);
- if (ra_used(ir)) {
- Reg r = ra_dest(as, ir, RSET_GPR);
- emit_getgl(as, r, cur_L);
- ra_evictk(as);
- }
+ UNUSED(as); /* Nothing to do. */
}
+/* -- Head of trace ------------------------------------------------------- */
+
/* Coalesce BASE register for a root trace. */
static void asm_head_root_base(ASMState *as)
{
- IRIns *ir;
- asm_head_lreg(as);
- ir = IR(REF_BASE);
- if (ra_hasreg(ir->r) && (rset_test(as->modset, ir->r) || irt_ismarked(ir->t)))
- ra_spill(as, ir);
- ra_destreg(as, ir, RID_BASE);
+ IRIns *ir = IR(REF_BASE);
+ Reg r = ir->r;
+ if (ra_hasreg(r)) {
+ ra_free(as, r);
+ if (rset_test(as->modset, r) || irt_ismarked(ir->t))
+ ir->r = RID_INIT; /* No inheritance for modified BASE register. */
+ if (r != RID_BASE)
+ emit_movrr(as, ir, r, RID_BASE);
+ }
}
/* Coalesce BASE register for a side trace. */
-static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
+static Reg asm_head_side_base(ASMState *as, IRIns *irp)
{
- IRIns *ir;
- asm_head_lreg(as);
- ir = IR(REF_BASE);
- if (ra_hasreg(ir->r) && (rset_test(as->modset, ir->r) || irt_ismarked(ir->t)))
- ra_spill(as, ir);
- if (ra_hasspill(irp->s)) {
- rset_clear(allow, ra_dest(as, ir, allow));
- } else {
- Reg r = irp->r;
- lua_assert(ra_hasreg(r));
- rset_clear(allow, r);
- if (r != ir->r && !rset_test(as->freeset, r))
- ra_restore(as, regcost_ref(as->cost[r]));
- ra_destreg(as, ir, r);
+ IRIns *ir = IR(REF_BASE);
+ Reg r = ir->r;
+ if (ra_hasreg(r)) {
+ ra_free(as, r);
+ if (rset_test(as->modset, r) || irt_ismarked(ir->t))
+ ir->r = RID_INIT; /* No inheritance for modified BASE register. */
+ if (irp->r == r) {
+ return r; /* Same BASE register already coalesced. */
+ } else if (ra_hasreg(irp->r) && rset_test(as->freeset, irp->r)) {
+ /* Move from coalesced parent reg. */
+ emit_movrr(as, ir, r, irp->r);
+ return irp->r;
+ } else {
+ emit_getgl(as, r, jit_base); /* Otherwise reload BASE. */
+ }
}
- return allow;
+ return RID_NONE;
}
/* -- Tail of trace ------------------------------------------------------- */
@@ -1915,12 +1927,12 @@
} else {
/* Patch stack adjustment. */
uint32_t k = emit_isk12(spadj);
- lua_assert(k);
+ lj_assertA(k, "stack adjustment %d does not fit in K12", spadj);
p[-2] = (A64I_ADDx^k) | A64F_D(RID_SP) | A64F_N(RID_SP);
}
/* Patch exit branch. */
target = lnk ? traceref(as->J, lnk)->mcode : (MCode *)lj_vm_exit_interp;
- p[-1] = A64I_B | (((target-p)+1)&0x03ffffffu);
+ p[-1] = A64I_B | A64F_S26((target-p)+1);
}
/* Prepare tail of code. */
@@ -1941,20 +1953,47 @@
/* Ensure there are enough stack slots for call arguments. */
static Reg asm_setup_call_slots(ASMState *as, IRIns *ir, const CCallInfo *ci)
{
- IRRef args[CCI_NARGS_MAX*2];
+#if LJ_HASFFI
uint32_t i, nargs = CCI_XNARGS(ci);
- int nslots = 0, ngpr = REGARG_NUMGPR, nfpr = REGARG_NUMFPR;
- asm_collectargs(as, ir, ci, args);
- for (i = 0; i < nargs; i++) {
- if (args[i] && irt_isfp(IR(args[i])->t)) {
- if (nfpr > 0) nfpr--; else nslots += 2;
- } else {
- if (ngpr > 0) ngpr--; else nslots += 2;
+ if (nargs > (REGARG_NUMGPR < REGARG_NUMFPR ? REGARG_NUMGPR : REGARG_NUMFPR) ||
+ (LJ_TARGET_OSX && (ci->flags & CCI_VARARG))) {
+ IRRef args[CCI_NARGS_MAX*2];
+ int ngpr = REGARG_NUMGPR, nfpr = REGARG_NUMFPR;
+ int spofs = 0, spalign = LJ_TARGET_OSX ? 0 : 7, nslots;
+ asm_collectargs(as, ir, ci, args);
+#if LJ_ABI_WIN
+ if ((ci->flags & CCI_VARARG)) nfpr = 0;
+#endif
+ for (i = 0; i < nargs; i++) {
+ int al = spalign;
+ if (!args[i]) {
+#if LJ_TARGET_OSX
+ /* Marker for start of varaargs. */
+ nfpr = 0;
+ ngpr = 0;
+ spalign = 7;
+#endif
+ } else if (irt_isfp(IR(args[i])->t)) {
+ if (nfpr > 0) { nfpr--; continue; }
+#if LJ_ABI_WIN
+ if ((ci->flags & CCI_VARARG) && ngpr > 0) { ngpr--; continue; }
+#elif LJ_TARGET_OSX
+ al |= irt_isnum(IR(args[i])->t) ? 7 : 3;
+#endif
+ } else {
+ if (ngpr > 0) { ngpr--; continue; }
+#if LJ_TARGET_OSX
+ al |= irt_size(IR(args[i])->t) - 1;
+#endif
+ }
+ spofs = (spofs + 2*al+1) & ~al; /* Align and bump stack pointer. */
}
+ nslots = (spofs + 3) >> 2;
+ if (nslots > as->evenspill) /* Leave room for args in stack slots. */
+ as->evenspill = nslots;
}
- if (nslots > as->evenspill) /* Leave room for args in stack slots. */
- as->evenspill = nslots;
- return REGSP_HINT(RID_RET);
+#endif
+ return REGSP_HINT(irt_isfp(ir->t) ? RID_FPRET : RID_RET);
}
static void asm_setup_target(ASMState *as)
@@ -1983,40 +2022,54 @@
{
MCode *p = T->mcode;
MCode *pe = (MCode *)((char *)p + T->szmcode);
- MCode *cstart = NULL, *cend = p;
+ MCode *cstart = NULL;
MCode *mcarea = lj_mcode_patch(J, p, 0);
MCode *px = exitstub_trace_addr(T, exitno);
+ int patchlong = 1;
+ /* Note: this assumes a trace exit is only ever patched once. */
for (; p < pe; p++) {
/* Look for exitstub branch, replace with branch to target. */
+ ptrdiff_t delta = target - p;
MCode ins = A64I_LE(*p);
if ((ins & 0xff000000u) == 0x54000000u &&
((ins ^ ((px-p)<<5)) & 0x00ffffe0u) == 0) {
- /* Patch bcc exitstub. */
- *p = A64I_LE((ins & 0xff00001fu) | (((target-p)<<5) & 0x00ffffe0u));
- cend = p+1;
- if (!cstart) cstart = p;
+ /* Patch bcc, if within range. */
+ if (A64F_S_OK(delta, 19)) {
+ *p = A64I_LE((ins & 0xff00001fu) | A64F_S19(delta));
+ if (!cstart) cstart = p;
+ }
} else if ((ins & 0xfc000000u) == 0x14000000u &&
((ins ^ (px-p)) & 0x03ffffffu) == 0) {
- /* Patch b exitstub. */
- *p = A64I_LE((ins & 0xfc000000u) | ((target-p) & 0x03ffffffu));
- cend = p+1;
+ /* Patch b. */
+ lj_assertJ(A64F_S_OK(delta, 26), "branch target out of range");
+ *p = A64I_LE((ins & 0xfc000000u) | A64F_S26(delta));
if (!cstart) cstart = p;
} else if ((ins & 0x7e000000u) == 0x34000000u &&
((ins ^ ((px-p)<<5)) & 0x00ffffe0u) == 0) {
- /* Patch cbz/cbnz exitstub. */
- *p = A64I_LE((ins & 0xff00001f) | (((target-p)<<5) & 0x00ffffe0u));
- cend = p+1;
- if (!cstart) cstart = p;
+ /* Patch cbz/cbnz, if within range. */
+ if (p[-1] == ARM64_NOPATCH_GC_CHECK) {
+ patchlong = 0;
+ } else if (A64F_S_OK(delta, 19)) {
+ *p = A64I_LE((ins & 0xff00001fu) | A64F_S19(delta));
+ if (!cstart) cstart = p;
+ }
} else if ((ins & 0x7e000000u) == 0x36000000u &&
((ins ^ ((px-p)<<5)) & 0x0007ffe0u) == 0) {
- /* Patch tbz/tbnz exitstub. */
- *p = A64I_LE((ins & 0xfff8001fu) | (((target-p)<<5) & 0x0007ffe0u));
- cend = p+1;
- if (!cstart) cstart = p;
+ /* Patch tbz/tbnz, if within range. */
+ if (A64F_S_OK(delta, 14)) {
+ *p = A64I_LE((ins & 0xfff8001fu) | A64F_S14(delta));
+ if (!cstart) cstart = p;
+ }
}
}
- lua_assert(cstart != NULL);
- lj_mcode_sync(cstart, cend);
+ /* Always patch long-range branch in exit stub itself. Except, if we can't. */
+ if (patchlong) {
+ ptrdiff_t delta = target - px;
+ lj_assertJ(A64F_S_OK(delta, 26), "branch target out of range");
+ *px = A64I_B | A64F_S26(delta);
+ if (!cstart) cstart = px;
+ }
+ if (cstart) lj_mcode_sync(cstart, px+1);
lj_mcode_patch(J, mcarea, 1);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_mips.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_mips.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_mips.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** MIPS IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Register allocator extensions --------------------------------------- */
@@ -23,7 +23,7 @@
{
Reg r = IR(ref)->r;
if (ra_noreg(r)) {
- if (!(allow & RSET_FPR) && irref_isk(ref) && get_kval(IR(ref)) == 0)
+ if (!(allow & RSET_FPR) && irref_isk(ref) && get_kval(as, ref) == 0)
return RID_ZERO;
r = ra_allocref(as, ref, allow);
} else {
@@ -64,20 +64,31 @@
/* Setup spare long-range jump slots per mcarea. */
static void asm_sparejump_setup(ASMState *as)
{
- MCode *mxp = as->mcbot;
- /* Assumes sizeof(MCLink) == 8. */
- if (((uintptr_t)mxp & (LJ_PAGESIZE-1)) == 8) {
- lua_assert(MIPSI_NOP == 0);
- memset(mxp+2, 0, MIPS_SPAREJUMP*8);
- mxp += MIPS_SPAREJUMP*2;
- lua_assert(mxp < as->mctop);
- lj_mcode_sync(as->mcbot, mxp);
- lj_mcode_commitbot(as->J, mxp);
- as->mcbot = mxp;
- as->mclim = as->mcbot + MCLIM_REDZONE;
+ MCode *mxp = as->mctop;
+ if ((char *)mxp == (char *)as->J->mcarea + as->J->szmcarea) {
+ mxp -= MIPS_SPAREJUMP*2;
+ lj_assertA(MIPSI_NOP == 0, "bad NOP");
+ memset(mxp, 0, MIPS_SPAREJUMP*2*sizeof(MCode));
+ as->mctop = mxp;
}
}
+static MCode *asm_sparejump_use(MCode *mcarea, MCode tjump)
+{
+ MCode *mxp = (MCode *)((char *)mcarea + ((MCLink *)mcarea)->size);
+ int slot = MIPS_SPAREJUMP;
+ while (slot--) {
+ mxp -= 2;
+ if (*mxp == tjump) {
+ return mxp;
+ } else if (*mxp == MIPSI_NOP) {
+ *mxp = tjump;
+ return mxp;
+ }
+ }
+ return NULL;
+}
+
/* Setup exit stub after the end of each trace. */
static void asm_exitstub_setup(ASMState *as)
{
@@ -85,7 +96,8 @@
/* sw TMP, 0(sp); j ->vm_exit_handler; li TMP, traceno */
*--mxp = MIPSI_LI|MIPSF_T(RID_TMP)|as->T->traceno;
*--mxp = MIPSI_J|((((uintptr_t)(void *)lj_vm_exit_handler)>>2)&0x03ffffffu);
- lua_assert(((uintptr_t)mxp ^ (uintptr_t)(void *)lj_vm_exit_handler)>>28 == 0);
+ lj_assertA(((uintptr_t)mxp ^ (uintptr_t)(void *)lj_vm_exit_handler)>>28 == 0,
+ "branch target out of range");
*--mxp = MIPSI_SW|MIPSF_T(RID_TMP)|MIPSF_S(RID_SP)|0;
as->mctop = mxp;
}
@@ -102,7 +114,12 @@
as->invmcp = NULL;
as->loopinv = 1;
as->mcp = p+1;
+#if !LJ_TARGET_MIPSR6
mi = mi ^ ((mi>>28) == 1 ? 0x04000000u : 0x00010000u); /* Invert cond. */
+#else
+ mi = mi ^ ((mi>>28) == 1 ? 0x04000000u :
+ (mi>>28) == 4 ? 0x00800000u : 0x00010000u); /* Invert cond. */
+#endif
target = p; /* Patch target later in asm_loop_fixup. */
}
emit_ti(as, MIPSI_LI, RID_TMP, as->snapno);
@@ -176,6 +193,9 @@
return ra_allock(as, ofs-(int16_t)ofs, allow);
}
}
+ } else if (ir->o == IR_TMPREF) {
+ *ofsp = (int32_t)(offsetof(global_State, tmptv)-32768);
+ return RID_JGL;
}
}
*ofsp = 0;
@@ -191,7 +211,7 @@
if (ra_noreg(ir->r) && canfuse(as, ir)) {
if (ir->o == IR_ADD) {
intptr_t ofs2;
- if (irref_isk(ir->op2) && (ofs2 = ofs + get_kval(IR(ir->op2)),
+ if (irref_isk(ir->op2) && (ofs2 = ofs + get_kval(as, ir->op2),
checki16(ofs2))) {
ref = ir->op1;
ofs = (int32_t)ofs2;
@@ -198,13 +218,13 @@
}
} else if (ir->o == IR_STRREF) {
intptr_t ofs2 = 65536;
- lua_assert(ofs == 0);
+ lj_assertA(ofs == 0, "bad usage");
ofs = (int32_t)sizeof(GCstr);
if (irref_isk(ir->op2)) {
- ofs2 = ofs + get_kval(IR(ir->op2));
+ ofs2 = ofs + get_kval(as, ir->op2);
ref = ir->op1;
} else if (irref_isk(ir->op1)) {
- ofs2 = ofs + get_kval(IR(ir->op1));
+ ofs2 = ofs + get_kval(as, ir->op1);
ref = ir->op2;
}
if (!checki16(ofs2)) {
@@ -248,7 +268,8 @@
#if !LJ_SOFTFP
if (irt_isfp(ir->t) && fpr <= REGARG_LASTFPR &&
!(ci->flags & CCI_VARARG)) {
- lua_assert(rset_test(as->freeset, fpr)); /* Already evicted. */
+ lj_assertA(rset_test(as->freeset, fpr),
+ "reg %d not free", fpr); /* Already evicted. */
ra_leftov(as, fpr, ref);
fpr += LJ_32 ? 2 : 1;
gpr += (LJ_32 && irt_isnum(ir->t)) ? 2 : 1;
@@ -260,7 +281,8 @@
#endif
if (LJ_32 && irt_isnum(ir->t)) gpr = (gpr+1) & ~1;
if (gpr <= REGARG_LASTGPR) {
- lua_assert(rset_test(as->freeset, gpr)); /* Already evicted. */
+ lj_assertA(rset_test(as->freeset, gpr),
+ "reg %d not free", gpr); /* Already evicted. */
#if !LJ_SOFTFP
if (irt_isfp(ir->t)) {
RegSet of = as->freeset;
@@ -273,7 +295,8 @@
#if LJ_32
emit_tg(as, MIPSI_MFC1, gpr+(LJ_BE?0:1), r+1);
emit_tg(as, MIPSI_MFC1, gpr+(LJ_BE?1:0), r);
- lua_assert(rset_test(as->freeset, gpr+1)); /* Already evicted. */
+ lj_assertA(rset_test(as->freeset, gpr+1),
+ "reg %d not free", gpr+1); /* Already evicted. */
gpr += 2;
#else
emit_tg(as, MIPSI_DMFC1, gpr, r);
@@ -291,7 +314,7 @@
{
ra_leftov(as, gpr, ref);
gpr++;
-#if LJ_64
+#if LJ_64 && !LJ_SOFTFP
fpr++;
#endif
}
@@ -302,7 +325,7 @@
emit_spstore(as, ir, r, ofs);
ofs += irt_isnum(ir->t) ? 8 : 4;
#else
- emit_spstore(as, ir, r, ofs + ((LJ_BE && (LJ_SOFTFP || r < RID_MAX_GPR) && !irt_is64(ir->t)) ? 4 : 0));
+ emit_spstore(as, ir, r, ofs + ((LJ_BE && !irt_isfp(ir->t) && !irt_is64(ir->t)) ? 4 : 0));
ofs += 8;
#endif
}
@@ -313,7 +336,7 @@
#endif
if (gpr <= REGARG_LASTGPR) {
gpr++;
-#if LJ_64
+#if LJ_64 && !LJ_SOFTFP
fpr++;
#endif
} else {
@@ -328,9 +351,7 @@
static void asm_setupresult(ASMState *as, IRIns *ir, const CCallInfo *ci)
{
RegSet drop = RSET_SCRATCH;
-#if LJ_32
int hiop = ((ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t));
-#endif
#if !LJ_SOFTFP
if ((ci->flags & CCI_NOFPRCLOBBER))
drop &= ~RSET_FPR;
@@ -337,13 +358,11 @@
#endif
if (ra_hasreg(ir->r))
rset_clear(drop, ir->r); /* Dest reg handled below. */
-#if LJ_32
if (hiop && ra_hasreg((ir+1)->r))
rset_clear(drop, (ir+1)->r); /* Dest reg handled below. */
-#endif
ra_evictset(as, drop); /* Evictions must be performed first. */
if (ra_used(ir)) {
- lua_assert(!irt_ispri(ir->t));
+ lj_assertA(!irt_ispri(ir->t), "PRI dest");
if (!LJ_SOFTFP && irt_isfp(ir->t)) {
if ((ci->flags & CCI_CASTU64)) {
int32_t ofs = sps_scale(ir->s);
@@ -369,10 +388,8 @@
} else {
ra_destreg(as, ir, RID_FPRET);
}
-#if LJ_32
} else if (hiop) {
ra_destpair(as, ir);
-#endif
} else {
ra_destreg(as, ir, RID_RET);
}
@@ -391,7 +408,7 @@
func = ir->op2; irf = IR(func);
if (irf->o == IR_CARG) { func = irf->op1; irf = IR(func); }
if (irref_isk(func)) { /* Call to constant address. */
- ci.func = (ASMFunction)(void *)get_kval(irf);
+ ci.func = (ASMFunction)(void *)get_kval(as, func);
} else { /* Need specific register for indirect calls. */
Reg r = ra_alloc1(as, func, RID2RSET(RID_CFUNCADDR));
MCode *p = as->mcp;
@@ -411,7 +428,11 @@
{
/* The modified regs must match with the *.dasc implementation. */
RegSet drop = RID2RSET(RID_R1)|RID2RSET(RID_R12)|RID2RSET(RID_FPRET)|
- RID2RSET(RID_F2)|RID2RSET(RID_F4)|RID2RSET(REGARG_FIRSTFPR);
+ RID2RSET(RID_F2)|RID2RSET(RID_F4)|RID2RSET(REGARG_FIRSTFPR)
+#if LJ_TARGET_MIPSR6
+ |RID2RSET(RID_F21)
+#endif
+ ;
if (ra_hasreg(ir->r)) rset_clear(drop, ir->r);
ra_evictset(as, drop);
ra_destreg(as, ir, RID_FPRET);
@@ -435,9 +456,30 @@
emit_addptr(as, base, -8*delta);
asm_guard(as, MIPSI_BNE, RID_TMP,
ra_allock(as, igcptr(pc), rset_exclude(RSET_GPR, base)));
- emit_tsi(as, MIPSI_AL, RID_TMP, base, -8);
+ emit_tsi(as, MIPSI_AL, RID_TMP, base, (LJ_BE || LJ_FR2) ? -8 : -4);
}
+/* -- Buffer operations --------------------------------------------------- */
+
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb)
+{
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irgc;
+ irgc.ot = IRT(0, IRT_PGC); /* GC type. */
+ emit_storeofs(as, &irgc, RID_TMP, sb, offsetof(SBuf, L));
+ if ((as->flags & JIT_F_MIPSXXR2)) {
+ emit_tsml(as, LJ_64 ? MIPSI_DINS : MIPSI_INS, RID_TMP, tmp,
+ lj_fls(SBUF_MASK_FLAG), 0);
+ } else {
+ emit_dst(as, MIPSI_OR, RID_TMP, RID_TMP, tmp);
+ emit_tsi(as, MIPSI_ANDI, tmp, tmp, SBUF_MASK_FLAG);
+ }
+ emit_getgl(as, RID_TMP, cur_L);
+ emit_loadofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+}
+#endif
+
/* -- Type conversions ---------------------------------------------------- */
#if !LJ_SOFTFP
@@ -445,8 +487,13 @@
{
Reg tmp = ra_scratch(as, rset_exclude(RSET_FPR, left));
Reg dest = ra_dest(as, ir, RSET_GPR);
+#if !LJ_TARGET_MIPSR6
asm_guard(as, MIPSI_BC1F, 0, 0);
emit_fgh(as, MIPSI_C_EQ_D, 0, tmp, left);
+#else
+ asm_guard(as, MIPSI_BC1EQZ, 0, (tmp&31));
+ emit_fgh(as, MIPSI_CMP_EQ_D, tmp, tmp, left);
+#endif
emit_fg(as, MIPSI_CVT_D_W, tmp, tmp);
emit_tg(as, MIPSI_MFC1, dest, tmp);
emit_fg(as, MIPSI_CVT_W_D, tmp, left);
@@ -462,12 +509,36 @@
emit_tg(as, MIPSI_MFC1, dest, tmp);
emit_fgh(as, MIPSI_ADD_D, tmp, left, right);
}
+#elif LJ_64 /* && LJ_SOFTFP */
+static void asm_tointg(ASMState *as, IRIns *ir, Reg r)
+{
+ /* The modified regs must match with the *.dasc implementation. */
+ RegSet drop = RID2RSET(REGARG_FIRSTGPR)|RID2RSET(RID_RET)|RID2RSET(RID_RET+1)|
+ RID2RSET(RID_R1)|RID2RSET(RID_R12);
+ if (ra_hasreg(ir->r)) rset_clear(drop, ir->r);
+ ra_evictset(as, drop);
+ /* Return values are in RID_RET (converted value) and RID_RET+1 (status). */
+ ra_destreg(as, ir, RID_RET);
+ asm_guard(as, MIPSI_BNE, RID_RET+1, RID_ZERO);
+ emit_call(as, (void *)lj_ir_callinfo[IRCALL_lj_vm_tointg].func, 0);
+ if (r == RID_NONE)
+ ra_leftov(as, REGARG_FIRSTGPR, ir->op1);
+ else if (r != REGARG_FIRSTGPR)
+ emit_move(as, REGARG_FIRSTGPR, r);
+}
+
+static void asm_tobit(ASMState *as, IRIns *ir)
+{
+ Reg dest = ra_dest(as, ir, RSET_GPR);
+ emit_dta(as, MIPSI_SLL, dest, dest, 0);
+ asm_callid(as, ir, IRCALL_lj_vm_tobit);
+}
#endif
static void asm_conv(ASMState *as, IRIns *ir)
{
IRType st = (IRType)(ir->op2 & IRCONV_SRCMASK);
-#if !LJ_SOFTFP
+#if !LJ_SOFTFP32
int stfp = (st == IRT_NUM || st == IRT_FLOAT);
#endif
#if LJ_64
@@ -475,15 +546,20 @@
#endif
IRRef lref = ir->op1;
#if LJ_32
- lua_assert(!(irt_isint64(ir->t) ||
- (st == IRT_I64 || st == IRT_U64))); /* Handled by SPLIT. */
+ /* 64 bit integer conversions are handled by SPLIT. */
+ lj_assertA(!(irt_isint64(ir->t) || (st == IRT_I64 || st == IRT_U64)),
+ "IR %04d has unsplit 64 bit type",
+ (int)(ir - as->ir) - REF_BIAS);
#endif
-#if LJ_32 && LJ_SOFTFP
+#if LJ_SOFTFP32
/* FP conversions are handled by SPLIT. */
- lua_assert(!irt_isfp(ir->t) && !(st == IRT_NUM || st == IRT_FLOAT));
+ lj_assertA(!irt_isfp(ir->t) && !(st == IRT_NUM || st == IRT_FLOAT),
+ "IR %04d has FP type",
+ (int)(ir - as->ir) - REF_BIAS);
/* Can't check for same types: SPLIT uses CONV int.int + BXOR for sfp NEG. */
#else
- lua_assert(irt_type(ir->t) != st);
+ lj_assertA(irt_type(ir->t) != st, "inconsistent types for CONV");
+#if !LJ_SOFTFP
if (irt_isfp(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
if (stfp) { /* FP to FP conversion. */
@@ -541,7 +617,8 @@
} else if (stfp) { /* FP to integer conversion. */
if (irt_isguard(ir->t)) {
/* Checked conversions are only supported from number to int. */
- lua_assert(irt_isint(ir->t) && st == IRT_NUM);
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -575,8 +652,13 @@
(void *)&as->J->k64[LJ_K64_M2P64],
rset_exclude(RSET_GPR, dest));
emit_fg(as, MIPSI_TRUNC_L_D, tmp, left); /* Delay slot. */
- emit_branch(as, MIPSI_BC1T, 0, 0, l_end);
- emit_fgh(as, MIPSI_C_OLT_D, 0, left, tmp);
+#if !LJ_TARGET_MIPSR6
+ emit_branch(as, MIPSI_BC1T, 0, 0, l_end);
+ emit_fgh(as, MIPSI_C_OLT_D, 0, left, tmp);
+#else
+ emit_branch(as, MIPSI_BC1NEZ, 0, (tmp&31), l_end);
+ emit_fgh(as, MIPSI_CMP_LT_D, tmp, left, tmp);
+#endif
emit_lsptr(as, MIPSI_LDC1, (tmp & 31),
(void *)&as->J->k64[LJ_K64_2P63],
rset_exclude(RSET_GPR, dest));
@@ -587,8 +669,13 @@
(void *)&as->J->k32[LJ_K32_M2P64],
rset_exclude(RSET_GPR, dest));
emit_fg(as, MIPSI_TRUNC_L_S, tmp, left); /* Delay slot. */
- emit_branch(as, MIPSI_BC1T, 0, 0, l_end);
- emit_fgh(as, MIPSI_C_OLT_S, 0, left, tmp);
+#if !LJ_TARGET_MIPSR6
+ emit_branch(as, MIPSI_BC1T, 0, 0, l_end);
+ emit_fgh(as, MIPSI_C_OLT_S, 0, left, tmp);
+#else
+ emit_branch(as, MIPSI_BC1NEZ, 0, (tmp&31), l_end);
+ emit_fgh(as, MIPSI_CMP_LT_S, tmp, left, tmp);
+#endif
emit_lsptr(as, MIPSI_LWC1, (tmp & 31),
(void *)&as->J->k32[LJ_K32_2P63],
rset_exclude(RSET_GPR, dest));
@@ -603,18 +690,55 @@
MIPSIns mi = irt_is64(ir->t) ?
(st == IRT_NUM ? MIPSI_TRUNC_L_D : MIPSI_TRUNC_L_S) :
(st == IRT_NUM ? MIPSI_TRUNC_W_D : MIPSI_TRUNC_W_S);
- emit_tg(as, irt_is64(ir->t) ? MIPSI_DMFC1 : MIPSI_MFC1, dest, left);
- emit_fg(as, mi, left, left);
+ emit_tg(as, irt_is64(ir->t) ? MIPSI_DMFC1 : MIPSI_MFC1, dest, tmp);
+ emit_fg(as, mi, tmp, left);
#endif
}
}
} else
+#else
+ if (irt_isfp(ir->t)) {
+#if LJ_64 && LJ_HASFFI
+ if (stfp) { /* FP to FP conversion. */
+ asm_callid(as, ir, irt_isnum(ir->t) ? IRCALL_softfp_f2d :
+ IRCALL_softfp_d2f);
+ } else { /* Integer to FP conversion. */
+ IRCallID cid = ((IRT_IS64 >> st) & 1) ?
+ (irt_isnum(ir->t) ?
+ (st == IRT_I64 ? IRCALL_fp64_l2d : IRCALL_fp64_ul2d) :
+ (st == IRT_I64 ? IRCALL_fp64_l2f : IRCALL_fp64_ul2f)) :
+ (irt_isnum(ir->t) ?
+ (st == IRT_INT ? IRCALL_softfp_i2d : IRCALL_softfp_ui2d) :
+ (st == IRT_INT ? IRCALL_softfp_i2f : IRCALL_softfp_ui2f));
+ asm_callid(as, ir, cid);
+ }
+#else
+ asm_callid(as, ir, IRCALL_softfp_i2d);
#endif
+ } else if (stfp) { /* FP to integer conversion. */
+ if (irt_isguard(ir->t)) {
+ /* Checked conversions are only supported from number to int. */
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
+ asm_tointg(as, ir, RID_NONE);
+ } else {
+ IRCallID cid = irt_is64(ir->t) ?
+ ((st == IRT_NUM) ?
+ (irt_isi64(ir->t) ? IRCALL_fp64_d2l : IRCALL_fp64_d2ul) :
+ (irt_isi64(ir->t) ? IRCALL_fp64_f2l : IRCALL_fp64_f2ul)) :
+ ((st == IRT_NUM) ?
+ (irt_isint(ir->t) ? IRCALL_softfp_d2i : IRCALL_softfp_d2ui) :
+ (irt_isint(ir->t) ? IRCALL_softfp_f2i : IRCALL_softfp_f2ui));
+ asm_callid(as, ir, cid);
+ }
+ } else
+#endif
+#endif
{
Reg dest = ra_dest(as, ir, RSET_GPR);
if (st >= IRT_I8 && st <= IRT_U16) { /* Extend to 32 bit integer. */
Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t), "bad type for CONV EXT");
if ((ir->op2 & IRCONV_SEXT)) {
if (LJ_64 || (as->flags & JIT_F_MIPSXXR2)) {
emit_dst(as, st == IRT_I8 ? MIPSI_SEB : MIPSI_SEH, dest, 0, left);
@@ -645,7 +769,7 @@
}
}
} else {
- if (st64) {
+ if (st64 && !(ir->op2 & IRCONV_NONE)) {
/* This is either a 32 bit reg/reg mov which zeroes the hiword
** or a load of the loword from a 64 bit address.
*/
@@ -666,7 +790,7 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_strscan_num];
IRRef args[2];
int32_t ofs = 0;
-#if LJ_SOFTFP
+#if LJ_SOFTFP32
ra_evictset(as, RSET_SCRATCH);
if (ra_used(ir)) {
if (ra_hasspill(ir->s) && ra_hasspill((ir+1)->s) &&
@@ -711,7 +835,8 @@
{
RegSet allow = rset_exclude(RSET_GPR, base);
IRIns *ir = IR(ref);
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t),
+ "store of IR type %d", irt_type(ir->t));
if (irref_isk(ref)) {
TValue k;
lj_ir_kvalue(as->J->L, &k, ir);
@@ -732,34 +857,63 @@
#endif
/* Get pointer to TValue. */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode)
{
- IRIns *ir = IR(ref);
- if (irt_isnum(ir->t)) {
- if (irref_isk(ref)) /* Use the number constant itself as a TValue. */
- ra_allockreg(as, igcptr(ir_knum(ir)), dest);
- else /* Otherwise force a spill and use the spill slot. */
- emit_tsi(as, MIPSI_AADDIU, dest, RID_SP, ra_spill(as, ir));
- } else {
- /* Otherwise use g->tmptv to hold the TValue. */
+ int32_t tmpofs = (int32_t)(offsetof(global_State, tmptv)-32768);
+ if ((mode & IRTMPREF_IN1)) {
+ IRIns *ir = IR(ref);
+ if (irt_isnum(ir->t)) {
+ if ((mode & IRTMPREF_OUT1)) {
+#if LJ_SOFTFP
+ emit_tsi(as, MIPSI_AADDIU, dest, RID_JGL, tmpofs);
+#if LJ_64
+ emit_setgl(as, ra_alloc1(as, ref, RSET_GPR), tmptv.u64);
+#else
+ lj_assertA(irref_isk(ref), "unsplit FP op");
+ emit_setgl(as,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.lo, RSET_GPR),
+ tmptv.u32.lo);
+ emit_setgl(as,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.hi, RSET_GPR),
+ tmptv.u32.hi);
+#endif
+#else
+ Reg src = ra_alloc1(as, ref, RSET_FPR);
+ emit_tsi(as, MIPSI_AADDIU, dest, RID_JGL, tmpofs);
+ emit_tsi(as, MIPSI_SDC1, (src & 31), RID_JGL, tmpofs);
+#endif
+ } else if (irref_isk(ref)) {
+ /* Use the number constant itself as a TValue. */
+ ra_allockreg(as, igcptr(ir_knum(ir)), dest);
+ } else {
+#if LJ_SOFTFP32
+ lj_assertA(0, "unsplit FP op");
+#else
+ /* Otherwise force a spill and use the spill slot. */
+ emit_tsi(as, MIPSI_AADDIU, dest, RID_SP, ra_spill(as, ir));
+#endif
+ }
+ } else {
+ /* Otherwise use g->tmptv to hold the TValue. */
#if LJ_32
- RegSet allow = rset_exclude(RSET_GPR, dest);
- Reg type;
- emit_tsi(as, MIPSI_ADDIU, dest, RID_JGL, (int32_t)(offsetof(global_State, tmptv)-32768));
- if (!irt_ispri(ir->t)) {
- Reg src = ra_alloc1(as, ref, allow);
- emit_setgl(as, src, tmptv.gcr);
- }
- if (LJ_SOFTFP && (ir+1)->o == IR_HIOP)
- type = ra_alloc1(as, ref+1, allow);
- else
- type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
- emit_setgl(as, type, tmptv.it);
+ Reg type;
+ emit_tsi(as, MIPSI_ADDIU, dest, RID_JGL, tmpofs);
+ if (!irt_ispri(ir->t)) {
+ Reg src = ra_alloc1(as, ref, RSET_GPR);
+ emit_setgl(as, src, tmptv.gcr);
+ }
+ if (LJ_SOFTFP && (ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t))
+ type = ra_alloc1(as, ref+1, RSET_GPR);
+ else
+ type = ra_allock(as, (int32_t)irt_toitype(ir->t), RSET_GPR);
+ emit_setgl(as, type, tmptv.it);
#else
- asm_tvstore64(as, dest, 0, ref);
- emit_tsi(as, MIPSI_DADDIU, dest, RID_JGL,
- (int32_t)(offsetof(global_State, tmptv)-32768));
+ asm_tvstore64(as, dest, 0, ref);
+ emit_tsi(as, MIPSI_DADDIU, dest, RID_JGL, tmpofs);
#endif
+ }
+ } else {
+ emit_tsi(as, MIPSI_AADDIU, dest, RID_JGL, tmpofs);
}
}
@@ -780,8 +934,12 @@
}
base = ra_alloc1(as, ir->op1, RSET_GPR);
idx = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, base));
+#if !LJ_TARGET_MIPSR6
emit_dst(as, MIPSI_AADDU, dest, RID_TMP, base);
emit_dta(as, MIPSI_SLL, RID_TMP, idx, 3);
+#else
+ emit_dst(as, MIPSI_ALSA | MIPSF_A(3-1), dest, idx, base);
+#endif
}
/* Inlined hash lookup. Specialized for key type and for const keys.
@@ -799,6 +957,9 @@
Reg dest = ra_dest(as, ir, allow);
Reg tab = ra_alloc1(as, ir->op1, rset_clear(allow, dest));
Reg key = RID_NONE, type = RID_NONE, tmpnum = RID_NONE, tmp1 = RID_TMP, tmp2;
+#if LJ_64
+ Reg cmp64 = RID_NONE;
+#endif
IRRef refkey = ir->op2;
IRIns *irkey = IR(refkey);
int isk = irref_isk(refkey);
@@ -807,11 +968,16 @@
MCLabel l_end, l_loop, l_next;
rset_clear(allow, tab);
-#if LJ_32 && LJ_SOFTFP
- if (!isk) {
- key = ra_alloc1(as, refkey, allow);
- rset_clear(allow, key);
- if (irkey[1].o == IR_HIOP) {
+ if (!LJ_SOFTFP && irt_isnum(kt)) {
+ key = ra_alloc1(as, refkey, RSET_FPR);
+ tmpnum = ra_scratch(as, rset_exclude(RSET_FPR, key));
+ } else {
+ if (!irt_ispri(kt)) {
+ key = ra_alloc1(as, refkey, allow);
+ rset_clear(allow, key);
+ }
+#if LJ_32
+ if (LJ_SOFTFP && irkey[1].o == IR_HIOP) {
if (ra_hasreg((irkey+1)->r)) {
type = tmpnum = (irkey+1)->r;
tmp1 = ra_scratch(as, allow);
@@ -822,25 +988,33 @@
}
rset_clear(allow, tmpnum);
} else {
- type = ra_allock(as, (int32_t)irt_toitype(irkey->t), allow);
+ type = ra_allock(as, (int32_t)irt_toitype(kt), allow);
rset_clear(allow, type);
}
- }
-#else
- if (irt_isnum(kt)) {
- key = ra_alloc1(as, refkey, RSET_FPR);
- tmpnum = ra_scratch(as, rset_exclude(RSET_FPR, key));
- } else if (!irt_ispri(kt)) {
- key = ra_alloc1(as, refkey, allow);
- rset_clear(allow, key);
-#if LJ_32
- type = ra_allock(as, (int32_t)irt_toitype(irkey->t), allow);
- rset_clear(allow, type);
#endif
}
-#endif
tmp2 = ra_scratch(as, allow);
rset_clear(allow, tmp2);
+#if LJ_64
+ if (LJ_SOFTFP || !irt_isnum(kt)) {
+ /* Allocate cmp64 register used for 64-bit comparisons */
+ if (LJ_SOFTFP && irt_isnum(kt)) {
+ cmp64 = key;
+ } else if (!isk && irt_isaddr(kt)) {
+ cmp64 = tmp2;
+ } else {
+ int64_t k;
+ if (isk && irt_isaddr(kt)) {
+ k = ((int64_t)irt_toitype(kt) << 47) | irkey[1].tv.u64;
+ } else {
+ lj_assertA(irt_ispri(kt) && !irt_isnil(kt), "bad HREF key type");
+ k = ~((int64_t)~irt_toitype(kt) << 47);
+ }
+ cmp64 = ra_allock(as, k, allow);
+ rset_clear(allow, cmp64);
+ }
+ }
+#endif
/* Key not found in chain: jump to exit (if merged) or load niltv. */
l_end = emit_label(as);
@@ -861,8 +1035,13 @@
l_end = asm_exitstub_addr(as);
}
if (!LJ_SOFTFP && irt_isnum(kt)) {
+#if !LJ_TARGET_MIPSR6
emit_branch(as, MIPSI_BC1T, 0, 0, l_end);
emit_fgh(as, MIPSI_C_EQ_D, 0, tmpnum, key);
+#else
+ emit_branch(as, MIPSI_BC1NEZ, 0, (tmpnum&31), l_end);
+ emit_fgh(as, MIPSI_CMP_EQ_D, tmpnum, tmpnum, key);
+#endif
*--as->mcp = MIPSI_NOP; /* Avoid NaN comparison overhead. */
emit_branch(as, MIPSI_BEQ, tmp1, RID_ZERO, l_next);
emit_tsi(as, MIPSI_SLTIU, tmp1, tmp1, (int32_t)LJ_TISNUM);
@@ -883,21 +1062,9 @@
emit_dta(as, MIPSI_DSRA32, tmp1, tmp1, 15);
emit_tg(as, MIPSI_DMTC1, tmp1, tmpnum);
emit_tsi(as, MIPSI_LD, tmp1, dest, (int32_t)offsetof(Node, key.u64));
- } else if (irt_isaddr(kt)) {
- Reg refk = tmp2;
- if (isk) {
- int64_t k = ((int64_t)irt_toitype(irkey->t) << 47) | irkey[1].tv.u64;
- refk = ra_allock(as, k, allow);
- rset_clear(allow, refk);
- }
- emit_branch(as, MIPSI_BEQ, tmp1, refk, l_end);
- emit_tsi(as, MIPSI_LD, tmp1, dest, offsetof(Node, key));
} else {
- Reg pri = ra_allock(as, ~((int64_t)~irt_toitype(ir->t) << 47), allow);
- rset_clear(allow, pri);
- lua_assert(irt_ispri(kt) && !irt_isnil(kt));
- emit_branch(as, MIPSI_BEQ, tmp1, pri, l_end);
- emit_tsi(as, MIPSI_LD, tmp1, dest, offsetof(Node, key));
+ emit_branch(as, MIPSI_BEQ, tmp1, cmp64, l_end);
+ emit_tsi(as, MIPSI_LD, tmp1, dest, (int32_t)offsetof(Node, key.u64));
}
*l_loop = MIPSI_BNE | MIPSF_S(tmp1) | ((as->mcp-l_loop-1) & 0xffffu);
if (!isk && irt_isaddr(kt)) {
@@ -908,7 +1075,7 @@
#endif
/* Load main position relative to tab->node into dest. */
- khash = isk ? ir_khash(irkey) : 1;
+ khash = isk ? ir_khash(as, irkey) : 1;
if (khash == 0) {
emit_tsi(as, MIPSI_AL, dest, tab, (int32_t)offsetof(GCtab, node));
} else {
@@ -916,7 +1083,7 @@
if (isk)
tmphash = ra_allock(as, khash, allow);
emit_dst(as, MIPSI_AADDU, dest, dest, tmp1);
- lua_assert(sizeof(Node) == 24);
+ lj_assertA(sizeof(Node) == 24, "bad Node size");
emit_dst(as, MIPSI_SUBU, tmp1, tmp2, tmp1);
emit_dta(as, MIPSI_SLL, tmp1, tmp1, 3);
emit_dta(as, MIPSI_SLL, tmp2, tmp1, 5);
@@ -926,7 +1093,7 @@
if (isk) {
/* Nothing to do. */
} else if (irt_isstr(kt)) {
- emit_tsi(as, MIPSI_LW, tmp1, key, (int32_t)offsetof(GCstr, hash));
+ emit_tsi(as, MIPSI_LW, tmp1, key, (int32_t)offsetof(GCstr, sid));
} else { /* Must match with hash*() in lj_tab.c. */
emit_dst(as, MIPSI_SUBU, tmp1, tmp1, tmp2);
emit_rotr(as, tmp2, tmp2, dest, (-HASH_ROT3)&31);
@@ -961,7 +1128,7 @@
emit_dta(as, MIPSI_ROTR, dest, tmp1, (-HASH_ROT1)&31);
if (irt_isnum(kt)) {
emit_dst(as, MIPSI_ADDU, tmp1, tmp1, tmp1);
- emit_dta(as, MIPSI_DSRA32, tmp1, tmp1, 0);
+ emit_dta(as, MIPSI_DSRA32, tmp1, LJ_SOFTFP ? key : tmp1, 0);
emit_dta(as, MIPSI_SLL, tmp2, LJ_SOFTFP ? key : tmp1, 0);
#if !LJ_SOFTFP
emit_tg(as, MIPSI_DMFC1, tmp1, key);
@@ -994,7 +1161,7 @@
Reg key = ra_scratch(as, allow);
int64_t k;
#endif
- lua_assert(ofs % sizeof(Node) == 0);
+ lj_assertA(ofs % sizeof(Node) == 0, "unaligned HREFK slot");
if (ofs > 32736) {
idx = dest;
rset_clear(allow, dest);
@@ -1023,7 +1190,7 @@
emit_tsi(as, MIPSI_LW, type, idx, kofs+(LJ_BE?0:4));
#else
if (irt_ispri(irkey->t)) {
- lua_assert(!irt_isnil(irkey->t));
+ lj_assertA(!irt_isnil(irkey->t), "bad HREFK key type");
k = ~((int64_t)~irt_toitype(irkey->t) << 47);
} else if (irt_isnum(irkey->t)) {
k = (int64_t)ir_knum(irkey)->u64;
@@ -1040,22 +1207,29 @@
static void asm_uref(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (irref_isk(ir->op1)) {
+ int guarded = (irt_t(ir->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC);
+ if (irref_isk(ir->op1) && !guarded) {
GCfunc *fn = ir_kfunc(IR(ir->op1));
MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
emit_lsptr(as, MIPSI_AL, dest, v, RSET_GPR);
} else {
- Reg uv = ra_scratch(as, RSET_GPR);
- Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
- if (ir->o == IR_UREFC) {
- asm_guard(as, MIPSI_BEQ, RID_TMP, RID_ZERO);
- emit_tsi(as, MIPSI_AADDIU, dest, uv, (int32_t)offsetof(GCupval, tv));
- emit_tsi(as, MIPSI_LBU, RID_TMP, uv, (int32_t)offsetof(GCupval, closed));
+ if (guarded)
+ asm_guard(as, ir->o == IR_UREFC ? MIPSI_BEQ : MIPSI_BNE, RID_TMP, RID_ZERO);
+ if (ir->o == IR_UREFC)
+ emit_tsi(as, MIPSI_AADDIU, dest, dest, (int32_t)offsetof(GCupval, tv));
+ else
+ emit_tsi(as, MIPSI_AL, dest, dest, (int32_t)offsetof(GCupval, v));
+ if (guarded)
+ emit_tsi(as, MIPSI_LBU, RID_TMP, dest, (int32_t)offsetof(GCupval, closed));
+ if (irref_isk(ir->op1)) {
+ GCfunc *fn = ir_kfunc(IR(ir->op1));
+ GCobj *o = gcref(fn->l.uvptr[(ir->op2 >> 8)]);
+ emit_loada(as, dest, o);
} else {
- emit_tsi(as, MIPSI_AL, dest, uv, (int32_t)offsetof(GCupval, v));
+ emit_tsi(as, MIPSI_AL, dest, ra_alloc1(as, ir->op1, RSET_GPR),
+ (int32_t)offsetof(GCfuncL, uvptr) +
+ (int32_t)sizeof(MRef) * (int32_t)(ir->op2 >> 8));
}
- emit_tsi(as, MIPSI_AL, uv, func, (int32_t)offsetof(GCfuncL, uvptr) +
- (int32_t)sizeof(MRef) * (int32_t)(ir->op2 >> 8));
}
}
@@ -1062,7 +1236,7 @@
static void asm_fref(ASMState *as, IRIns *ir)
{
UNUSED(as); UNUSED(ir);
- lua_assert(!ra_used(ir));
+ lj_assertA(!ra_used(ir), "unfused FREF");
}
static void asm_strref(ASMState *as, IRIns *ir)
@@ -1117,26 +1291,36 @@
/* -- Loads and stores ---------------------------------------------------- */
-static MIPSIns asm_fxloadins(IRIns *ir)
+static MIPSIns asm_fxloadins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: return MIPSI_LB;
case IRT_U8: return MIPSI_LBU;
case IRT_I16: return MIPSI_LH;
case IRT_U16: return MIPSI_LHU;
- case IRT_NUM: lua_assert(!LJ_SOFTFP); return MIPSI_LDC1;
+ case IRT_NUM:
+ lj_assertA(!LJ_SOFTFP32, "unsplit FP op");
+ if (!LJ_SOFTFP) return MIPSI_LDC1;
+ /* fallthrough */
case IRT_FLOAT: if (!LJ_SOFTFP) return MIPSI_LWC1;
+ /* fallthrough */
default: return (LJ_64 && irt_is64(ir->t)) ? MIPSI_LD : MIPSI_LW;
}
}
-static MIPSIns asm_fxstoreins(IRIns *ir)
+static MIPSIns asm_fxstoreins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: case IRT_U8: return MIPSI_SB;
case IRT_I16: case IRT_U16: return MIPSI_SH;
- case IRT_NUM: lua_assert(!LJ_SOFTFP); return MIPSI_SDC1;
+ case IRT_NUM:
+ lj_assertA(!LJ_SOFTFP32, "unsplit FP op");
+ if (!LJ_SOFTFP) return MIPSI_SDC1;
+ /* fallthrough */
case IRT_FLOAT: if (!LJ_SOFTFP) return MIPSI_SWC1;
+ /* fallthrough */
default: return (LJ_64 && irt_is64(ir->t)) ? MIPSI_SD : MIPSI_SW;
}
}
@@ -1144,10 +1328,10 @@
static void asm_fload(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- MIPSIns mi = asm_fxloadins(ir);
+ MIPSIns mi = asm_fxloadins(as, ir);
Reg idx;
int32_t ofs;
- if (ir->op1 == REF_NIL) {
+ if (ir->op1 == REF_NIL) { /* FLOAD from GG_State with offset. */
idx = RID_JGL;
ofs = (ir->op2 << 2) - 32768 - GG_OFS(g);
} else {
@@ -1160,8 +1344,8 @@
}
}
ofs = field_ofs[ir->op2];
+ lj_assertA(!irt_isfp(ir->t), "bad FP FLOAD");
}
- lua_assert(!irt_isfp(ir->t));
emit_tsi(as, mi, dest, idx, ofs);
}
@@ -1172,8 +1356,8 @@
IRIns *irf = IR(ir->op1);
Reg idx = ra_alloc1(as, irf->op1, rset_exclude(RSET_GPR, src));
int32_t ofs = field_ofs[irf->op2];
- MIPSIns mi = asm_fxstoreins(ir);
- lua_assert(!irt_isfp(ir->t));
+ MIPSIns mi = asm_fxstoreins(as, ir);
+ lj_assertA(!irt_isfp(ir->t), "bad FP FSTORE");
emit_tsi(as, mi, src, idx, ofs);
}
}
@@ -1182,8 +1366,9 @@
{
Reg dest = ra_dest(as, ir,
(!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
- lua_assert(!(ir->op2 & IRXLOAD_UNALIGNED));
- asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR, 0);
+ lj_assertA(LJ_TARGET_UNALIGNED || !(ir->op2 & IRXLOAD_UNALIGNED),
+ "unaligned XLOAD");
+ asm_fusexref(as, asm_fxloadins(as, ir), dest, ir->op1, RSET_GPR, 0);
}
static void asm_xstore_(ASMState *as, IRIns *ir, int32_t ofs)
@@ -1191,7 +1376,7 @@
if (ir->r != RID_SINK) {
Reg src = ra_alloc1z(as, ir->op2,
(!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
- asm_fusexref(as, asm_fxstoreins(ir), src, ir->op1,
+ asm_fusexref(as, asm_fxstoreins(as, ir), src, ir->op1,
rset_exclude(RSET_GPR, src), ofs);
}
}
@@ -1200,7 +1385,7 @@
static void asm_ahuvload(ASMState *as, IRIns *ir)
{
- int hiop = (LJ_32 && LJ_SOFTFP && (ir+1)->o == IR_HIOP);
+ int hiop = (LJ_SOFTFP32 && (ir+1)->o == IR_HIOP);
Reg dest = RID_NONE, type = RID_TMP, idx;
RegSet allow = RSET_GPR;
int32_t ofs = 0;
@@ -1213,8 +1398,9 @@
}
}
if (ra_used(ir)) {
- lua_assert((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
- irt_isint(ir->t) || irt_isaddr(ir->t));
+ lj_assertA((LJ_SOFTFP32 ? 0 : irt_isnum(ir->t)) ||
+ irt_isint(ir->t) || irt_isaddr(ir->t),
+ "bad load type %d", irt_type(ir->t));
dest = ra_dest(as, ir, (!LJ_SOFTFP && irt_isnum(t)) ? RSET_FPR : allow);
rset_clear(allow, dest);
#if LJ_64
@@ -1225,6 +1411,7 @@
#endif
}
idx = asm_fuseahuref(as, ir->op1, &ofs, allow);
+ if (ir->o == IR_VLOAD) ofs += 8 * ir->op2;
rset_clear(allow, idx);
if (irt_isnum(t)) {
asm_guard(as, MIPSI_BEQ, RID_TMP, RID_ZERO);
@@ -1262,10 +1449,10 @@
int32_t ofs = 0;
if (ir->r == RID_SINK)
return;
- if (!LJ_SOFTFP && irt_isnum(ir->t)) {
- src = ra_alloc1(as, ir->op2, RSET_FPR);
+ if (!LJ_SOFTFP32 && irt_isnum(ir->t)) {
+ src = ra_alloc1(as, ir->op2, LJ_SOFTFP ? RSET_GPR : RSET_FPR);
idx = asm_fuseahuref(as, ir->op1, &ofs, allow);
- emit_hsi(as, MIPSI_SDC1, src, idx, ofs);
+ emit_hsi(as, LJ_SOFTFP ? MIPSI_SD : MIPSI_SDC1, src, idx, ofs);
} else {
#if LJ_32
if (!irt_ispri(ir->t)) {
@@ -1313,16 +1500,19 @@
IRType1 t = ir->t;
#if LJ_32
int32_t ofs = 8*((int32_t)ir->op1-1) + ((ir->op2 & IRSLOAD_FRAME) ? 4 : 0);
- int hiop = (LJ_32 && LJ_SOFTFP && (ir+1)->o == IR_HIOP);
+ int hiop = (LJ_SOFTFP32 && (ir+1)->o == IR_HIOP);
if (hiop)
t.irt = IRT_NUM;
#else
int32_t ofs = 8*((int32_t)ir->op1-2);
#endif
- lua_assert(!(ir->op2 & IRSLOAD_PARENT)); /* Handled by asm_head_side(). */
- lua_assert(irt_isguard(ir->t) || !(ir->op2 & IRSLOAD_TYPECHECK));
-#if LJ_32 && LJ_SOFTFP
- lua_assert(!(ir->op2 & IRSLOAD_CONVERT)); /* Handled by LJ_SOFTFP SPLIT. */
+ lj_assertA(!(ir->op2 & IRSLOAD_PARENT),
+ "bad parent SLOAD"); /* Handled by asm_head_side(). */
+ lj_assertA(irt_isguard(ir->t) || !(ir->op2 & IRSLOAD_TYPECHECK),
+ "inconsistent SLOAD variant");
+#if LJ_SOFTFP32
+ lj_assertA(!(ir->op2 & IRSLOAD_CONVERT),
+ "unsplit SLOAD convert"); /* Handled by LJ_SOFTFP SPLIT. */
if (hiop && ra_used(ir+1)) {
type = ra_dest(as, ir+1, allow);
rset_clear(allow, type);
@@ -1329,29 +1519,45 @@
}
#else
if ((ir->op2 & IRSLOAD_CONVERT) && irt_isguard(t) && irt_isint(t)) {
- dest = ra_scratch(as, RSET_FPR);
+ dest = ra_scratch(as, LJ_SOFTFP ? allow : RSET_FPR);
asm_tointg(as, ir, dest);
t.irt = IRT_NUM; /* Continue with a regular number type check. */
} else
#endif
if (ra_used(ir)) {
- lua_assert((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
- irt_isint(ir->t) || irt_isaddr(ir->t));
+ lj_assertA((LJ_SOFTFP32 ? 0 : irt_isnum(ir->t)) ||
+ irt_isint(ir->t) || irt_isaddr(ir->t),
+ "bad SLOAD type %d", irt_type(ir->t));
dest = ra_dest(as, ir, (!LJ_SOFTFP && irt_isnum(t)) ? RSET_FPR : allow);
rset_clear(allow, dest);
base = ra_alloc1(as, REF_BASE, allow);
rset_clear(allow, base);
- if (!LJ_SOFTFP && (ir->op2 & IRSLOAD_CONVERT)) {
+ if (!LJ_SOFTFP32 && (ir->op2 & IRSLOAD_CONVERT)) {
if (irt_isint(t)) {
- Reg tmp = ra_scratch(as, RSET_FPR);
+ Reg tmp = ra_scratch(as, LJ_SOFTFP ? RSET_GPR : RSET_FPR);
+#if LJ_SOFTFP
+ ra_evictset(as, rset_exclude(RSET_SCRATCH, dest));
+ ra_destreg(as, ir, RID_RET);
+ emit_call(as, (void *)lj_ir_callinfo[IRCALL_softfp_d2i].func, 0);
+ if (tmp != REGARG_FIRSTGPR)
+ emit_move(as, REGARG_FIRSTGPR, tmp);
+#else
emit_tg(as, MIPSI_MFC1, dest, tmp);
emit_fg(as, MIPSI_TRUNC_W_D, tmp, tmp);
+#endif
dest = tmp;
t.irt = IRT_NUM; /* Check for original type. */
} else {
Reg tmp = ra_scratch(as, RSET_GPR);
+#if LJ_SOFTFP
+ ra_evictset(as, rset_exclude(RSET_SCRATCH, dest));
+ ra_destreg(as, ir, RID_RET);
+ emit_call(as, (void *)lj_ir_callinfo[IRCALL_softfp_i2d].func, 0);
+ emit_dta(as, MIPSI_SLL, REGARG_FIRSTGPR, tmp, 0);
+#else
emit_fg(as, MIPSI_CVT_D_W, dest, dest);
emit_tg(as, MIPSI_MTC1, tmp, dest);
+#endif
dest = tmp;
t.irt = IRT_INT; /* Check for original type. */
}
@@ -1378,7 +1584,7 @@
asm_guard(as, MIPSI_BEQ, RID_TMP, RID_ZERO);
emit_tsi(as, MIPSI_SLTIU, RID_TMP, type, (int32_t)LJ_TISNUM);
} else {
- Reg ktype = ra_allock(as, irt_toitype(t), allow);
+ Reg ktype = ra_allock(as, (ir->op2 & IRSLOAD_KEYINDEX) ? LJ_KEYINDEX : irt_toitype(t), allow);
asm_guard(as, MIPSI_BNE, type, ktype);
}
}
@@ -1396,11 +1602,15 @@
if (irt_ispri(t)) {
asm_guard(as, MIPSI_BNE, type,
ra_allock(as, ~((int64_t)~irt_toitype(t) << 47) , allow));
+ } else if ((ir->op2 & IRSLOAD_KEYINDEX)) {
+ asm_guard(as, MIPSI_BNE, RID_TMP,
+ ra_allock(as, (int32_t)LJ_KEYINDEX, allow));
+ emit_dta(as, MIPSI_DSRA32, RID_TMP, type, 0);
} else {
if (irt_isnum(t)) {
asm_guard(as, MIPSI_BEQ, RID_TMP, RID_ZERO);
emit_tsi(as, MIPSI_SLTIU, RID_TMP, RID_TMP, (int32_t)LJ_TISNUM);
- if (ra_hasreg(dest))
+ if (!LJ_SOFTFP && ra_hasreg(dest))
emit_hsi(as, MIPSI_LDC1, dest, base, ofs);
} else {
asm_guard(as, MIPSI_BNE, RID_TMP,
@@ -1410,7 +1620,7 @@
}
emit_tsi(as, MIPSI_LD, type, base, ofs);
} else if (ra_hasreg(dest)) {
- if (irt_isnum(t))
+ if (!LJ_SOFTFP && irt_isnum(t))
emit_hsi(as, MIPSI_LDC1, dest, base, ofs);
else
emit_tsi(as, irt_isint(t) ? MIPSI_LW : MIPSI_LD, dest, base,
@@ -1431,7 +1641,8 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
IRRef args[4];
RegSet drop = RSET_SCRATCH;
- lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
+ lj_assertA(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL),
+ "bad CNEW/CNEWI operands");
as->gcsteps++;
if (ra_hasreg(ir->r))
@@ -1447,7 +1658,7 @@
int32_t ofs = sizeof(GCcdata);
if (sz == 8) {
ofs += 4;
- lua_assert((ir+1)->o == IR_HIOP);
+ lj_assertA((ir+1)->o == IR_HIOP, "expected HIOP for CNEWI");
if (LJ_LE) ir++;
}
for (;;) {
@@ -1458,10 +1669,10 @@
ofs -= 4; if (LJ_BE) ir++; else ir--;
}
#else
- emit_tsi(as, MIPSI_SD, ra_alloc1(as, ir->op2, allow),
+ emit_tsi(as, sz == 8 ? MIPSI_SD : MIPSI_SW, ra_alloc1(as, ir->op2, allow),
RID_RET, sizeof(GCcdata));
#endif
- lua_assert(sz == 4 || sz == 8);
+ lj_assertA(sz == 4 || sz == 8, "bad CNEWI size %d", sz);
} else if (ir->op2 != REF_NIL) { /* Create VLA/VLS/aligned cdata. */
ci = &lj_ir_callinfo[IRCALL_lj_cdata_newv];
args[0] = ASMREF_L; /* lua_State *L */
@@ -1484,8 +1695,6 @@
ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
ra_releasetmp(as, ASMREF_TMP1));
}
-#else
-#define asm_cnew(as, ir) ((void)0)
#endif
/* -- Write barriers ------------------------------------------------------ */
@@ -1513,7 +1722,7 @@
MCLabel l_end;
Reg obj, val, tmp;
/* No need for other object barriers (yet). */
- lua_assert(IR(ir->op1)->o == IR_UREFC);
+ lj_assertA(IR(ir->op1)->o == IR_UREFC, "bad OBAR type");
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
args[0] = ASMREF_TMP1; /* global_State *g */
@@ -1549,33 +1758,46 @@
Reg left = ra_hintalloc(as, ir->op1, dest, RSET_FPR);
emit_fg(as, mi, dest, left);
}
+#endif
+#if !LJ_SOFTFP32
static void asm_fpmath(ASMState *as, IRIns *ir)
{
- if (ir->op2 == IRFPM_EXP2 && asm_fpjoin_pow(as, ir))
- return;
+#if !LJ_SOFTFP
if (ir->op2 <= IRFPM_TRUNC)
asm_callround(as, ir, IRCALL_lj_vm_floor + ir->op2);
else if (ir->op2 == IRFPM_SQRT)
asm_fpunary(as, ir, MIPSI_SQRT_D);
else
+#endif
asm_callid(as, ir, IRCALL_lj_vm_floor + ir->op2);
}
#endif
+#if !LJ_SOFTFP
+#define asm_fpadd(as, ir) asm_fparith(as, ir, MIPSI_ADD_D)
+#define asm_fpsub(as, ir) asm_fparith(as, ir, MIPSI_SUB_D)
+#define asm_fpmul(as, ir) asm_fparith(as, ir, MIPSI_MUL_D)
+#elif LJ_64 /* && LJ_SOFTFP */
+#define asm_fpadd(as, ir) asm_callid(as, ir, IRCALL_softfp_add)
+#define asm_fpsub(as, ir) asm_callid(as, ir, IRCALL_softfp_sub)
+#define asm_fpmul(as, ir) asm_callid(as, ir, IRCALL_softfp_mul)
+#endif
+
static void asm_add(ASMState *as, IRIns *ir)
{
IRType1 t = ir->t;
-#if !LJ_SOFTFP
+#if !LJ_SOFTFP32
if (irt_isnum(t)) {
- asm_fparith(as, ir, MIPSI_ADD_D);
+ asm_fpadd(as, ir);
} else
#endif
{
+ /* TODO MIPSR6: Fuse ADD(BSHL(a,1-4),b) or ADD(ADD(a,a),b) to MIPSI_ALSA. */
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg right, left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
if (irref_isk(ir->op2)) {
- intptr_t k = get_kval(IR(ir->op2));
+ intptr_t k = get_kval(as, ir->op2);
if (checki16(k)) {
emit_tsi(as, (LJ_64 && irt_is64(t)) ? MIPSI_DADDIU : MIPSI_ADDIU, dest,
left, k);
@@ -1590,9 +1812,9 @@
static void asm_sub(ASMState *as, IRIns *ir)
{
-#if !LJ_SOFTFP
+#if !LJ_SOFTFP32
if (irt_isnum(ir->t)) {
- asm_fparith(as, ir, MIPSI_SUB_D);
+ asm_fpsub(as, ir);
} else
#endif
{
@@ -1606,9 +1828,9 @@
static void asm_mul(ASMState *as, IRIns *ir)
{
-#if !LJ_SOFTFP
+#if !LJ_SOFTFP32
if (irt_isnum(ir->t)) {
- asm_fparith(as, ir, MIPSI_MUL_D);
+ asm_fpmul(as, ir);
} else
#endif
{
@@ -1616,8 +1838,12 @@
Reg right, left = ra_alloc2(as, ir, RSET_GPR);
right = (left >> 8); left &= 255;
if (LJ_64 && irt_is64(ir->t)) {
+#if !LJ_TARGET_MIPSR6
emit_dst(as, MIPSI_MFLO, dest, 0, 0);
emit_dst(as, MIPSI_DMULT, 0, left, right);
+#else
+ emit_dst(as, MIPSI_DMUL, dest, left, right);
+#endif
} else {
emit_dst(as, MIPSI_MUL, dest, left, right);
}
@@ -1624,40 +1850,16 @@
}
}
-static void asm_mod(ASMState *as, IRIns *ir)
+#if !LJ_SOFTFP32
+static void asm_fpdiv(ASMState *as, IRIns *ir)
{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isint(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_modi64 :
- IRCALL_lj_carith_modu64);
- else
-#endif
- asm_callid(as, ir, IRCALL_lj_vm_modi);
-}
-
#if !LJ_SOFTFP
-static void asm_pow(ASMState *as, IRIns *ir)
-{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_powi64 :
- IRCALL_lj_carith_powu64);
- else
+ asm_fparith(as, ir, MIPSI_DIV_D);
+#else
+ asm_callid(as, ir, IRCALL_softfp_div);
#endif
- asm_callid(as, ir, IRCALL_lj_vm_powi);
}
-
-static void asm_div(ASMState *as, IRIns *ir)
-{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_divi64 :
- IRCALL_lj_carith_divu64);
- else
#endif
- asm_fparith(as, ir, MIPSI_DIV_D);
-}
-#endif
static void asm_neg(ASMState *as, IRIns *ir)
{
@@ -1665,6 +1867,13 @@
if (irt_isnum(ir->t)) {
asm_fpunary(as, ir, MIPSI_NEG_D);
} else
+#elif LJ_64 /* && LJ_SOFTFP */
+ if (irt_isnum(ir->t)) {
+ Reg dest = ra_dest(as, ir, RSET_GPR);
+ Reg left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
+ emit_dst(as, MIPSI_XOR, dest, left,
+ ra_allock(as, 0x8000000000000000ll, rset_exclude(RSET_GPR, dest)));
+ } else
#endif
{
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -1674,17 +1883,25 @@
}
}
+#if !LJ_SOFTFP
#define asm_abs(as, ir) asm_fpunary(as, ir, MIPSI_ABS_D)
-#define asm_atan2(as, ir) asm_callid(as, ir, IRCALL_atan2)
-#define asm_ldexp(as, ir) asm_callid(as, ir, IRCALL_ldexp)
+#elif LJ_64 /* && LJ_SOFTFP */
+static void asm_abs(ASMState *as, IRIns *ir)
+{
+ Reg dest = ra_dest(as, ir, RSET_GPR);
+ Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
+ emit_tsml(as, MIPSI_DEXTM, dest, left, 30, 0);
+}
+#endif
static void asm_arithov(ASMState *as, IRIns *ir)
{
+ /* TODO MIPSR6: bovc/bnvc. Caveat: no delay slot to load RID_TMP. */
Reg right, left, tmp, dest = ra_dest(as, ir, RSET_GPR);
- lua_assert(!irt_is64(ir->t));
+ lj_assertA(!irt_is64(ir->t), "bad usage");
if (irref_isk(ir->op2)) {
int k = IR(ir->op2)->i;
- if (ir->o == IR_SUBOV) k = -k;
+ if (ir->o == IR_SUBOV) k = (int)(~(unsigned int)k+1u);
if (checki16(k)) { /* (dest < left) == (k >= 0 ? 1 : 0) */
left = ra_alloc1(as, ir->op1, RSET_GPR);
asm_guard(as, k >= 0 ? MIPSI_BNE : MIPSI_BEQ, RID_TMP, RID_ZERO);
@@ -1724,9 +1941,14 @@
right), dest));
asm_guard(as, MIPSI_BNE, RID_TMP, tmp);
emit_dta(as, MIPSI_SRA, RID_TMP, dest, 31);
+#if !LJ_TARGET_MIPSR6
emit_dst(as, MIPSI_MFHI, tmp, 0, 0);
emit_dst(as, MIPSI_MFLO, dest, 0, 0);
emit_dst(as, MIPSI_MULT, 0, left, right);
+#else
+ emit_dst(as, MIPSI_MUL, dest, left, right);
+ emit_dst(as, MIPSI_MUH, tmp, left, right);
+#endif
}
#if LJ_32 && LJ_HASFFI
@@ -1863,7 +2085,7 @@
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg right, left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
if (irref_isk(ir->op2)) {
- intptr_t k = get_kval(IR(ir->op2));
+ intptr_t k = get_kval(as, ir->op2);
if (checku16(k)) {
emit_tsi(as, mik, dest, left, k);
return;
@@ -1896,7 +2118,7 @@
#define asm_bshl(as, ir) asm_bitshift(as, ir, MIPSI_SLLV, MIPSI_SLL)
#define asm_bshr(as, ir) asm_bitshift(as, ir, MIPSI_SRLV, MIPSI_SRL)
#define asm_bsar(as, ir) asm_bitshift(as, ir, MIPSI_SRAV, MIPSI_SRA)
-#define asm_brol(as, ir) lua_assert(0)
+#define asm_brol(as, ir) lj_assertA(0, "unexpected BROL")
static void asm_bror(ASMState *as, IRIns *ir)
{
@@ -1919,15 +2141,21 @@
}
}
-#if LJ_32 && LJ_SOFTFP
+#if LJ_SOFTFP
static void asm_sfpmin_max(ASMState *as, IRIns *ir)
{
CCallInfo ci = lj_ir_callinfo[(IROp)ir->o == IR_MIN ? IRCALL_lj_vm_sfmin : IRCALL_lj_vm_sfmax];
+#if LJ_64
+ IRRef args[2];
+ args[0] = ir->op1;
+ args[1] = ir->op2;
+#else
IRRef args[4];
args[0^LJ_BE] = ir->op1;
args[1^LJ_BE] = (ir+1)->op1;
args[2^LJ_BE] = ir->op2;
args[3^LJ_BE] = (ir+1)->op2;
+#endif
asm_setupresult(as, ir, &ci);
emit_call(as, (void *)ci.func, 0);
ci.func = NULL;
@@ -1937,29 +2165,52 @@
static void asm_min_max(ASMState *as, IRIns *ir, int ismax)
{
- if (!LJ_SOFTFP && irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP32 && irt_isnum(ir->t)) {
+#if LJ_SOFTFP
+ asm_sfpmin_max(as, ir);
+#else
Reg dest = ra_dest(as, ir, RSET_FPR);
Reg right, left = ra_alloc2(as, ir, RSET_FPR);
right = (left >> 8); left &= 255;
+#if !LJ_TARGET_MIPSR6
if (dest == left) {
- emit_fg(as, MIPSI_MOVT_D, dest, right);
+ emit_fg(as, MIPSI_MOVF_D, dest, right);
} else {
- emit_fg(as, MIPSI_MOVF_D, dest, left);
+ emit_fg(as, MIPSI_MOVT_D, dest, left);
if (dest != right) emit_fg(as, MIPSI_MOV_D, dest, right);
}
- emit_fgh(as, MIPSI_C_OLT_D, 0, ismax ? left : right, ismax ? right : left);
+ emit_fgh(as, MIPSI_C_OLT_D, 0, ismax ? right : left, ismax ? left : right);
+#else
+ emit_fgh(as, ismax ? MIPSI_MAX_D : MIPSI_MIN_D, dest, left, right);
+#endif
+#endif
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg right, left = ra_alloc2(as, ir, RSET_GPR);
right = (left >> 8); left &= 255;
- if (dest == left) {
- emit_dst(as, MIPSI_MOVN, dest, right, RID_TMP);
+ if (left == right) {
+ if (dest != left) emit_move(as, dest, left);
} else {
- emit_dst(as, MIPSI_MOVZ, dest, left, RID_TMP);
- if (dest != right) emit_move(as, dest, right);
+#if !LJ_TARGET_MIPSR6
+ if (dest == left) {
+ emit_dst(as, MIPSI_MOVN, dest, right, RID_TMP);
+ } else {
+ emit_dst(as, MIPSI_MOVZ, dest, left, RID_TMP);
+ if (dest != right) emit_move(as, dest, right);
+ }
+#else
+ emit_dst(as, MIPSI_OR, dest, dest, RID_TMP);
+ if (dest != right) {
+ emit_dst(as, MIPSI_SELNEZ, RID_TMP, right, RID_TMP);
+ emit_dst(as, MIPSI_SELEQZ, dest, left, RID_TMP);
+ } else {
+ emit_dst(as, MIPSI_SELEQZ, RID_TMP, left, RID_TMP);
+ emit_dst(as, MIPSI_SELNEZ, dest, right, RID_TMP);
+ }
+#endif
+ emit_dst(as, MIPSI_SLT, RID_TMP,
+ ismax ? left : right, ismax ? right : left);
}
- emit_dst(as, MIPSI_SLT, RID_TMP,
- ismax ? left : right, ismax ? right : left);
}
}
@@ -1968,7 +2219,7 @@
/* -- Comparisons --------------------------------------------------------- */
-#if LJ_32 && LJ_SOFTFP
+#if LJ_SOFTFP
/* SFP comparisons. */
static void asm_sfpcomp(ASMState *as, IRIns *ir)
{
@@ -1975,11 +2226,17 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_softfp_cmp];
RegSet drop = RSET_SCRATCH;
Reg r;
+#if LJ_64
+ IRRef args[2];
+ args[0] = ir->op1;
+ args[1] = ir->op2;
+#else
IRRef args[4];
args[LJ_LE ? 0 : 1] = ir->op1; args[LJ_LE ? 1 : 0] = (ir+1)->op1;
args[LJ_LE ? 2 : 3] = ir->op2; args[LJ_LE ? 3 : 2] = (ir+1)->op2;
+#endif
- for (r = REGARG_FIRSTGPR; r <= REGARG_FIRSTGPR+3; r++) {
+ for (r = REGARG_FIRSTGPR; r <= REGARG_FIRSTGPR+(LJ_64?1:3); r++) {
if (!rset_test(as->freeset, r) &&
regcost_ref(as->cost[r]) == args[r-REGARG_FIRSTGPR])
rset_clear(drop, r);
@@ -2033,21 +2290,33 @@
{
/* ORDER IR: LT GE LE GT ULT UGE ULE UGT. */
IROp op = ir->o;
- if (!LJ_SOFTFP && irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP32 && irt_isnum(ir->t)) {
+#if LJ_SOFTFP
+ asm_sfpcomp(as, ir);
+#else
+#if !LJ_TARGET_MIPSR6
Reg right, left = ra_alloc2(as, ir, RSET_FPR);
right = (left >> 8); left &= 255;
asm_guard(as, (op&1) ? MIPSI_BC1T : MIPSI_BC1F, 0, 0);
emit_fgh(as, MIPSI_C_OLT_D + ((op&3) ^ ((op>>2)&1)), 0, left, right);
+#else
+ Reg tmp, right, left = ra_alloc2(as, ir, RSET_FPR);
+ right = (left >> 8); left &= 255;
+ tmp = ra_scratch(as, rset_exclude(rset_exclude(RSET_FPR, left), right));
+ asm_guard(as, (op&1) ? MIPSI_BC1NEZ : MIPSI_BC1EQZ, 0, (tmp&31));
+ emit_fgh(as, MIPSI_CMP_LT_D + ((op&3) ^ ((op>>2)&1)), tmp, left, right);
+#endif
+#endif
} else {
Reg right, left = ra_alloc1(as, ir->op1, RSET_GPR);
if (op == IR_ABC) op = IR_UGT;
- if ((op&4) == 0 && irref_isk(ir->op2) && get_kval(IR(ir->op2)) == 0) {
+ if ((op&4) == 0 && irref_isk(ir->op2) && get_kval(as, ir->op2) == 0) {
MIPSIns mi = (op&2) ? ((op&1) ? MIPSI_BLEZ : MIPSI_BGTZ) :
((op&1) ? MIPSI_BLTZ : MIPSI_BGEZ);
asm_guard(as, mi, left, 0);
} else {
if (irref_isk(ir->op2)) {
- intptr_t k = get_kval(IR(ir->op2));
+ intptr_t k = get_kval(as, ir->op2);
if ((op&2)) k++;
if (checki16(k)) {
asm_guard(as, (op&1) ? MIPSI_BNE : MIPSI_BEQ, RID_TMP, RID_ZERO);
@@ -2069,9 +2338,17 @@
Reg right, left = ra_alloc2(as, ir, (!LJ_SOFTFP && irt_isnum(ir->t)) ?
RSET_FPR : RSET_GPR);
right = (left >> 8); left &= 255;
- if (!LJ_SOFTFP && irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP32 && irt_isnum(ir->t)) {
+#if LJ_SOFTFP
+ asm_sfpcomp(as, ir);
+#elif !LJ_TARGET_MIPSR6
asm_guard(as, (ir->o & 1) ? MIPSI_BC1T : MIPSI_BC1F, 0, 0);
emit_fgh(as, MIPSI_C_EQ_D, 0, left, right);
+#else
+ Reg tmp = ra_scratch(as, rset_exclude(rset_exclude(RSET_FPR, left), right));
+ asm_guard(as, (ir->o & 1) ? MIPSI_BC1NEZ : MIPSI_BC1EQZ, 0, (tmp&31));
+ emit_fgh(as, MIPSI_CMP_EQ_D, tmp, left, right);
+#endif
} else {
asm_guard(as, (ir->o & 1) ? MIPSI_BEQ : MIPSI_BNE, left, right);
}
@@ -2114,15 +2391,15 @@
}
#endif
-/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
+/* -- Split register ops -------------------------------------------------- */
-/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
+/* Hiword op of a split 32/32 or 64/64 bit op. Previous op is the loword op. */
static void asm_hiop(ASMState *as, IRIns *ir)
{
-#if LJ_32 && (LJ_HASFFI || LJ_SOFTFP)
/* HIOP is marked as a store because it needs its own DCE logic. */
int uselo = ra_used(ir-1), usehi = ra_used(ir); /* Loword/hiword used? */
if (LJ_UNLIKELY(!(as->flags & JIT_F_OPT_DCE))) uselo = usehi = 1;
+#if LJ_32 && (LJ_HASFFI || LJ_SOFTFP)
if ((ir-1)->o == IR_CONV) { /* Conversions to/from 64 bit. */
as->curins--; /* Always skip the CONV. */
#if LJ_HASFFI && !LJ_SOFTFP
@@ -2169,37 +2446,33 @@
}
return;
}
+#endif
if (!usehi) return; /* Skip unused hiword op for all remaining ops. */
switch ((ir-1)->o) {
-#if LJ_HASFFI
+#if LJ_32 && LJ_HASFFI
case IR_ADD: as->curins--; asm_add64(as, ir); break;
case IR_SUB: as->curins--; asm_sub64(as, ir); break;
case IR_NEG: as->curins--; asm_neg64(as, ir); break;
+ case IR_CNEWI:
+ /* Nothing to do here. Handled by lo op itself. */
+ break;
#endif
-#if LJ_SOFTFP
+#if LJ_32 && LJ_SOFTFP
case IR_SLOAD: case IR_ALOAD: case IR_HLOAD: case IR_ULOAD: case IR_VLOAD:
case IR_STRTO:
if (!uselo)
ra_allocref(as, ir->op1, RSET_GPR); /* Mark lo op as used. */
break;
+ case IR_ASTORE: case IR_HSTORE: case IR_USTORE: case IR_TOSTR: case IR_TMPREF:
+ /* Nothing to do here. Handled by lo op itself. */
+ break;
#endif
- case IR_CALLN:
- case IR_CALLS:
- case IR_CALLXS:
+ case IR_CALLN: case IR_CALLL: case IR_CALLS: case IR_CALLXS:
if (!uselo)
ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
break;
-#if LJ_SOFTFP
- case IR_ASTORE: case IR_HSTORE: case IR_USTORE: case IR_TOSTR:
-#endif
- case IR_CNEWI:
- /* Nothing to do here. Handled by lo op itself. */
- break;
- default: lua_assert(0); break;
+ default: lj_assertA(0, "bad HIOP for op %d", (ir-1)->o); break;
}
-#else
- UNUSED(as); UNUSED(ir); lua_assert(0); /* Unused without FFI. */
-#endif
}
/* -- Profiling ----------------------------------------------------------- */
@@ -2264,15 +2537,19 @@
if ((sn & SNAP_NORESTORE))
continue;
if (irt_isnum(ir->t)) {
-#if LJ_SOFTFP
+#if LJ_SOFTFP32
Reg tmp;
RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
- lua_assert(irref_isk(ref)); /* LJ_SOFTFP: must be a number constant. */
+ /* LJ_SOFTFP: must be a number constant. */
+ lj_assertA(irref_isk(ref), "unsplit FP op");
tmp = ra_allock(as, (int32_t)ir_knum(ir)->u32.lo, allow);
emit_tsi(as, MIPSI_SW, tmp, RID_BASE, ofs+(LJ_BE?4:0));
if (rset_test(as->freeset, tmp+1)) allow = RID2RSET(tmp+1);
tmp = ra_allock(as, (int32_t)ir_knum(ir)->u32.hi, allow);
emit_tsi(as, MIPSI_SW, tmp, RID_BASE, ofs+(LJ_BE?0:4));
+#elif LJ_SOFTFP /* && LJ_64 */
+ Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, RID_BASE));
+ emit_tsi(as, MIPSI_SD, src, RID_BASE, ofs);
#else
Reg src = ra_alloc1(as, ref, RSET_FPR);
emit_hsi(as, MIPSI_SDC1, src, RID_BASE, ofs);
@@ -2281,7 +2558,8 @@
#if LJ_32
RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
Reg type;
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t),
+ "restore of IR type %d", irt_type(ir->t));
if (!irt_ispri(ir->t)) {
Reg src = ra_alloc1(as, ref, allow);
rset_clear(allow, src);
@@ -2294,21 +2572,41 @@
} else if ((sn & SNAP_SOFTFPNUM)) {
type = ra_alloc1(as, ref+1, rset_exclude(RSET_GPR, RID_BASE));
#endif
+ } else if ((sn & SNAP_KEYINDEX)) {
+ type = ra_allock(as, (int32_t)LJ_KEYINDEX, allow);
} else {
type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
}
emit_tsi(as, MIPSI_SW, type, RID_BASE, ofs+(LJ_BE?0:4));
#else
- asm_tvstore64(as, RID_BASE, ofs, ref);
+ if ((sn & SNAP_KEYINDEX)) {
+ RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
+ int64_t kki = (int64_t)LJ_KEYINDEX << 32;
+ if (irref_isk(ref)) {
+ emit_tsi(as, MIPSI_SD,
+ ra_allock(as, kki | (int64_t)(uint32_t)ir->i, allow),
+ RID_BASE, ofs);
+ } else {
+ Reg src = ra_alloc1(as, ref, allow);
+ Reg rki = ra_allock(as, kki, rset_exclude(allow, src));
+ emit_tsi(as, MIPSI_SD, RID_TMP, RID_BASE, ofs);
+ emit_dst(as, MIPSI_DADDU, RID_TMP, src, rki);
+ }
+ } else {
+ asm_tvstore64(as, RID_BASE, ofs, ref);
+ }
#endif
}
checkmclim(as);
}
- lua_assert(map + nent == flinks);
+ lj_assertA(map + nent == flinks, "inconsistent frames in snapshot");
}
/* -- GC handling --------------------------------------------------------- */
+/* Marker to prevent patching the GC check exit. */
+#define MIPS_NOPATCH_GC_CHECK MIPSI_OR
+
/* Check GC threshold and do one or more GC steps. */
static void asm_gc_check(ASMState *as)
{
@@ -2324,6 +2622,7 @@
args[0] = ASMREF_TMP1; /* global_State *g */
args[1] = ASMREF_TMP2; /* MSize steps */
asm_gencall(as, ci, args);
+ l_end[-3] = MIPS_NOPATCH_GC_CHECK; /* Replace the nop after the call. */
emit_tsi(as, MIPSI_AADDIU, ra_releasetmp(as, ASMREF_TMP1), RID_JGL, -32768);
tmp = ra_releasetmp(as, ASMREF_TMP2);
emit_loadi(as, tmp, as->gcsteps);
@@ -2352,6 +2651,12 @@
}
}
+/* Fixup the tail of the loop. */
+static void asm_loop_tail_fixup(ASMState *as)
+{
+ if (as->loopinv) as->mctop--;
+}
+
/* -- Head of trace ------------------------------------------------------- */
/* Coalesce BASE register for a root trace. */
@@ -2359,7 +2664,6 @@
{
IRIns *ir = IR(REF_BASE);
Reg r = ir->r;
- if (as->loopinv) as->mctop--;
if (ra_hasreg(r)) {
ra_free(as, r);
if (rset_test(as->modset, r) || irt_ismarked(ir->t))
@@ -2370,25 +2674,24 @@
}
/* Coalesce BASE register for a side trace. */
-static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
+static Reg asm_head_side_base(ASMState *as, IRIns *irp)
{
IRIns *ir = IR(REF_BASE);
Reg r = ir->r;
- if (as->loopinv) as->mctop--;
if (ra_hasreg(r)) {
ra_free(as, r);
if (rset_test(as->modset, r) || irt_ismarked(ir->t))
ir->r = RID_INIT; /* No inheritance for modified BASE register. */
if (irp->r == r) {
- rset_clear(allow, r); /* Mark same BASE register as coalesced. */
+ return r; /* Same BASE register already coalesced. */
} else if (ra_hasreg(irp->r) && rset_test(as->freeset, irp->r)) {
- rset_clear(allow, irp->r);
emit_move(as, r, irp->r); /* Move from coalesced parent reg. */
+ return irp->r;
} else {
emit_getgl(as, r, jit_base); /* Otherwise reload BASE. */
}
}
- return allow;
+ return RID_NONE;
}
/* -- Tail of trace ------------------------------------------------------- */
@@ -2466,25 +2769,32 @@
MCode tjump = MIPSI_J|(((uintptr_t)target>>2)&0x03ffffffu);
for (p++; p < pe; p++) {
if (*p == exitload) { /* Look for load of exit number. */
- if (((p[-1] ^ (px-p)) & 0xffffu) == 0) { /* Look for exitstub branch. */
+ /* Look for exitstub branch. Yes, this covers all used branch variants. */
+ if (((p[-1] ^ (px-p)) & 0xffffu) == 0 &&
+ ((p[-1] & 0xf0000000u) == MIPSI_BEQ ||
+ (p[-1] & 0xfc1e0000u) == MIPSI_BLTZ ||
+#if !LJ_TARGET_MIPSR6
+ (p[-1] & 0xffe00000u) == MIPSI_BC1F
+#else
+ (p[-1] & 0xff600000u) == MIPSI_BC1EQZ
+#endif
+ ) && p[-2] != MIPS_NOPATCH_GC_CHECK) {
ptrdiff_t delta = target - p;
if (((delta + 0x8000) >> 16) == 0) { /* Patch in-range branch. */
patchbranch:
p[-1] = (p[-1] & 0xffff0000u) | (delta & 0xffffu);
*p = MIPSI_NOP; /* Replace the load of the exit number. */
- cstop = p;
+ cstop = p+1;
if (!cstart) cstart = p-1;
} else { /* Branch out of range. Use spare jump slot in mcarea. */
- int i;
- for (i = 2; i < 2+MIPS_SPAREJUMP*2; i += 2) {
- if (mcarea[i] == tjump) {
- delta = mcarea+i - p;
+ MCode *mcjump = asm_sparejump_use(mcarea, tjump);
+ if (mcjump) {
+ lj_mcode_sync(mcjump, mcjump+1);
+ delta = mcjump - p;
+ if (((delta + 0x8000) >> 16) == 0) {
goto patchbranch;
- } else if (mcarea[i] == MIPSI_NOP) {
- mcarea[i] = tjump;
- cstart = mcarea+i;
- delta = mcarea+i - p;
- goto patchbranch;
+ } else {
+ lj_assertJ(0, "spare jump out of range: -Osizemcode too big");
}
}
/* Ignore jump slot overflow. Child trace is simply not attached. */
@@ -2491,7 +2801,7 @@
}
} else if (p+1 == pe) {
/* Patch NOP after code for inverted loop branch. Use of J is ok. */
- lua_assert(p[1] == MIPSI_NOP);
+ lj_assertJ(p[1] == MIPSI_NOP, "expected NOP");
p[1] = tjump;
*p = MIPSI_NOP; /* Replace the load of the exit number. */
cstop = p+2;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_ppc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_ppc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_ppc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** PPC IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Register allocator extensions --------------------------------------- */
@@ -156,6 +156,9 @@
return ra_allock(as, ofs-(int16_t)ofs, allow);
}
}
+ } else if (ir->o == IR_TMPREF) {
+ *ofsp = (int32_t)(offsetof(global_State, tmptv)-32768);
+ return RID_JGL;
}
}
*ofsp = 0;
@@ -181,7 +184,7 @@
return;
}
} else if (ir->o == IR_STRREF) {
- lua_assert(ofs == 0);
+ lj_assertA(ofs == 0, "bad usage");
ofs = (int32_t)sizeof(GCstr);
if (irref_isk(ir->op2)) {
ofs += IR(ir->op2)->i;
@@ -226,12 +229,14 @@
emit_tab(as, pi, rt, left, right);
}
+#if !LJ_SOFTFP
/* Fuse to multiply-add/sub instruction. */
static int asm_fusemadd(ASMState *as, IRIns *ir, PPCIns pi, PPCIns pir)
{
IRRef lref = ir->op1, rref = ir->op2;
IRIns *irm;
- if (lref != rref &&
+ if ((as->flags & JIT_F_OPT_FMA) &&
+ lref != rref &&
((mayfuse(as, lref) && (irm = IR(lref), irm->o == IR_MUL) &&
ra_noreg(irm->r)) ||
(mayfuse(as, rref) && (irm = IR(rref), irm->o == IR_MUL) &&
@@ -245,6 +250,7 @@
}
return 0;
}
+#endif
/* -- Calls --------------------------------------------------------------- */
@@ -253,7 +259,10 @@
{
uint32_t n, nargs = CCI_XNARGS(ci);
int32_t ofs = 8;
- Reg gpr = REGARG_FIRSTGPR, fpr = REGARG_FIRSTFPR;
+ Reg gpr = REGARG_FIRSTGPR;
+#if !LJ_SOFTFP
+ Reg fpr = REGARG_FIRSTFPR;
+#endif
if ((void *)ci->func)
emit_call(as, (void *)ci->func);
for (n = 0; n < nargs; n++) { /* Setup args. */
@@ -260,9 +269,11 @@
IRRef ref = args[n];
if (ref) {
IRIns *ir = IR(ref);
+#if !LJ_SOFTFP
if (irt_isfp(ir->t)) {
if (fpr <= REGARG_LASTFPR) {
- lua_assert(rset_test(as->freeset, fpr)); /* Already evicted. */
+ lj_assertA(rset_test(as->freeset, fpr),
+ "reg %d not free", fpr); /* Already evicted. */
ra_leftov(as, fpr, ref);
fpr++;
} else {
@@ -271,9 +282,12 @@
emit_spstore(as, ir, r, ofs);
ofs += irt_isnum(ir->t) ? 8 : 4;
}
- } else {
+ } else
+#endif
+ {
if (gpr <= REGARG_LASTGPR) {
- lua_assert(rset_test(as->freeset, gpr)); /* Already evicted. */
+ lj_assertA(rset_test(as->freeset, gpr),
+ "reg %d not free", gpr); /* Already evicted. */
ra_leftov(as, gpr, ref);
gpr++;
} else {
@@ -290,8 +304,10 @@
}
checkmclim(as);
}
+#if !LJ_SOFTFP
if ((ci->flags & CCI_VARARG)) /* Vararg calls need to know about FPR use. */
emit_tab(as, fpr == REGARG_FIRSTFPR ? PPCI_CRXOR : PPCI_CREQV, 6, 6, 6);
+#endif
}
/* Setup result reg/sp for call. Evict scratch regs. */
@@ -299,8 +315,10 @@
{
RegSet drop = RSET_SCRATCH;
int hiop = ((ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t));
+#if !LJ_SOFTFP
if ((ci->flags & CCI_NOFPRCLOBBER))
drop &= ~RSET_FPR;
+#endif
if (ra_hasreg(ir->r))
rset_clear(drop, ir->r); /* Dest reg handled below. */
if (hiop && ra_hasreg((ir+1)->r))
@@ -307,8 +325,8 @@
rset_clear(drop, (ir+1)->r); /* Dest reg handled below. */
ra_evictset(as, drop); /* Evictions must be performed first. */
if (ra_used(ir)) {
- lua_assert(!irt_ispri(ir->t));
- if (irt_isfp(ir->t)) {
+ lj_assertA(!irt_ispri(ir->t), "PRI dest");
+ if (!LJ_SOFTFP && irt_isfp(ir->t)) {
if ((ci->flags & CCI_CASTU64)) {
/* Use spill slot or temp slots. */
int32_t ofs = ir->s ? sps_scale(ir->s) : SPOFS_TMP;
@@ -323,10 +341,8 @@
} else {
ra_destreg(as, ir, RID_FPRET);
}
-#if LJ_32
} else if (hiop) {
ra_destpair(as, ir);
-#endif
} else {
ra_destreg(as, ir, RID_RET);
}
@@ -375,8 +391,24 @@
emit_tai(as, PPCI_LWZ, RID_TMP, base, -8);
}
+/* -- Buffer operations --------------------------------------------------- */
+
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb)
+{
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irgc;
+ irgc.ot = IRT(0, IRT_PGC); /* GC type. */
+ emit_storeofs(as, &irgc, RID_TMP, sb, offsetof(SBuf, L));
+ emit_rot(as, PPCI_RLWIMI, RID_TMP, tmp, 0, 31-lj_fls(SBUF_MASK_FLAG), 31);
+ emit_getgl(as, RID_TMP, cur_L);
+ emit_loadofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+}
+#endif
+
/* -- Type conversions ---------------------------------------------------- */
+#if !LJ_SOFTFP
static void asm_tointg(ASMState *as, IRIns *ir, Reg left)
{
RegSet allow = RSET_FPR;
@@ -409,15 +441,27 @@
emit_fai(as, PPCI_STFD, tmp, RID_SP, SPOFS_TMP);
emit_fab(as, PPCI_FADD, tmp, left, right);
}
+#endif
static void asm_conv(ASMState *as, IRIns *ir)
{
IRType st = (IRType)(ir->op2 & IRCONV_SRCMASK);
+#if !LJ_SOFTFP
int stfp = (st == IRT_NUM || st == IRT_FLOAT);
+#endif
IRRef lref = ir->op1;
- lua_assert(irt_type(ir->t) != st);
- lua_assert(!(irt_isint64(ir->t) ||
- (st == IRT_I64 || st == IRT_U64))); /* Handled by SPLIT. */
+ /* 64 bit integer conversions are handled by SPLIT. */
+ lj_assertA(!(irt_isint64(ir->t) || (st == IRT_I64 || st == IRT_U64)),
+ "IR %04d has unsplit 64 bit type",
+ (int)(ir - as->ir) - REF_BIAS);
+#if LJ_SOFTFP
+ /* FP conversions are handled by SPLIT. */
+ lj_assertA(!irt_isfp(ir->t) && !(st == IRT_NUM || st == IRT_FLOAT),
+ "IR %04d has FP type",
+ (int)(ir - as->ir) - REF_BIAS);
+ /* Can't check for same types: SPLIT uses CONV int.int + BXOR for sfp NEG. */
+#else
+ lj_assertA(irt_type(ir->t) != st, "inconsistent types for CONV");
if (irt_isfp(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
if (stfp) { /* FP to FP conversion. */
@@ -446,7 +490,8 @@
} else if (stfp) { /* FP to integer conversion. */
if (irt_isguard(ir->t)) {
/* Checked conversions are only supported from number to int. */
- lua_assert(irt_isint(ir->t) && st == IRT_NUM);
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -476,11 +521,13 @@
emit_fb(as, PPCI_FCTIWZ, tmp, left);
}
}
- } else {
+ } else
+#endif
+ {
Reg dest = ra_dest(as, ir, RSET_GPR);
if (st >= IRT_I8 && st <= IRT_U16) { /* Extend to 32 bit integer. */
Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t), "bad type for CONV EXT");
if ((ir->op2 & IRCONV_SEXT))
emit_as(as, st == IRT_I8 ? PPCI_EXTSB : PPCI_EXTSH, dest, left);
else
@@ -496,10 +543,35 @@
{
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_strscan_num];
IRRef args[2];
- int32_t ofs;
+ int32_t ofs = SPOFS_TMP;
+#if LJ_SOFTFP
+ ra_evictset(as, RSET_SCRATCH);
+ if (ra_used(ir)) {
+ if (ra_hasspill(ir->s) && ra_hasspill((ir+1)->s) &&
+ (ir->s & 1) == LJ_BE && (ir->s ^ 1) == (ir+1)->s) {
+ int i;
+ for (i = 0; i < 2; i++) {
+ Reg r = (ir+i)->r;
+ if (ra_hasreg(r)) {
+ ra_free(as, r);
+ ra_modified(as, r);
+ emit_spload(as, ir+i, r, sps_scale((ir+i)->s));
+ }
+ }
+ ofs = sps_scale(ir->s & ~1);
+ } else {
+ Reg rhi = ra_dest(as, ir+1, RSET_GPR);
+ Reg rlo = ra_dest(as, ir, rset_exclude(RSET_GPR, rhi));
+ emit_tai(as, PPCI_LWZ, rhi, RID_SP, ofs);
+ emit_tai(as, PPCI_LWZ, rlo, RID_SP, ofs+4);
+ }
+ }
+#else
RegSet drop = RSET_SCRATCH;
if (ra_hasreg(ir->r)) rset_set(drop, ir->r); /* Spill dest reg (if any). */
ra_evictset(as, drop);
+ if (ir->s) ofs = sps_scale(ir->s);
+#endif
asm_guardcc(as, CC_EQ);
emit_ai(as, PPCI_CMPWI, RID_RET, 0); /* Test return status. */
args[0] = ir->op1; /* GCstr *str */
@@ -506,7 +578,6 @@
args[1] = ASMREF_TMP1; /* TValue *n */
asm_gencall(as, ci, args);
/* Store the result to the spill slot or temp slots. */
- ofs = ir->s ? sps_scale(ir->s) : SPOFS_TMP;
emit_tai(as, PPCI_ADDI, ra_releasetmp(as, ASMREF_TMP1), RID_SP, ofs);
}
@@ -513,25 +584,54 @@
/* -- Memory references --------------------------------------------------- */
/* Get pointer to TValue. */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode)
{
- IRIns *ir = IR(ref);
- if (irt_isnum(ir->t)) {
- if (irref_isk(ref)) /* Use the number constant itself as a TValue. */
- ra_allockreg(as, i32ptr(ir_knum(ir)), dest);
- else /* Otherwise force a spill and use the spill slot. */
- emit_tai(as, PPCI_ADDI, dest, RID_SP, ra_spill(as, ir));
+ int32_t tmpofs = (int32_t)(offsetof(global_State, tmptv)-32768);
+ if ((mode & IRTMPREF_IN1)) {
+ IRIns *ir = IR(ref);
+ if (irt_isnum(ir->t)) {
+ if ((mode & IRTMPREF_OUT1)) {
+#if LJ_SOFTFP
+ lj_assertA(irref_isk(ref), "unsplit FP op");
+ emit_tai(as, PPCI_ADDI, dest, RID_JGL, tmpofs);
+ emit_setgl(as,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.lo, RSET_GPR),
+ tmptv.u32.lo);
+ emit_setgl(as,
+ ra_allock(as, (int32_t)ir_knum(ir)->u32.hi, RSET_GPR),
+ tmptv.u32.hi);
+#else
+ Reg src = ra_alloc1(as, ref, RSET_FPR);
+ emit_tai(as, PPCI_ADDI, dest, RID_JGL, tmpofs);
+ emit_fai(as, PPCI_STFD, src, RID_JGL, tmpofs);
+#endif
+ } else if (irref_isk(ref)) {
+ /* Use the number constant itself as a TValue. */
+ ra_allockreg(as, i32ptr(ir_knum(ir)), dest);
+ } else {
+#if LJ_SOFTFP
+ lj_assertA(0, "unsplit FP op");
+#else
+ /* Otherwise force a spill and use the spill slot. */
+ emit_tai(as, PPCI_ADDI, dest, RID_SP, ra_spill(as, ir));
+#endif
+ }
+ } else {
+ /* Otherwise use g->tmptv to hold the TValue. */
+ Reg type;
+ emit_tai(as, PPCI_ADDI, dest, RID_JGL, tmpofs);
+ if (!irt_ispri(ir->t)) {
+ Reg src = ra_alloc1(as, ref, RSET_GPR);
+ emit_setgl(as, src, tmptv.gcr);
+ }
+ if (LJ_SOFTFP && (ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t))
+ type = ra_alloc1(as, ref+1, RSET_GPR);
+ else
+ type = ra_allock(as, irt_toitype(ir->t), RSET_GPR);
+ emit_setgl(as, type, tmptv.it);
+ }
} else {
- /* Otherwise use g->tmptv to hold the TValue. */
- RegSet allow = rset_exclude(RSET_GPR, dest);
- Reg type;
- emit_tai(as, PPCI_ADDI, dest, RID_JGL, (int32_t)offsetof(global_State, tmptv)-32768);
- if (!irt_ispri(ir->t)) {
- Reg src = ra_alloc1(as, ref, allow);
- emit_setgl(as, src, tmptv.gcr);
- }
- type = ra_allock(as, irt_toitype(ir->t), allow);
- emit_setgl(as, type, tmptv.it);
+ emit_tai(as, PPCI_ADDI, dest, RID_JGL, tmpofs);
}
}
@@ -574,11 +674,27 @@
Reg tisnum = RID_NONE, tmpnum = RID_NONE;
IRRef refkey = ir->op2;
IRIns *irkey = IR(refkey);
+ int isk = irref_isk(refkey);
IRType1 kt = irkey->t;
uint32_t khash;
MCLabel l_end, l_loop, l_next;
rset_clear(allow, tab);
+#if LJ_SOFTFP
+ if (!isk) {
+ key = ra_alloc1(as, refkey, allow);
+ rset_clear(allow, key);
+ if (irkey[1].o == IR_HIOP) {
+ if (ra_hasreg((irkey+1)->r)) {
+ tmpnum = (irkey+1)->r;
+ ra_noweak(as, tmpnum);
+ } else {
+ tmpnum = ra_allocref(as, refkey+1, allow);
+ }
+ rset_clear(allow, tmpnum);
+ }
+ }
+#else
if (irt_isnum(kt)) {
key = ra_alloc1(as, refkey, RSET_FPR);
tmpnum = ra_scratch(as, rset_exclude(RSET_FPR, key));
@@ -588,6 +704,7 @@
key = ra_alloc1(as, refkey, allow);
rset_clear(allow, key);
}
+#endif
tmp2 = ra_scratch(as, allow);
rset_clear(allow, tmp2);
@@ -610,7 +727,7 @@
asm_guardcc(as, CC_EQ);
else
emit_condbranch(as, PPCI_BC|PPCF_Y, CC_EQ, l_end);
- if (irt_isnum(kt)) {
+ if (!LJ_SOFTFP && irt_isnum(kt)) {
emit_fab(as, PPCI_FCMPU, 0, tmpnum, key);
emit_condbranch(as, PPCI_BC, CC_GE, l_next);
emit_ab(as, PPCI_CMPLW, tmp1, tisnum);
@@ -620,7 +737,10 @@
emit_ab(as, PPCI_CMPW, tmp2, key);
emit_condbranch(as, PPCI_BC, CC_NE, l_next);
}
- emit_ai(as, PPCI_CMPWI, tmp1, irt_toitype(irkey->t));
+ if (LJ_SOFTFP && ra_hasreg(tmpnum))
+ emit_ab(as, PPCI_CMPW, tmp1, tmpnum);
+ else
+ emit_ai(as, PPCI_CMPWI, tmp1, irt_toitype(irkey->t));
if (!irt_ispri(kt))
emit_tai(as, PPCI_LWZ, tmp2, dest, (int32_t)offsetof(Node, key.gcr));
}
@@ -629,12 +749,12 @@
(((char *)as->mcp-(char *)l_loop) & 0xffffu);
/* Load main position relative to tab->node into dest. */
- khash = irref_isk(refkey) ? ir_khash(irkey) : 1;
+ khash = isk ? ir_khash(as, irkey) : 1;
if (khash == 0) {
emit_tai(as, PPCI_LWZ, dest, tab, (int32_t)offsetof(GCtab, node));
} else {
Reg tmphash = tmp1;
- if (irref_isk(refkey))
+ if (isk)
tmphash = ra_allock(as, khash, allow);
emit_tab(as, PPCI_ADD, dest, dest, tmp1);
emit_tai(as, PPCI_MULLI, tmp1, tmp1, sizeof(Node));
@@ -641,10 +761,10 @@
emit_asb(as, PPCI_AND, tmp1, tmp2, tmphash);
emit_tai(as, PPCI_LWZ, dest, tab, (int32_t)offsetof(GCtab, node));
emit_tai(as, PPCI_LWZ, tmp2, tab, (int32_t)offsetof(GCtab, hmask));
- if (irref_isk(refkey)) {
+ if (isk) {
/* Nothing to do. */
} else if (irt_isstr(kt)) {
- emit_tai(as, PPCI_LWZ, tmp1, key, (int32_t)offsetof(GCstr, hash));
+ emit_tai(as, PPCI_LWZ, tmp1, key, (int32_t)offsetof(GCstr, sid));
} else { /* Must match with hash*() in lj_tab.c. */
emit_tab(as, PPCI_SUBF, tmp1, tmp2, tmp1);
emit_rotlwi(as, tmp2, tmp2, HASH_ROT3);
@@ -651,7 +771,12 @@
emit_asb(as, PPCI_XOR, tmp1, tmp1, tmp2);
emit_rotlwi(as, tmp1, tmp1, (HASH_ROT2+HASH_ROT1)&31);
emit_tab(as, PPCI_SUBF, tmp2, dest, tmp2);
- if (irt_isnum(kt)) {
+ if (LJ_SOFTFP ? (irkey[1].o == IR_HIOP) : irt_isnum(kt)) {
+#if LJ_SOFTFP
+ emit_asb(as, PPCI_XOR, tmp2, key, tmp1);
+ emit_rotlwi(as, dest, tmp1, HASH_ROT1);
+ emit_tab(as, PPCI_ADD, tmp1, tmpnum, tmpnum);
+#else
int32_t ofs = ra_spill(as, irkey);
emit_asb(as, PPCI_XOR, tmp2, tmp2, tmp1);
emit_rotlwi(as, dest, tmp1, HASH_ROT1);
@@ -658,6 +783,7 @@
emit_tab(as, PPCI_ADD, tmp1, tmp1, tmp1);
emit_tai(as, PPCI_LWZ, tmp2, RID_SP, ofs+4);
emit_tai(as, PPCI_LWZ, tmp1, RID_SP, ofs);
+#endif
} else {
emit_asb(as, PPCI_XOR, tmp2, key, tmp1);
emit_rotlwi(as, dest, tmp1, HASH_ROT1);
@@ -678,7 +804,7 @@
Reg node = ra_alloc1(as, ir->op1, RSET_GPR);
Reg key = RID_NONE, type = RID_TMP, idx = node;
RegSet allow = rset_exclude(RSET_GPR, node);
- lua_assert(ofs % sizeof(Node) == 0);
+ lj_assertA(ofs % sizeof(Node) == 0, "unaligned HREFK slot");
if (ofs > 32736) {
idx = dest;
rset_clear(allow, dest);
@@ -714,23 +840,30 @@
static void asm_uref(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (irref_isk(ir->op1)) {
+ int guarded = (irt_t(ir->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC);
+ if (irref_isk(ir->op1) && !guarded) {
GCfunc *fn = ir_kfunc(IR(ir->op1));
MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
emit_lsptr(as, PPCI_LWZ, dest, v, RSET_GPR);
} else {
- Reg uv = ra_scratch(as, RSET_GPR);
- Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
- if (ir->o == IR_UREFC) {
- asm_guardcc(as, CC_NE);
+ if (guarded) {
+ asm_guardcc(as, ir->o == IR_UREFC ? CC_NE : CC_EQ);
emit_ai(as, PPCI_CMPWI, RID_TMP, 1);
- emit_tai(as, PPCI_ADDI, dest, uv, (int32_t)offsetof(GCupval, tv));
- emit_tai(as, PPCI_LBZ, RID_TMP, uv, (int32_t)offsetof(GCupval, closed));
+ }
+ if (ir->o == IR_UREFC)
+ emit_tai(as, PPCI_ADDI, dest, dest, (int32_t)offsetof(GCupval, tv));
+ else
+ emit_tai(as, PPCI_LWZ, dest, dest, (int32_t)offsetof(GCupval, v));
+ if (guarded)
+ emit_tai(as, PPCI_LBZ, RID_TMP, dest, (int32_t)offsetof(GCupval, closed));
+ if (irref_isk(ir->op1)) {
+ GCfunc *fn = ir_kfunc(IR(ir->op1));
+ int32_t k = (int32_t)gcrefu(fn->l.uvptr[(ir->op2 >> 8)]);
+ emit_loadi(as, dest, k);
} else {
- emit_tai(as, PPCI_LWZ, dest, uv, (int32_t)offsetof(GCupval, v));
+ emit_tai(as, PPCI_LWZ, dest, ra_alloc1(as, ir->op1, RSET_GPR),
+ (int32_t)offsetof(GCfuncL, uvptr) + 4*(int32_t)(ir->op2 >> 8));
}
- emit_tai(as, PPCI_LWZ, uv, func,
- (int32_t)offsetof(GCfuncL, uvptr) + 4*(int32_t)(ir->op2 >> 8));
}
}
@@ -737,7 +870,7 @@
static void asm_fref(ASMState *as, IRIns *ir)
{
UNUSED(as); UNUSED(ir);
- lua_assert(!ra_used(ir));
+ lj_assertA(!ra_used(ir), "unfused FREF");
}
static void asm_strref(ASMState *as, IRIns *ir)
@@ -777,26 +910,28 @@
/* -- Loads and stores ---------------------------------------------------- */
-static PPCIns asm_fxloadins(IRIns *ir)
+static PPCIns asm_fxloadins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: return PPCI_LBZ; /* Needs sign-extension. */
case IRT_U8: return PPCI_LBZ;
case IRT_I16: return PPCI_LHA;
case IRT_U16: return PPCI_LHZ;
- case IRT_NUM: return PPCI_LFD;
- case IRT_FLOAT: return PPCI_LFS;
+ case IRT_NUM: lj_assertA(!LJ_SOFTFP, "unsplit FP op"); return PPCI_LFD;
+ case IRT_FLOAT: if (!LJ_SOFTFP) return PPCI_LFS;
default: return PPCI_LWZ;
}
}
-static PPCIns asm_fxstoreins(IRIns *ir)
+static PPCIns asm_fxstoreins(ASMState *as, IRIns *ir)
{
+ UNUSED(as);
switch (irt_type(ir->t)) {
case IRT_I8: case IRT_U8: return PPCI_STB;
case IRT_I16: case IRT_U16: return PPCI_STH;
- case IRT_NUM: return PPCI_STFD;
- case IRT_FLOAT: return PPCI_STFS;
+ case IRT_NUM: lj_assertA(!LJ_SOFTFP, "unsplit FP op"); return PPCI_STFD;
+ case IRT_FLOAT: if (!LJ_SOFTFP) return PPCI_STFS;
default: return PPCI_STW;
}
}
@@ -804,12 +939,12 @@
static void asm_fload(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- PPCIns pi = asm_fxloadins(ir);
+ PPCIns pi = asm_fxloadins(as, ir);
Reg idx;
int32_t ofs;
- if (ir->op1 == REF_NIL) {
+ if (ir->op1 == REF_NIL) { /* FLOAD from GG_State with offset. */
idx = RID_JGL;
- ofs = (ir->op2 << 2) - 32768;
+ ofs = (ir->op2 << 2) - 32768 - GG_OFS(g);
} else {
idx = ra_alloc1(as, ir->op1, RSET_GPR);
if (ir->op2 == IRFL_TAB_ARRAY) {
@@ -821,7 +956,7 @@
}
ofs = field_ofs[ir->op2];
}
- lua_assert(!irt_isi8(ir->t));
+ lj_assertA(!irt_isi8(ir->t), "unsupported FLOAD I8");
emit_tai(as, pi, dest, idx, ofs);
}
@@ -832,7 +967,7 @@
IRIns *irf = IR(ir->op1);
Reg idx = ra_alloc1(as, irf->op1, rset_exclude(RSET_GPR, src));
int32_t ofs = field_ofs[irf->op2];
- PPCIns pi = asm_fxstoreins(ir);
+ PPCIns pi = asm_fxstoreins(as, ir);
emit_tai(as, pi, src, idx, ofs);
}
}
@@ -839,11 +974,12 @@
static void asm_xload(ASMState *as, IRIns *ir)
{
- Reg dest = ra_dest(as, ir, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR);
- lua_assert(!(ir->op2 & IRXLOAD_UNALIGNED));
+ Reg dest = ra_dest(as, ir,
+ (!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
+ lj_assertA(!(ir->op2 & IRXLOAD_UNALIGNED), "unaligned XLOAD");
if (irt_isi8(ir->t))
emit_as(as, PPCI_EXTSB, dest, dest);
- asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR, 0);
+ asm_fusexref(as, asm_fxloadins(as, ir), dest, ir->op1, RSET_GPR, 0);
}
static void asm_xstore_(ASMState *as, IRIns *ir, int32_t ofs)
@@ -857,8 +993,9 @@
Reg src = ra_alloc1(as, irb->op1, RSET_GPR);
asm_fusexrefx(as, PPCI_STWBRX, src, ir->op1, rset_exclude(RSET_GPR, src));
} else {
- Reg src = ra_alloc1(as, ir->op2, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR);
- asm_fusexref(as, asm_fxstoreins(ir), src, ir->op1,
+ Reg src = ra_alloc1(as, ir->op2,
+ (!LJ_SOFTFP && irt_isfp(ir->t)) ? RSET_FPR : RSET_GPR);
+ asm_fusexref(as, asm_fxstoreins(as, ir), src, ir->op1,
rset_exclude(RSET_GPR, src), ofs);
}
}
@@ -871,24 +1008,39 @@
Reg dest = RID_NONE, type = RID_TMP, tmp = RID_TMP, idx;
RegSet allow = RSET_GPR;
int32_t ofs = AHUREF_LSX;
+ if (LJ_SOFTFP && (ir+1)->o == IR_HIOP) {
+ t.irt = IRT_NUM;
+ if (ra_used(ir+1)) {
+ type = ra_dest(as, ir+1, allow);
+ rset_clear(allow, type);
+ }
+ ofs = 0;
+ }
if (ra_used(ir)) {
- lua_assert(irt_isnum(t) || irt_isint(t) || irt_isaddr(t));
- if (!irt_isnum(t)) ofs = 0;
- dest = ra_dest(as, ir, irt_isnum(t) ? RSET_FPR : RSET_GPR);
+ lj_assertA((LJ_SOFTFP ? 0 : irt_isnum(ir->t)) ||
+ irt_isint(ir->t) || irt_isaddr(ir->t),
+ "bad load type %d", irt_type(ir->t));
+ if (LJ_SOFTFP || !irt_isnum(t)) ofs = 0;
+ dest = ra_dest(as, ir, (!LJ_SOFTFP && irt_isnum(t)) ? RSET_FPR : allow);
rset_clear(allow, dest);
}
idx = asm_fuseahuref(as, ir->op1, &ofs, allow);
+ if (ir->o == IR_VLOAD) {
+ ofs = ofs != AHUREF_LSX ? ofs + 8 * ir->op2 :
+ ir->op2 ? 8 * ir->op2 : AHUREF_LSX;
+ }
if (irt_isnum(t)) {
Reg tisnum = ra_allock(as, (int32_t)LJ_TISNUM, rset_exclude(allow, idx));
asm_guardcc(as, CC_GE);
emit_ab(as, PPCI_CMPLW, type, tisnum);
if (ra_hasreg(dest)) {
- if (ofs == AHUREF_LSX) {
+ if (!LJ_SOFTFP && ofs == AHUREF_LSX) {
tmp = ra_scratch(as, rset_exclude(rset_exclude(RSET_GPR,
(idx&255)), (idx>>8)));
emit_fab(as, PPCI_LFDX, dest, (idx&255), tmp);
} else {
- emit_fai(as, PPCI_LFD, dest, idx, ofs);
+ emit_fai(as, LJ_SOFTFP ? PPCI_LWZ : PPCI_LFD, dest, idx,
+ ofs+4*LJ_SOFTFP);
}
}
} else {
@@ -911,7 +1063,7 @@
int32_t ofs = AHUREF_LSX;
if (ir->r == RID_SINK)
return;
- if (irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP && irt_isnum(ir->t)) {
src = ra_alloc1(as, ir->op2, RSET_FPR);
} else {
if (!irt_ispri(ir->t)) {
@@ -919,11 +1071,14 @@
rset_clear(allow, src);
ofs = 0;
}
- type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
+ if (LJ_SOFTFP && (ir+1)->o == IR_HIOP)
+ type = ra_alloc1(as, (ir+1)->op2, allow);
+ else
+ type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
rset_clear(allow, type);
}
idx = asm_fuseahuref(as, ir->op1, &ofs, allow);
- if (irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP && irt_isnum(ir->t)) {
if (ofs == AHUREF_LSX) {
emit_fab(as, PPCI_STFDX, src, (idx&255), RID_TMP);
emit_slwi(as, RID_TMP, (idx>>8), 3);
@@ -948,21 +1103,39 @@
IRType1 t = ir->t;
Reg dest = RID_NONE, type = RID_NONE, base;
RegSet allow = RSET_GPR;
- lua_assert(!(ir->op2 & IRSLOAD_PARENT)); /* Handled by asm_head_side(). */
- lua_assert(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK));
- lua_assert(LJ_DUALNUM ||
- !irt_isint(t) || (ir->op2 & (IRSLOAD_CONVERT|IRSLOAD_FRAME)));
+ int hiop = (LJ_SOFTFP && (ir+1)->o == IR_HIOP);
+ if (hiop)
+ t.irt = IRT_NUM;
+ lj_assertA(!(ir->op2 & IRSLOAD_PARENT),
+ "bad parent SLOAD"); /* Handled by asm_head_side(). */
+ lj_assertA(irt_isguard(ir->t) || !(ir->op2 & IRSLOAD_TYPECHECK),
+ "inconsistent SLOAD variant");
+ lj_assertA(LJ_DUALNUM ||
+ !irt_isint(t) ||
+ (ir->op2 & (IRSLOAD_CONVERT|IRSLOAD_FRAME|IRSLOAD_KEYINDEX)),
+ "bad SLOAD type");
+#if LJ_SOFTFP
+ lj_assertA(!(ir->op2 & IRSLOAD_CONVERT),
+ "unsplit SLOAD convert"); /* Handled by LJ_SOFTFP SPLIT. */
+ if (hiop && ra_used(ir+1)) {
+ type = ra_dest(as, ir+1, allow);
+ rset_clear(allow, type);
+ }
+#else
if ((ir->op2 & IRSLOAD_CONVERT) && irt_isguard(t) && irt_isint(t)) {
dest = ra_scratch(as, RSET_FPR);
asm_tointg(as, ir, dest);
t.irt = IRT_NUM; /* Continue with a regular number type check. */
- } else if (ra_used(ir)) {
- lua_assert(irt_isnum(t) || irt_isint(t) || irt_isaddr(t));
- dest = ra_dest(as, ir, irt_isnum(t) ? RSET_FPR : RSET_GPR);
+ } else
+#endif
+ if (ra_used(ir)) {
+ lj_assertA(irt_isnum(t) || irt_isint(t) || irt_isaddr(t),
+ "bad SLOAD type %d", irt_type(ir->t));
+ dest = ra_dest(as, ir, (!LJ_SOFTFP && irt_isnum(t)) ? RSET_FPR : allow);
rset_clear(allow, dest);
base = ra_alloc1(as, REF_BASE, allow);
rset_clear(allow, base);
- if ((ir->op2 & IRSLOAD_CONVERT)) {
+ if (!LJ_SOFTFP && (ir->op2 & IRSLOAD_CONVERT)) {
if (irt_isint(t)) {
emit_tai(as, PPCI_LWZ, dest, RID_SP, SPOFS_TMPLO);
dest = ra_scratch(as, RSET_FPR);
@@ -994,14 +1167,22 @@
if ((ir->op2 & IRSLOAD_TYPECHECK)) {
Reg tisnum = ra_allock(as, (int32_t)LJ_TISNUM, allow);
asm_guardcc(as, CC_GE);
- emit_ab(as, PPCI_CMPLW, RID_TMP, tisnum);
+#if !LJ_SOFTFP
type = RID_TMP;
+#endif
+ emit_ab(as, PPCI_CMPLW, type, tisnum);
}
- if (ra_hasreg(dest)) emit_fai(as, PPCI_LFD, dest, base, ofs-4);
+ if (ra_hasreg(dest)) emit_fai(as, LJ_SOFTFP ? PPCI_LWZ : PPCI_LFD, dest,
+ base, ofs-(LJ_SOFTFP?0:4));
} else {
if ((ir->op2 & IRSLOAD_TYPECHECK)) {
asm_guardcc(as, CC_NE);
- emit_ai(as, PPCI_CMPWI, RID_TMP, irt_toitype(t));
+ if ((ir->op2 & IRSLOAD_KEYINDEX)) {
+ emit_ai(as, PPCI_CMPWI, RID_TMP, (LJ_KEYINDEX & 0xffff));
+ emit_asi(as, PPCI_XORIS, RID_TMP, RID_TMP, (LJ_KEYINDEX >> 16));
+ } else {
+ emit_ai(as, PPCI_CMPWI, RID_TMP, irt_toitype(t));
+ }
type = RID_TMP;
}
if (ra_hasreg(dest)) emit_tai(as, PPCI_LWZ, dest, base, ofs);
@@ -1021,7 +1202,8 @@
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
IRRef args[4];
RegSet drop = RSET_SCRATCH;
- lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
+ lj_assertA(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL),
+ "bad CNEW/CNEWI operands");
as->gcsteps++;
if (ra_hasreg(ir->r))
@@ -1034,10 +1216,10 @@
if (ir->o == IR_CNEWI) {
RegSet allow = (RSET_GPR & ~RSET_SCRATCH);
int32_t ofs = sizeof(GCcdata);
- lua_assert(sz == 4 || sz == 8);
+ lj_assertA(sz == 4 || sz == 8, "bad CNEWI size %d", sz);
if (sz == 8) {
ofs += 4;
- lua_assert((ir+1)->o == IR_HIOP);
+ lj_assertA((ir+1)->o == IR_HIOP, "expected HIOP for CNEWI");
}
for (;;) {
Reg r = ra_alloc1(as, ir->op2, allow);
@@ -1068,8 +1250,6 @@
ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
ra_releasetmp(as, ASMREF_TMP1));
}
-#else
-#define asm_cnew(as, ir) ((void)0)
#endif
/* -- Write barriers ------------------------------------------------------ */
@@ -1083,7 +1263,7 @@
emit_tai(as, PPCI_STW, link, tab, (int32_t)offsetof(GCtab, gclist));
emit_tai(as, PPCI_STB, mark, tab, (int32_t)offsetof(GCtab, marked));
emit_setgl(as, tab, gc.grayagain);
- lua_assert(LJ_GC_BLACK == 0x04);
+ lj_assertA(LJ_GC_BLACK == 0x04, "bad LJ_GC_BLACK");
emit_rot(as, PPCI_RLWINM, mark, mark, 0, 30, 28); /* Clear black bit. */
emit_getgl(as, link, gc.grayagain);
emit_condbranch(as, PPCI_BC|PPCF_Y, CC_EQ, l_end);
@@ -1098,7 +1278,7 @@
MCLabel l_end;
Reg obj, val, tmp;
/* No need for other object barriers (yet). */
- lua_assert(IR(ir->op1)->o == IR_UREFC);
+ lj_assertA(IR(ir->op1)->o == IR_UREFC, "bad OBAR type");
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
args[0] = ASMREF_TMP1; /* global_State *g */
@@ -1119,6 +1299,7 @@
/* -- Arithmetic and logic operations ------------------------------------- */
+#if !LJ_SOFTFP
static void asm_fparith(ASMState *as, IRIns *ir, PPCIns pi)
{
Reg dest = ra_dest(as, ir, RSET_FPR);
@@ -1139,20 +1320,22 @@
static void asm_fpmath(ASMState *as, IRIns *ir)
{
- if (ir->op2 == IRFPM_EXP2 && asm_fpjoin_pow(as, ir))
- return;
if (ir->op2 == IRFPM_SQRT && (as->flags & JIT_F_SQRT))
asm_fpunary(as, ir, PPCI_FSQRT);
else
asm_callid(as, ir, IRCALL_lj_vm_floor + ir->op2);
}
+#endif
static void asm_add(ASMState *as, IRIns *ir)
{
+#if !LJ_SOFTFP
if (irt_isnum(ir->t)) {
if (!asm_fusemadd(as, ir, PPCI_FMADD, PPCI_FMADD))
asm_fparith(as, ir, PPCI_FADD);
- } else {
+ } else
+#endif
+ {
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg right, left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
PPCIns pi;
@@ -1191,10 +1374,13 @@
static void asm_sub(ASMState *as, IRIns *ir)
{
+#if !LJ_SOFTFP
if (irt_isnum(ir->t)) {
if (!asm_fusemadd(as, ir, PPCI_FMSUB, PPCI_FNMSUB))
asm_fparith(as, ir, PPCI_FSUB);
- } else {
+ } else
+#endif
+ {
PPCIns pi = PPCI_SUBF;
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg left, right;
@@ -1220,9 +1406,12 @@
static void asm_mul(ASMState *as, IRIns *ir)
{
+#if !LJ_SOFTFP
if (irt_isnum(ir->t)) {
asm_fparith(as, ir, PPCI_FMUL);
- } else {
+ } else
+#endif
+ {
PPCIns pi = PPCI_MULLW;
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg right, left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
@@ -1244,15 +1433,16 @@
}
}
-#define asm_div(as, ir) asm_fparith(as, ir, PPCI_FDIV)
-#define asm_mod(as, ir) asm_callid(as, ir, IRCALL_lj_vm_modi)
-#define asm_pow(as, ir) asm_callid(as, ir, IRCALL_lj_vm_powi)
+#define asm_fpdiv(as, ir) asm_fparith(as, ir, PPCI_FDIV)
static void asm_neg(ASMState *as, IRIns *ir)
{
+#if !LJ_SOFTFP
if (irt_isnum(ir->t)) {
asm_fpunary(as, ir, PPCI_FNEG);
- } else {
+ } else
+#endif
+ {
Reg dest, left;
PPCIns pi = PPCI_NEG;
if (as->flagmcp == as->mcp) {
@@ -1267,8 +1457,6 @@
}
#define asm_abs(as, ir) asm_fpunary(as, ir, PPCI_FABS)
-#define asm_atan2(as, ir) asm_callid(as, ir, IRCALL_atan2)
-#define asm_ldexp(as, ir) asm_callid(as, ir, IRCALL_ldexp)
static void asm_arithov(ASMState *as, IRIns *ir, PPCIns pi)
{
@@ -1561,11 +1749,42 @@
#define asm_brol(as, ir) \
asm_bitshift(as, ir, PPCI_RLWNM|PPCF_MB(0)|PPCF_ME(31), \
PPCI_RLWINM|PPCF_MB(0)|PPCF_ME(31))
-#define asm_bror(as, ir) lua_assert(0)
+#define asm_bror(as, ir) lj_assertA(0, "unexpected BROR")
+#if LJ_SOFTFP
+static void asm_sfpmin_max(ASMState *as, IRIns *ir)
+{
+ CCallInfo ci = lj_ir_callinfo[IRCALL_softfp_cmp];
+ IRRef args[4];
+ MCLabel l_right, l_end;
+ Reg desthi = ra_dest(as, ir, RSET_GPR), destlo = ra_dest(as, ir+1, RSET_GPR);
+ Reg righthi, lefthi = ra_alloc2(as, ir, RSET_GPR);
+ Reg rightlo, leftlo = ra_alloc2(as, ir+1, RSET_GPR);
+ PPCCC cond = (IROp)ir->o == IR_MIN ? CC_EQ : CC_NE;
+ righthi = (lefthi >> 8); lefthi &= 255;
+ rightlo = (leftlo >> 8); leftlo &= 255;
+ args[0^LJ_BE] = ir->op1; args[1^LJ_BE] = (ir+1)->op1;
+ args[2^LJ_BE] = ir->op2; args[3^LJ_BE] = (ir+1)->op2;
+ l_end = emit_label(as);
+ if (desthi != righthi) emit_mr(as, desthi, righthi);
+ if (destlo != rightlo) emit_mr(as, destlo, rightlo);
+ l_right = emit_label(as);
+ if (l_end != l_right) emit_jmp(as, l_end);
+ if (desthi != lefthi) emit_mr(as, desthi, lefthi);
+ if (destlo != leftlo) emit_mr(as, destlo, leftlo);
+ if (l_right == as->mcp+1) {
+ cond ^= 4; l_right = l_end; ++as->mcp;
+ }
+ emit_condbranch(as, PPCI_BC, cond, l_right);
+ ra_evictset(as, RSET_SCRATCH);
+ emit_cmpi(as, RID_RET, 1);
+ asm_gencall(as, &ci, args);
+}
+#endif
+
static void asm_min_max(ASMState *as, IRIns *ir, int ismax)
{
- if (irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP && irt_isnum(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
Reg tmp = dest;
Reg right, left = ra_alloc2(as, ir, RSET_FPR);
@@ -1573,9 +1792,8 @@
if (tmp == left || tmp == right)
tmp = ra_scratch(as, rset_exclude(rset_exclude(rset_exclude(RSET_FPR,
dest), left), right));
- emit_facb(as, PPCI_FSEL, dest, tmp,
- ismax ? left : right, ismax ? right : left);
- emit_fab(as, PPCI_FSUB, tmp, left, right);
+ emit_facb(as, PPCI_FSEL, dest, tmp, left, right);
+ emit_fab(as, PPCI_FSUB, tmp, ismax ? left : right, ismax ? right : left);
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
Reg tmp1 = RID_TMP, tmp2 = dest;
@@ -1653,7 +1871,7 @@
static void asm_comp(ASMState *as, IRIns *ir)
{
PPCCC cc = asm_compmap[ir->o];
- if (irt_isnum(ir->t)) {
+ if (!LJ_SOFTFP && irt_isnum(ir->t)) {
Reg right, left = ra_alloc2(as, ir, RSET_FPR);
right = (left >> 8); left &= 255;
asm_guardcc(as, (cc >> 4));
@@ -1674,6 +1892,44 @@
#define asm_equal(as, ir) asm_comp(as, ir)
+#if LJ_SOFTFP
+/* SFP comparisons. */
+static void asm_sfpcomp(ASMState *as, IRIns *ir)
+{
+ const CCallInfo *ci = &lj_ir_callinfo[IRCALL_softfp_cmp];
+ RegSet drop = RSET_SCRATCH;
+ Reg r;
+ IRRef args[4];
+ args[0^LJ_BE] = ir->op1; args[1^LJ_BE] = (ir+1)->op1;
+ args[2^LJ_BE] = ir->op2; args[3^LJ_BE] = (ir+1)->op2;
+
+ for (r = REGARG_FIRSTGPR; r <= REGARG_FIRSTGPR+3; r++) {
+ if (!rset_test(as->freeset, r) &&
+ regcost_ref(as->cost[r]) == args[r-REGARG_FIRSTGPR])
+ rset_clear(drop, r);
+ }
+ ra_evictset(as, drop);
+ asm_setupresult(as, ir, ci);
+ switch ((IROp)ir->o) {
+ case IR_ULT:
+ asm_guardcc(as, CC_EQ);
+ emit_ai(as, PPCI_CMPWI, RID_RET, 0);
+ case IR_ULE:
+ asm_guardcc(as, CC_EQ);
+ emit_ai(as, PPCI_CMPWI, RID_RET, 1);
+ break;
+ case IR_GE: case IR_GT:
+ asm_guardcc(as, CC_EQ);
+ emit_ai(as, PPCI_CMPWI, RID_RET, 2);
+ default:
+ asm_guardcc(as, (asm_compmap[ir->o] & 0xf));
+ emit_ai(as, PPCI_CMPWI, RID_RET, 0);
+ break;
+ }
+ asm_gencall(as, ci, args);
+}
+#endif
+
#if LJ_HASFFI
/* 64 bit integer comparisons. */
static void asm_comp64(ASMState *as, IRIns *ir)
@@ -1698,24 +1954,41 @@
}
#endif
-/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
+/* -- Split register ops -------------------------------------------------- */
-/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
+/* Hiword op of a split 32/32 bit op. Previous op is be the loword op. */
static void asm_hiop(ASMState *as, IRIns *ir)
{
-#if LJ_HASFFI
/* HIOP is marked as a store because it needs its own DCE logic. */
int uselo = ra_used(ir-1), usehi = ra_used(ir); /* Loword/hiword used? */
if (LJ_UNLIKELY(!(as->flags & JIT_F_OPT_DCE))) uselo = usehi = 1;
+#if LJ_HASFFI || LJ_SOFTFP
if ((ir-1)->o == IR_CONV) { /* Conversions to/from 64 bit. */
as->curins--; /* Always skip the CONV. */
+#if LJ_HASFFI && !LJ_SOFTFP
if (usehi || uselo)
asm_conv64(as, ir);
return;
+#endif
} else if ((ir-1)->o <= IR_NE) { /* 64 bit integer comparisons. ORDER IR. */
as->curins--; /* Always skip the loword comparison. */
+#if LJ_SOFTFP
+ if (!irt_isint(ir->t)) {
+ asm_sfpcomp(as, ir-1);
+ return;
+ }
+#endif
+#if LJ_HASFFI
asm_comp64(as, ir);
+#endif
return;
+#if LJ_SOFTFP
+ } else if ((ir-1)->o == IR_MIN || (ir-1)->o == IR_MAX) {
+ as->curins--; /* Always skip the loword min/max. */
+ if (uselo || usehi)
+ asm_sfpmin_max(as, ir-1);
+ return;
+#endif
} else if ((ir-1)->o == IR_XSTORE) {
as->curins--; /* Handle both stores here. */
if ((ir-1)->r != RID_SINK) {
@@ -1724,24 +1997,33 @@
}
return;
}
+#endif
if (!usehi) return; /* Skip unused hiword op for all remaining ops. */
switch ((ir-1)->o) {
+#if LJ_HASFFI
case IR_ADD: as->curins--; asm_add64(as, ir); break;
case IR_SUB: as->curins--; asm_sub64(as, ir); break;
case IR_NEG: as->curins--; asm_neg64(as, ir); break;
- case IR_CALLN:
- case IR_CALLXS:
+ case IR_CNEWI:
+ /* Nothing to do here. Handled by lo op itself. */
+ break;
+#endif
+#if LJ_SOFTFP
+ case IR_SLOAD: case IR_ALOAD: case IR_HLOAD: case IR_ULOAD: case IR_VLOAD:
+ case IR_STRTO:
if (!uselo)
- ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
+ ra_allocref(as, ir->op1, RSET_GPR); /* Mark lo op as used. */
break;
- case IR_CNEWI:
+ case IR_ASTORE: case IR_HSTORE: case IR_USTORE: case IR_TOSTR: case IR_TMPREF:
/* Nothing to do here. Handled by lo op itself. */
break;
- default: lua_assert(0); break;
+#endif
+ case IR_CALLN: case IR_CALLL: case IR_CALLS: case IR_CALLXS:
+ if (!uselo)
+ ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
+ break;
+ default: lj_assertA(0, "bad HIOP for op %d", (ir-1)->o); break;
}
-#else
- UNUSED(as); UNUSED(ir); lua_assert(0); /* Unused without FFI. */
-#endif
}
/* -- Profiling ----------------------------------------------------------- */
@@ -1797,12 +2079,25 @@
if ((sn & SNAP_NORESTORE))
continue;
if (irt_isnum(ir->t)) {
+#if LJ_SOFTFP
+ Reg tmp;
+ RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
+ /* LJ_SOFTFP: must be a number constant. */
+ lj_assertA(irref_isk(ref), "unsplit FP op");
+ tmp = ra_allock(as, (int32_t)ir_knum(ir)->u32.lo, allow);
+ emit_tai(as, PPCI_STW, tmp, RID_BASE, ofs+(LJ_BE?4:0));
+ if (rset_test(as->freeset, tmp+1)) allow = RID2RSET(tmp+1);
+ tmp = ra_allock(as, (int32_t)ir_knum(ir)->u32.hi, allow);
+ emit_tai(as, PPCI_STW, tmp, RID_BASE, ofs+(LJ_BE?0:4));
+#else
Reg src = ra_alloc1(as, ref, RSET_FPR);
emit_fai(as, PPCI_STFD, src, RID_BASE, ofs);
+#endif
} else {
Reg type;
RegSet allow = rset_exclude(RSET_GPR, RID_BASE);
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t),
+ "restore of IR type %d", irt_type(ir->t));
if (!irt_ispri(ir->t)) {
Reg src = ra_alloc1(as, ref, allow);
rset_clear(allow, src);
@@ -1811,6 +2106,12 @@
if ((sn & (SNAP_CONT|SNAP_FRAME))) {
if (s == 0) continue; /* Do not overwrite link to previous frame. */
type = ra_allock(as, (int32_t)(*flinks--), allow);
+#if LJ_SOFTFP
+ } else if ((sn & SNAP_SOFTFPNUM)) {
+ type = ra_alloc1(as, ref+1, rset_exclude(RSET_GPR, RID_BASE));
+#endif
+ } else if ((sn & SNAP_KEYINDEX)) {
+ type = ra_allock(as, (int32_t)LJ_KEYINDEX, allow);
} else {
type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
}
@@ -1818,11 +2119,14 @@
}
checkmclim(as);
}
- lua_assert(map + nent == flinks);
+ lj_assertA(map + nent == flinks, "inconsistent frames in snapshot");
}
/* -- GC handling --------------------------------------------------------- */
+/* Marker to prevent patching the GC check exit. */
+#define PPC_NOPATCH_GC_CHECK PPCI_ORIS
+
/* Check GC threshold and do one or more GC steps. */
static void asm_gc_check(ASMState *as)
{
@@ -1834,6 +2138,7 @@
l_end = emit_label(as);
/* Exit trace if in GCSatomic or GCSfinalize. Avoids syncing GC objects. */
asm_guardcc(as, CC_NE); /* Assumes asm_snap_prep() already done. */
+ *--as->mcp = PPC_NOPATCH_GC_CHECK;
emit_ai(as, PPCI_CMPWI, RID_RET, 0);
args[0] = ASMREF_TMP1; /* global_State *g */
args[1] = ASMREF_TMP2; /* MSize steps */
@@ -1865,6 +2170,12 @@
}
}
+/* Fixup the tail of the loop. */
+static void asm_loop_tail_fixup(ASMState *as)
+{
+ UNUSED(as); /* Nothing to do. */
+}
+
/* -- Head of trace ------------------------------------------------------- */
/* Coalesce BASE register for a root trace. */
@@ -1882,7 +2193,7 @@
}
/* Coalesce BASE register for a side trace. */
-static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
+static Reg asm_head_side_base(ASMState *as, IRIns *irp)
{
IRIns *ir = IR(REF_BASE);
Reg r = ir->r;
@@ -1891,15 +2202,15 @@
if (rset_test(as->modset, r) || irt_ismarked(ir->t))
ir->r = RID_INIT; /* No inheritance for modified BASE register. */
if (irp->r == r) {
- rset_clear(allow, r); /* Mark same BASE register as coalesced. */
+ return r; /* Same BASE register already coalesced. */
} else if (ra_hasreg(irp->r) && rset_test(as->freeset, irp->r)) {
- rset_clear(allow, irp->r);
emit_mr(as, r, irp->r); /* Move from coalesced parent reg. */
+ return irp->r;
} else {
emit_getgl(as, r, jit_base); /* Otherwise reload BASE. */
}
}
- return allow;
+ return RID_NONE;
}
/* -- Tail of trace ------------------------------------------------------- */
@@ -1916,7 +2227,7 @@
as->mctop = p;
} else {
/* Patch stack adjustment. */
- lua_assert(checki16(CFRAME_SIZE+spadj));
+ lj_assertA(checki16(CFRAME_SIZE+spadj), "stack adjustment out of range");
p[-3] = PPCI_ADDI | PPCF_T(RID_TMP) | PPCF_A(RID_SP) | (CFRAME_SIZE+spadj);
p[-2] = PPCI_STWU | PPCF_T(RID_TMP) | PPCF_A(RID_SP) | spadj;
}
@@ -1947,7 +2258,7 @@
int nslots = 2, ngpr = REGARG_NUMGPR, nfpr = REGARG_NUMFPR;
asm_collectargs(as, ir, ci, args);
for (i = 0; i < nargs; i++)
- if (args[i] && irt_isfp(IR(args[i])->t)) {
+ if (!LJ_SOFTFP && args[i] && irt_isfp(IR(args[i])->t)) {
if (nfpr > 0) nfpr--; else nslots = (nslots+3) & ~1;
} else {
if (ngpr > 0) ngpr--; else nslots++;
@@ -1954,7 +2265,8 @@
}
if (nslots > as->evenspill) /* Leave room for args in stack slots. */
as->evenspill = nslots;
- return irt_isfp(ir->t) ? REGSP_HINT(RID_FPRET) : REGSP_HINT(RID_RET);
+ return (!LJ_SOFTFP && irt_isfp(ir->t)) ? REGSP_HINT(RID_FPRET) :
+ REGSP_HINT(RID_RET);
}
static void asm_setup_target(ASMState *as)
@@ -1972,7 +2284,7 @@
MCode *px = exitstub_trace_addr(T, exitno);
MCode *cstart = NULL;
MCode *mcarea = lj_mcode_patch(J, p, 0);
- int clearso = 0;
+ int clearso = 0, patchlong = 1;
for (; p < pe; p++) {
/* Look for exitstub branch, try to replace with branch to target. */
uint32_t ins = *p;
@@ -1984,7 +2296,9 @@
delta -= sizeof(MCode);
}
/* Many, but not all short-range branches can be patched directly. */
- if (((delta + 0x8000) >> 16) == 0) {
+ if (p[-1] == PPC_NOPATCH_GC_CHECK) {
+ patchlong = 0;
+ } else if (((delta + 0x8000) >> 16) == 0) {
*p = (ins & 0xffdf0000u) | ((uint32_t)delta & 0xffffu) |
((delta & 0x8000) * (PPCF_Y/0x8000));
if (!cstart) cstart = p;
@@ -1992,14 +2306,17 @@
} else if ((ins & 0xfc000000u) == PPCI_B &&
((ins ^ ((char *)px-(char *)p)) & 0x03ffffffu) == 0) {
ptrdiff_t delta = (char *)target - (char *)p;
- lua_assert(((delta + 0x02000000) >> 26) == 0);
+ lj_assertJ(((delta + 0x02000000) >> 26) == 0,
+ "branch target out of range");
*p = PPCI_B | ((uint32_t)delta & 0x03ffffffu);
if (!cstart) cstart = p;
}
}
- { /* Always patch long-range branch in exit stub itself. */
+ /* Always patch long-range branch in exit stub itself. Except, if we can't. */
+ if (patchlong) {
ptrdiff_t delta = (char *)target - (char *)px - clearso;
- lua_assert(((delta + 0x02000000) >> 26) == 0);
+ lj_assertJ(((delta + 0x02000000) >> 26) == 0,
+ "branch target out of range");
*px = PPCI_B | ((uint32_t)delta & 0x03ffffffu);
}
if (!cstart) cstart = px;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_x86.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_x86.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_asm_x86.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** x86/x64 IR assembler (SSA IR -> machine code).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Guard handling ------------------------------------------------------ */
@@ -31,7 +31,7 @@
#endif
/* Jump to exit handler which fills in the ExitState. */
*mxp++ = XI_JMP; mxp += 4;
- *((int32_t *)(mxp-4)) = jmprel(mxp, (MCode *)(void *)lj_vm_exit_handler);
+ *((int32_t *)(mxp-4)) = jmprel(as->J, mxp, (MCode *)(void *)lj_vm_exit_handler);
/* Commit the code for this group (even if assembly fails later on). */
lj_mcode_commitbot(as->J, mxp);
as->mcbot = mxp;
@@ -60,7 +60,7 @@
MCode *p = as->mcp;
if (LJ_UNLIKELY(p == as->invmcp)) {
as->loopinv = 1;
- *(int32_t *)(p+1) = jmprel(p+5, target);
+ *(int32_t *)(p+1) = jmprel(as->J, p+5, target);
target = p;
cc ^= 1;
if (as->realign) {
@@ -109,7 +109,7 @@
/* Check if there's no conflicting instruction between curins and ref.
** Also avoid fusing loads if there are multiple references.
*/
-static int noconflict(ASMState *as, IRRef ref, IROp conflict, int noload)
+static int noconflict(ASMState *as, IRRef ref, IROp conflict, int check)
{
IRIns *ir = as->ir;
IRRef i = as->curins;
@@ -118,8 +118,10 @@
while (--i > ref) {
if (ir[i].o == conflict)
return 0; /* Conflict found. */
- else if (!noload && (ir[i].op1 == ref || ir[i].op2 == ref))
+ else if ((check & 1) && (ir[i].o == IR_NEWREF || ir[i].o == IR_CALLS))
return 0;
+ else if ((check & 2) && (ir[i].op1 == ref || ir[i].op2 == ref))
+ return 0;
}
return 1; /* Ok, no conflict. */
}
@@ -131,16 +133,17 @@
as->mrm.ofs = 0;
if (irb->o == IR_FLOAD) {
IRIns *ira = IR(irb->op1);
- lua_assert(irb->op2 == IRFL_TAB_ARRAY);
+ lj_assertA(irb->op2 == IRFL_TAB_ARRAY, "expected FLOAD TAB_ARRAY");
/* We can avoid the FLOAD of t->array for colocated arrays. */
if (ira->o == IR_TNEW && ira->op1 <= LJ_MAX_COLOSIZE &&
- !neverfuse(as) && noconflict(as, irb->op1, IR_NEWREF, 1)) {
+ !neverfuse(as) && noconflict(as, irb->op1, IR_NEWREF, 0)) {
as->mrm.ofs = (int32_t)sizeof(GCtab); /* Ofs to colocated array. */
return irb->op1; /* Table obj. */
}
} else if (irb->o == IR_ADD && irref_isk(irb->op2)) {
/* Fuse base offset (vararg load). */
- as->mrm.ofs = IR(irb->op2)->i;
+ IRIns *irk = IR(irb->op2);
+ as->mrm.ofs = irk->o == IR_KINT ? irk->i : (int32_t)ir_kint64(irk)->u64;
return irb->op1;
}
return ref; /* Otherwise use the given array base. */
@@ -150,7 +153,7 @@
static void asm_fusearef(ASMState *as, IRIns *ir, RegSet allow)
{
IRIns *irx;
- lua_assert(ir->o == IR_AREF);
+ lj_assertA(ir->o == IR_AREF, "expected AREF");
as->mrm.base = (uint8_t)ra_alloc1(as, asm_fuseabase(as, ir->op1), allow);
irx = IR(ir->op2);
if (irref_isk(ir->op2)) {
@@ -216,9 +219,17 @@
#endif
}
break;
+ case IR_TMPREF:
+#if LJ_GC64
+ as->mrm.ofs = (int32_t)dispofs(as, &J2G(as->J)->tmptv);
+ as->mrm.base = RID_DISPATCH;
+ as->mrm.idx = RID_NONE;
+#else
+ as->mrm.ofs = igcptr(&J2G(as->J)->tmptv);
+ as->mrm.base = as->mrm.idx = RID_NONE;
+#endif
+ return;
default:
- lua_assert(ir->o == IR_HREF || ir->o == IR_NEWREF || ir->o == IR_UREFO ||
- ir->o == IR_KKPTR);
break;
}
}
@@ -230,9 +241,10 @@
/* Fuse FLOAD/FREF reference into memory operand. */
static void asm_fusefref(ASMState *as, IRIns *ir, RegSet allow)
{
- lua_assert(ir->o == IR_FLOAD || ir->o == IR_FREF);
+ lj_assertA(ir->o == IR_FLOAD || ir->o == IR_FREF,
+ "bad IR op %d", ir->o);
as->mrm.idx = RID_NONE;
- if (ir->op1 == REF_NIL) {
+ if (ir->op1 == REF_NIL) { /* FLOAD from GG_State with offset. */
#if LJ_GC64
as->mrm.ofs = (int32_t)(ir->op2 << 2) - GG_OFS(dispatch);
as->mrm.base = RID_DISPATCH;
@@ -271,7 +283,7 @@
static void asm_fusestrref(ASMState *as, IRIns *ir, RegSet allow)
{
IRIns *irr;
- lua_assert(ir->o == IR_STRREF);
+ lj_assertA(ir->o == IR_STRREF, "bad IR op %d", ir->o);
as->mrm.base = as->mrm.idx = RID_NONE;
as->mrm.scale = XM_SCALE1;
as->mrm.ofs = sizeof(GCstr);
@@ -378,9 +390,10 @@
checki32(mctopofs(as, k)) && checki32(mctopofs(as, k+1))) {
as->mrm.ofs = (int32_t)mcpofs(as, k);
as->mrm.base = RID_RIP;
- } else {
+ } else { /* Intern 64 bit constant at bottom of mcode. */
if (ir->i) {
- lua_assert(*k == *(uint64_t*)(as->mctop - ir->i));
+ lj_assertA(*k == *(uint64_t*)(as->mctop - ir->i),
+ "bad interned 64 bit constant");
} else {
while ((uintptr_t)as->mcbot & 7) *as->mcbot++ = XI_INT3;
*(uint64_t*)as->mcbot = *k;
@@ -387,6 +400,7 @@
ir->i = (int32_t)(as->mctop - as->mcbot);
as->mcbot += 8;
as->mclim = as->mcbot + MCLIM_REDZONE;
+ lj_mcode_commitbot(as->J, as->mcbot);
}
as->mrm.ofs = (int32_t)mcpofs(as, as->mctop - ir->i);
as->mrm.base = RID_RIP;
@@ -419,12 +433,12 @@
}
if (ir->o == IR_KNUM) {
RegSet avail = as->freeset & ~as->modset & RSET_FPR;
- lua_assert(allow != RSET_EMPTY);
+ lj_assertA(allow != RSET_EMPTY, "no register allowed");
if (!(avail & (avail-1))) /* Fuse if less than two regs available. */
return asm_fuseloadk64(as, ir);
} else if (ref == REF_BASE || ir->o == IR_KINT64) {
RegSet avail = as->freeset & ~as->modset & RSET_GPR;
- lua_assert(allow != RSET_EMPTY);
+ lj_assertA(allow != RSET_EMPTY, "no register allowed");
if (!(avail & (avail-1))) { /* Fuse if less than two regs available. */
if (ref == REF_BASE) {
#if LJ_GC64
@@ -444,7 +458,7 @@
RegSet xallow = (allow & RSET_GPR) ? allow : RSET_GPR;
if (ir->o == IR_SLOAD) {
if (!(ir->op2 & (IRSLOAD_PARENT|IRSLOAD_CONVERT)) &&
- noconflict(as, ref, IR_RETF, 0) &&
+ noconflict(as, ref, IR_RETF, 2) &&
!(LJ_GC64 && irt_isaddr(ir->t))) {
as->mrm.base = (uint8_t)ra_alloc1(as, REF_BASE, xallow);
as->mrm.ofs = 8*((int32_t)ir->op1-1-LJ_FR2) +
@@ -455,12 +469,12 @@
} else if (ir->o == IR_FLOAD) {
/* Generic fusion is only ok for 32 bit operand (but see asm_comp). */
if ((irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t)) &&
- noconflict(as, ref, IR_FSTORE, 0)) {
+ noconflict(as, ref, IR_FSTORE, 2)) {
asm_fusefref(as, ir, xallow);
return RID_MRM;
}
} else if (ir->o == IR_ALOAD || ir->o == IR_HLOAD || ir->o == IR_ULOAD) {
- if (noconflict(as, ref, ir->o + IRDELTA_L2S, 0) &&
+ if (noconflict(as, ref, ir->o + IRDELTA_L2S, 2+(ir->o != IR_ULOAD)) &&
!(LJ_GC64 && irt_isaddr(ir->t))) {
asm_fuseahuref(as, ir->op1, xallow);
return RID_MRM;
@@ -470,12 +484,14 @@
** Fusing unaligned memory operands is ok on x86 (except for SIMD types).
*/
if ((!irt_typerange(ir->t, IRT_I8, IRT_U16)) &&
- noconflict(as, ref, IR_XSTORE, 0)) {
+ noconflict(as, ref, IR_XSTORE, 2)) {
asm_fusexref(as, ir->op1, xallow);
return RID_MRM;
}
- } else if (ir->o == IR_VLOAD && !(LJ_GC64 && irt_isaddr(ir->t))) {
+ } else if (ir->o == IR_VLOAD && IR(ir->op1)->o == IR_AREF &&
+ !(LJ_GC64 && irt_isaddr(ir->t))) {
asm_fuseahuref(as, ir->op1, xallow);
+ as->mrm.ofs += 8 * ir->op2;
return RID_MRM;
}
}
@@ -605,7 +621,8 @@
#endif
emit_loadi(as, r, ir->i);
} else {
- lua_assert(rset_test(as->freeset, r)); /* Must have been evicted. */
+ /* Must have been evicted. */
+ lj_assertA(rset_test(as->freeset, r), "reg %d not free", r);
if (ra_hasreg(ir->r)) {
ra_noweak(as, ir->r);
emit_movrr(as, ir, r, ir->r);
@@ -614,7 +631,8 @@
}
}
} else if (irt_isfp(ir->t)) { /* FP argument is on stack. */
- lua_assert(!(irt_isfloat(ir->t) && irref_isk(ref))); /* No float k. */
+ lj_assertA(!(irt_isfloat(ir->t) && irref_isk(ref)),
+ "unexpected float constant");
if (LJ_32 && (ofs & 4) && irref_isk(ref)) {
/* Split stores for unaligned FP consts. */
emit_movmroi(as, RID_ESP, ofs, (int32_t)ir_knum(ir)->u32.lo);
@@ -645,7 +663,7 @@
static void asm_setupresult(ASMState *as, IRIns *ir, const CCallInfo *ci)
{
RegSet drop = RSET_SCRATCH;
- int hiop = (LJ_32 && (ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t));
+ int hiop = ((ir+1)->o == IR_HIOP && !irt_isnil((ir+1)->t));
if ((ci->flags & CCI_NOFPRCLOBBER))
drop &= ~RSET_FPR;
if (ra_hasreg(ir->r))
@@ -685,12 +703,10 @@
irt_isnum(ir->t) ? XOg_FSTPq : XOg_FSTPd, RID_ESP, ofs);
}
#endif
-#if LJ_32
} else if (hiop) {
ra_destpair(as, ir);
-#endif
} else {
- lua_assert(!irt_ispri(ir->t));
+ lj_assertA(!irt_ispri(ir->t), "PRI dest");
ra_destreg(as, ir, RID_RET);
}
} else if (LJ_32 && irt_isfp(ir->t) && !(ci->flags & CCI_CASTU64)) {
@@ -775,6 +791,21 @@
#endif
}
+/* -- Buffer operations --------------------------------------------------- */
+
+#if LJ_HASBUFFER
+static void asm_bufhdr_write(ASMState *as, Reg sb)
+{
+ Reg tmp = ra_scratch(as, rset_exclude(RSET_GPR, sb));
+ IRIns irgc;
+ irgc.ot = IRT(0, IRT_PGC); /* GC type. */
+ emit_storeofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+ emit_opgl(as, XO_ARITH(XOg_OR), tmp|REX_GC64, cur_L);
+ emit_gri(as, XG_ARITHi(XOg_AND), tmp, SBUF_MASK_FLAG);
+ emit_loadofs(as, &irgc, tmp, sb, offsetof(SBuf, L));
+}
+#endif
+
/* -- Type conversions ---------------------------------------------------- */
static void asm_tointg(ASMState *as, IRIns *ir, Reg left)
@@ -786,6 +817,7 @@
emit_rr(as, XO_UCOMISD, left, tmp);
emit_rr(as, XO_CVTSI2SD, tmp, dest);
emit_rr(as, XO_XORPS, tmp, tmp); /* Avoid partial register stall. */
+ checkmclim(as);
emit_rr(as, XO_CVTTSD2SI, dest, left);
/* Can't fuse since left is needed twice. */
}
@@ -809,8 +841,10 @@
int st64 = (st == IRT_I64 || st == IRT_U64 || (LJ_64 && st == IRT_P64));
int stfp = (st == IRT_NUM || st == IRT_FLOAT);
IRRef lref = ir->op1;
- lua_assert(irt_type(ir->t) != st);
- lua_assert(!(LJ_32 && (irt_isint64(ir->t) || st64))); /* Handled by SPLIT. */
+ lj_assertA(irt_type(ir->t) != st, "inconsistent types for CONV");
+ lj_assertA(!(LJ_32 && (irt_isint64(ir->t) || st64)),
+ "IR %04d has unsplit 64 bit type",
+ (int)(ir - as->ir) - REF_BIAS);
if (irt_isfp(ir->t)) {
Reg dest = ra_dest(as, ir, RSET_FPR);
if (stfp) { /* FP to FP conversion. */
@@ -826,6 +860,7 @@
emit_rr(as, XO_SUBSD, dest, bias); /* Subtract 2^52+2^51 bias. */
emit_rr(as, XO_XORPS, dest, bias); /* Merge bias and integer. */
emit_rma(as, XO_MOVSD, bias, k);
+ checkmclim(as);
emit_mrm(as, XO_MOVD, dest, asm_fuseload(as, lref, RSET_GPR));
return;
} else { /* Integer to FP conversion. */
@@ -846,7 +881,8 @@
} else if (stfp) { /* FP to integer conversion. */
if (irt_isguard(ir->t)) {
/* Checked conversions are only supported from number to int. */
- lua_assert(irt_isint(ir->t) && st == IRT_NUM);
+ lj_assertA(irt_isint(ir->t) && st == IRT_NUM,
+ "bad type for checked CONV");
asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -881,7 +917,7 @@
Reg left, dest = ra_dest(as, ir, RSET_GPR);
RegSet allow = RSET_GPR;
x86Op op;
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t), "bad type for CONV EXT");
if (st == IRT_I8) {
op = XO_MOVSXb; allow = RSET_GPR8; dest |= FORCE_REX;
} else if (st == IRT_U8) {
@@ -915,7 +951,7 @@
}
} else {
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (st64) {
+ if (st64 && !(ir->op2 & IRCONV_NONE)) {
Reg left = asm_fuseload(as, lref, RSET_GPR);
/* This is either a 32 bit reg/reg mov which zeroes the hiword
** or a load of the loword from a 64 bit address.
@@ -952,7 +988,7 @@
emit_sjcc(as, CC_NS, l_end);
emit_rr(as, XO_TEST, hi, hi); /* Check if u64 >= 2^63. */
} else {
- lua_assert(((ir-1)->op2 & IRCONV_SRCMASK) == IRT_I64);
+ lj_assertA(((ir-1)->op2 & IRCONV_SRCMASK) == IRT_I64, "bad type for CONV");
}
emit_rmro(as, XO_FILDq, XOg_FILDq, RID_ESP, 0);
/* NYI: Avoid narrow-to-wide store-to-load forwarding stall. */
@@ -966,8 +1002,8 @@
IRType st = (IRType)((ir-1)->op2 & IRCONV_SRCMASK);
IRType dt = (((ir-1)->op2 & IRCONV_DSTMASK) >> IRCONV_DSH);
Reg lo, hi;
- lua_assert(st == IRT_NUM || st == IRT_FLOAT);
- lua_assert(dt == IRT_I64 || dt == IRT_U64);
+ lj_assertA(st == IRT_NUM || st == IRT_FLOAT, "bad type for CONV");
+ lj_assertA(dt == IRT_I64 || dt == IRT_U64, "bad type for CONV");
hi = ra_dest(as, ir, RSET_GPR);
lo = ra_dest(as, ir-1, rset_exclude(RSET_GPR, hi));
if (ra_used(ir-1)) emit_rmro(as, XO_MOV, lo, RID_ESP, 0);
@@ -1041,47 +1077,48 @@
/* -- Memory references --------------------------------------------------- */
/* Get pointer to TValue. */
-static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref, MSize mode)
{
- IRIns *ir = IR(ref);
- if (irt_isnum(ir->t)) {
- /* For numbers use the constant itself or a spill slot as a TValue. */
- if (irref_isk(ref))
- emit_loada(as, dest, ir_knum(ir));
- else
- emit_rmro(as, XO_LEA, dest|REX_64, RID_ESP, ra_spill(as, ir));
- } else {
- /* Otherwise use g->tmptv to hold the TValue. */
+ if ((mode & IRTMPREF_IN1)) {
+ IRIns *ir = IR(ref);
+ if (irt_isnum(ir->t)) {
+ if (irref_isk(ref) && !(mode & IRTMPREF_OUT1)) {
+ /* Use the number constant itself as a TValue. */
+ emit_loada(as, dest, ir_knum(ir));
+ return;
+ }
+ emit_rmro(as, XO_MOVSDto, ra_alloc1(as, ref, RSET_FPR), dest, 0);
+ } else {
#if LJ_GC64
- if (irref_isk(ref)) {
- TValue k;
- lj_ir_kvalue(as->J->L, &k, ir);
- emit_movmroi(as, dest, 4, k.u32.hi);
- emit_movmroi(as, dest, 0, k.u32.lo);
- } else {
- /* TODO: 64 bit store + 32 bit load-modify-store is suboptimal. */
- Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, dest));
- if (irt_is64(ir->t)) {
- emit_u32(as, irt_toitype(ir->t) << 15);
- emit_rmro(as, XO_ARITHi, XOg_OR, dest, 4);
+ if (irref_isk(ref)) {
+ TValue k;
+ lj_ir_kvalue(as->J->L, &k, ir);
+ emit_movmroi(as, dest, 4, k.u32.hi);
+ emit_movmroi(as, dest, 0, k.u32.lo);
} else {
- /* Currently, no caller passes integers that might end up here. */
- emit_movmroi(as, dest, 4, (irt_toitype(ir->t) << 15));
+ /* TODO: 64 bit store + 32 bit load-modify-store is suboptimal. */
+ Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, dest));
+ if (irt_is64(ir->t)) {
+ emit_u32(as, irt_toitype(ir->t) << 15);
+ emit_rmro(as, XO_ARITHi, XOg_OR, dest, 4);
+ } else {
+ emit_movmroi(as, dest, 4, (irt_toitype(ir->t) << 15));
+ }
+ emit_movtomro(as, REX_64IR(ir, src), dest, 0);
}
- emit_movtomro(as, REX_64IR(ir, src), dest, 0);
- }
#else
- if (!irref_isk(ref)) {
- Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, dest));
- emit_movtomro(as, REX_64IR(ir, src), dest, 0);
- } else if (!irt_ispri(ir->t)) {
- emit_movmroi(as, dest, 0, ir->i);
+ if (!irref_isk(ref)) {
+ Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, dest));
+ emit_movtomro(as, REX_64IR(ir, src), dest, 0);
+ } else if (!irt_ispri(ir->t)) {
+ emit_movmroi(as, dest, 0, ir->i);
+ }
+ if (!(LJ_64 && irt_islightud(ir->t)))
+ emit_movmroi(as, dest, 4, irt_toitype(ir->t));
+#endif
}
- if (!(LJ_64 && irt_islightud(ir->t)))
- emit_movmroi(as, dest, 4, irt_toitype(ir->t));
-#endif
- emit_loada(as, dest, &J2G(as->J)->tmptv);
}
+ emit_loada(as, dest, &J2G(as->J)->tmptv); /* g->tmptv holds the TValue(s). */
}
static void asm_aref(ASMState *as, IRIns *ir)
@@ -1140,6 +1177,7 @@
asm_guardcc(as, CC_E);
else
emit_sjcc(as, CC_E, l_end);
+ checkmclim(as);
if (irt_isnum(kt)) {
if (isk) {
/* Assumes -0.0 is already canonicalized to +0.0. */
@@ -1179,13 +1217,13 @@
emit_rmro(as, XO_CMP, tmp|REX_64, dest, offsetof(Node, key.u64));
}
} else {
- lua_assert(irt_ispri(kt) && !irt_isnil(kt));
+ lj_assertA(irt_ispri(kt) && !irt_isnil(kt), "bad HREF key type");
emit_u32(as, (irt_toitype(kt)<<15)|0x7fff);
emit_rmro(as, XO_ARITHi, XOg_CMP, dest, offsetof(Node, key.it));
#else
} else {
if (!irt_ispri(kt)) {
- lua_assert(irt_isaddr(kt));
+ lj_assertA(irt_isaddr(kt), "bad HREF key type");
if (isk)
emit_gmroi(as, XG_ARITHi(XOg_CMP), dest, offsetof(Node, key.gcr),
ptr2addr(ir_kgc(irkey)));
@@ -1193,13 +1231,12 @@
emit_rmro(as, XO_CMP, key, dest, offsetof(Node, key.gcr));
emit_sjcc(as, CC_NE, l_next);
}
- lua_assert(!irt_isnil(kt));
+ lj_assertA(!irt_isnil(kt), "bad HREF key type");
emit_i8(as, irt_toitype(kt));
emit_rmro(as, XO_ARITHi8, XOg_CMP, dest, offsetof(Node, key.it));
#endif
}
emit_sfixup(as, l_loop);
- checkmclim(as);
#if LJ_GC64
if (!isk && irt_isaddr(kt)) {
emit_rr(as, XO_OR, tmp|REX_64, key);
@@ -1208,23 +1245,18 @@
#endif
/* Load main position relative to tab->node into dest. */
- khash = isk ? ir_khash(irkey) : 1;
+ khash = isk ? ir_khash(as, irkey) : 1;
if (khash == 0) {
emit_rmro(as, XO_MOV, dest|REX_GC64, tab, offsetof(GCtab, node));
} else {
emit_rmro(as, XO_ARITH(XOg_ADD), dest|REX_GC64, tab, offsetof(GCtab,node));
- if ((as->flags & JIT_F_PREFER_IMUL)) {
- emit_i8(as, sizeof(Node));
- emit_rr(as, XO_IMULi8, dest, dest);
- } else {
- emit_shifti(as, XOg_SHL, dest, 3);
- emit_rmrxo(as, XO_LEA, dest, dest, dest, XM_SCALE2, 0);
- }
+ emit_shifti(as, XOg_SHL, dest, 3);
+ emit_rmrxo(as, XO_LEA, dest, dest, dest, XM_SCALE2, 0);
if (isk) {
emit_gri(as, XG_ARITHi(XOg_AND), dest, (int32_t)khash);
emit_rmro(as, XO_MOV, dest, tab, offsetof(GCtab, hmask));
} else if (irt_isstr(kt)) {
- emit_rmro(as, XO_ARITH(XOg_AND), dest, key, offsetof(GCstr, hash));
+ emit_rmro(as, XO_ARITH(XOg_AND), dest, key, offsetof(GCstr, sid));
emit_rmro(as, XO_MOV, dest, tab, offsetof(GCtab, hmask));
} else { /* Must match with hashrot() in lj_tab.c. */
emit_rmro(as, XO_ARITH(XOg_AND), dest, tab, offsetof(GCtab, hmask));
@@ -1231,6 +1263,7 @@
emit_rr(as, XO_ARITH(XOg_SUB), dest, tmp);
emit_shifti(as, XOg_ROL, tmp, HASH_ROT3);
emit_rr(as, XO_ARITH(XOg_XOR), dest, tmp);
+ checkmclim(as);
emit_shifti(as, XOg_ROL, dest, HASH_ROT2);
emit_rr(as, XO_ARITH(XOg_SUB), tmp, dest);
emit_shifti(as, XOg_ROL, dest, HASH_ROT1);
@@ -1248,7 +1281,6 @@
} else {
emit_rr(as, XO_MOV, tmp, key);
#if LJ_GC64
- checkmclim(as);
emit_gri(as, XG_ARITHi(XOg_XOR), dest, irt_toitype(kt) << 15);
if ((as->flags & JIT_F_BMI2)) {
emit_i8(as, 32);
@@ -1275,10 +1307,10 @@
#if !LJ_64
MCLabel l_exit;
#endif
- lua_assert(ofs % sizeof(Node) == 0);
+ lj_assertA(ofs % sizeof(Node) == 0, "unaligned HREFK slot");
if (ra_hasreg(dest)) {
if (ofs != 0) {
- if (dest == node && !(as->flags & JIT_F_LEA_AGU))
+ if (dest == node)
emit_gri(as, XG_ARITHi(XOg_ADD), dest|REX_GC64, ofs);
else
emit_rmro(as, XO_LEA, dest|REX_GC64, node, ofs);
@@ -1292,7 +1324,8 @@
Reg key = ra_scratch(as, rset_exclude(RSET_GPR, node));
emit_rmro(as, XO_CMP, key|REX_64, node,
ofs + (int32_t)offsetof(Node, key.u64));
- lua_assert(irt_isnum(irkey->t) || irt_isgcv(irkey->t));
+ lj_assertA(irt_isnum(irkey->t) || irt_isgcv(irkey->t),
+ "bad HREFK key type");
/* Assumes -0.0 is already canonicalized to +0.0. */
emit_loadu64(as, key, irt_isnum(irkey->t) ? ir_knum(irkey)->u64 :
#if LJ_GC64
@@ -1303,7 +1336,7 @@
(uint64_t)(uint32_t)ptr2addr(ir_kgc(irkey)));
#endif
} else {
- lua_assert(!irt_isnil(irkey->t));
+ lj_assertA(!irt_isnil(irkey->t), "bad HREFK key type");
#if LJ_GC64
emit_i32(as, (irt_toitype(irkey->t)<<15)|0x7fff);
emit_rmro(as, XO_ARITHi, XOg_CMP, node,
@@ -1327,13 +1360,13 @@
(int32_t)ir_knum(irkey)->u32.hi);
} else {
if (!irt_ispri(irkey->t)) {
- lua_assert(irt_isgcv(irkey->t));
+ lj_assertA(irt_isgcv(irkey->t), "bad HREFK key type");
emit_gmroi(as, XG_ARITHi(XOg_CMP), node,
ofs + (int32_t)offsetof(Node, key.gcr),
ptr2addr(ir_kgc(irkey)));
emit_sjcc(as, CC_NE, l_exit);
}
- lua_assert(!irt_isnil(irkey->t));
+ lj_assertA(!irt_isnil(irkey->t), "bad HREFK key type");
emit_i8(as, irt_toitype(irkey->t));
emit_rmro(as, XO_ARITHi8, XOg_CMP, node,
ofs + (int32_t)offsetof(Node, key.it));
@@ -1344,24 +1377,31 @@
static void asm_uref(ASMState *as, IRIns *ir)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
- if (irref_isk(ir->op1)) {
+ int guarded = (irt_t(ir->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC);
+ if (irref_isk(ir->op1) && !guarded) {
GCfunc *fn = ir_kfunc(IR(ir->op1));
MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
emit_rma(as, XO_MOV, dest|REX_GC64, v);
} else {
Reg uv = ra_scratch(as, RSET_GPR);
- Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
- if (ir->o == IR_UREFC) {
+ if (ir->o == IR_UREFC)
emit_rmro(as, XO_LEA, dest|REX_GC64, uv, offsetof(GCupval, tv));
- asm_guardcc(as, CC_NE);
- emit_i8(as, 1);
+ else
+ emit_rmro(as, XO_MOV, dest|REX_GC64, uv, offsetof(GCupval, v));
+ if (guarded) {
+ asm_guardcc(as, ir->o == IR_UREFC ? CC_E : CC_NE);
+ emit_i8(as, 0);
emit_rmro(as, XO_ARITHib, XOg_CMP, uv, offsetof(GCupval, closed));
+ }
+ if (irref_isk(ir->op1)) {
+ GCfunc *fn = ir_kfunc(IR(ir->op1));
+ GCobj *o = gcref(fn->l.uvptr[(ir->op2 >> 8)]);
+ emit_loada(as, uv, o);
} else {
- emit_rmro(as, XO_MOV, dest|REX_GC64, uv, offsetof(GCupval, v));
+ emit_rmro(as, XO_MOV, uv|REX_GC64, ra_alloc1(as, ir->op1, RSET_GPR),
+ (int32_t)offsetof(GCfuncL, uvptr) +
+ (int32_t)sizeof(MRef) * (int32_t)(ir->op2 >> 8));
}
- emit_rmro(as, XO_MOV, uv|REX_GC64, func,
- (int32_t)offsetof(GCfuncL, uvptr) +
- (int32_t)sizeof(MRef) * (int32_t)(ir->op2 >> 8));
}
}
@@ -1406,7 +1446,8 @@
if (LJ_64 && irt_is64(ir->t))
dest |= REX_64;
else
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t),
+ "unsplit 64 bit load");
xo = XO_MOV;
break;
}
@@ -1451,13 +1492,16 @@
case IRT_NUM: xo = XO_MOVSDto; break;
case IRT_FLOAT: xo = XO_MOVSSto; break;
#if LJ_64 && !LJ_GC64
- case IRT_LIGHTUD: lua_assert(0); /* NYI: mask 64 bit lightuserdata. */
+ case IRT_LIGHTUD:
+ /* NYI: mask 64 bit lightuserdata. */
+ lj_assertA(0, "store of lightuserdata");
#endif
default:
if (LJ_64 && irt_is64(ir->t))
src |= REX_64;
else
- lua_assert(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t));
+ lj_assertA(irt_isint(ir->t) || irt_isu32(ir->t) || irt_isaddr(ir->t),
+ "unsplit 64 bit store");
xo = XO_MOVto;
break;
}
@@ -1471,8 +1515,8 @@
emit_i8(as, k);
emit_mrm(as, XO_MOVmib, 0, RID_MRM);
} else {
- lua_assert(irt_is64(ir->t) || irt_isint(ir->t) || irt_isu32(ir->t) ||
- irt_isaddr(ir->t));
+ lj_assertA(irt_is64(ir->t) || irt_isint(ir->t) || irt_isu32(ir->t) ||
+ irt_isaddr(ir->t), "bad store type");
emit_i32(as, k);
emit_mrm(as, XO_MOVmi, REX_64IR(ir, 0), RID_MRM);
}
@@ -1507,13 +1551,16 @@
#if LJ_GC64
Reg tmp = RID_NONE;
#endif
- lua_assert(irt_isnum(ir->t) || irt_ispri(ir->t) || irt_isaddr(ir->t) ||
- (LJ_DUALNUM && irt_isint(ir->t)));
+ lj_assertA(irt_isnum(ir->t) || irt_ispri(ir->t) || irt_isaddr(ir->t) ||
+ (LJ_DUALNUM && irt_isint(ir->t)),
+ "bad load type %d", irt_type(ir->t));
#if LJ_64 && !LJ_GC64
if (irt_islightud(ir->t)) {
Reg dest = asm_load_lightud64(as, ir, 1);
if (ra_hasreg(dest)) {
+ checkmclim(as);
asm_fuseahuref(as, ir->op1, RSET_GPR);
+ if (ir->o == IR_VLOAD) as->mrm.ofs += 8 * ir->op2;
emit_mrm(as, XO_MOV, dest|REX_64, RID_MRM);
}
return;
@@ -1523,6 +1570,7 @@
RegSet allow = irt_isnum(ir->t) ? RSET_FPR : RSET_GPR;
Reg dest = ra_dest(as, ir, allow);
asm_fuseahuref(as, ir->op1, RSET_GPR);
+ if (ir->o == IR_VLOAD) as->mrm.ofs += 8 * ir->op2;
#if LJ_GC64
if (irt_isaddr(ir->t)) {
emit_shifti(as, XOg_SHR|REX_64, dest, 17);
@@ -1550,12 +1598,15 @@
}
#endif
asm_fuseahuref(as, ir->op1, gpr);
+ if (ir->o == IR_VLOAD) as->mrm.ofs += 8 * ir->op2;
}
/* Always do the type check, even if the load result is unused. */
as->mrm.ofs += 4;
asm_guardcc(as, irt_isnum(ir->t) ? CC_AE : CC_NE);
if (LJ_64 && irt_type(ir->t) >= IRT_NUM) {
- lua_assert(irt_isinteger(ir->t) || irt_isnum(ir->t));
+ lj_assertA(irt_isinteger(ir->t) || irt_isnum(ir->t),
+ "bad load type %d", irt_type(ir->t));
+ checkmclim(as);
#if LJ_GC64
emit_u32(as, LJ_TISNUM << 15);
#else
@@ -1637,13 +1688,14 @@
#endif
emit_mrm(as, XO_MOVto, src, RID_MRM);
} else if (!irt_ispri(irr->t)) {
- lua_assert(irt_isaddr(ir->t) || (LJ_DUALNUM && irt_isinteger(ir->t)));
+ lj_assertA(irt_isaddr(ir->t) || (LJ_DUALNUM && irt_isinteger(ir->t)),
+ "bad store type");
emit_i32(as, irr->i);
emit_mrm(as, XO_MOVmi, 0, RID_MRM);
}
as->mrm.ofs += 4;
#if LJ_GC64
- lua_assert(LJ_DUALNUM && irt_isinteger(ir->t));
+ lj_assertA(LJ_DUALNUM && irt_isinteger(ir->t), "bad store type");
emit_i32(as, LJ_TNUMX << 15);
#else
emit_i32(as, (int32_t)irt_toitype(ir->t));
@@ -1658,10 +1710,14 @@
(!LJ_FR2 && (ir->op2 & IRSLOAD_FRAME) ? 4 : 0);
IRType1 t = ir->t;
Reg base;
- lua_assert(!(ir->op2 & IRSLOAD_PARENT)); /* Handled by asm_head_side(). */
- lua_assert(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK));
- lua_assert(LJ_DUALNUM ||
- !irt_isint(t) || (ir->op2 & (IRSLOAD_CONVERT|IRSLOAD_FRAME)));
+ lj_assertA(!(ir->op2 & IRSLOAD_PARENT),
+ "bad parent SLOAD"); /* Handled by asm_head_side(). */
+ lj_assertA(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK),
+ "inconsistent SLOAD variant");
+ lj_assertA(LJ_DUALNUM ||
+ !irt_isint(t) ||
+ (ir->op2 & (IRSLOAD_CONVERT|IRSLOAD_FRAME|IRSLOAD_KEYINDEX)),
+ "bad SLOAD type");
if ((ir->op2 & IRSLOAD_CONVERT) && irt_isguard(t) && irt_isint(t)) {
Reg left = ra_scratch(as, RSET_FPR);
asm_tointg(as, ir, left); /* Frees dest reg. Do this before base alloc. */
@@ -1681,7 +1737,8 @@
RegSet allow = irt_isnum(t) ? RSET_FPR : RSET_GPR;
Reg dest = ra_dest(as, ir, allow);
base = ra_alloc1(as, REF_BASE, RSET_GPR);
- lua_assert(irt_isnum(t) || irt_isint(t) || irt_isaddr(t));
+ lj_assertA(irt_isnum(t) || irt_isint(t) || irt_isaddr(t),
+ "bad SLOAD type %d", irt_type(t));
if ((ir->op2 & IRSLOAD_CONVERT)) {
t.irt = irt_isint(t) ? IRT_NUM : IRT_INT; /* Check for original type. */
emit_rmro(as, irt_isint(t) ? XO_CVTSI2SD : XO_CVTTSD2SI, dest, base, ofs);
@@ -1726,13 +1783,11 @@
if ((ir->op2 & IRSLOAD_TYPECHECK)) {
/* Need type check, even if the load result is unused. */
asm_guardcc(as, irt_isnum(t) ? CC_AE : CC_NE);
- if (LJ_64 && irt_type(t) >= IRT_NUM) {
- lua_assert(irt_isinteger(t) || irt_isnum(t));
-#if LJ_GC64
- emit_u32(as, LJ_TISNUM << 15);
-#else
- emit_u32(as, LJ_TISNUM);
-#endif
+ if ((LJ_64 && irt_type(t) >= IRT_NUM) || (ir->op2 & IRSLOAD_KEYINDEX)) {
+ lj_assertA(irt_isinteger(t) || irt_isnum(t),
+ "bad SLOAD type %d", irt_type(t));
+ emit_u32(as, (ir->op2 & IRSLOAD_KEYINDEX) ? LJ_KEYINDEX :
+ LJ_GC64 ? (LJ_TISNUM << 15) : LJ_TISNUM);
emit_rmro(as, XO_ARITHi, XOg_CMP, base, ofs+4);
#if LJ_GC64
} else if (irt_isnil(t)) {
@@ -1758,7 +1813,7 @@
emit_i8(as, irt_toitype(t));
emit_rr(as, XO_ARITHi8, XOg_CMP, tmp);
emit_shifti(as, XOg_SAR|REX_64, tmp, 47);
- emit_rmro(as, XO_MOV, tmp|REX_64, base, ofs+4);
+ emit_rmro(as, XO_MOV, tmp|REX_64, base, ofs);
#else
} else {
emit_i8(as, irt_toitype(t));
@@ -1779,7 +1834,8 @@
CTInfo info = lj_ctype_info(cts, id, &sz);
const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
IRRef args[4];
- lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
+ lj_assertA(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL),
+ "bad CNEW/CNEWI operands");
as->gcsteps++;
asm_setupresult(as, ir, ci); /* GCcdata * */
@@ -1809,7 +1865,7 @@
int32_t ofs = sizeof(GCcdata);
if (sz == 8) {
ofs += 4; ir++;
- lua_assert(ir->o == IR_HIOP);
+ lj_assertA(ir->o == IR_HIOP, "missing CNEWI HIOP");
}
do {
if (irref_isk(ir->op2)) {
@@ -1823,7 +1879,7 @@
ofs -= 4; ir--;
} while (1);
#endif
- lua_assert(sz == 4 || sz == 8);
+ lj_assertA(sz == 4 || sz == 8, "bad CNEWI size %d", sz);
} else if (ir->op2 != REF_NIL) { /* Create VLA/VLS/aligned cdata. */
ci = &lj_ir_callinfo[IRCALL_lj_cdata_newv];
args[0] = ASMREF_L; /* lua_State *L */
@@ -1847,8 +1903,6 @@
asm_gencall(as, ci, args);
emit_loadi(as, ra_releasetmp(as, ASMREF_TMP1), (int32_t)(sz+sizeof(GCcdata)));
}
-#else
-#define asm_cnew(as, ir) ((void)0)
#endif
/* -- Write barriers ------------------------------------------------------ */
@@ -1875,7 +1929,7 @@
MCLabel l_end;
Reg obj;
/* No need for other object barriers (yet). */
- lua_assert(IR(ir->op1)->o == IR_UREFC);
+ lj_assertA(IR(ir->op1)->o == IR_UREFC, "bad OBAR type");
ra_evictset(as, RSET_SCRATCH);
l_end = emit_label(as);
args[0] = ASMREF_TMP1; /* global_State *g */
@@ -1954,15 +2008,11 @@
fpm == IRFPM_CEIL ? lj_vm_ceil_sse : lj_vm_trunc_sse);
ra_left(as, RID_XMM0, ir->op1);
}
- } else if (fpm == IRFPM_EXP2 && asm_fpjoin_pow(as, ir)) {
- /* Rejoined to pow(). */
} else {
asm_callid(as, ir, IRCALL_lj_vm_floor + fpm);
}
}
-#define asm_atan2(as, ir) asm_callid(as, ir, IRCALL_atan2)
-
static void asm_ldexp(ASMState *as, IRIns *ir)
{
int32_t ofs = sps_scale(ir->s); /* Use spill slot or temp slots. */
@@ -1979,35 +2029,11 @@
asm_x87load(as, ir->op2);
}
-static void asm_fppowi(ASMState *as, IRIns *ir)
-{
- /* The modified regs must match with the *.dasc implementation. */
- RegSet drop = RSET_RANGE(RID_XMM0, RID_XMM1+1)|RID2RSET(RID_EAX);
- if (ra_hasreg(ir->r))
- rset_clear(drop, ir->r); /* Dest reg handled below. */
- ra_evictset(as, drop);
- ra_destreg(as, ir, RID_XMM0);
- emit_call(as, lj_vm_powi_sse);
- ra_left(as, RID_XMM0, ir->op1);
- ra_left(as, RID_EAX, ir->op2);
-}
-
-static void asm_pow(ASMState *as, IRIns *ir)
-{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_powi64 :
- IRCALL_lj_carith_powu64);
- else
-#endif
- asm_fppowi(as, ir);
-}
-
static int asm_swapops(ASMState *as, IRIns *ir)
{
IRIns *irl = IR(ir->op1);
IRIns *irr = IR(ir->op2);
- lua_assert(ra_noreg(irr->r));
+ lj_assertA(ra_noreg(irr->r), "bad usage");
if (!irm_iscomm(lj_ir_mode[ir->o]))
return 0; /* Can't swap non-commutative operations. */
if (irref_isk(ir->op2))
@@ -2060,8 +2086,9 @@
int32_t k = 0;
if (as->flagmcp == as->mcp) { /* Drop test r,r instruction. */
MCode *p = as->mcp + ((LJ_64 && *as->mcp < XI_TESTb) ? 3 : 2);
- if ((p[1] & 15) < 14) {
- if ((p[1] & 15) >= 12) p[1] -= 4; /* L <->S, NL <-> NS */
+ MCode *q = p[0] == 0x0f ? p+1 : p;
+ if ((*q & 15) < 14) {
+ if ((*q & 15) >= 12) *q -= 4; /* L <->S, NL <-> NS */
as->flagmcp = NULL;
as->mcp = p;
} /* else: cannot transform LE/NLE to cc without use of OF. */
@@ -2178,8 +2205,7 @@
{
if (irt_isnum(ir->t))
asm_fparith(as, ir, XO_ADDSD);
- else if ((as->flags & JIT_F_LEA_AGU) || as->flagmcp == as->mcp ||
- irt_is64(ir->t) || !asm_lea(as, ir))
+ else if (as->flagmcp == as->mcp || irt_is64(ir->t) || !asm_lea(as, ir))
asm_intarith(as, ir, XOg_ADD);
}
@@ -2199,28 +2225,8 @@
asm_intarith(as, ir, XOg_X_IMUL);
}
-static void asm_div(ASMState *as, IRIns *ir)
-{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isnum(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_divi64 :
- IRCALL_lj_carith_divu64);
- else
-#endif
- asm_fparith(as, ir, XO_DIVSD);
-}
+#define asm_fpdiv(as, ir) asm_fparith(as, ir, XO_DIVSD)
-static void asm_mod(ASMState *as, IRIns *ir)
-{
-#if LJ_64 && LJ_HASFFI
- if (!irt_isint(ir->t))
- asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_modi64 :
- IRCALL_lj_carith_modu64);
- else
-#endif
- asm_callid(as, ir, IRCALL_lj_vm_modi);
-}
-
static void asm_neg_not(ASMState *as, IRIns *ir, x86Group3 xg)
{
Reg dest = ra_dest(as, ir, RSET_GPR);
@@ -2319,7 +2325,7 @@
dest = ra_dest(as, ir, rset_exclude(RSET_GPR, RID_ECX));
if (dest == RID_ECX) {
dest = ra_scratch(as, rset_exclude(RSET_GPR, RID_ECX));
- emit_rr(as, XO_MOV, RID_ECX, dest);
+ emit_rr(as, XO_MOV, REX_64IR(ir, RID_ECX), dest);
}
right = irr->r;
if (ra_noreg(right))
@@ -2417,8 +2423,9 @@
IROp leftop = (IROp)(IR(lref)->o);
Reg r64 = REX_64IR(ir, 0);
int32_t imm = 0;
- lua_assert(irt_is64(ir->t) || irt_isint(ir->t) ||
- irt_isu32(ir->t) || irt_isaddr(ir->t) || irt_isu8(ir->t));
+ lj_assertA(irt_is64(ir->t) || irt_isint(ir->t) ||
+ irt_isu32(ir->t) || irt_isaddr(ir->t) || irt_isu8(ir->t),
+ "bad comparison data type %d", irt_type(ir->t));
/* Swap constants (only for ABC) and fusable loads to the right. */
if (irref_isk(lref) || (!irref_isk(rref) && opisfusableload(leftop))) {
if ((cc & 0xc) == 0xc) cc ^= 0x53; /* L <-> G, LE <-> GE */
@@ -2500,7 +2507,7 @@
/* Use test r,r instead of cmp r,0. */
x86Op xo = XO_TEST;
if (irt_isu8(ir->t)) {
- lua_assert(ir->o == IR_EQ || ir->o == IR_NE);
+ lj_assertA(ir->o == IR_EQ || ir->o == IR_NE, "bad usage");
xo = XO_TESTb;
if (!rset_test(RSET_RANGE(RID_EAX, RID_EBX+1), left)) {
if (LJ_64) {
@@ -2602,15 +2609,15 @@
}
#endif
-/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
+/* -- Split register ops -------------------------------------------------- */
-/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
+/* Hiword op of a split 32/32 or 64/64 bit op. Previous op is the loword op. */
static void asm_hiop(ASMState *as, IRIns *ir)
{
-#if LJ_32 && LJ_HASFFI
/* HIOP is marked as a store because it needs its own DCE logic. */
int uselo = ra_used(ir-1), usehi = ra_used(ir); /* Loword/hiword used? */
if (LJ_UNLIKELY(!(as->flags & JIT_F_OPT_DCE))) uselo = usehi = 1;
+#if LJ_32 && LJ_HASFFI
if ((ir-1)->o == IR_CONV) { /* Conversions to/from 64 bit. */
as->curins--; /* Always skip the CONV. */
if (usehi || uselo)
@@ -2624,8 +2631,10 @@
asm_fxstore(as, ir);
return;
}
+#endif
if (!usehi) return; /* Skip unused hiword op for all remaining ops. */
switch ((ir-1)->o) {
+#if LJ_32 && LJ_HASFFI
case IR_ADD:
as->flagmcp = NULL;
as->curins--;
@@ -2648,19 +2657,16 @@
asm_neg_not(as, ir-1, XOg_NEG);
break;
}
- case IR_CALLN:
- case IR_CALLXS:
+ case IR_CNEWI:
+ /* Nothing to do here. Handled by CNEWI itself. */
+ break;
+#endif
+ case IR_CALLN: case IR_CALLL: case IR_CALLS: case IR_CALLXS:
if (!uselo)
ra_allocref(as, ir->op1, RID2RSET(RID_RETLO)); /* Mark lo op as used. */
break;
- case IR_CNEWI:
- /* Nothing to do here. Handled by CNEWI itself. */
- break;
- default: lua_assert(0); break;
+ default: lj_assertA(0, "bad HIOP for op %d", (ir-1)->o); break;
}
-#else
- UNUSED(as); UNUSED(ir); lua_assert(0); /* Unused on x64 or without FFI. */
-#endif
}
/* -- Profiling ----------------------------------------------------------- */
@@ -2721,12 +2727,21 @@
IRIns *ir = IR(ref);
if ((sn & SNAP_NORESTORE))
continue;
- if (irt_isnum(ir->t)) {
+ if ((sn & SNAP_KEYINDEX)) {
+ emit_movmroi(as, RID_BASE, ofs+4, LJ_KEYINDEX);
+ if (irref_isk(ref)) {
+ emit_movmroi(as, RID_BASE, ofs, ir->i);
+ } else {
+ Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, RID_BASE));
+ emit_movtomro(as, src, RID_BASE, ofs);
+ }
+ } else if (irt_isnum(ir->t)) {
Reg src = ra_alloc1(as, ref, RSET_FPR);
emit_rmro(as, XO_MOVSDto, src, RID_BASE, ofs);
} else {
- lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) ||
- (LJ_DUALNUM && irt_isinteger(ir->t)));
+ lj_assertA(irt_ispri(ir->t) || irt_isaddr(ir->t) ||
+ (LJ_DUALNUM && irt_isinteger(ir->t)),
+ "restore of IR type %d", irt_type(ir->t));
if (!irref_isk(ref)) {
Reg src = ra_alloc1(as, ref, rset_exclude(RSET_GPR, RID_BASE));
#if LJ_GC64
@@ -2771,7 +2786,7 @@
}
checkmclim(as);
}
- lua_assert(map + nent == flinks);
+ lj_assertA(map + nent == flinks, "inconsistent frames in snapshot");
}
/* -- GC handling --------------------------------------------------------- */
@@ -2815,16 +2830,16 @@
MCode *target = as->mcp;
if (as->realign) { /* Realigned loops use short jumps. */
as->realign = NULL; /* Stop another retry. */
- lua_assert(((intptr_t)target & 15) == 0);
+ lj_assertA(((intptr_t)target & 15) == 0, "loop realign failed");
if (as->loopinv) { /* Inverted loop branch? */
p -= 5;
p[0] = XI_JMP;
- lua_assert(target - p >= -128);
+ lj_assertA(target - p >= -128, "loop realign failed");
p[-1] = (MCode)(target - p); /* Patch sjcc. */
if (as->loopinv == 2)
p[-3] = (MCode)(target - p + 2); /* Patch opt. short jp. */
} else {
- lua_assert(target - p >= -128);
+ lj_assertA(target - p >= -128, "loop realign failed");
p[-1] = (MCode)(int8_t)(target - p); /* Patch short jmp. */
p[-2] = XI_JMPs;
}
@@ -2853,6 +2868,12 @@
}
}
+/* Fixup the tail of the loop. */
+static void asm_loop_tail_fixup(ASMState *as)
+{
+ UNUSED(as); /* Nothing to do. */
+}
+
/* -- Head of trace ------------------------------------------------------- */
/* Coalesce BASE register for a root trace. */
@@ -2870,7 +2891,7 @@
}
/* Coalesce or reload BASE register for a side trace. */
-static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
+static Reg asm_head_side_base(ASMState *as, IRIns *irp)
{
IRIns *ir = IR(REF_BASE);
Reg r = ir->r;
@@ -2879,16 +2900,16 @@
if (rset_test(as->modset, r) || irt_ismarked(ir->t))
ir->r = RID_INIT; /* No inheritance for modified BASE register. */
if (irp->r == r) {
- rset_clear(allow, r); /* Mark same BASE register as coalesced. */
+ return r; /* Same BASE register already coalesced. */
} else if (ra_hasreg(irp->r) && rset_test(as->freeset, irp->r)) {
/* Move from coalesced parent reg. */
- rset_clear(allow, irp->r);
emit_rr(as, XO_MOV, r|REX_GC64, irp->r);
+ return irp->r;
} else {
emit_getgl(as, r, jit_base); /* Otherwise reload BASE. */
}
}
- return allow;
+ return RID_NONE;
}
/* -- Tail of trace ------------------------------------------------------- */
@@ -2901,7 +2922,7 @@
MCode *target, *q;
int32_t spadj = as->T->spadjust;
if (spadj == 0) {
- p -= ((as->flags & JIT_F_LEA_AGU) ? 7 : 6) + (LJ_64 ? 1 : 0);
+ p -= LJ_64 ? 7 : 6;
} else {
MCode *p1;
/* Patch stack adjustment. */
@@ -2913,24 +2934,15 @@
p1 = p-9;
*(int32_t *)p1 = spadj;
}
- if ((as->flags & JIT_F_LEA_AGU)) {
#if LJ_64
- p1[-4] = 0x48;
+ p1[-3] = 0x48;
#endif
- p1[-3] = (MCode)XI_LEA;
- p1[-2] = MODRM(checki8(spadj) ? XM_OFS8 : XM_OFS32, RID_ESP, RID_ESP);
- p1[-1] = MODRM(XM_SCALE1, RID_ESP, RID_ESP);
- } else {
-#if LJ_64
- p1[-3] = 0x48;
-#endif
- p1[-2] = (MCode)(checki8(spadj) ? XI_ARITHi8 : XI_ARITHi);
- p1[-1] = MODRM(XM_REG, XOg_ADD, RID_ESP);
- }
+ p1[-2] = (MCode)(checki8(spadj) ? XI_ARITHi8 : XI_ARITHi);
+ p1[-1] = MODRM(XM_REG, XOg_ADD, RID_ESP);
}
/* Patch exit branch. */
target = lnk ? traceref(as->J, lnk)->mcode : (MCode *)lj_vm_exit_interp;
- *(int32_t *)(p-4) = jmprel(p, target);
+ *(int32_t *)(p-4) = jmprel(as->J, p, target);
p[-5] = XI_JMP;
/* Drop unused mcode tail. Fill with NOPs to make the prefetcher happy. */
for (q = as->mctop-1; q >= p; q--)
@@ -2957,7 +2969,7 @@
as->invmcp = as->mcp = p;
} else {
/* Leave room for ESP adjustment: add esp, imm or lea esp, [esp+imm] */
- as->mcp = p - (((as->flags & JIT_F_LEA_AGU) ? 7 : 6) + (LJ_64 ? 1 : 0));
+ as->mcp = p - (LJ_64 ? 7 : 6);
as->invmcp = NULL;
}
}
@@ -3097,6 +3109,7 @@
MSize len = T->szmcode;
MCode *px = exitstub_addr(J, exitno) - 6;
MCode *pe = p+len-6;
+ MCode *pgc = NULL;
#if LJ_GC64
uint32_t statei = (uint32_t)(GG_OFS(g.vmstate) - GG_OFS(dispatch));
#else
@@ -3103,7 +3116,7 @@
uint32_t statei = u32ptr(&J2G(J)->vmstate);
#endif
if (len > 5 && p[len-5] == XI_JMP && p+len-6 + *(int32_t *)(p+len-4) == px)
- *(int32_t *)(p+len-4) = jmprel(p+len, target);
+ *(int32_t *)(p+len-4) = jmprel(J, p+len, target);
/* Do not patch parent exit for a stack check. Skip beyond vmstate update. */
for (; p < pe; p += asm_x86_inslen(p)) {
intptr_t ofs = LJ_GC64 ? (p[0] & 0xf0) == 0x40 : LJ_64;
@@ -3110,10 +3123,16 @@
if (*(uint32_t *)(p+2+ofs) == statei && p[ofs+LJ_GC64-LJ_64] == XI_MOVmi)
break;
}
- lua_assert(p < pe);
- for (; p < pe; p += asm_x86_inslen(p))
- if ((*(uint16_t *)p & 0xf0ff) == 0x800f && p + *(int32_t *)(p+2) == px)
- *(int32_t *)(p+2) = jmprel(p+6, target);
+ lj_assertJ(p < pe, "instruction length decoder failed");
+ for (; p < pe; p += asm_x86_inslen(p)) {
+ if ((*(uint16_t *)p & 0xf0ff) == 0x800f && p + *(int32_t *)(p+2) == px &&
+ p != pgc) {
+ *(int32_t *)(p+2) = jmprel(J, p+6, target);
+ } else if (*p == XI_CALL &&
+ (void *)(p+5+*(int32_t *)(p+1)) == (void *)lj_gc_step_jit) {
+ pgc = p+7; /* Do not patch GC check exit. */
+ }
+ }
lj_mcode_sync(T->mcode, T->mcode + T->szmcode);
lj_mcode_patch(J, mcarea, 1);
}
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_assert.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_assert.c (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_assert.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,28 @@
+/*
+** Internal assertions.
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
+*/
+
+#define lj_assert_c
+#define LUA_CORE
+
+#if defined(LUA_USE_ASSERT) || defined(LUA_USE_APICHECK)
+
+#include <stdio.h>
+
+#include "lj_obj.h"
+
+void lj_assert_fail(global_State *g, const char *file, int line,
+ const char *func, const char *fmt, ...)
+{
+ va_list argp;
+ va_start(argp, fmt);
+ fprintf(stderr, "LuaJIT ASSERT %s:%d: %s: ", file, line, func);
+ vfprintf(stderr, fmt, argp);
+ fputc('\n', stderr);
+ va_end(argp);
+ UNUSED(g); /* May be NULL. TODO: optionally dump state. */
+ abort();
+}
+
+#endif
Property changes on: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_assert.c
___________________________________________________________________
Added: svn:eol-style
## -0,0 +1 ##
+native
\ No newline at end of property
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bytecode instruction modes.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_bc_c
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bytecode instruction format.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_BC_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcdump.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcdump.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcdump.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bytecode dump definitions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_BCDUMP_H
@@ -46,6 +46,8 @@
#define BCDUMP_F_KNOWN (BCDUMP_F_FR2*2-1)
+#define BCDUMP_F_DETERMINISTIC 0x80000000
+
/* Type codes for the GC constants of a prototype. Plus length for strings. */
enum {
BCDUMP_KGC_CHILD, BCDUMP_KGC_TAB, BCDUMP_KGC_I64, BCDUMP_KGC_U64,
@@ -61,7 +63,7 @@
/* -- Bytecode reader/writer ---------------------------------------------- */
LJ_FUNC int lj_bcwrite(lua_State *L, GCproto *pt, lua_Writer writer,
- void *data, int strip);
+ void *data, uint32_t flags);
LJ_FUNC GCproto *lj_bcread_proto(LexState *ls);
LJ_FUNC GCproto *lj_bcread(LexState *ls);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcread.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcread.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcread.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bytecode reader.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_bcread_c
@@ -47,17 +47,17 @@
/* Refill buffer. */
static LJ_NOINLINE void bcread_fill(LexState *ls, MSize len, int need)
{
- lua_assert(len != 0);
+ lj_assertLS(len != 0, "empty refill");
if (len > LJ_MAX_BUF || ls->c < 0)
bcread_error(ls, LJ_ERR_BCBAD);
do {
const char *buf;
size_t sz;
- char *p = sbufB(&ls->sb);
+ char *p = ls->sb.b;
MSize n = (MSize)(ls->pe - ls->p);
if (n) { /* Copy remainder to buffer. */
if (sbuflen(&ls->sb)) { /* Move down in buffer. */
- lua_assert(ls->pe == sbufP(&ls->sb));
+ lj_assertLS(ls->pe == ls->sb.w, "bad buffer pointer");
if (ls->p != p) memmove(p, ls->p, n);
} else { /* Copy from buffer provided by reader. */
p = lj_buf_need(&ls->sb, len);
@@ -66,7 +66,7 @@
ls->p = p;
ls->pe = p + n;
}
- setsbufP(&ls->sb, p + n);
+ ls->sb.w = p + n;
buf = ls->rfunc(ls->L, ls->rdata, &sz); /* Get more data from reader. */
if (buf == NULL || sz == 0) { /* EOF? */
if (need) bcread_error(ls, LJ_ERR_BCBAD);
@@ -73,11 +73,12 @@
ls->c = -1; /* Only bad if we get called again. */
break;
}
+ if (sz >= LJ_MAX_BUF - n) lj_err_mem(ls->L);
if (n) { /* Append to buffer. */
n += (MSize)sz;
p = lj_buf_need(&ls->sb, n < len ? len : n);
- memcpy(sbufP(&ls->sb), buf, sz);
- setsbufP(&ls->sb, p + n);
+ memcpy(ls->sb.w, buf, sz);
+ ls->sb.w = p + n;
ls->p = p;
ls->pe = p + n;
} else { /* Return buffer provided by reader. */
@@ -84,13 +85,13 @@
ls->p = buf;
ls->pe = buf + sz;
}
- } while (ls->p + len > ls->pe);
+ } while ((MSize)(ls->pe - ls->p) < len);
}
/* Need a certain number of bytes. */
static LJ_AINLINE void bcread_need(LexState *ls, MSize len)
{
- if (LJ_UNLIKELY(ls->p + len > ls->pe))
+ if (LJ_UNLIKELY((MSize)(ls->pe - ls->p) < len))
bcread_fill(ls, len, 1);
}
@@ -97,7 +98,7 @@
/* Want to read up to a certain number of bytes, but may need less. */
static LJ_AINLINE void bcread_want(LexState *ls, MSize len)
{
- if (LJ_UNLIKELY(ls->p + len > ls->pe))
+ if (LJ_UNLIKELY((MSize)(ls->pe - ls->p) < len))
bcread_fill(ls, len, 0);
}
@@ -106,7 +107,7 @@
{
uint8_t *p = (uint8_t *)ls->p;
ls->p += len;
- lua_assert(ls->p <= ls->pe);
+ lj_assertLS(ls->p <= ls->pe, "buffer read overflow");
return p;
}
@@ -119,7 +120,7 @@
/* Read byte from buffer. */
static LJ_AINLINE uint32_t bcread_byte(LexState *ls)
{
- lua_assert(ls->p < ls->pe);
+ lj_assertLS(ls->p < ls->pe, "buffer read overflow");
return (uint32_t)(uint8_t)*ls->p++;
}
@@ -127,7 +128,7 @@
static LJ_AINLINE uint32_t bcread_uleb128(LexState *ls)
{
uint32_t v = lj_buf_ruleb128(&ls->p);
- lua_assert(ls->p <= ls->pe);
+ lj_assertLS(ls->p <= ls->pe, "buffer read overflow");
return v;
}
@@ -144,7 +145,7 @@
} while (*p++ >= 0x80);
}
ls->p = (char *)p;
- lua_assert(ls->p <= ls->pe);
+ lj_assertLS(ls->p <= ls->pe, "buffer read overflow");
return v;
}
@@ -191,7 +192,7 @@
o->u32.lo = bcread_uleb128(ls);
o->u32.hi = bcread_uleb128(ls);
} else {
- lua_assert(tp <= BCDUMP_KTAB_TRUE);
+ lj_assertLS(tp <= BCDUMP_KTAB_TRUE, "bad constant type %d", tp);
setpriV(o, ~tp);
}
}
@@ -213,7 +214,7 @@
for (i = 0; i < nhash; i++) {
TValue key;
bcread_ktabk(ls, &key);
- lua_assert(!tvisnil(&key));
+ lj_assertLS(!tvisnil(&key), "nil key");
bcread_ktabk(ls, lj_tab_set(ls->L, t, &key));
}
}
@@ -250,7 +251,7 @@
#endif
} else {
lua_State *L = ls->L;
- lua_assert(tp == BCDUMP_KGC_CHILD);
+ lj_assertLS(tp == BCDUMP_KGC_CHILD, "bad constant type %d", tp);
if (L->top <= bcread_oldtop(L, ls)) /* Stack underflow? */
bcread_error(ls, LJ_ERR_BCBAD);
L->top--;
@@ -280,8 +281,11 @@
static void bcread_bytecode(LexState *ls, GCproto *pt, MSize sizebc)
{
BCIns *bc = proto_bc(pt);
- bc[0] = BCINS_AD((pt->flags & PROTO_VARARG) ? BC_FUNCV : BC_FUNCF,
- pt->framesize, 0);
+ BCIns op;
+ if (ls->fr2 != LJ_FR2) op = BC_NOT; /* Mark non-native prototype. */
+ else if ((pt->flags & PROTO_VARARG)) op = BC_FUNCV;
+ else op = BC_FUNCF;
+ bc[0] = BCINS_AD(op, pt->framesize, 0);
bcread_block(ls, bc+1, (sizebc-1)*(MSize)sizeof(BCIns));
/* Swap bytecode instructions if the endianess differs. */
if (bcread_swap(ls)) {
@@ -394,21 +398,17 @@
bcread_byte(ls) != BCDUMP_VERSION) return 0;
bcread_flags(ls) = flags = bcread_uleb128(ls);
if ((flags & ~(BCDUMP_F_KNOWN)) != 0) return 0;
- if ((flags & BCDUMP_F_FR2) != LJ_FR2*BCDUMP_F_FR2) return 0;
+ if ((flags & BCDUMP_F_FR2) != (uint32_t)ls->fr2*BCDUMP_F_FR2) return 0;
if ((flags & BCDUMP_F_FFI)) {
#if LJ_HASFFI
lua_State *L = ls->L;
- if (!ctype_ctsG(G(L))) {
- ptrdiff_t oldtop = savestack(L, L->top);
- luaopen_ffi(L); /* Load FFI library on-demand. */
- L->top = restorestack(L, oldtop);
- }
+ ctype_loadffi(L);
#else
return 0;
#endif
}
if ((flags & BCDUMP_F_STRIP)) {
- ls->chunkname = lj_str_newz(ls->L, ls->chunkarg);
+ ls->chunkname = lj_str_newz(ls->L, *ls->chunkarg == BCDUMP_HEAD1 ? "=?" : ls->chunkarg);
} else {
MSize len = bcread_uleb128(ls);
bcread_need(ls, len);
@@ -421,7 +421,7 @@
GCproto *lj_bcread(LexState *ls)
{
lua_State *L = ls->L;
- lua_assert(ls->c == BCDUMP_HEAD1);
+ lj_assertLS(ls->c == BCDUMP_HEAD1, "bad bytecode header");
bcread_savetop(L, ls, L->top);
lj_buf_reset(&ls->sb);
/* Check for a valid bytecode dump header. */
@@ -447,8 +447,7 @@
setprotoV(L, L->top, pt);
incr_top(L);
}
- if ((int32_t)(2*(uint32_t)(ls->pe - ls->p)) > 0 ||
- L->top-1 != bcread_oldtop(L, ls))
+ if ((ls->pe != ls->p && !ls->endmark) || L->top-1 != bcread_oldtop(L, ls))
bcread_error(ls, LJ_ERR_BCBAD);
/* Pop off last prototype. */
L->top--;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcwrite.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcwrite.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_bcwrite.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Bytecode writer.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_bcwrite_c
@@ -27,10 +27,21 @@
GCproto *pt; /* Root prototype. */
lua_Writer wfunc; /* Writer callback. */
void *wdata; /* Writer callback data. */
- int strip; /* Strip debug info. */
+ TValue **heap; /* Heap used for deterministic sorting. */
+ uint32_t heapsz; /* Size of heap. */
+ uint32_t flags; /* BCDUMP_F_* flags. */
int status; /* Status from writer callback. */
+#ifdef LUA_USE_ASSERT
+ global_State *g;
+#endif
} BCWriteCtx;
+#ifdef LUA_USE_ASSERT
+#define lj_assertBCW(c, ...) lj_assertG_(ctx->g, (c), __VA_ARGS__)
+#else
+#define lj_assertBCW(c, ...) ((void)ctx)
+#endif
+
/* -- Bytecode writer ----------------------------------------------------- */
/* Write a single constant key/value of a template table. */
@@ -53,7 +64,7 @@
if (num == (lua_Number)k) { /* -0 is never a constant. */
*p++ = BCDUMP_KTAB_INT;
p = lj_strfmt_wuleb128(p, k);
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
return;
}
}
@@ -61,12 +72,81 @@
p = lj_strfmt_wuleb128(p, o->u32.lo);
p = lj_strfmt_wuleb128(p, o->u32.hi);
} else {
- lua_assert(tvispri(o));
+ lj_assertBCW(tvispri(o), "unhandled type %d", itype(o));
*p++ = BCDUMP_KTAB_NIL+~itype(o);
}
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
}
+/* Compare two template table keys. */
+static LJ_AINLINE int bcwrite_ktabk_lt(TValue *a, TValue *b)
+{
+ uint32_t at = itype(a), bt = itype(b);
+ if (at != bt) { /* This also handles false and true keys. */
+ return at < bt;
+ } else if (at == LJ_TSTR) {
+ return lj_str_cmp(strV(a), strV(b)) < 0;
+ } else {
+ return a->u64 < b->u64; /* This works for numbers and integers. */
+ }
+}
+
+/* Insert key into a sorted heap. */
+static void bcwrite_ktabk_heap_insert(TValue **heap, MSize idx, MSize end,
+ TValue *key)
+{
+ MSize child;
+ while ((child = idx * 2 + 1) < end) {
+ /* Find lower of the two children. */
+ TValue *c0 = heap[child];
+ if (child + 1 < end) {
+ TValue *c1 = heap[child + 1];
+ if (bcwrite_ktabk_lt(c1, c0)) {
+ c0 = c1;
+ child++;
+ }
+ }
+ if (bcwrite_ktabk_lt(key, c0)) break; /* Key lower? Found our position. */
+ heap[idx] = c0; /* Move lower child up. */
+ idx = child; /* Descend. */
+ }
+ heap[idx] = key; /* Insert key here. */
+}
+
+/* Resize heap, dropping content. */
+static void bcwrite_heap_resize(BCWriteCtx *ctx, uint32_t nsz)
+{
+ lua_State *L = sbufL(&ctx->sb);
+ if (ctx->heapsz) {
+ lj_mem_freevec(G(L), ctx->heap, ctx->heapsz, TValue *);
+ ctx->heapsz = 0;
+ }
+ if (nsz) {
+ ctx->heap = lj_mem_newvec(L, nsz, TValue *);
+ ctx->heapsz = nsz;
+ }
+}
+
+/* Write hash part of template table in sorted order. */
+static void bcwrite_ktab_sorted_hash(BCWriteCtx *ctx, Node *node, MSize nhash)
+{
+ TValue **heap = ctx->heap;
+ MSize i = nhash;
+ for (;; node--) { /* Build heap. */
+ if (!tvisnil(&node->key)) {
+ bcwrite_ktabk_heap_insert(heap, --i, nhash, &node->key);
+ if (i == 0) break;
+ }
+ }
+ do { /* Drain heap. */
+ TValue *key = heap[0]; /* Output lowest key from top. */
+ bcwrite_ktabk(ctx, key, 0);
+ bcwrite_ktabk(ctx, (TValue *)((char *)key - offsetof(Node, key)), 1);
+ key = heap[--nhash]; /* Remove last key. */
+ bcwrite_ktabk_heap_insert(heap, 0, nhash, key); /* Re-insert. */
+ } while (nhash);
+}
+
/* Write a template table. */
static void bcwrite_ktab(BCWriteCtx *ctx, char *p, const GCtab *t)
{
@@ -83,12 +163,12 @@
MSize i, hmask = t->hmask;
Node *node = noderef(t->node);
for (i = 0; i <= hmask; i++)
- nhash += !tvisnil(&node[i].val);
+ nhash += !tvisnil(&node[i].key);
}
/* Write number of array slots and hash slots. */
p = lj_strfmt_wuleb128(p, narray);
p = lj_strfmt_wuleb128(p, nhash);
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
if (narray) { /* Write array entries (may contain nil). */
MSize i;
TValue *o = tvref(t->array);
@@ -96,14 +176,20 @@
bcwrite_ktabk(ctx, o, 1);
}
if (nhash) { /* Write hash entries. */
- MSize i = nhash;
Node *node = noderef(t->node) + t->hmask;
- for (;; node--)
- if (!tvisnil(&node->val)) {
- bcwrite_ktabk(ctx, &node->key, 0);
- bcwrite_ktabk(ctx, &node->val, 1);
- if (--i == 0) break;
- }
+ if ((ctx->flags & BCDUMP_F_DETERMINISTIC) && nhash > 1) {
+ if (ctx->heapsz < nhash)
+ bcwrite_heap_resize(ctx, t->hmask + 1);
+ bcwrite_ktab_sorted_hash(ctx, node, nhash);
+ } else {
+ MSize i = nhash;
+ for (;; node--)
+ if (!tvisnil(&node->key)) {
+ bcwrite_ktabk(ctx, &node->key, 0);
+ bcwrite_ktabk(ctx, &node->val, 1);
+ if (--i == 0) break;
+ }
+ }
}
}
@@ -121,7 +207,7 @@
tp = BCDUMP_KGC_STR + gco2str(o)->len;
need = 5+gco2str(o)->len;
} else if (o->gch.gct == ~LJ_TPROTO) {
- lua_assert((pt->flags & PROTO_CHILD));
+ lj_assertBCW((pt->flags & PROTO_CHILD), "prototype has unexpected child");
tp = BCDUMP_KGC_CHILD;
#if LJ_HASFFI
} else if (o->gch.gct == ~LJ_TCDATA) {
@@ -132,12 +218,14 @@
} else if (id == CTID_UINT64) {
tp = BCDUMP_KGC_U64;
} else {
- lua_assert(id == CTID_COMPLEX_DOUBLE);
+ lj_assertBCW(id == CTID_COMPLEX_DOUBLE,
+ "bad cdata constant CTID %d", id);
tp = BCDUMP_KGC_COMPLEX;
}
#endif
} else {
- lua_assert(o->gch.gct == ~LJ_TTAB);
+ lj_assertBCW(o->gch.gct == ~LJ_TTAB,
+ "bad constant GC type %d", o->gch.gct);
tp = BCDUMP_KGC_TAB;
need = 1+2*5;
}
@@ -161,7 +249,7 @@
}
#endif
}
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
}
}
@@ -178,7 +266,8 @@
goto save_int;
} else {
/* Write a 33 bit ULEB128 for the int (lsb=0) or loword (lsb=1). */
- if (!LJ_DUALNUM) { /* Narrow number constants to integers. */
+ if (!LJ_DUALNUM && o->u32.hi != LJ_KEYINDEX) {
+ /* Narrow number constants to integers. */
lua_Number num = numV(o);
k = lj_num2int(num);
if (num == (lua_Number)k) { /* -0 is never a constant. */
@@ -195,7 +284,7 @@
p = lj_strfmt_wuleb128(p, o->u32.hi);
}
}
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
}
/* Write bytecode instructions. */
@@ -219,10 +308,7 @@
q[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_IFORL+BC_FORL);
} else if (op == BC_JFORL || op == BC_JITERL || op == BC_JLOOP) {
BCReg rd = q[LJ_ENDIAN_SELECT(2, 1)] + (q[LJ_ENDIAN_SELECT(3, 0)] << 8);
- BCIns ins = traceref(J, rd)->startins;
- q[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_JFORL+BC_FORL);
- q[LJ_ENDIAN_SELECT(2, 1)] = bc_c(ins);
- q[LJ_ENDIAN_SELECT(3, 0)] = bc_b(ins);
+ memcpy(q, &traceref(J, rd)->startins, 4);
}
}
}
@@ -260,7 +346,7 @@
p = lj_strfmt_wuleb128(p, pt->sizekgc);
p = lj_strfmt_wuleb128(p, pt->sizekn);
p = lj_strfmt_wuleb128(p, pt->sizebc-1);
- if (!ctx->strip) {
+ if (!(ctx->flags & BCDUMP_F_STRIP)) {
if (proto_lineinfo(pt))
sizedbg = pt->sizept - (MSize)((char *)proto_lineinfo(pt) - (char *)pt);
p = lj_strfmt_wuleb128(p, sizedbg);
@@ -273,7 +359,7 @@
/* Write bytecode instructions and upvalue refs. */
p = bcwrite_bytecode(ctx, p, pt);
p = lj_buf_wmem(p, proto_uv(pt), pt->sizeuv*2);
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
/* Write constants. */
bcwrite_kgc(ctx, pt);
@@ -283,7 +369,7 @@
if (sizedbg) {
p = lj_buf_more(&ctx->sb, sizedbg);
p = lj_buf_wmem(p, proto_lineinfo(pt), sizedbg);
- setsbufP(&ctx->sb, p);
+ ctx->sb.w = p;
}
/* Pass buffer to writer function. */
@@ -290,9 +376,9 @@
if (ctx->status == 0) {
MSize n = sbuflen(&ctx->sb) - 5;
MSize nn = (lj_fls(n)+8)*9 >> 6;
- char *q = sbufB(&ctx->sb) + (5 - nn);
+ char *q = ctx->sb.b + (5 - nn);
p = lj_strfmt_wuleb128(q, n); /* Fill in final size. */
- lua_assert(p == sbufB(&ctx->sb) + 5);
+ lj_assertBCW(p == ctx->sb.b + 5, "bad ULEB128 write");
ctx->status = ctx->wfunc(sbufL(&ctx->sb), q, nn+n, ctx->wdata);
}
}
@@ -308,16 +394,15 @@
*p++ = BCDUMP_HEAD2;
*p++ = BCDUMP_HEAD3;
*p++ = BCDUMP_VERSION;
- *p++ = (ctx->strip ? BCDUMP_F_STRIP : 0) +
+ *p++ = (ctx->flags & (BCDUMP_F_STRIP | BCDUMP_F_FR2)) +
LJ_BE*BCDUMP_F_BE +
- ((ctx->pt->flags & PROTO_FFI) ? BCDUMP_F_FFI : 0) +
- LJ_FR2*BCDUMP_F_FR2;
- if (!ctx->strip) {
+ ((ctx->pt->flags & PROTO_FFI) ? BCDUMP_F_FFI : 0);
+ if (!(ctx->flags & BCDUMP_F_STRIP)) {
p = lj_strfmt_wuleb128(p, len);
p = lj_buf_wmem(p, name, len);
}
- ctx->status = ctx->wfunc(sbufL(&ctx->sb), sbufB(&ctx->sb),
- (MSize)(p - sbufB(&ctx->sb)), ctx->wdata);
+ ctx->status = ctx->wfunc(sbufL(&ctx->sb), ctx->sb.b,
+ (MSize)(p - ctx->sb.b), ctx->wdata);
}
/* Write footer of bytecode dump. */
@@ -343,7 +428,7 @@
/* Write bytecode for a prototype. */
int lj_bcwrite(lua_State *L, GCproto *pt, lua_Writer writer, void *data,
- int strip)
+ uint32_t flags)
{
BCWriteCtx ctx;
int status;
@@ -350,12 +435,18 @@
ctx.pt = pt;
ctx.wfunc = writer;
ctx.wdata = data;
- ctx.strip = strip;
+ ctx.heapsz = 0;
+ if ((bc_op(proto_bc(pt)[0]) != BC_NOT) == LJ_FR2) flags |= BCDUMP_F_FR2;
+ ctx.flags = flags;
ctx.status = 0;
+#ifdef LUA_USE_ASSERT
+ ctx.g = G(L);
+#endif
lj_buf_init(L, &ctx.sb);
status = lj_vm_cpcall(L, NULL, &ctx, cpwriter);
if (status == 0) status = ctx.status;
lj_buf_free(G(sbufL(&ctx.sb)), &ctx.sb);
+ bcwrite_heap_resize(&ctx, 0);
return status;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Buffer handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_buf_c
@@ -20,44 +20,83 @@
{
MSize osz = sbufsz(sb), len = sbuflen(sb), nsz = osz;
char *b;
+ GCSize flag;
if (nsz < LJ_MIN_SBUF) nsz = LJ_MIN_SBUF;
while (nsz < sz) nsz += nsz;
- b = (char *)lj_mem_realloc(sbufL(sb), sbufB(sb), osz, nsz);
- setmref(sb->b, b);
- setmref(sb->p, b + len);
- setmref(sb->e, b + nsz);
+ flag = sbufflag(sb);
+ if ((flag & SBUF_FLAG_COW)) { /* Copy-on-write semantics. */
+ lj_assertG_(G(sbufL(sb)), sb->w == sb->e, "bad SBuf COW");
+ b = (char *)lj_mem_new(sbufL(sb), nsz);
+ setsbufflag(sb, flag & ~(GCSize)SBUF_FLAG_COW);
+ setgcrefnull(sbufX(sb)->cowref);
+ memcpy(b, sb->b, osz);
+ } else {
+ b = (char *)lj_mem_realloc(sbufL(sb), sb->b, osz, nsz);
+ }
+ if ((flag & SBUF_FLAG_EXT)) {
+ sbufX(sb)->r = sbufX(sb)->r - sb->b + b; /* Adjust read pointer, too. */
+ }
+ /* Adjust buffer pointers. */
+ sb->b = b;
+ sb->w = b + len;
+ sb->e = b + nsz;
+ if ((flag & SBUF_FLAG_BORROW)) { /* Adjust borrowed buffer pointers. */
+ SBuf *bsb = mref(sbufX(sb)->bsb, SBuf);
+ bsb->b = b;
+ bsb->w = b + len;
+ bsb->e = b + nsz;
+ }
}
LJ_NOINLINE char *LJ_FASTCALL lj_buf_need2(SBuf *sb, MSize sz)
{
- lua_assert(sz > sbufsz(sb));
+ lj_assertG_(G(sbufL(sb)), sz > sbufsz(sb), "SBuf overflow");
if (LJ_UNLIKELY(sz > LJ_MAX_BUF))
lj_err_mem(sbufL(sb));
buf_grow(sb, sz);
- return sbufB(sb);
+ return sb->b;
}
LJ_NOINLINE char *LJ_FASTCALL lj_buf_more2(SBuf *sb, MSize sz)
{
- MSize len = sbuflen(sb);
- lua_assert(sz > sbufleft(sb));
- if (LJ_UNLIKELY(sz > LJ_MAX_BUF || len + sz > LJ_MAX_BUF))
- lj_err_mem(sbufL(sb));
- buf_grow(sb, len + sz);
- return sbufP(sb);
+ if (sbufisext(sb)) {
+ SBufExt *sbx = (SBufExt *)sb;
+ MSize len = sbufxlen(sbx);
+ if (LJ_UNLIKELY(sz > LJ_MAX_BUF || len + sz > LJ_MAX_BUF))
+ lj_err_mem(sbufL(sbx));
+ if (len + sz > sbufsz(sbx)) { /* Must grow. */
+ buf_grow((SBuf *)sbx, len + sz);
+ } else if (sbufiscow(sb) || sbufxslack(sbx) < (sbufsz(sbx) >> 3)) {
+ /* Also grow to avoid excessive compactions, if slack < size/8. */
+ buf_grow((SBuf *)sbx, sbuflen(sbx) + sz); /* Not sbufxlen! */
+ return sbx->w;
+ }
+ if (sbx->r != sbx->b) { /* Compact by moving down. */
+ memmove(sbx->b, sbx->r, len);
+ sbx->r = sbx->b;
+ sbx->w = sbx->b + len;
+ lj_assertG_(G(sbufL(sbx)), len + sz <= sbufsz(sbx), "bad SBuf compact");
+ }
+ } else {
+ MSize len = sbuflen(sb);
+ lj_assertG_(G(sbufL(sb)), sz > sbufleft(sb), "SBuf overflow");
+ if (LJ_UNLIKELY(sz > LJ_MAX_BUF || len + sz > LJ_MAX_BUF))
+ lj_err_mem(sbufL(sb));
+ buf_grow(sb, len + sz);
+ }
+ return sb->w;
}
void LJ_FASTCALL lj_buf_shrink(lua_State *L, SBuf *sb)
{
- char *b = sbufB(sb);
- MSize osz = (MSize)(sbufE(sb) - b);
+ char *b = sb->b;
+ MSize osz = (MSize)(sb->e - b);
if (osz > 2*LJ_MIN_SBUF) {
- MSize n = (MSize)(sbufP(sb) - b);
b = lj_mem_realloc(L, b, osz, (osz >> 1));
- setmref(sb->b, b);
- setmref(sb->p, b + n);
- setmref(sb->e, b + (osz >> 1));
+ sb->w = sb->b = b; /* Not supposed to keep data across shrinks. */
+ sb->e = b + (osz >> 1);
}
+ lj_assertG_(G(sbufL(sb)), !sbufisext(sb), "YAGNI shrink SBufExt");
}
char * LJ_FASTCALL lj_buf_tmp(lua_State *L, MSize sz)
@@ -67,30 +106,62 @@
return lj_buf_need(sb, sz);
}
+#if LJ_HASBUFFER && LJ_HASJIT
+void lj_bufx_set(SBufExt *sbx, const char *p, MSize len, GCobj *ref)
+{
+ lua_State *L = sbufL(sbx);
+ lj_bufx_free(L, sbx);
+ lj_bufx_set_cow(L, sbx, p, len);
+ setgcref(sbx->cowref, ref);
+ lj_gc_objbarrier(L, (GCudata *)sbx - 1, ref);
+}
+
+#if LJ_HASFFI
+MSize LJ_FASTCALL lj_bufx_more(SBufExt *sbx, MSize sz)
+{
+ lj_buf_more((SBuf *)sbx, sz);
+ return sbufleft(sbx);
+}
+#endif
+#endif
+
/* -- Low-level buffer put operations ------------------------------------- */
SBuf *lj_buf_putmem(SBuf *sb, const void *q, MSize len)
{
- char *p = lj_buf_more(sb, len);
- p = lj_buf_wmem(p, q, len);
- setsbufP(sb, p);
+ char *w = lj_buf_more(sb, len);
+ w = lj_buf_wmem(w, q, len);
+ sb->w = w;
return sb;
}
-SBuf * LJ_FASTCALL lj_buf_putchar(SBuf *sb, int c)
+#if LJ_HASJIT || LJ_HASFFI
+static LJ_NOINLINE SBuf * LJ_FASTCALL lj_buf_putchar2(SBuf *sb, int c)
{
- char *p = lj_buf_more(sb, 1);
- *p++ = (char)c;
- setsbufP(sb, p);
+ char *w = lj_buf_more2(sb, 1);
+ *w++ = (char)c;
+ sb->w = w;
return sb;
}
+SBuf * LJ_FASTCALL lj_buf_putchar(SBuf *sb, int c)
+{
+ char *w = sb->w;
+ if (LJ_LIKELY(w < sb->e)) {
+ *w++ = (char)c;
+ sb->w = w;
+ return sb;
+ }
+ return lj_buf_putchar2(sb, c);
+}
+#endif
+
SBuf * LJ_FASTCALL lj_buf_putstr(SBuf *sb, GCstr *s)
{
MSize len = s->len;
- char *p = lj_buf_more(sb, len);
- p = lj_buf_wmem(p, strdata(s), len);
- setsbufP(sb, p);
+ char *w = lj_buf_more(sb, len);
+ w = lj_buf_wmem(w, strdata(s), len);
+ sb->w = w;
return sb;
}
@@ -99,11 +170,11 @@
SBuf * LJ_FASTCALL lj_buf_putstr_reverse(SBuf *sb, GCstr *s)
{
MSize len = s->len;
- char *p = lj_buf_more(sb, len), *e = p+len;
+ char *w = lj_buf_more(sb, len), *e = w+len;
const char *q = strdata(s)+len-1;
- while (p < e)
- *p++ = *q--;
- setsbufP(sb, p);
+ while (w < e)
+ *w++ = *q--;
+ sb->w = w;
return sb;
}
@@ -110,18 +181,18 @@
SBuf * LJ_FASTCALL lj_buf_putstr_lower(SBuf *sb, GCstr *s)
{
MSize len = s->len;
- char *p = lj_buf_more(sb, len), *e = p+len;
+ char *w = lj_buf_more(sb, len), *e = w+len;
const char *q = strdata(s);
- for (; p < e; p++, q++) {
+ for (; w < e; w++, q++) {
uint32_t c = *(unsigned char *)q;
#if LJ_TARGET_PPC
- *p = c + ((c >= 'A' && c <= 'Z') << 5);
+ *w = c + ((c >= 'A' && c <= 'Z') << 5);
#else
if (c >= 'A' && c <= 'Z') c += 0x20;
- *p = c;
+ *w = c;
#endif
}
- setsbufP(sb, p);
+ sb->w = w;
return sb;
}
@@ -128,18 +199,18 @@
SBuf * LJ_FASTCALL lj_buf_putstr_upper(SBuf *sb, GCstr *s)
{
MSize len = s->len;
- char *p = lj_buf_more(sb, len), *e = p+len;
+ char *w = lj_buf_more(sb, len), *e = w+len;
const char *q = strdata(s);
- for (; p < e; p++, q++) {
+ for (; w < e; w++, q++) {
uint32_t c = *(unsigned char *)q;
#if LJ_TARGET_PPC
- *p = c - ((c >= 'a' && c <= 'z') << 5);
+ *w = c - ((c >= 'a' && c <= 'z') << 5);
#else
if (c >= 'a' && c <= 'z') c -= 0x20;
- *p = c;
+ *w = c;
#endif
}
- setsbufP(sb, p);
+ sb->w = w;
return sb;
}
@@ -148,21 +219,21 @@
MSize len = s->len;
if (rep > 0 && len) {
uint64_t tlen = (uint64_t)rep * len;
- char *p;
+ char *w;
if (LJ_UNLIKELY(tlen > LJ_MAX_STR))
lj_err_mem(sbufL(sb));
- p = lj_buf_more(sb, (MSize)tlen);
+ w = lj_buf_more(sb, (MSize)tlen);
if (len == 1) { /* Optimize a common case. */
uint32_t c = strdata(s)[0];
- do { *p++ = c; } while (--rep > 0);
+ do { *w++ = c; } while (--rep > 0);
} else {
const char *e = strdata(s) + len;
do {
const char *q = strdata(s);
- do { *p++ = *q++; } while (q < e);
+ do { *w++ = *q++; } while (q < e);
} while (--rep > 0);
}
- setsbufP(sb, p);
+ sb->w = w;
}
return sb;
}
@@ -173,27 +244,27 @@
if (i <= e) {
for (;;) {
cTValue *o = lj_tab_getint(t, i);
- char *p;
+ char *w;
if (!o) {
badtype: /* Error: bad element type. */
- setsbufP(sb, (void *)(intptr_t)i); /* Store failing index. */
+ sb->w = (char *)(intptr_t)i; /* Store failing index. */
return NULL;
} else if (tvisstr(o)) {
MSize len = strV(o)->len;
- p = lj_buf_wmem(lj_buf_more(sb, len + seplen), strVdata(o), len);
+ w = lj_buf_wmem(lj_buf_more(sb, len + seplen), strVdata(o), len);
} else if (tvisint(o)) {
- p = lj_strfmt_wint(lj_buf_more(sb, STRFMT_MAXBUF_INT+seplen), intV(o));
+ w = lj_strfmt_wint(lj_buf_more(sb, STRFMT_MAXBUF_INT+seplen), intV(o));
} else if (tvisnum(o)) {
- p = lj_buf_more(lj_strfmt_putfnum(sb, STRFMT_G14, numV(o)), seplen);
+ w = lj_buf_more(lj_strfmt_putfnum(sb, STRFMT_G14, numV(o)), seplen);
} else {
goto badtype;
}
if (i++ == e) {
- setsbufP(sb, p);
+ sb->w = w;
break;
}
- if (seplen) p = lj_buf_wmem(p, strdata(sep), seplen);
- setsbufP(sb, p);
+ if (seplen) w = lj_buf_wmem(w, strdata(sep), seplen);
+ sb->w = w;
}
}
return sb;
@@ -203,7 +274,7 @@
GCstr * LJ_FASTCALL lj_buf_tostr(SBuf *sb)
{
- return lj_str_new(sbufL(sb), sbufB(sb), sbuflen(sb));
+ return lj_str_new(sbufL(sb), sb->b, sbuflen(sb));
}
/* Concatenate two strings. */
@@ -219,14 +290,14 @@
/* Read ULEB128 from buffer. */
uint32_t LJ_FASTCALL lj_buf_ruleb128(const char **pp)
{
- const uint8_t *p = (const uint8_t *)*pp;
- uint32_t v = *p++;
+ const uint8_t *w = (const uint8_t *)*pp;
+ uint32_t v = *w++;
if (LJ_UNLIKELY(v >= 0x80)) {
int sh = 0;
v &= 0x7f;
- do { v |= ((*p & 0x7f) << (sh += 7)); } while (*p++ >= 0x80);
+ do { v |= ((*w & 0x7f) << (sh += 7)); } while (*w++ >= 0x80);
}
- *pp = (const char *)p;
+ *pp = (const char *)w;
return v;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_buf.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Buffer handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_BUF_H
@@ -10,17 +10,61 @@
#include "lj_gc.h"
#include "lj_str.h"
-/* Resizable string buffers. Struct definition in lj_obj.h. */
-#define sbufB(sb) (mref((sb)->b, char))
-#define sbufP(sb) (mref((sb)->p, char))
-#define sbufE(sb) (mref((sb)->e, char))
-#define sbufL(sb) (mref((sb)->L, lua_State))
-#define sbufsz(sb) ((MSize)(sbufE((sb)) - sbufB((sb))))
-#define sbuflen(sb) ((MSize)(sbufP((sb)) - sbufB((sb))))
-#define sbufleft(sb) ((MSize)(sbufE((sb)) - sbufP((sb))))
-#define setsbufP(sb, q) (setmref((sb)->p, (q)))
-#define setsbufL(sb, l) (setmref((sb)->L, (l)))
+/* Resizable string buffers. */
+/* The SBuf struct definition is in lj_obj.h:
+** char *w; Write pointer.
+** char *e; End pointer.
+** char *b; Base pointer.
+** MRef L; lua_State, used for buffer resizing. Extension bits in 3 LSB.
+*/
+
+/* Extended string buffer. */
+typedef struct SBufExt {
+ SBufHeader;
+ union {
+ GCRef cowref; /* Copy-on-write object reference. */
+ MRef bsb; /* Borrowed string buffer. */
+ };
+ char *r; /* Read pointer. */
+ GCRef dict_str; /* Serialization string dictionary table. */
+ GCRef dict_mt; /* Serialization metatable dictionary table. */
+ int depth; /* Remaining recursion depth. */
+} SBufExt;
+
+#define sbufsz(sb) ((MSize)((sb)->e - (sb)->b))
+#define sbuflen(sb) ((MSize)((sb)->w - (sb)->b))
+#define sbufleft(sb) ((MSize)((sb)->e - (sb)->w))
+#define sbufxlen(sbx) ((MSize)((sbx)->w - (sbx)->r))
+#define sbufxslack(sbx) ((MSize)((sbx)->r - (sbx)->b))
+
+#define SBUF_MASK_FLAG (7)
+#define SBUF_MASK_L (~(GCSize)SBUF_MASK_FLAG)
+#define SBUF_FLAG_EXT 1 /* Extended string buffer. */
+#define SBUF_FLAG_COW 2 /* Copy-on-write buffer. */
+#define SBUF_FLAG_BORROW 4 /* Borrowed string buffer. */
+
+#define sbufL(sb) \
+ ((lua_State *)(void *)(uintptr_t)(mrefu((sb)->L) & SBUF_MASK_L))
+#define setsbufL(sb, l) (setmref((sb)->L, (l)))
+#define setsbufXL(sb, l, flag) \
+ (setmrefu((sb)->L, (GCSize)(uintptr_t)(void *)(l) + (flag)))
+#define setsbufXL_(sb, l) \
+ (setmrefu((sb)->L, (GCSize)(uintptr_t)(void *)(l) | (mrefu((sb)->L) & SBUF_MASK_FLAG)))
+
+#define sbufflag(sb) (mrefu((sb)->L))
+#define sbufisext(sb) (sbufflag((sb)) & SBUF_FLAG_EXT)
+#define sbufiscow(sb) (sbufflag((sb)) & SBUF_FLAG_COW)
+#define sbufisborrow(sb) (sbufflag((sb)) & SBUF_FLAG_BORROW)
+#define sbufiscoworborrow(sb) (sbufflag((sb)) & (SBUF_FLAG_COW|SBUF_FLAG_BORROW))
+#define sbufX(sb) \
+ (lj_assertG_(G(sbufL(sb)), sbufisext(sb), "not an SBufExt"), (SBufExt *)(sb))
+#define setsbufflag(sb, flag) (setmrefu((sb)->L, (flag)))
+
+#define tvisbuf(o) \
+ (LJ_HASBUFFER && tvisudata(o) && udataV(o)->udtype == UDTYPE_BUFFER)
+#define bufV(o) check_exp(tvisbuf(o), ((SBufExt *)uddata(udataV(o))))
+
/* Buffer management */
LJ_FUNC char *LJ_FASTCALL lj_buf_need2(SBuf *sb, MSize sz);
LJ_FUNC char *LJ_FASTCALL lj_buf_more2(SBuf *sb, MSize sz);
@@ -30,12 +74,12 @@
static LJ_AINLINE void lj_buf_init(lua_State *L, SBuf *sb)
{
setsbufL(sb, L);
- setmref(sb->p, NULL); setmref(sb->e, NULL); setmref(sb->b, NULL);
+ sb->w = sb->e = sb->b = NULL;
}
static LJ_AINLINE void lj_buf_reset(SBuf *sb)
{
- setmrefr(sb->p, sb->b);
+ sb->w = sb->b;
}
static LJ_AINLINE SBuf *lj_buf_tmp_(lua_State *L)
@@ -48,7 +92,8 @@
static LJ_AINLINE void lj_buf_free(global_State *g, SBuf *sb)
{
- lj_mem_free(g, sbufB(sb), sbufsz(sb));
+ lj_assertG(!sbufisext(sb), "bad free of SBufExt");
+ lj_mem_free(g, sb->b, sbufsz(sb));
}
static LJ_AINLINE char *lj_buf_need(SBuf *sb, MSize sz)
@@ -55,7 +100,7 @@
{
if (LJ_UNLIKELY(sz > sbufsz(sb)))
return lj_buf_need2(sb, sz);
- return sbufB(sb);
+ return sb->b;
}
static LJ_AINLINE char *lj_buf_more(SBuf *sb, MSize sz)
@@ -62,12 +107,62 @@
{
if (LJ_UNLIKELY(sz > sbufleft(sb)))
return lj_buf_more2(sb, sz);
- return sbufP(sb);
+ return sb->w;
}
+/* Extended buffer management */
+static LJ_AINLINE void lj_bufx_init(lua_State *L, SBufExt *sbx)
+{
+ memset(sbx, 0, sizeof(SBufExt));
+ setsbufXL(sbx, L, SBUF_FLAG_EXT);
+}
+
+static LJ_AINLINE void lj_bufx_set_borrow(lua_State *L, SBufExt *sbx, SBuf *sb)
+{
+ setsbufXL(sbx, L, SBUF_FLAG_EXT | SBUF_FLAG_BORROW);
+ setmref(sbx->bsb, sb);
+ sbx->r = sbx->w = sbx->b = sb->b;
+ sbx->e = sb->e;
+}
+
+static LJ_AINLINE void lj_bufx_set_cow(lua_State *L, SBufExt *sbx,
+ const char *p, MSize len)
+{
+ setsbufXL(sbx, L, SBUF_FLAG_EXT | SBUF_FLAG_COW);
+ sbx->r = sbx->b = (char *)p;
+ sbx->w = sbx->e = (char *)p + len;
+}
+
+static LJ_AINLINE void lj_bufx_reset(SBufExt *sbx)
+{
+ if (sbufiscow(sbx)) {
+ setmrefu(sbx->L, (mrefu(sbx->L) & ~(GCSize)SBUF_FLAG_COW));
+ setgcrefnull(sbx->cowref);
+ sbx->b = sbx->e = NULL;
+ }
+ sbx->r = sbx->w = sbx->b;
+}
+
+static LJ_AINLINE void lj_bufx_free(lua_State *L, SBufExt *sbx)
+{
+ if (!sbufiscoworborrow(sbx)) lj_mem_free(G(L), sbx->b, sbufsz(sbx));
+ setsbufXL(sbx, L, SBUF_FLAG_EXT);
+ setgcrefnull(sbx->cowref);
+ sbx->r = sbx->w = sbx->b = sbx->e = NULL;
+}
+
+#if LJ_HASBUFFER && LJ_HASJIT
+LJ_FUNC void lj_bufx_set(SBufExt *sbx, const char *p, MSize len, GCobj *o);
+#if LJ_HASFFI
+LJ_FUNC MSize LJ_FASTCALL lj_bufx_more(SBufExt *sbx, MSize sz);
+#endif
+#endif
+
/* Low-level buffer put operations */
LJ_FUNC SBuf *lj_buf_putmem(SBuf *sb, const void *q, MSize len);
+#if LJ_HASJIT || LJ_HASFFI
LJ_FUNC SBuf * LJ_FASTCALL lj_buf_putchar(SBuf *sb, int c);
+#endif
LJ_FUNC SBuf * LJ_FASTCALL lj_buf_putstr(SBuf *sb, GCstr *s);
static LJ_AINLINE char *lj_buf_wmem(char *p, const void *q, MSize len)
@@ -77,9 +172,9 @@
static LJ_AINLINE void lj_buf_putb(SBuf *sb, int c)
{
- char *p = lj_buf_more(sb, 1);
- *p++ = (char)c;
- setsbufP(sb, p);
+ char *w = lj_buf_more(sb, 1);
+ *w++ = (char)c;
+ sb->w = w;
}
/* High-level buffer put operations */
@@ -97,7 +192,7 @@
static LJ_AINLINE GCstr *lj_buf_str(lua_State *L, SBuf *sb)
{
- return lj_str_new(L, sbufB(sb), sbuflen(sb));
+ return lj_str_new(L, sb->b, sbuflen(sb));
}
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C data arithmetic.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -44,9 +44,13 @@
p = (uint8_t *)cdata_getptr(p, ct->size);
if (ctype_isref(ct->info)) ct = ctype_rawchild(cts, ct);
} else if (ctype_isfunc(ct->info)) {
+ CTypeID id0 = i ? ctype_typeid(cts, ca->ct[0]) : 0;
p = (uint8_t *)*(void **)p;
ct = ctype_get(cts,
lj_ctype_intern(cts, CTINFO(CT_PTR, CTALIGN_PTR|id), CTSIZE_PTR));
+ if (i) { /* cts->tab may have been reallocated. */
+ ca->ct[0] = ctype_get(cts, id0);
+ }
}
if (ctype_isenum(ct->info)) ct = ctype_child(cts, ct);
ca->ct[i] = ct;
@@ -122,7 +126,7 @@
setboolV(L->top-1, ((uintptr_t)pp < (uintptr_t)pp2));
return 1;
} else {
- lua_assert(mm == MM_le);
+ lj_assertL(mm == MM_le, "bad metamethod %d", mm);
setboolV(L->top-1, ((uintptr_t)pp <= (uintptr_t)pp2));
return 1;
}
@@ -207,8 +211,10 @@
else
*up = lj_carith_powu64(u0, u1);
break;
- case MM_unm: *up = (uint64_t)-(int64_t)u0; break;
- default: lua_assert(0); break;
+ case MM_unm: *up = ~u0+1u; break;
+ default:
+ lj_assertL(0, "bad metamethod %d", mm);
+ break;
}
lj_gc_check(L);
return 1;
@@ -265,7 +271,7 @@
{
CTState *cts = ctype_cts(L);
CDArith ca;
- if (carith_checkarg(L, cts, &ca)) {
+ if (carith_checkarg(L, cts, &ca) && mm != MM_len && mm != MM_concat) {
if (carith_int64(L, cts, &ca, mm) || carith_ptr(L, cts, &ca, mm)) {
copyTV(L, &G(L)->tmptv2, L->top-1); /* Remember for trace recorder. */
return 1;
@@ -301,7 +307,9 @@
case IR_BSAR-IR_BSHL: x = lj_carith_sar64(x, sh); break;
case IR_BROL-IR_BSHL: x = lj_carith_rol64(x, sh); break;
case IR_BROR-IR_BSHL: x = lj_carith_ror64(x, sh); break;
- default: lua_assert(0); break;
+ default:
+ lj_assertX(0, "bad shift op %d", op);
+ break;
}
return x;
}
@@ -341,13 +349,10 @@
if (LJ_LIKELY(tvisint(o))) {
return (uint32_t)intV(o);
} else {
- int32_t i = lj_num2bit(numV(o));
- if (LJ_DUALNUM) setintV(o, i);
- return (uint32_t)i;
+ return (uint32_t)lj_num2bit(numV(o));
}
}
-
/* -- 64 bit integer arithmetic helpers ----------------------------------- */
#if LJ_32 && LJ_HASJIT
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_carith.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C data arithmetic.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CARITH_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C call handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -20,12 +20,15 @@
#if LJ_TARGET_X86
/* -- x86 calling conventions --------------------------------------------- */
+#define CCALL_PUSH(arg) \
+ *(GPRArg *)((uint8_t *)cc->stack + nsp) = (GPRArg)(arg), nsp += CTSIZE_PTR
+
#if LJ_ABI_WIN
#define CCALL_HANDLE_STRUCTRET \
/* Return structs bigger than 8 by reference (on stack only). */ \
cc->retref = (sz > 8); \
- if (cc->retref) cc->stack[nsp++] = (GPRArg)dp;
+ if (cc->retref) CCALL_PUSH(dp);
#define CCALL_HANDLE_COMPLEXRET CCALL_HANDLE_STRUCTRET
@@ -40,7 +43,7 @@
if (ngpr < maxgpr) \
cc->gpr[ngpr++] = (GPRArg)dp; \
else \
- cc->stack[nsp++] = (GPRArg)dp; \
+ CCALL_PUSH(dp); \
} else { /* Struct with single FP field ends up in FPR. */ \
cc->resx87 = ccall_classify_struct(cts, ctr); \
}
@@ -56,7 +59,7 @@
if (ngpr < maxgpr) \
cc->gpr[ngpr++] = (GPRArg)dp; \
else \
- cc->stack[nsp++] = (GPRArg)dp;
+ CCALL_PUSH(dp);
#endif
@@ -67,7 +70,7 @@
if (ngpr < maxgpr) \
cc->gpr[ngpr++] = (GPRArg)dp; \
else \
- cc->stack[nsp++] = (GPRArg)dp; \
+ CCALL_PUSH(dp); \
}
#endif
@@ -278,8 +281,8 @@
if (ngpr < maxgpr) { \
dp = &cc->gpr[ngpr]; \
if (ngpr + n > maxgpr) { \
- nsp += ngpr + n - maxgpr; /* Assumes contiguous gpr/stack fields. */ \
- if (nsp > CCALL_MAXSTACK) goto err_nyi; /* Too many arguments. */ \
+ nsp += (ngpr + n - maxgpr) * CTSIZE_PTR; /* Assumes contiguous gpr/stack fields. */ \
+ if (nsp > CCALL_SIZE_STACK) goto err_nyi; /* Too many arguments. */ \
ngpr = maxgpr; \
} else { \
ngpr += n; \
@@ -334,10 +337,11 @@
isfp = sz == 2*sizeof(float) ? 2 : 1;
#define CCALL_HANDLE_REGARG \
- if (LJ_TARGET_IOS && isva) { \
+ if (LJ_TARGET_OSX && isva) { \
/* IOS: All variadic arguments are on the stack. */ \
} else if (isfp) { /* Try to pass argument in FPRs. */ \
- int n2 = ctype_isvector(d->info) ? 1 : n*isfp; \
+ int n2 = ctype_isvector(d->info) ? 1 : \
+ isfp == 1 ? n : (d->size >> (4-isfp)); \
if (nfpr + n2 <= CCALL_NARG_FPR) { \
dp = &cc->fpr[nfpr]; \
nfpr += n2; \
@@ -344,10 +348,9 @@
goto done; \
} else { \
nfpr = CCALL_NARG_FPR; /* Prevent reordering. */ \
- if (LJ_TARGET_IOS && d->size < 8) goto err_nyi; \
} \
} else { /* Try to pass argument in GPRs. */ \
- if (!LJ_TARGET_IOS && (d->info & CTF_ALIGN) > CTALIGN_PTR) \
+ if (!LJ_TARGET_OSX && (d->info & CTF_ALIGN) > CTALIGN_PTR) \
ngpr = (ngpr + 1u) & ~1u; /* Align to regpair. */ \
if (ngpr + n <= maxgpr) { \
dp = &cc->gpr[ngpr]; \
@@ -355,7 +358,6 @@
goto done; \
} else { \
ngpr = maxgpr; /* Prevent reordering. */ \
- if (LJ_TARGET_IOS && d->size < 8) goto err_nyi; \
} \
}
@@ -387,6 +389,25 @@
#define CCALL_HANDLE_COMPLEXARG \
/* Pass complex by value in 2 or 4 GPRs. */
+#define CCALL_HANDLE_GPR \
+ /* Try to pass argument in GPRs. */ \
+ if (n > 1) { \
+ /* int64_t or complex (float). */ \
+ lj_assertL(n == 2 || n == 4, "bad GPR size %d", n); \
+ if (ctype_isinteger(d->info) || ctype_isfp(d->info)) \
+ ngpr = (ngpr + 1u) & ~1u; /* Align int64_t to regpair. */ \
+ else if (ngpr + n > maxgpr) \
+ ngpr = maxgpr; /* Prevent reordering. */ \
+ } \
+ if (ngpr + n <= maxgpr) { \
+ dp = &cc->gpr[ngpr]; \
+ ngpr += n; \
+ goto done; \
+ } \
+
+#if LJ_ABI_SOFTFP
+#define CCALL_HANDLE_REGARG CCALL_HANDLE_GPR
+#else
#define CCALL_HANDLE_REGARG \
if (isfp) { /* Try to pass argument in FPRs. */ \
if (nfpr + 1 <= CCALL_NARG_FPR) { \
@@ -395,24 +416,16 @@
d = ctype_get(cts, CTID_DOUBLE); /* FPRs always hold doubles. */ \
goto done; \
} \
- } else { /* Try to pass argument in GPRs. */ \
- if (n > 1) { \
- lua_assert(n == 2 || n == 4); /* int64_t or complex (float). */ \
- if (ctype_isinteger(d->info)) \
- ngpr = (ngpr + 1u) & ~1u; /* Align int64_t to regpair. */ \
- else if (ngpr + n > maxgpr) \
- ngpr = maxgpr; /* Prevent reordering. */ \
- } \
- if (ngpr + n <= maxgpr) { \
- dp = &cc->gpr[ngpr]; \
- ngpr += n; \
- goto done; \
- } \
+ } else { \
+ CCALL_HANDLE_GPR \
}
+#endif
+#if !LJ_ABI_SOFTFP
#define CCALL_HANDLE_RET \
if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
ctr = ctype_get(cts, CTID_DOUBLE); /* FPRs always hold doubles. */
+#endif
#elif LJ_TARGET_MIPS32
/* -- MIPS o32 calling conventions ---------------------------------------- */
@@ -459,8 +472,8 @@
if (ngpr < maxgpr) { \
dp = &cc->gpr[ngpr]; \
if (ngpr + n > maxgpr) { \
- nsp += ngpr + n - maxgpr; /* Assumes contiguous gpr/stack fields. */ \
- if (nsp > CCALL_MAXSTACK) goto err_nyi; /* Too many arguments. */ \
+ nsp += (ngpr + n - maxgpr) * CTSIZE_PTR; /* Assumes contiguous gpr/stack fields. */ \
+ if (nsp > CCALL_SIZE_STACK) goto err_nyi; /* Too many arguments. */ \
ngpr = maxgpr; \
} else { \
ngpr += n; \
@@ -553,8 +566,8 @@
if (ngpr < maxgpr) { \
dp = &cc->gpr[ngpr]; \
if (ngpr + n > maxgpr) { \
- nsp += ngpr + n - maxgpr; /* Assumes contiguous gpr/stack fields. */ \
- if (nsp > CCALL_MAXSTACK) goto err_nyi; /* Too many arguments. */ \
+ nsp += (ngpr + n - maxgpr) * CTSIZE_PTR; /* Assumes contiguous gpr/stack fields. */ \
+ if (nsp > CCALL_SIZE_STACK) goto err_nyi; /* Too many arguments. */ \
ngpr = maxgpr; \
} else { \
ngpr += n; \
@@ -631,7 +644,8 @@
ccall_classify_struct(cts, ct, rcl, ofs);
} else {
int cl = ctype_isfp(ct->info) ? CCALL_RCL_SSE : CCALL_RCL_INT;
- lua_assert(ctype_hassize(ct->info));
+ lj_assertCTS(ctype_hassize(ct->info),
+ "classify ctype %08x without size", ct->info);
if ((ofs & (ct->size-1))) cl = CCALL_RCL_MEM; /* Unaligned. */
rcl[(ofs >= 8)] |= cl;
}
@@ -656,12 +670,13 @@
}
/* Try to split up a small struct into registers. */
-static int ccall_struct_reg(CCallState *cc, GPRArg *dp, int *rcl)
+static int ccall_struct_reg(CCallState *cc, CTState *cts, GPRArg *dp, int *rcl)
{
MSize ngpr = cc->ngpr, nfpr = cc->nfpr;
uint32_t i;
+ UNUSED(cts);
for (i = 0; i < 2; i++) {
- lua_assert(!(rcl[i] & CCALL_RCL_MEM));
+ lj_assertCTS(!(rcl[i] & CCALL_RCL_MEM), "pass mem struct in reg");
if ((rcl[i] & CCALL_RCL_INT)) { /* Integer class takes precedence. */
if (ngpr >= CCALL_NARG_GPR) return 1; /* Register overflow. */
cc->gpr[ngpr++] = dp[i];
@@ -682,11 +697,13 @@
dp[0] = dp[1] = 0;
/* Convert to temp. struct. */
lj_cconv_ct_tv(cts, d, (uint8_t *)dp, o, CCF_ARG(narg));
- if (ccall_struct_reg(cc, dp, rcl)) { /* Register overflow? Pass on stack. */
- MSize nsp = cc->nsp, n = rcl[1] ? 2 : 1;
- if (nsp + n > CCALL_MAXSTACK) return 1; /* Too many arguments. */
- cc->nsp = nsp + n;
- memcpy(&cc->stack[nsp], dp, n*CTSIZE_PTR);
+ if (ccall_struct_reg(cc, cts, dp, rcl)) {
+ /* Register overflow? Pass on stack. */
+ MSize nsp = cc->nsp, sz = rcl[1] ? 2*CTSIZE_PTR : CTSIZE_PTR;
+ if (nsp + sz > CCALL_SIZE_STACK)
+ return 1; /* Too many arguments. */
+ cc->nsp = nsp + sz;
+ memcpy((uint8_t *)cc->stack + nsp, dp, sz);
}
return 0; /* Ok. */
}
@@ -838,7 +855,8 @@
return 0; /* Struct is in GPRs. */
}
-void ccall_copy_struct(CCallState *cc, CType *ctr, void *dp, void *sp, int ft)
+static void ccall_copy_struct(CCallState *cc, CType *ctr, void *dp, void *sp,
+ int ft)
{
if (LJ_ABI_SOFTFP ? ft :
((ft & 3) == FTYPE_FLOAT || (ft >> 2) == FTYPE_FLOAT)) {
@@ -967,6 +985,14 @@
fid = ctf->sib;
}
+#if LJ_TARGET_ARM64 && LJ_ABI_WIN
+ if ((ct->info & CTF_VARARG)) {
+ nsp -= maxgpr * CTSIZE_PTR; /* May end up with negative nsp. */
+ ngpr = maxgpr;
+ nfpr = CCALL_NARG_FPR;
+ }
+#endif
+
/* Walk through all passed arguments. */
for (o = L->base+1, narg = 1; o < top; o++, narg++) {
CTypeID did;
@@ -978,7 +1004,7 @@
if (fid) { /* Get argument type from field. */
CType *ctf = ctype_get(cts, fid);
fid = ctf->sib;
- lua_assert(ctype_isfield(ctf->info));
+ lj_assertL(ctype_isfield(ctf->info), "field expected");
did = ctype_cid(ctf->info);
} else {
if (!(ct->info & CTF_VARARG))
@@ -1003,25 +1029,31 @@
CCALL_HANDLE_STRUCTARG
} else if (ctype_iscomplex(d->info)) {
CCALL_HANDLE_COMPLEXARG
- } else {
+ } else if (!(CCALL_PACK_STACKARG && ctype_isenum(d->info))) {
sz = CTSIZE_PTR;
}
- sz = (sz + CTSIZE_PTR-1) & ~(CTSIZE_PTR-1);
- n = sz / CTSIZE_PTR; /* Number of GPRs or stack slots needed. */
+ n = (sz + CTSIZE_PTR-1) / CTSIZE_PTR; /* Number of GPRs or stack slots needed. */
CCALL_HANDLE_REGARG /* Handle register arguments. */
/* Otherwise pass argument on stack. */
- if (CCALL_ALIGN_STACKARG && !rp && (d->info & CTF_ALIGN) > CTALIGN_PTR) {
- MSize align = (1u << ctype_align(d->info-CTALIGN_PTR)) -1;
- nsp = (nsp + align) & ~align; /* Align argument on stack. */
+ if (CCALL_ALIGN_STACKARG) { /* Align argument on stack. */
+ MSize align = (1u << ctype_align(d->info)) - 1;
+ if (rp || (CCALL_PACK_STACKARG && isva && align < CTSIZE_PTR-1))
+ align = CTSIZE_PTR-1;
+ nsp = (nsp + align) & ~align;
}
- if (nsp + n > CCALL_MAXSTACK) { /* Too many arguments. */
+#if LJ_TARGET_ARM64 && LJ_ABI_WIN
+ /* A negative nsp points into cc->gpr. Blame MS for their messy ABI. */
+ dp = ((uint8_t *)cc->stack) + (int32_t)nsp;
+#else
+ dp = ((uint8_t *)cc->stack) + nsp;
+#endif
+ nsp += CCALL_PACK_STACKARG ? sz : n * CTSIZE_PTR;
+ if ((int32_t)nsp > CCALL_SIZE_STACK) { /* Too many arguments. */
err_nyi:
lj_err_caller(L, LJ_ERR_FFI_NYICALL);
}
- dp = &cc->stack[nsp];
- nsp += n;
isva = 0;
done:
@@ -1032,7 +1064,8 @@
}
lj_cconv_ct_tv(cts, d, (uint8_t *)dp, o, CCF_ARG(narg));
/* Extend passed integers to 32 bits at least. */
- if (ctype_isinteger_or_bool(d->info) && d->size < 4) {
+ if (ctype_isinteger_or_bool(d->info) && d->size < 4 &&
+ (!CCALL_PACK_STACKARG || !((uintptr_t)dp & 3))) { /* Assumes LJ_LE. */
if (d->info & CTF_UNSIGNED)
*(uint32_t *)dp = d->size == 1 ? (uint32_t)*(uint8_t *)dp :
(uint32_t)*(uint16_t *)dp;
@@ -1079,14 +1112,17 @@
#endif
}
if (fid) lj_err_caller(L, LJ_ERR_FFI_NUMARG); /* Too few arguments. */
+#if LJ_TARGET_ARM64 && LJ_ABI_WIN
+ if ((int32_t)nsp < 0) nsp = 0;
+#endif
-#if LJ_TARGET_X64 || LJ_TARGET_PPC
+#if LJ_TARGET_X64 || (LJ_TARGET_PPC && !LJ_ABI_SOFTFP)
cc->nfpr = nfpr; /* Required for vararg functions. */
#endif
- cc->nsp = nsp;
- cc->spadj = (CCALL_SPS_FREE + CCALL_SPS_EXTRA)*CTSIZE_PTR;
- if (nsp > CCALL_SPS_FREE)
- cc->spadj += (((nsp-CCALL_SPS_FREE)*CTSIZE_PTR + 15u) & ~15u);
+ cc->nsp = (nsp + CTSIZE_PTR-1) & ~(CTSIZE_PTR-1);
+ cc->spadj = (CCALL_SPS_FREE + CCALL_SPS_EXTRA) * CTSIZE_PTR;
+ if (cc->nsp > CCALL_SPS_FREE * CTSIZE_PTR)
+ cc->spadj += (((cc->nsp - CCALL_SPS_FREE * CTSIZE_PTR) + 15u) & ~15u);
return gcsteps;
}
@@ -1126,7 +1162,8 @@
CCALL_HANDLE_RET
#endif
/* No reference types end up here, so there's no need for the CTypeID. */
- lua_assert(!(ctype_isrefarray(ctr->info) || ctype_isstruct(ctr->info)));
+ lj_assertL(!(ctype_isrefarray(ctr->info) || ctype_isstruct(ctr->info)),
+ "unexpected reference ctype");
return lj_cconv_tv_ct(cts, ctr, 0, L->top-1, sp);
}
@@ -1150,7 +1187,7 @@
lj_vm_ffi_call(&cc);
if (cts->cb.slot != ~0u) { /* Blacklist function that called a callback. */
TValue tv;
- setlightudV(&tv, (void *)cc.func);
+ tv.u64 = ((uintptr_t)(void *)cc.func >> 2) | U64x(800000000, 00000000);
setboolV(lj_tab_set(L, cts->miscmap, &tv), 1);
}
ct = (CType *)((intptr_t)ct+(intptr_t)cts->tab); /* May be reallocated. */
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccall.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C call handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CCALL_H
@@ -75,6 +75,9 @@
#define CCALL_NARG_FPR 8
#define CCALL_NRET_FPR 4
#define CCALL_SPS_FREE 0
+#if LJ_TARGET_OSX
+#define CCALL_PACK_STACKARG 1
+#endif
typedef intptr_t GPRArg;
typedef union FPRArg {
@@ -86,9 +89,9 @@
#elif LJ_TARGET_PPC
#define CCALL_NARG_GPR 8
-#define CCALL_NARG_FPR 8
+#define CCALL_NARG_FPR (LJ_ABI_SOFTFP ? 0 : 8)
#define CCALL_NRET_GPR 4 /* For complex double. */
-#define CCALL_NRET_FPR 1
+#define CCALL_NRET_FPR (LJ_ABI_SOFTFP ? 0 : 1)
#define CCALL_SPS_EXTRA 4
#define CCALL_SPS_FREE 0
@@ -139,6 +142,9 @@
#ifndef CCALL_ALIGN_STACKARG
#define CCALL_ALIGN_STACKARG 1
#endif
+#ifndef CCALL_PACK_STACKARG
+#define CCALL_PACK_STACKARG 0
+#endif
#ifndef CCALL_ALIGN_CALLSTATE
#define CCALL_ALIGN_CALLSTATE 8
#endif
@@ -152,7 +158,8 @@
LJ_STATIC_ASSERT(CCALL_NUM_GPR <= CCALL_MAX_GPR);
LJ_STATIC_ASSERT(CCALL_NUM_FPR <= CCALL_MAX_FPR);
-#define CCALL_MAXSTACK 32
+#define CCALL_NUM_STACK 31
+#define CCALL_SIZE_STACK (CCALL_NUM_STACK * CTSIZE_PTR)
/* -- C call state -------------------------------------------------------- */
@@ -159,7 +166,7 @@
typedef LJ_ALIGN(CCALL_ALIGN_CALLSTATE) struct CCallState {
void (*func)(void); /* Pointer to called function. */
uint32_t spadj; /* Stack pointer adjustment. */
- uint8_t nsp; /* Number of stack slots. */
+ uint8_t nsp; /* Number of bytes on stack. */
uint8_t retref; /* Return value by reference. */
#if LJ_TARGET_X64
uint8_t ngpr; /* Number of arguments in GPRs. */
@@ -178,7 +185,7 @@
FPRArg fpr[CCALL_NUM_FPR]; /* Arguments/results in FPRs. */
#endif
GPRArg gpr[CCALL_NUM_GPR]; /* Arguments/results in GPRs. */
- GPRArg stack[CCALL_MAXSTACK]; /* Stack slots. */
+ GPRArg stack[CCALL_NUM_STACK]; /* Stack slots. */
} CCallState;
/* -- C call handling ----------------------------------------------------- */
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C callback handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -107,9 +107,9 @@
/* Initialize machine code for callback function pointers. */
#if LJ_OS_NOJIT
/* Disabled callback support. */
-#define callback_mcode_init(g, p) UNUSED(p)
+#define callback_mcode_init(g, p) (p)
#elif LJ_TARGET_X86ORX64
-static void callback_mcode_init(global_State *g, uint8_t *page)
+static void *callback_mcode_init(global_State *g, uint8_t *page)
{
uint8_t *p = page;
uint8_t *target = (uint8_t *)(void *)lj_vm_ffi_callback;
@@ -143,10 +143,10 @@
*p++ = XI_JMPs; *p++ = (uint8_t)((2+2)*(31-(slot&31)) - 2);
}
}
- lua_assert(p - page <= CALLBACK_MCODE_SIZE);
+ return p;
}
#elif LJ_TARGET_ARM
-static void callback_mcode_init(global_State *g, uint32_t *page)
+static void *callback_mcode_init(global_State *g, uint32_t *page)
{
uint32_t *p = page;
void *target = (void *)lj_vm_ffi_callback;
@@ -165,19 +165,19 @@
*p = ARMI_B | ((page-p-2) & 0x00ffffffu);
p++;
}
- lua_assert(p - page <= CALLBACK_MCODE_SIZE);
+ return p;
}
#elif LJ_TARGET_ARM64
-static void callback_mcode_init(global_State *g, uint32_t *page)
+static void *callback_mcode_init(global_State *g, uint32_t *page)
{
uint32_t *p = page;
- void *target = (void *)lj_vm_ffi_callback;
+ ASMFunction target = lj_vm_ffi_callback;
MSize slot;
*p++ = A64I_LE(A64I_LDRLx | A64F_D(RID_X11) | A64F_S19(4));
*p++ = A64I_LE(A64I_LDRLx | A64F_D(RID_X10) | A64F_S19(5));
- *p++ = A64I_LE(A64I_BR | A64F_N(RID_X11));
+ *p++ = A64I_LE(A64I_BR_AUTH | A64F_N(RID_X11));
*p++ = A64I_LE(A64I_NOP);
- ((void **)p)[0] = target;
+ ((ASMFunction *)p)[0] = target;
((void **)p)[1] = g;
p += 4;
for (slot = 0; slot < CALLBACK_MAX_SLOT; slot++) {
@@ -185,10 +185,10 @@
*p = A64I_LE(A64I_B | A64F_S26((page-p) & 0x03ffffffu));
p++;
}
- lua_assert(p - page <= CALLBACK_MCODE_SIZE);
+ return p;
}
#elif LJ_TARGET_PPC
-static void callback_mcode_init(global_State *g, uint32_t *page)
+static void *callback_mcode_init(global_State *g, uint32_t *page)
{
uint32_t *p = page;
void *target = (void *)lj_vm_ffi_callback;
@@ -204,10 +204,10 @@
*p = PPCI_B | (((page-p) & 0x00ffffffu) << 2);
p++;
}
- lua_assert(p - page <= CALLBACK_MCODE_SIZE);
+ return p;
}
#elif LJ_TARGET_MIPS
-static void callback_mcode_init(global_State *g, uint32_t *page)
+static void *callback_mcode_init(global_State *g, uint32_t *page)
{
uint32_t *p = page;
uintptr_t target = (uintptr_t)(void *)lj_vm_ffi_callback;
@@ -236,11 +236,11 @@
p++;
*p++ = MIPSI_LI | MIPSF_T(RID_R1) | slot;
}
- lua_assert(p - page <= CALLBACK_MCODE_SIZE);
+ return p;
}
#else
/* Missing support for this architecture. */
-#define callback_mcode_init(g, p) UNUSED(p)
+#define callback_mcode_init(g, p) (p)
#endif
/* -- Machine code management --------------------------------------------- */
@@ -256,6 +256,11 @@
#ifndef MAP_ANONYMOUS
#define MAP_ANONYMOUS MAP_ANON
#endif
+#ifdef PROT_MPROTECT
+#define CCPROT_CREATE (PROT_MPROTECT(PROT_EXEC))
+#else
+#define CCPROT_CREATE 0
+#endif
#endif
@@ -263,15 +268,15 @@
static void callback_mcode_new(CTState *cts)
{
size_t sz = (size_t)CALLBACK_MCODE_SIZE;
- void *p;
+ void *p, *pe;
if (CALLBACK_MAX_SLOT == 0)
lj_err_caller(cts->L, LJ_ERR_FFI_CBACKOV);
#if LJ_TARGET_WINDOWS
- p = VirtualAlloc(NULL, sz, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
+ p = LJ_WIN_VALLOC(NULL, sz, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
if (!p)
lj_err_caller(cts->L, LJ_ERR_FFI_CBACKOV);
#elif LJ_TARGET_POSIX
- p = mmap(NULL, sz, (PROT_READ|PROT_WRITE), MAP_PRIVATE|MAP_ANONYMOUS,
+ p = mmap(NULL, sz, (PROT_READ|PROT_WRITE|CCPROT_CREATE), MAP_PRIVATE|MAP_ANONYMOUS,
-1, 0);
if (p == MAP_FAILED)
lj_err_caller(cts->L, LJ_ERR_FFI_CBACKOV);
@@ -280,12 +285,15 @@
p = lj_mem_new(cts->L, sz);
#endif
cts->cb.mcode = p;
- callback_mcode_init(cts->g, p);
+ pe = callback_mcode_init(cts->g, p);
+ UNUSED(pe);
+ lj_assertCTS((size_t)((char *)pe - (char *)p) <= sz,
+ "miscalculated CALLBACK_MAX_SLOT");
lj_mcode_sync(p, (char *)p + sz);
#if LJ_TARGET_WINDOWS
{
DWORD oprot;
- VirtualProtect(p, sz, PAGE_EXECUTE_READ, &oprot);
+ LJ_WIN_VPROTECT(p, sz, PAGE_EXECUTE_READ, &oprot);
}
#elif LJ_TARGET_POSIX
mprotect(p, sz, (PROT_READ|PROT_EXEC));
@@ -406,7 +414,7 @@
nfpr = CCALL_NARG_FPR; /* Prevent reordering. */ \
} \
} else { \
- if (!LJ_TARGET_IOS && n > 1) \
+ if (!LJ_TARGET_OSX && n > 1) \
ngpr = (ngpr + 1u) & ~1u; /* Align to regpair. */ \
if (ngpr + n <= maxgpr) { \
sp = &cts->cb.gpr[ngpr]; \
@@ -419,7 +427,25 @@
#elif LJ_TARGET_PPC
+#define CALLBACK_HANDLE_GPR \
+ if (n > 1) { \
+ lj_assertCTS(((LJ_ABI_SOFTFP && ctype_isnum(cta->info)) || /* double. */ \
+ ctype_isinteger(cta->info)) && n == 2, /* int64_t. */ \
+ "bad GPR type"); \
+ ngpr = (ngpr + 1u) & ~1u; /* Align int64_t to regpair. */ \
+ } \
+ if (ngpr + n <= maxgpr) { \
+ sp = &cts->cb.gpr[ngpr]; \
+ ngpr += n; \
+ goto done; \
+ }
+
+#if LJ_ABI_SOFTFP
#define CALLBACK_HANDLE_REGARG \
+ CALLBACK_HANDLE_GPR \
+ UNUSED(isfp);
+#else
+#define CALLBACK_HANDLE_REGARG \
if (isfp) { \
if (nfpr + 1 <= CCALL_NARG_FPR) { \
sp = &cts->cb.fpr[nfpr++]; \
@@ -427,20 +453,15 @@
goto done; \
} \
} else { /* Try to pass argument in GPRs. */ \
- if (n > 1) { \
- lua_assert(ctype_isinteger(cta->info) && n == 2); /* int64_t. */ \
- ngpr = (ngpr + 1u) & ~1u; /* Align int64_t to regpair. */ \
- } \
- if (ngpr + n <= maxgpr) { \
- sp = &cts->cb.gpr[ngpr]; \
- ngpr += n; \
- goto done; \
- } \
+ CALLBACK_HANDLE_GPR \
}
+#endif
+#if !LJ_ABI_SOFTFP
#define CALLBACK_HANDLE_RET \
if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
*(double *)dp = *(float *)dp; /* FPRs always hold doubles. */
+#endif
#elif LJ_TARGET_MIPS32
@@ -533,7 +554,6 @@
if (LJ_FR2) {
(o++)->u64 = LJ_CONT_FFI_CALLBACK;
(o++)->u64 = rid;
- o++;
} else {
o->u32.lo = LJ_CONT_FFI_CALLBACK;
o->u32.hi = rid;
@@ -540,6 +560,7 @@
o++;
}
setframe_gc(o, obj2gco(fn), fntp);
+ if (LJ_FR2) o++;
setframe_ftsz(o, ((char *)(o+1) - (char *)L->base) + FRAME_CONT);
L->top = L->base = ++o;
if (!ct)
@@ -567,7 +588,7 @@
CTSize sz;
int isfp;
MSize n;
- lua_assert(ctype_isfield(ctf->info));
+ lj_assertCTS(ctype_isfield(ctf->info), "field expected");
cta = ctype_rawchild(cts, ctf);
isfp = ctype_isfp(cta->info);
sz = (cta->size + CTSIZE_PTR-1) & ~(CTSIZE_PTR-1);
@@ -659,7 +680,7 @@
{
lua_State *L = cts->L;
global_State *g = cts->g;
- lua_assert(L != NULL);
+ lj_assertG(L != NULL, "uninitialized cts->L in callback");
if (tvref(g->jit_base)) {
setstrV(L, L->top++, lj_err_str(L, LJ_ERR_FFI_BADCBACK));
if (g->panic) g->panic(L);
@@ -744,7 +765,7 @@
CType *ctf = ctype_get(cts, fid);
if (!ctype_isattrib(ctf->info)) {
CType *cta;
- lua_assert(ctype_isfield(ctf->info));
+ lj_assertCTS(ctype_isfield(ctf->info), "field expected");
cta = ctype_rawchild(cts, ctf);
if (!(ctype_isenum(cta->info) || ctype_isptr(cta->info) ||
(ctype_isnum(cta->info) && cta->size <= 8)) ||
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ccallback.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C callback handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CCALLBACK_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C type conversions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -8,6 +8,7 @@
#if LJ_HASFFI
#include "lj_err.h"
+#include "lj_buf.h"
#include "lj_tab.h"
#include "lj_ctype.h"
#include "lj_cdata.h"
@@ -122,19 +123,25 @@
CTInfo dinfo = d->info, sinfo = s->info;
void *tmpptr;
- lua_assert(!ctype_isenum(dinfo) && !ctype_isenum(sinfo));
- lua_assert(!ctype_isattrib(dinfo) && !ctype_isattrib(sinfo));
+ lj_assertCTS(!ctype_isenum(dinfo) && !ctype_isenum(sinfo),
+ "unresolved enum");
+ lj_assertCTS(!ctype_isattrib(dinfo) && !ctype_isattrib(sinfo),
+ "unstripped attribute");
if (ctype_type(dinfo) > CT_MAYCONVERT || ctype_type(sinfo) > CT_MAYCONVERT)
goto err_conv;
/* Some basic sanity checks. */
- lua_assert(!ctype_isnum(dinfo) || dsize > 0);
- lua_assert(!ctype_isnum(sinfo) || ssize > 0);
- lua_assert(!ctype_isbool(dinfo) || dsize == 1 || dsize == 4);
- lua_assert(!ctype_isbool(sinfo) || ssize == 1 || ssize == 4);
- lua_assert(!ctype_isinteger(dinfo) || (1u<<lj_fls(dsize)) == dsize);
- lua_assert(!ctype_isinteger(sinfo) || (1u<<lj_fls(ssize)) == ssize);
+ lj_assertCTS(!ctype_isnum(dinfo) || dsize > 0, "bad size for number type");
+ lj_assertCTS(!ctype_isnum(sinfo) || ssize > 0, "bad size for number type");
+ lj_assertCTS(!ctype_isbool(dinfo) || dsize == 1 || dsize == 4,
+ "bad size for bool type");
+ lj_assertCTS(!ctype_isbool(sinfo) || ssize == 1 || ssize == 4,
+ "bad size for bool type");
+ lj_assertCTS(!ctype_isinteger(dinfo) || (1u<<lj_fls(dsize)) == dsize,
+ "bad size for integer type");
+ lj_assertCTS(!ctype_isinteger(sinfo) || (1u<<lj_fls(ssize)) == ssize,
+ "bad size for integer type");
switch (cconv_idx2(dinfo, sinfo)) {
/* Destination is a bool. */
@@ -357,7 +364,7 @@
if ((flags & CCF_CAST) || (d->info & CTF_VLA) || d != s)
goto err_conv; /* Must be exact same type. */
copyval: /* Copy value. */
- lua_assert(dsize == ssize);
+ lj_assertCTS(dsize == ssize, "value copy with different sizes");
memcpy(dp, sp, dsize);
break;
@@ -389,7 +396,7 @@
lj_cconv_ct_ct(cts, ctype_get(cts, CTID_DOUBLE), s,
(uint8_t *)&o->n, sp, 0);
/* Numbers are NOT canonicalized here! Beware of uninitialized data. */
- lua_assert(tvisnum(o));
+ lj_assertCTS(tvisnum(o), "non-canonical NaN passed");
}
} else {
uint32_t b = s->size == 1 ? (*sp != 0) : (*(int *)sp != 0);
@@ -406,7 +413,7 @@
CTSize sz;
copyval: /* Copy value. */
sz = s->size;
- lua_assert(sz != CTSIZE_INVALID);
+ lj_assertCTS(sz != CTSIZE_INVALID, "value copy with invalid size");
/* Attributes are stripped, qualifiers are kept (but mostly ignored). */
cd = lj_cdata_new(cts, ctype_typeid(cts, s), sz);
setcdataV(cts->L, o, cd);
@@ -421,19 +428,22 @@
CTInfo info = s->info;
CTSize pos, bsz;
uint32_t val;
- lua_assert(ctype_isbitfield(info));
+ lj_assertCTS(ctype_isbitfield(info), "bitfield expected");
/* NYI: packed bitfields may cause misaligned reads. */
switch (ctype_bitcsz(info)) {
case 4: val = *(uint32_t *)sp; break;
case 2: val = *(uint16_t *)sp; break;
case 1: val = *(uint8_t *)sp; break;
- default: lua_assert(0); val = 0; break;
+ default:
+ lj_assertCTS(0, "bad bitfield container size %d", ctype_bitcsz(info));
+ val = 0;
+ break;
}
/* Check if a packed bitfield crosses a container boundary. */
pos = ctype_bitpos(info);
bsz = ctype_bitbsz(info);
- lua_assert(pos < 8*ctype_bitcsz(info));
- lua_assert(bsz > 0 && bsz <= 8*ctype_bitcsz(info));
+ lj_assertCTS(pos < 8*ctype_bitcsz(info), "bad bitfield position");
+ lj_assertCTS(bsz > 0 && bsz <= 8*ctype_bitcsz(info), "bad bitfield size");
if (pos + bsz > 8*ctype_bitcsz(info))
lj_err_caller(cts->L, LJ_ERR_FFI_NYIPACKBIT);
if (!(info & CTF_BOOL)) {
@@ -449,7 +459,7 @@
}
} else {
uint32_t b = (val >> pos) & 1;
- lua_assert(bsz == 1);
+ lj_assertCTS(bsz == 1, "bad bool bitfield size");
setboolV(o, b);
setboolV(&cts->g->tmptv2, b); /* Remember for trace recorder. */
}
@@ -553,13 +563,15 @@
sid = cdataV(o)->ctypeid;
s = ctype_get(cts, sid);
if (ctype_isref(s->info)) { /* Resolve reference for value. */
- lua_assert(s->size == CTSIZE_PTR);
+ lj_assertCTS(s->size == CTSIZE_PTR, "ref is not pointer-sized");
sp = *(void **)sp;
sid = ctype_cid(s->info);
}
s = ctype_raw(cts, sid);
if (ctype_isfunc(s->info)) {
+ CTypeID did = ctype_typeid(cts, d);
sid = lj_ctype_intern(cts, CTINFO(CT_PTR, CTALIGN_PTR|sid), CTSIZE_PTR);
+ d = ctype_get(cts, did); /* cts->tab may have been reallocated. */
} else {
if (ctype_isenum(s->info)) s = ctype_child(cts, s);
goto doconv;
@@ -571,7 +583,7 @@
CType *cct = lj_ctype_getfield(cts, d, str, &ofs);
if (!cct || !ctype_isconstval(cct->info))
goto err_conv;
- lua_assert(d->size == 4);
+ lj_assertCTS(d->size == 4, "only 32 bit enum supported"); /* NYI */
sp = (uint8_t *)&cct->size;
sid = ctype_cid(cct->info);
} else if (ctype_isrefarray(d->info)) { /* Copy string to array. */
@@ -610,8 +622,10 @@
tmpptr = uddata(ud);
if (ud->udtype == UDTYPE_IO_FILE)
tmpptr = *(void **)tmpptr;
+ else if (ud->udtype == UDTYPE_BUFFER)
+ tmpptr = ((SBufExt *)tmpptr)->r;
} else if (tvislightud(o)) {
- tmpptr = lightudV(o);
+ tmpptr = lightudV(cts->g, o);
} else if (tvisfunc(o)) {
void *p = lj_ccallback_new(cts, d, funcV(o));
if (p) {
@@ -635,10 +649,10 @@
CTInfo info = d->info;
CTSize pos, bsz;
uint32_t val, mask;
- lua_assert(ctype_isbitfield(info));
+ lj_assertCTS(ctype_isbitfield(info), "bitfield expected");
if ((info & CTF_BOOL)) {
uint8_t tmpbool;
- lua_assert(ctype_bitbsz(info) == 1);
+ lj_assertCTS(ctype_bitbsz(info) == 1, "bad bool bitfield size");
lj_cconv_ct_tv(cts, ctype_get(cts, CTID_BOOL), &tmpbool, o, 0);
val = tmpbool;
} else {
@@ -647,8 +661,8 @@
}
pos = ctype_bitpos(info);
bsz = ctype_bitbsz(info);
- lua_assert(pos < 8*ctype_bitcsz(info));
- lua_assert(bsz > 0 && bsz <= 8*ctype_bitcsz(info));
+ lj_assertCTS(pos < 8*ctype_bitcsz(info), "bad bitfield position");
+ lj_assertCTS(bsz > 0 && bsz <= 8*ctype_bitcsz(info), "bad bitfield size");
/* Check if a packed bitfield crosses a container boundary. */
if (pos + bsz > 8*ctype_bitcsz(info))
lj_err_caller(cts->L, LJ_ERR_FFI_NYIPACKBIT);
@@ -659,7 +673,9 @@
case 4: *(uint32_t *)dp = (*(uint32_t *)dp & ~mask) | (uint32_t)val; break;
case 2: *(uint16_t *)dp = (*(uint16_t *)dp & ~mask) | (uint16_t)val; break;
case 1: *(uint8_t *)dp = (*(uint8_t *)dp & ~mask) | (uint8_t)val; break;
- default: lua_assert(0); break;
+ default:
+ lj_assertCTS(0, "bad bitfield container size %d", ctype_bitcsz(info));
+ break;
}
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cconv.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C type conversions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CCONV_H
@@ -27,13 +27,14 @@
static LJ_AINLINE uint32_t cconv_idx(CTInfo info)
{
uint32_t idx = ((info >> 26) & 15u); /* Dispatch bits. */
- lua_assert(ctype_type(info) <= CT_MAYCONVERT);
+ lj_assertX(ctype_type(info) <= CT_MAYCONVERT,
+ "cannot convert ctype %08x", info);
#if LJ_64
idx = ((uint32_t)(U64x(f436fff5,fff7f021) >> 4*idx) & 15u);
#else
idx = (((idx < 8 ? 0xfff7f021u : 0xf436fff5) >> 4*(idx & 7u)) & 15u);
#endif
- lua_assert(idx < 8);
+ lj_assertX(idx < 8, "cannot convert ctype %08x", info);
return idx;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C data management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -35,7 +35,7 @@
uintptr_t adata = (uintptr_t)p + sizeof(GCcdataVar) + sizeof(GCcdata);
uintptr_t almask = (1u << align) - 1u;
GCcdata *cd = (GCcdata *)(((adata + almask) & ~almask) - sizeof(GCcdata));
- lua_assert((char *)cd - p < 65536);
+ lj_assertL((char *)cd - p < 65536, "excessive cdata alignment");
cdatav(cd)->offset = (uint16_t)((char *)cd - p);
cdatav(cd)->extra = extra;
cdatav(cd)->len = sz;
@@ -76,8 +76,8 @@
} else if (LJ_LIKELY(!cdataisv(cd))) {
CType *ct = ctype_raw(ctype_ctsG(g), cd->ctypeid);
CTSize sz = ctype_hassize(ct->info) ? ct->size : CTSIZE_PTR;
- lua_assert(ctype_hassize(ct->info) || ctype_isfunc(ct->info) ||
- ctype_isextern(ct->info));
+ lj_assertG(ctype_hassize(ct->info) || ctype_isfunc(ct->info) ||
+ ctype_isextern(ct->info), "free of ctype without a size");
lj_mem_free(g, cd, sizeof(GCcdata) + sz);
} else {
lj_mem_free(g, memcdatav(cd), sizecdatav(cd));
@@ -86,7 +86,7 @@
void lj_cdata_setfin(lua_State *L, GCcdata *cd, GCobj *obj, uint32_t it)
{
- GCtab *t = ctype_ctsG(G(L))->finalizer;
+ GCtab *t = tabref(G(L)->gcroot[GCROOT_FFI_FIN]);
if (gcref(t->metatable)) {
/* Add cdata to finalizer table, if still enabled. */
TValue *tv, tmp;
@@ -115,7 +115,7 @@
/* Resolve reference for cdata object. */
if (ctype_isref(ct->info)) {
- lua_assert(ct->size == CTSIZE_PTR);
+ lj_assertCTS(ct->size == CTSIZE_PTR, "ref is not pointer-sized");
p = *(uint8_t **)p;
ct = ctype_child(cts, ct);
}
@@ -126,7 +126,8 @@
if (ctype_attrib(ct->info) == CTA_QUAL) *qual |= ct->size;
ct = ctype_child(cts, ct);
}
- lua_assert(!ctype_isref(ct->info)); /* Interning rejects refs to refs. */
+ /* Interning rejects refs to refs. */
+ lj_assertCTS(!ctype_isref(ct->info), "bad ref of ref");
if (tvisint(key)) {
idx = (ptrdiff_t)intV(key);
@@ -212,7 +213,8 @@
static void cdata_getconst(CTState *cts, TValue *o, CType *ct)
{
CType *ctt = ctype_child(cts, ct);
- lua_assert(ctype_isinteger(ctt->info) && ctt->size <= 4);
+ lj_assertCTS(ctype_isinteger(ctt->info) && ctt->size <= 4,
+ "only 32 bit const supported"); /* NYI */
/* Constants are already zero-extended/sign-extended to 32 bits. */
if ((ctt->info & CTF_UNSIGNED) && (int32_t)ct->size < 0)
setnumV(o, (lua_Number)(uint32_t)ct->size);
@@ -233,13 +235,14 @@
}
/* Get child type of pointer/array/field. */
- lua_assert(ctype_ispointer(s->info) || ctype_isfield(s->info));
+ lj_assertCTS(ctype_ispointer(s->info) || ctype_isfield(s->info),
+ "pointer or field expected");
sid = ctype_cid(s->info);
s = ctype_get(cts, sid);
/* Resolve reference for field. */
if (ctype_isref(s->info)) {
- lua_assert(s->size == CTSIZE_PTR);
+ lj_assertCTS(s->size == CTSIZE_PTR, "ref is not pointer-sized");
sp = *(uint8_t **)sp;
sid = ctype_cid(s->info);
s = ctype_get(cts, sid);
@@ -266,12 +269,13 @@
}
/* Get child type of pointer/array/field. */
- lua_assert(ctype_ispointer(d->info) || ctype_isfield(d->info));
+ lj_assertCTS(ctype_ispointer(d->info) || ctype_isfield(d->info),
+ "pointer or field expected");
d = ctype_child(cts, d);
/* Resolve reference for field. */
if (ctype_isref(d->info)) {
- lua_assert(d->size == CTSIZE_PTR);
+ lj_assertCTS(d->size == CTSIZE_PTR, "ref is not pointer-sized");
dp = *(uint8_t **)dp;
d = ctype_child(cts, d);
}
@@ -286,7 +290,8 @@
d = ctype_child(cts, d);
}
- lua_assert(ctype_hassize(d->info) && !ctype_isvoid(d->info));
+ lj_assertCTS(ctype_hassize(d->info), "store to ctype without size");
+ lj_assertCTS(!ctype_isvoid(d->info), "store to void type");
if (((d->info|qual) & CTF_CONST)) {
err_const:
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cdata.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C data management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CDATA_H
@@ -18,7 +18,7 @@
if (LJ_64 && sz == 4) { /* Support 32 bit pointers on 64 bit targets. */
return ((void *)(uintptr_t)*(uint32_t *)p);
} else {
- lua_assert(sz == CTSIZE_PTR);
+ lj_assertX(sz == CTSIZE_PTR, "bad pointer size %d", sz);
return *(void **)p;
}
}
@@ -29,7 +29,7 @@
if (LJ_64 && sz == 4) { /* Support 32 bit pointers on 64 bit targets. */
*(uint32_t *)p = (uint32_t)(uintptr_t)v;
} else {
- lua_assert(sz == CTSIZE_PTR);
+ lj_assertX(sz == CTSIZE_PTR, "bad pointer size %d", sz);
*(void **)p = (void *)v;
}
}
@@ -40,7 +40,8 @@
GCcdata *cd;
#ifdef LUA_USE_ASSERT
CType *ct = ctype_raw(cts, id);
- lua_assert((ctype_hassize(ct->info) ? ct->size : CTSIZE_PTR) == sz);
+ lj_assertCTS((ctype_hassize(ct->info) ? ct->size : CTSIZE_PTR) == sz,
+ "inconsistent size of fixed-size cdata alloc");
#endif
cd = (GCcdata *)lj_mem_newgco(cts->L, sizeof(GCcdata) + sz);
cd->gct = ~LJ_TCDATA;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C library loader.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -25,7 +25,7 @@
#include <dlfcn.h>
#include <stdio.h>
-#if defined(RTLD_DEFAULT)
+#if defined(RTLD_DEFAULT) && !defined(NO_RTLD_DEFAULT)
#define CLIB_DEFHANDLE RTLD_DEFAULT
#elif LJ_TARGET_OSX || LJ_TARGET_BSD
#define CLIB_DEFHANDLE ((void *)(intptr_t)-2)
@@ -119,12 +119,13 @@
RTLD_LAZY | (global?RTLD_GLOBAL:RTLD_LOCAL));
if (!h) {
const char *e, *err = dlerror();
- if (*err == '/' && (e = strchr(err, ':')) &&
+ if (err && *err == '/' && (e = strchr(err, ':')) &&
(name = clib_resolve_lds(L, strdata(lj_str_new(L, err, e-err))))) {
h = dlopen(name, RTLD_LAZY | (global?RTLD_GLOBAL:RTLD_LOCAL));
if (h) return h;
err = dlerror();
}
+ if (!err) err = "dlopen failed";
lj_err_callermsg(L, err);
}
return h;
@@ -158,11 +159,13 @@
/* Default libraries. */
enum {
CLIB_HANDLE_EXE,
+#if !LJ_TARGET_UWP
CLIB_HANDLE_DLL,
CLIB_HANDLE_CRT,
CLIB_HANDLE_KERNEL32,
CLIB_HANDLE_USER32,
CLIB_HANDLE_GDI32,
+#endif
CLIB_HANDLE_MAX
};
@@ -208,7 +211,7 @@
static void *clib_loadlib(lua_State *L, const char *name, int global)
{
DWORD oldwerr = GetLastError();
- void *h = (void *)LoadLibraryExA(clib_extname(L, name), NULL, 0);
+ void *h = LJ_WIN_LOADLIBA(clib_extname(L, name));
if (!h) clib_error(L, "cannot load module " LUA_QS ": %s", name);
SetLastError(oldwerr);
UNUSED(global);
@@ -218,6 +221,7 @@
static void clib_unloadlib(CLibrary *cl)
{
if (cl->handle == CLIB_DEFHANDLE) {
+#if !LJ_TARGET_UWP
MSize i;
for (i = CLIB_HANDLE_KERNEL32; i < CLIB_HANDLE_MAX; i++) {
void *h = clib_def_handle[i];
@@ -226,11 +230,16 @@
FreeLibrary((HINSTANCE)h);
}
}
+#endif
} else if (cl->handle) {
FreeLibrary((HINSTANCE)cl->handle);
}
}
+#if LJ_TARGET_UWP
+EXTERN_C IMAGE_DOS_HEADER __ImageBase;
+#endif
+
static void *clib_getsym(CLibrary *cl, const char *name)
{
void *p = NULL;
@@ -239,6 +248,9 @@
for (i = 0; i < CLIB_HANDLE_MAX; i++) {
HINSTANCE h = (HINSTANCE)clib_def_handle[i];
if (!(void *)h) { /* Resolve default library handles (once). */
+#if LJ_TARGET_UWP
+ h = (HINSTANCE)&__ImageBase;
+#else
switch (i) {
case CLIB_HANDLE_EXE: GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT, NULL, &h); break;
case CLIB_HANDLE_DLL:
@@ -249,11 +261,12 @@
GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS|GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT,
(const char *)&_fmode, &h);
break;
- case CLIB_HANDLE_KERNEL32: h = LoadLibraryExA("kernel32.dll", NULL, 0); break;
- case CLIB_HANDLE_USER32: h = LoadLibraryExA("user32.dll", NULL, 0); break;
- case CLIB_HANDLE_GDI32: h = LoadLibraryExA("gdi32.dll", NULL, 0); break;
+ case CLIB_HANDLE_KERNEL32: h = LJ_WIN_LOADLIBA("kernel32.dll"); break;
+ case CLIB_HANDLE_USER32: h = LJ_WIN_LOADLIBA("user32.dll"); break;
+ case CLIB_HANDLE_GDI32: h = LJ_WIN_LOADLIBA("gdi32.dll"); break;
}
if (!h) continue;
+#endif
clib_def_handle[i] = (void *)h;
}
p = (void *)GetProcAddress(h, name);
@@ -337,7 +350,8 @@
lj_err_callerv(L, LJ_ERR_FFI_NODECL, strdata(name));
if (ctype_isconstval(ct->info)) {
CType *ctt = ctype_child(cts, ct);
- lua_assert(ctype_isinteger(ctt->info) && ctt->size <= 4);
+ lj_assertCTS(ctype_isinteger(ctt->info) && ctt->size <= 4,
+ "only 32 bit const supported"); /* NYI */
if ((ctt->info & CTF_UNSIGNED) && (int32_t)ct->size < 0)
setnumV(tv, (lua_Number)(uint32_t)ct->size);
else
@@ -349,7 +363,8 @@
#endif
void *p = clib_getsym(cl, sym);
GCcdata *cd;
- lua_assert(ctype_isfunc(ct->info) || ctype_isextern(ct->info));
+ lj_assertCTS(ctype_isfunc(ct->info) || ctype_isextern(ct->info),
+ "unexpected ctype %08x in clib", ct->info);
#if LJ_TARGET_X86 && LJ_ABI_WIN
/* Retry with decorated name for fastcall/stdcall functions. */
if (!p && ctype_isfunc(ct->info)) {
@@ -372,6 +387,7 @@
cd = lj_cdata_new(cts, id, CTSIZE_PTR);
*(void **)cdataptr(cd) = p;
setcdataV(L, tv, cd);
+ lj_gc_anybarriert(L, cl->cache);
}
}
return tv;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_clib.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** FFI C library loader.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CLIB_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C declaration parser.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -28,6 +28,30 @@
** If in doubt, please check the input against your favorite C compiler.
*/
+#ifdef LUA_USE_ASSERT
+#define lj_assertCP(c, ...) (lj_assertG_(G(cp->L), (c), __VA_ARGS__))
+#else
+#define lj_assertCP(c, ...) ((void)cp)
+#endif
+
+/* -- Miscellaneous ------------------------------------------------------- */
+
+/* Match string against a C literal. */
+#define cp_str_is(str, k) \
+ ((str)->len == sizeof(k)-1 && !memcmp(strdata(str), k, sizeof(k)-1))
+
+/* Check string against a linear list of matches. */
+int lj_cparse_case(GCstr *str, const char *match)
+{
+ MSize len;
+ int n;
+ for (n = 0; (len = (MSize)*match++); n++, match += len) {
+ if (str->len == len && !memcmp(match, strdata(str), len))
+ return n;
+ }
+ return -1;
+}
+
/* -- C lexer ------------------------------------------------------------- */
/* C lexer token names. */
@@ -43,7 +67,7 @@
static const char *cp_tok2str(CPState *cp, CPToken tok)
{
- lua_assert(tok < CTOK_FIRSTDECL);
+ lj_assertCP(tok < CTOK_FIRSTDECL, "bad CPToken %d", tok);
if (tok > CTOK_OFS)
return ctoknames[tok-CTOK_OFS-1];
else if (!lj_char_iscntrl(tok))
@@ -109,9 +133,9 @@
tokstr = NULL;
} else if (tok == CTOK_IDENT || tok == CTOK_INTEGER || tok == CTOK_STRING ||
tok >= CTOK_FIRSTDECL) {
- if (sbufP(&cp->sb) == sbufB(&cp->sb)) cp_save(cp, '$');
+ if (cp->sb.w == cp->sb.b) cp_save(cp, '$');
cp_save(cp, '\0');
- tokstr = sbufB(&cp->sb);
+ tokstr = cp->sb.b;
} else {
tokstr = cp_tok2str(cp, tok);
}
@@ -151,7 +175,8 @@
TValue o;
do { cp_save(cp, cp->c); } while (lj_char_isident(cp_get(cp)));
cp_save(cp, '\0');
- fmt = lj_strscan_scan((const uint8_t *)sbufB(&cp->sb), &o, STRSCAN_OPT_C);
+ fmt = lj_strscan_scan((const uint8_t *)(cp->sb.b), sbuflen(&cp->sb)-1,
+ &o, STRSCAN_OPT_C);
if (fmt == STRSCAN_INT) cp->val.id = CTID_INT32;
else if (fmt == STRSCAN_U32) cp->val.id = CTID_UINT32;
else if (!(cp->mode & CPARSE_MODE_SKIP))
@@ -254,7 +279,7 @@
return CTOK_STRING;
} else {
if (sbuflen(&cp->sb) != 1) cp_err_token(cp, '\'');
- cp->val.i32 = (int32_t)(char)*sbufB(&cp->sb);
+ cp->val.i32 = (int32_t)(char)*cp->sb.b;
cp->val.id = CTID_INT32;
return CTOK_INTEGER;
}
@@ -373,7 +398,7 @@
cp->curpack = 0;
cp->packstack[0] = 255;
lj_buf_init(cp->L, &cp->sb);
- lua_assert(cp->p != NULL);
+ lj_assertCP(cp->p != NULL, "uninitialized cp->p");
cp_get(cp); /* Read-ahead first char. */
cp->tok = 0;
cp->tmask = CPNS_DEFAULT;
@@ -443,7 +468,7 @@
} else {
cp_expr_unary(cp, k);
}
- info = lj_ctype_info(cp->cts, k->id, &sz);
+ info = lj_ctype_info_raw(cp->cts, k->id, &sz);
if (wantsz) {
if (sz != CTSIZE_INVALID)
k->u32 = sz;
@@ -463,7 +488,7 @@
} else if (cp_opt(cp, '+')) {
cp_expr_unary(cp, k); /* Nothing to do (well, integer promotion). */
} else if (cp_opt(cp, '-')) {
- cp_expr_unary(cp, k); k->i32 = -k->i32;
+ cp_expr_unary(cp, k); k->i32 = (int32_t)(~(uint32_t)k->i32+1);
} else if (cp_opt(cp, '~')) {
cp_expr_unary(cp, k); k->i32 = ~k->i32;
} else if (cp_opt(cp, '!')) {
@@ -576,28 +601,34 @@
k->id = k2.id > k3.id ? k2.id : k3.id;
continue;
}
+ /* fallthrough */
case 1:
if (cp_opt(cp, CTOK_OROR)) {
cp_expr_sub(cp, &k2, 2); k->i32 = k->u32 || k2.u32; k->id = CTID_INT32;
continue;
}
+ /* fallthrough */
case 2:
if (cp_opt(cp, CTOK_ANDAND)) {
cp_expr_sub(cp, &k2, 3); k->i32 = k->u32 && k2.u32; k->id = CTID_INT32;
continue;
}
+ /* fallthrough */
case 3:
if (cp_opt(cp, '|')) {
cp_expr_sub(cp, &k2, 4); k->u32 = k->u32 | k2.u32; goto arith_result;
}
+ /* fallthrough */
case 4:
if (cp_opt(cp, '^')) {
cp_expr_sub(cp, &k2, 5); k->u32 = k->u32 ^ k2.u32; goto arith_result;
}
+ /* fallthrough */
case 5:
if (cp_opt(cp, '&')) {
cp_expr_sub(cp, &k2, 6); k->u32 = k->u32 & k2.u32; goto arith_result;
}
+ /* fallthrough */
case 6:
if (cp_opt(cp, CTOK_EQ)) {
cp_expr_sub(cp, &k2, 7); k->i32 = k->u32 == k2.u32; k->id = CTID_INT32;
@@ -606,6 +637,7 @@
cp_expr_sub(cp, &k2, 7); k->i32 = k->u32 != k2.u32; k->id = CTID_INT32;
continue;
}
+ /* fallthrough */
case 7:
if (cp_opt(cp, '<')) {
cp_expr_sub(cp, &k2, 8);
@@ -640,6 +672,7 @@
k->id = CTID_INT32;
continue;
}
+ /* fallthrough */
case 8:
if (cp_opt(cp, CTOK_SHL)) {
cp_expr_sub(cp, &k2, 9); k->u32 = k->u32 << k2.u32;
@@ -652,6 +685,7 @@
k->u32 = k->u32 >> k2.u32;
continue;
}
+ /* fallthrough */
case 9:
if (cp_opt(cp, '+')) {
cp_expr_sub(cp, &k2, 10); k->u32 = k->u32 + k2.u32;
@@ -661,6 +695,7 @@
} else if (cp_opt(cp, '-')) {
cp_expr_sub(cp, &k2, 10); k->u32 = k->u32 - k2.u32; goto arith_result;
}
+ /* fallthrough */
case 10:
if (cp_opt(cp, '*')) {
cp_expr_unary(cp, &k2); k->u32 = k->u32 * k2.u32; goto arith_result;
@@ -824,12 +859,13 @@
/* The cid is already part of info for copies of pointers/functions. */
idx = ct->next;
if (ctype_istypedef(info)) {
- lua_assert(id == 0);
+ lj_assertCP(id == 0, "typedef not at toplevel");
id = ctype_cid(info);
/* Always refetch info/size, since struct/enum may have been completed. */
cinfo = ctype_get(cp->cts, id)->info;
csize = ctype_get(cp->cts, id)->size;
- lua_assert(ctype_isstruct(cinfo) || ctype_isenum(cinfo));
+ lj_assertCP(ctype_isstruct(cinfo) || ctype_isenum(cinfo),
+ "typedef of bad type");
} else if (ctype_isfunc(info)) { /* Intern function. */
CType *fct;
CTypeID fid;
@@ -862,7 +898,7 @@
/* Inherit csize/cinfo from original type. */
} else {
if (ctype_isnum(info)) { /* Handle mode/vector-size attributes. */
- lua_assert(id == 0);
+ lj_assertCP(id == 0, "number not at toplevel");
if (!(info & CTF_BOOL)) {
CTSize msize = ctype_msizeP(decl->attr);
CTSize vsize = ctype_vsizeP(decl->attr);
@@ -917,7 +953,7 @@
info = (info & ~CTF_ALIGN) | (cinfo & CTF_ALIGN);
info |= (cinfo & CTF_QUAL); /* Inherit qual. */
} else {
- lua_assert(ctype_isvoid(info));
+ lj_assertCP(ctype_isvoid(info), "bad ctype %08x", info);
}
csize = size;
cinfo = info+id;
@@ -929,8 +965,6 @@
/* -- C declaration parser ------------------------------------------------ */
-#define H_(le, be) LJ_ENDIAN_SELECT(0x##le, 0x##be)
-
/* Reset declaration state to declaration specifier. */
static void cp_decl_reset(CPDecl *decl)
{
@@ -1059,17 +1093,30 @@
if (cp->tok == CTOK_IDENT) {
GCstr *attrstr = cp->str;
cp_next(cp);
- switch (attrstr->hash) {
- case H_(64a9208e,8ce14319): case H_(8e6331b2,95a282af): /* aligned */
+ switch (lj_cparse_case(attrstr,
+ "\007aligned" "\013__aligned__"
+ "\006packed" "\012__packed__"
+ "\004mode" "\010__mode__"
+ "\013vector_size" "\017__vector_size__"
+#if LJ_TARGET_X86
+ "\007regparm" "\013__regparm__"
+ "\005cdecl" "\011__cdecl__"
+ "\010thiscall" "\014__thiscall__"
+ "\010fastcall" "\014__fastcall__"
+ "\007stdcall" "\013__stdcall__"
+ "\012sseregparm" "\016__sseregparm__"
+#endif
+ )) {
+ case 0: case 1: /* aligned */
cp_decl_align(cp, decl);
break;
- case H_(42eb47de,f0ede26c): case H_(29f48a09,cf383e0c): /* packed */
+ case 2: case 3: /* packed */
decl->attr |= CTFP_PACKED;
break;
- case H_(0a84eef6,8dfab04c): case H_(995cf92c,d5696591): /* mode */
+ case 4: case 5: /* mode */
cp_decl_mode(cp, decl);
break;
- case H_(0ab31997,2d5213fa): case H_(bf875611,200e9990): /* vector_size */
+ case 6: case 7: /* vector_size */
{
CTSize vsize = cp_decl_sizeattr(cp);
if (vsize) CTF_INSERT(decl->attr, VSIZEP, lj_fls(vsize));
@@ -1076,27 +1123,27 @@
}
break;
#if LJ_TARGET_X86
- case H_(5ad22db8,c689b848): case H_(439150fa,65ea78cb): /* regparm */
+ case 8: case 9: /* regparm */
CTF_INSERT(decl->fattr, REGPARM, cp_decl_sizeattr(cp));
decl->fattr |= CTFP_CCONV;
break;
- case H_(18fc0b98,7ff4c074): case H_(4e62abed,0a747424): /* cdecl */
+ case 10: case 11: /* cdecl */
CTF_INSERT(decl->fattr, CCONV, CTCC_CDECL);
decl->fattr |= CTFP_CCONV;
break;
- case H_(72b2e41b,494c5a44): case H_(f2356d59,f25fc9bd): /* thiscall */
+ case 12: case 13: /* thiscall */
CTF_INSERT(decl->fattr, CCONV, CTCC_THISCALL);
decl->fattr |= CTFP_CCONV;
break;
- case H_(0d0ffc42,ab746f88): case H_(21c54ba1,7f0ca7e3): /* fastcall */
+ case 14: case 15: /* fastcall */
CTF_INSERT(decl->fattr, CCONV, CTCC_FASTCALL);
decl->fattr |= CTFP_CCONV;
break;
- case H_(ef76b040,9412e06a): case H_(de56697b,c750e6e1): /* stdcall */
+ case 16: case 17: /* stdcall */
CTF_INSERT(decl->fattr, CCONV, CTCC_STDCALL);
decl->fattr |= CTFP_CCONV;
break;
- case H_(ea78b622,f234bd8e): case H_(252ffb06,8d50f34b): /* sseregparm */
+ case 18: case 19: /* sseregparm */
decl->fattr |= CTF_SSEREGPARM;
decl->fattr |= CTFP_CCONV;
break;
@@ -1128,16 +1175,13 @@
while (cp->tok == CTOK_IDENT) {
GCstr *attrstr = cp->str;
cp_next(cp);
- switch (attrstr->hash) {
- case H_(bc2395fa,98f267f8): /* align */
+ if (cp_str_is(attrstr, "align")) {
cp_decl_align(cp, decl);
- break;
- default: /* Ignore all other attributes. */
+ } else { /* Ignore all other attributes. */
if (cp_opt(cp, '(')) {
while (cp->tok != ')' && cp->tok != CTOK_EOF) cp_next(cp);
cp_check(cp, ')');
}
- break;
}
}
cp_check(cp, ')');
@@ -1548,7 +1592,7 @@
cp_errmsg(cp, cp->tok, LJ_ERR_FFI_DECLSPEC);
sz = sizeof(int);
}
- lua_assert(sz != 0);
+ lj_assertCP(sz != 0, "basic ctype with zero size");
info += CTALIGN(lj_fls(sz)); /* Use natural alignment. */
info += (decl->attr & CTF_QUAL); /* Merge qualifiers. */
cp_push(decl, info, sz);
@@ -1717,17 +1761,18 @@
static void cp_pragma(CPState *cp, BCLine pragmaline)
{
cp_next(cp);
- if (cp->tok == CTOK_IDENT &&
- cp->str->hash == H_(e79b999f,42ca3e85)) { /* pack */
+ if (cp->tok == CTOK_IDENT && cp_str_is(cp->str, "pack")) {
cp_next(cp);
cp_check(cp, '(');
if (cp->tok == CTOK_IDENT) {
- if (cp->str->hash == H_(738e923c,a1b65954)) { /* push */
- if (cp->curpack < CPARSE_MAX_PACKSTACK) {
+ if (cp_str_is(cp->str, "push")) {
+ if (cp->curpack < CPARSE_MAX_PACKSTACK-1) {
cp->packstack[cp->curpack+1] = cp->packstack[cp->curpack];
cp->curpack++;
+ } else {
+ cp_errmsg(cp, cp->tok, LJ_ERR_XLEVELS);
}
- } else if (cp->str->hash == H_(6c71cf27,6c71cf27)) { /* pop */
+ } else if (cp_str_is(cp->str, "pop")) {
if (cp->curpack > 0) cp->curpack--;
} else {
cp_errmsg(cp, cp->tok, LJ_ERR_XSYMBOL);
@@ -1776,13 +1821,11 @@
if (tok == CTOK_INTEGER) {
cp_line(cp, hashline);
continue;
- } else if (tok == CTOK_IDENT &&
- cp->str->hash == H_(187aab88,fcb60b42)) { /* line */
+ } else if (tok == CTOK_IDENT && cp_str_is(cp->str, "line")) {
if (cp_next(cp) != CTOK_INTEGER) cp_err_token(cp, tok);
cp_line(cp, hashline);
continue;
- } else if (tok == CTOK_IDENT &&
- cp->str->hash == H_(f5e6b4f8,1d509107)) { /* pragma */
+ } else if (tok == CTOK_IDENT && cp_str_is(cp->str, "pragma")) {
cp_pragma(cp, hashline);
continue;
} else {
@@ -1811,7 +1854,7 @@
/* Treat both static and extern function declarations as extern. */
ct = ctype_get(cp->cts, ctypeid);
/* We always get new anonymous functions (typedefs are copied). */
- lua_assert(gcref(ct->name) == NULL);
+ lj_assertCP(gcref(ct->name) == NULL, "unexpected named function");
id = ctypeid; /* Just name it. */
} else if ((scl & CDF_STATIC)) { /* Accept static constants. */
id = cp_decl_constinit(cp, &ct, ctypeid);
@@ -1853,8 +1896,6 @@
if (cp->tok != CTOK_EOF) cp_err_token(cp, CTOK_EOF);
}
-#undef H_
-
/* ------------------------------------------------------------------------ */
/* Protected callback for C parser. */
@@ -1870,7 +1911,7 @@
cp_decl_single(cp);
if (cp->param && cp->param != cp->L->top)
cp_err(cp, LJ_ERR_FFI_NUMPARAM);
- lua_assert(cp->depth == 0);
+ lj_assertCP(cp->depth == 0, "unbalanced cparser declaration depth");
return NULL;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_cparse.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C declaration parser.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CPARSE_H
@@ -60,6 +60,8 @@
LJ_FUNC int lj_cparse(CPState *cp);
+LJ_FUNC int lj_cparse_case(GCstr *str, const char *match);
+
#endif
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Trace recorder for C data operations.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_ffrecord_c
@@ -32,6 +32,7 @@
#include "lj_crecord.h"
#include "lj_dispatch.h"
#include "lj_strfmt.h"
+#include "lj_strscan.h"
/* Some local macros to save typing. Undef'd at the end. */
#define IR(ref) (&J->cur.ir[(ref)])
@@ -61,7 +62,8 @@
static CTypeID crec_constructor(jit_State *J, GCcdata *cd, TRef tr)
{
CTypeID id;
- lua_assert(tref_iscdata(tr) && cd->ctypeid == CTID_CTYPEID);
+ lj_assertJ(tref_iscdata(tr) && cd->ctypeid == CTID_CTYPEID,
+ "expected CTypeID cdata");
id = *(CTypeID *)cdataptr(cd);
tr = emitir(IRT(IR_FLOAD, IRT_INT), tr, IRFL_CDATA_INT);
emitir(IRTG(IR_EQ, IRT_INT), tr, lj_ir_kint(J, (int32_t)id));
@@ -77,7 +79,7 @@
/* Specialize to the string containing the C type declaration. */
emitir(IRTG(IR_EQ, IRT_STR), tr, lj_ir_kstr(J, s));
cp.L = J->L;
- cp.cts = ctype_ctsG(J2G(J));
+ cp.cts = ctype_cts(J->L);
oldtop = cp.cts->top;
cp.srcname = strdata(s);
cp.p = strdata(s);
@@ -212,7 +214,7 @@
ml[i].trval = emitir(IRT(IR_XLOAD, ml[i].tp), trsptr, 0);
ml[i].trofs = trofs;
i++;
- rwin += (LJ_SOFTFP && ml[i].tp == IRT_NUM) ? 2 : 1;
+ rwin += (LJ_SOFTFP32 && ml[i].tp == IRT_NUM) ? 2 : 1;
if (rwin >= CREC_COPY_REGWIN || i >= mlp) { /* Flush buffered stores. */
rwin = 0;
for ( ; j < i; j++) {
@@ -237,13 +239,14 @@
if (len > CREC_COPY_MAXLEN) goto fallback;
if (ct) {
CTState *cts = ctype_ctsG(J2G(J));
- lua_assert(ctype_isarray(ct->info) || ctype_isstruct(ct->info));
+ lj_assertJ(ctype_isarray(ct->info) || ctype_isstruct(ct->info),
+ "copy of non-aggregate");
if (ctype_isarray(ct->info)) {
CType *cct = ctype_rawchild(cts, ct);
tp = crec_ct2irt(cts, cct);
if (tp == IRT_CDATA) goto rawcopy;
step = lj_ir_type_size[tp];
- lua_assert((len & (step-1)) == 0);
+ lj_assertJ((len & (step-1)) == 0, "copy of fractional size");
} else if ((ct->info & CTF_UNION)) {
step = (1u << ctype_align(ct->info));
goto rawcopy;
@@ -614,10 +617,12 @@
sp = lj_ir_kptr(J, NULL);
} else if (tref_isudata(sp)) {
GCudata *ud = udataV(sval);
- if (ud->udtype == UDTYPE_IO_FILE) {
+ if (ud->udtype == UDTYPE_IO_FILE || ud->udtype == UDTYPE_BUFFER) {
TRef tr = emitir(IRT(IR_FLOAD, IRT_U8), sp, IRFL_UDATA_UDTYPE);
- emitir(IRTGI(IR_EQ), tr, lj_ir_kint(J, UDTYPE_IO_FILE));
- sp = emitir(IRT(IR_FLOAD, IRT_PTR), sp, IRFL_UDATA_FILE);
+ emitir(IRTGI(IR_EQ), tr, lj_ir_kint(J, ud->udtype));
+ sp = emitir(IRT(IR_FLOAD, IRT_PTR), sp,
+ ud->udtype == UDTYPE_IO_FILE ? IRFL_UDATA_FILE :
+ IRFL_SBUF_R);
} else {
sp = emitir(IRT(IR_ADD, IRT_PTR), sp, lj_ir_kintp(J, sizeof(GCudata)));
}
@@ -629,7 +634,8 @@
/* Specialize to the name of the enum constant. */
emitir(IRTG(IR_EQ, IRT_STR), sp, lj_ir_kstr(J, str));
if (cct && ctype_isconstval(cct->info)) {
- lua_assert(ctype_child(cts, cct)->size == 4);
+ lj_assertJ(ctype_child(cts, cct)->size == 4,
+ "only 32 bit const supported"); /* NYI */
svisnz = (void *)(intptr_t)(ofs != 0);
sp = lj_ir_kint(J, (int32_t)ofs);
sid = ctype_cid(cct->info);
@@ -643,8 +649,7 @@
}
} else if (tref_islightud(sp)) {
#if LJ_64
- sp = emitir(IRT(IR_BAND, IRT_P64), sp,
- lj_ir_kint64(J, U64x(00007fff,ffffffff)));
+ lj_trace_err(J, LJ_TRERR_NYICONV);
#endif
} else { /* NYI: tref_istab(sp). */
IRType t;
@@ -757,7 +762,7 @@
IRType t = IRT_I8 + 2*lj_fls(ctype_bitcsz(info)) + ((info&CTF_UNSIGNED)?1:0);
TRef tr = emitir(IRT(IR_XLOAD, t), ptr, 0);
CTSize pos = ctype_bitpos(info), bsz = ctype_bitbsz(info), shift = 32 - bsz;
- lua_assert(t <= IRT_U32); /* NYI: 64 bit bitfields. */
+ lj_assertJ(t <= IRT_U32, "only 32 bit bitfields supported"); /* NYI */
if (rd->data == 0) { /* __index metamethod. */
if ((info & CTF_BOOL)) {
tr = emitir(IRTI(IR_BAND), tr, lj_ir_kint(J, (int32_t)((1u << pos))));
@@ -769,7 +774,7 @@
tr = emitir(IRTI(IR_BSHL), tr, lj_ir_kint(J, shift - pos));
tr = emitir(IRTI(IR_BSAR), tr, lj_ir_kint(J, shift));
} else {
- lua_assert(bsz < 32); /* Full-size fields cannot end up here. */
+ lj_assertJ(bsz < 32, "unexpected full bitfield index");
tr = emitir(IRTI(IR_BSHR), tr, lj_ir_kint(J, pos));
tr = emitir(IRTI(IR_BAND), tr, lj_ir_kint(J, (int32_t)((1u << bsz)-1)));
/* We can omit the U32 to NUM conversion, since bsz < 32. */
@@ -884,7 +889,7 @@
crec_index_bf(J, rd, ptr, fct->info);
return;
} else {
- lua_assert(ctype_isfield(fct->info));
+ lj_assertJ(ctype_isfield(fct->info), "field expected");
sid = ctype_cid(fct->info);
}
}
@@ -1022,8 +1027,26 @@
crec_ct_tv(J, dc, dp, sp, sval);
}
} else if (ctype_isstruct(d->info)) {
- CTypeID fid = d->sib;
+ CTypeID fid;
MSize i = 1;
+ if (!J->base[1]) { /* Handle zero-fill of struct-of-NYI. */
+ fid = d->sib;
+ while (fid) {
+ CType *df = ctype_get(cts, fid);
+ fid = df->sib;
+ if (ctype_isfield(df->info)) {
+ CType *dc;
+ if (!gcref(df->name)) continue; /* Ignore unnamed fields. */
+ dc = ctype_rawchild(cts, df); /* Field type. */
+ if (!(ctype_isnum(dc->info) || ctype_isptr(dc->info) ||
+ ctype_isenum(dc->info)))
+ goto special;
+ } else if (!ctype_isconstval(df->info)) {
+ goto special;
+ }
+ }
+ }
+ fid = d->sib;
while (fid) {
CType *df = ctype_get(cts, fid);
fid = df->sib;
@@ -1048,6 +1071,11 @@
dp = emitir(IRT(IR_ADD, IRT_PTR), trcd,
lj_ir_kintp(J, df->size + sizeof(GCcdata)));
crec_ct_tv(J, dc, dp, sp, sval);
+ if ((d->info & CTF_UNION)) {
+ if (d->size != dc->size) /* NYI: partial init of union. */
+ lj_trace_err(J, LJ_TRERR_NYICONV);
+ break;
+ }
} else if (!ctype_isconstval(df->info)) {
/* NYI: init bitfields and sub-structures. */
lj_trace_err(J, LJ_TRERR_NYICONV);
@@ -1091,6 +1119,8 @@
ngpr = 1;
else if (ctype_cconv(ct->info) == CTCC_FASTCALL)
ngpr = 2;
+#elif LJ_TARGET_ARM64 && LJ_TARGET_OSX
+ int ngpr = CCALL_NARG_GPR;
#endif
/* Skip initial attributes. */
@@ -1111,11 +1141,19 @@
if (fid) { /* Get argument type from field. */
CType *ctf = ctype_get(cts, fid);
fid = ctf->sib;
- lua_assert(ctype_isfield(ctf->info));
+ lj_assertJ(ctype_isfield(ctf->info), "field expected");
did = ctype_cid(ctf->info);
} else {
if (!(ct->info & CTF_VARARG))
lj_trace_err(J, LJ_TRERR_NYICALL); /* Too many arguments. */
+#if LJ_TARGET_ARM64 && LJ_TARGET_OSX
+ if (ngpr >= 0) {
+ ngpr = -1;
+ args[n++] = TREF_NIL; /* Marker for start of varargs. */
+ if (n >= CCI_NARGS_MAX)
+ lj_trace_err(J, LJ_TRERR_NYICALL);
+ }
+#endif
did = lj_ccall_ctid_vararg(cts, o); /* Infer vararg type. */
}
d = ctype_raw(cts, did);
@@ -1124,6 +1162,15 @@
lj_trace_err(J, LJ_TRERR_NYICALL);
tr = crec_ct_tv(J, d, 0, *base, o);
if (ctype_isinteger_or_bool(d->info)) {
+#if LJ_TARGET_ARM64 && LJ_TARGET_OSX
+ if (!ngpr) {
+ /* Fixed args passed on the stack use their unpromoted size. */
+ if (d->size != lj_ir_type_size[tref_type(tr)]) {
+ lj_assertJ(d->size == 1 || d->size==2, "unexpected size %d", d->size);
+ tr = emitconv(tr, d->size==1 ? IRT_U8 : IRT_U16, tref_type(tr), 0);
+ }
+ } else
+#endif
if (d->size < 4) {
if ((d->info & CTF_UNSIGNED))
tr = emitconv(tr, IRT_INT, d->size==1 ? IRT_U8 : IRT_U16, 0);
@@ -1130,7 +1177,7 @@
else
tr = emitconv(tr, IRT_INT, d->size==1 ? IRT_I8 : IRT_I16,IRCONV_SEXT);
}
- } else if (LJ_SOFTFP && ctype_isfp(d->info) && d->size > 4) {
+ } else if (LJ_SOFTFP32 && ctype_isfp(d->info) && d->size > 4) {
lj_needsplit(J);
}
#if LJ_TARGET_X86
@@ -1161,6 +1208,10 @@
}
}
#endif
+#elif LJ_TARGET_ARM64 && LJ_TARGET_OSX
+ if (!ctype_isfp(d->info) && ngpr) {
+ ngpr--;
+ }
#endif
args[n] = tr;
}
@@ -1209,8 +1260,7 @@
TRef tr;
TValue tv;
/* Check for blacklisted C functions that might call a callback. */
- setlightudV(&tv,
- cdata_getptr(cdataptr(cd), (LJ_64 && tp == IRT_P64) ? 8 : 4));
+ tv.u64 = ((uintptr_t)cdata_getptr(cdataptr(cd), (LJ_64 && tp == IRT_P64) ? 8 : 4) >> 2) | U64x(800000000, 00000000);
if (tvistrue(lj_tab_get(J->L, cts->miscmap, &tv)))
lj_trace_err(J, LJ_TRERR_BLACKL);
if (ctype_isvoid(ctr->info)) {
@@ -1449,7 +1499,8 @@
void LJ_FASTCALL recff_cdata_arith(jit_State *J, RecordFFData *rd)
{
- CTState *cts = ctype_ctsG(J2G(J));
+ CTState *cts = ctype_cts(J->L);
+ MMS mm = (MMS)rd->data;
TRef sp[2];
CType *s[2];
MSize i;
@@ -1478,9 +1529,13 @@
if (ctype_isenum(ct->info)) ct = ctype_child(cts, ct);
goto ok;
} else if (ctype_isfunc(ct->info)) {
+ CTypeID id0 = i ? ctype_typeid(cts, s[0]) : 0;
tr = emitir(IRT(IR_FLOAD, IRT_PTR), tr, IRFL_CDATA_PTR);
ct = ctype_get(cts,
lj_ctype_intern(cts, CTINFO(CT_PTR, CTALIGN_PTR|id), CTSIZE_PTR));
+ if (i) {
+ s[0] = ctype_get(cts, id0); /* cts->tab may have been reallocated. */
+ }
goto ok;
} else {
tr = emitir(IRT(IR_ADD, IRT_PTR), tr, lj_ir_kintp(J, sizeof(GCcdata)));
@@ -1495,6 +1550,8 @@
}
}
} else if (tref_isnil(tr)) {
+ if (!(mm == MM_len || mm == MM_eq || mm == MM_lt || mm == MM_le))
+ lj_trace_err(J, LJ_TRERR_BADTYPE);
tr = lj_ir_kptr(J, NULL);
ct = ctype_get(cts, CTID_P_VOID);
} else if (tref_isinteger(tr)) {
@@ -1513,12 +1570,12 @@
ct = ctype_child(cts, cct);
tr = lj_ir_kint(J, (int32_t)ofs);
} else { /* Interpreter will throw or return false. */
- ct = ctype_get(cts, CTID_P_VOID);
+ lj_trace_err(J, LJ_TRERR_BADTYPE);
}
} else if (ctype_isptr(ct->info)) {
tr = emitir(IRT(IR_ADD, IRT_PTR), tr, lj_ir_kintp(J, sizeof(GCstr)));
} else {
- ct = ctype_get(cts, CTID_P_VOID);
+ lj_trace_err(J, LJ_TRERR_BADTYPE);
}
} else if (!tref_isnum(tr)) {
tr = 0;
@@ -1530,8 +1587,9 @@
}
{
TRef tr;
- if (!(tr = crec_arith_int64(J, sp, s, (MMS)rd->data)) &&
- !(tr = crec_arith_ptr(J, sp, s, (MMS)rd->data)) &&
+ if ((mm == MM_len || mm == MM_concat ||
+ (!(tr = crec_arith_int64(J, sp, s, mm)) &&
+ !(tr = crec_arith_ptr(J, sp, s, mm)))) &&
!(tr = crec_arith_meta(J, sp, s, cts, rd)))
return;
J->base[0] = tr;
@@ -1750,11 +1808,21 @@
return 0; /* Use regular 32 bit ops. */
}
+static TRef crec_bit64_arg(jit_State *J, CType *d, TRef sp, TValue *sval)
+{
+ if (LJ_UNLIKELY(tref_isstr(sp))) {
+ if (lj_strscan_num(strV(sval), sval)) {
+ sp = emitir(IRTG(IR_STRTO, IRT_NUM), sp, 0);
+ } /* else: interpreter will throw. */
+ }
+ return crec_ct_tv(J, d, 0, sp, sval);
+}
+
void LJ_FASTCALL recff_bit64_tobit(jit_State *J, RecordFFData *rd)
{
CTState *cts = ctype_ctsG(J2G(J));
- TRef tr = crec_ct_tv(J, ctype_get(cts, CTID_INT64), 0,
- J->base[0], &rd->argv[0]);
+ TRef tr = crec_bit64_arg(J, ctype_get(cts, CTID_INT64),
+ J->base[0], &rd->argv[0]);
if (!tref_isinteger(tr))
tr = emitconv(tr, IRT_INT, tref_type(tr), 0);
J->base[0] = tr;
@@ -1765,7 +1833,7 @@
CTState *cts = ctype_ctsG(J2G(J));
CTypeID id = crec_bit64_type(cts, &rd->argv[0]);
if (id) {
- TRef tr = crec_ct_tv(J, ctype_get(cts, id), 0, J->base[0], &rd->argv[0]);
+ TRef tr = crec_bit64_arg(J, ctype_get(cts, id), J->base[0], &rd->argv[0]);
tr = emitir(IRT(rd->data, id-CTID_INT64+IRT_I64), tr, 0);
J->base[0] = emitir(IRTG(IR_CNEWI, IRT_CDATA), lj_ir_kint(J, id), tr);
return 1;
@@ -1785,9 +1853,9 @@
if (id) {
CType *ct = ctype_get(cts, id);
uint32_t ot = IRT(rd->data, id-CTID_INT64+IRT_I64);
- TRef tr = crec_ct_tv(J, ct, 0, J->base[0], &rd->argv[0]);
+ TRef tr = crec_bit64_arg(J, ct, J->base[0], &rd->argv[0]);
for (i = 1; J->base[i] != 0; i++) {
- TRef tr2 = crec_ct_tv(J, ct, 0, J->base[i], &rd->argv[i]);
+ TRef tr2 = crec_bit64_arg(J, ct, J->base[i], &rd->argv[i]);
tr = emitir(ot, tr, tr2);
}
J->base[0] = emitir(IRTG(IR_CNEWI, IRT_CDATA), lj_ir_kint(J, id), tr);
@@ -1802,8 +1870,8 @@
CTypeID id;
TRef tsh = 0;
if (J->base[0] && tref_iscdata(J->base[1])) {
- tsh = crec_ct_tv(J, ctype_get(cts, CTID_INT64), 0,
- J->base[1], &rd->argv[1]);
+ tsh = crec_bit64_arg(J, ctype_get(cts, CTID_INT64),
+ J->base[1], &rd->argv[1]);
if (!tref_isinteger(tsh))
tsh = emitconv(tsh, IRT_INT, tref_type(tsh), 0);
J->base[1] = tsh;
@@ -1810,7 +1878,7 @@
}
id = crec_bit64_type(cts, &rd->argv[0]);
if (id) {
- TRef tr = crec_ct_tv(J, ctype_get(cts, id), 0, J->base[0], &rd->argv[0]);
+ TRef tr = crec_bit64_arg(J, ctype_get(cts, id), J->base[0], &rd->argv[0]);
uint32_t op = rd->data;
if (!tsh) tsh = lj_opt_narrow_tobit(J, J->base[1]);
if (!(op < IR_BROL ? LJ_TARGET_MASKSHIFT : LJ_TARGET_MASKROT) &&
@@ -1840,7 +1908,7 @@
CTypeID id2 = 0;
n = (int32_t)lj_carith_check64(J->L, 2, &id2);
if (id2)
- trsf = crec_ct_tv(J, ctype_get(cts, CTID_INT32), 0, trsf, &rd->argv[1]);
+ trsf = crec_bit64_arg(J, ctype_get(cts, CTID_INT32), trsf, &rd->argv[1]);
else
trsf = lj_opt_narrow_tobit(J, trsf);
emitir(IRTGI(IR_EQ), trsf, lj_ir_kint(J, n)); /* Specialize to n. */
@@ -1847,10 +1915,11 @@
} else {
n = id ? 16 : 8;
}
- if (n < 0) { n = -n; sf |= STRFMT_F_UPPER; }
+ if (n < 0) { n = (int32_t)(~n+1u); sf |= STRFMT_F_UPPER; }
+ if ((uint32_t)n > 254) n = 254;
sf |= ((SFormat)((n+1)&255) << STRFMT_SH_PREC);
if (id) {
- tr = crec_ct_tv(J, ctype_get(cts, id), 0, J->base[0], &rd->argv[0]);
+ tr = crec_bit64_arg(J, ctype_get(cts, id), J->base[0], &rd->argv[0]);
if (n < 16)
tr = emitir(IRT(IR_BAND, IRT_U64), tr,
lj_ir_kint64(J, ((uint64_t)1 << 4*n)-1));
@@ -1879,10 +1948,36 @@
d = ctype_get(cts, CTID_DOUBLE);
J->base[0] = crec_ct_tv(J, d, 0, J->base[0], &rd->argv[0]);
} else {
+ /* Specialize to the ctype that couldn't be converted. */
+ argv2cdata(J, J->base[0], &rd->argv[0]);
J->base[0] = TREF_NIL;
}
}
+TRef lj_crecord_loadiu64(jit_State *J, TRef tr, cTValue *o)
+{
+ CTypeID id = argv2cdata(J, tr, o)->ctypeid;
+ if (!(id == CTID_INT64 || id == CTID_UINT64))
+ lj_trace_err(J, LJ_TRERR_BADTYPE);
+ lj_needsplit(J);
+ return emitir(IRT(IR_FLOAD, id == CTID_INT64 ? IRT_I64 : IRT_U64), tr,
+ IRFL_CDATA_INT64);
+}
+
+#if LJ_HASBUFFER
+TRef lj_crecord_topcvoid(jit_State *J, TRef tr, cTValue *o)
+{
+ CTState *cts = ctype_ctsG(J2G(J));
+ if (!tref_iscdata(tr)) lj_trace_err(J, LJ_TRERR_BADTYPE);
+ return crec_ct_tv(J, ctype_get(cts, CTID_P_CVOID), 0, tr, o);
+}
+
+TRef lj_crecord_topuint8(jit_State *J, TRef tr)
+{
+ return emitir(IRTG(IR_CNEWI, IRT_CDATA), lj_ir_kint(J, CTID_P_UINT8), tr);
+}
+#endif
+
#undef IR
#undef emitir
#undef emitconv
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_crecord.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Trace recorder for C data operations.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CRECORD_H
@@ -33,6 +33,11 @@
LJ_FUNC TRef recff_bit64_tohex(jit_State *J, RecordFFData *rd, TRef hdr);
LJ_FUNC void LJ_FASTCALL lj_crecord_tonumber(jit_State *J, RecordFFData *rd);
+LJ_FUNC TRef lj_crecord_loadiu64(jit_State *J, TRef tr, cTValue *o);
+#if LJ_HASBUFFER
+LJ_FUNC TRef lj_crecord_topcvoid(jit_State *J, TRef tr, cTValue *o);
+LJ_FUNC TRef lj_crecord_topuint8(jit_State *J, TRef tr);
#endif
+#endif
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C type management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include "lj_obj.h"
@@ -153,7 +153,7 @@
{
CTypeID id = cts->top;
CType *ct;
- lua_assert(cts->L);
+ lj_assertCTS(cts->L, "uninitialized cts->L");
if (LJ_UNLIKELY(id >= cts->sizetab)) {
if (id >= CTID_MAX) lj_err_msg(cts->L, LJ_ERR_TABOV);
#ifdef LUAJIT_CTYPE_CHECK_ANCHOR
@@ -182,7 +182,7 @@
{
uint32_t h = ct_hashtype(info, size);
CTypeID id = cts->hash[h];
- lua_assert(cts->L);
+ lj_assertCTS(cts->L, "uninitialized cts->L");
while (id) {
CType *ct = ctype_get(cts, id);
if (ct->info == info && ct->size == size)
@@ -191,8 +191,20 @@
}
id = cts->top;
if (LJ_UNLIKELY(id >= cts->sizetab)) {
+#ifdef LUAJIT_CTYPE_CHECK_ANCHOR
+ CType *ct;
+#endif
if (id >= CTID_MAX) lj_err_msg(cts->L, LJ_ERR_TABOV);
+#ifdef LUAJIT_CTYPE_CHECK_ANCHOR
+ ct = lj_mem_newvec(cts->L, id+1, CType);
+ memcpy(ct, cts->tab, id*sizeof(CType));
+ memset(cts->tab, 0, id*sizeof(CType));
+ lj_mem_freevec(cts->g, cts->tab, cts->sizetab, CType);
+ cts->tab = ct;
+ cts->sizetab = id+1;
+#else
lj_mem_growvec(cts->L, cts->tab, cts->sizetab, CTID_MAX, CType);
+#endif
}
cts->top = id+1;
cts->tab[id].info = info;
@@ -298,9 +310,9 @@
}
ct = ctype_raw(cts, arrid);
}
- lua_assert(ctype_isvlarray(ct->info)); /* Must be a VLA. */
+ lj_assertCTS(ctype_isvlarray(ct->info), "VLA expected");
ct = ctype_rawchild(cts, ct); /* Get array element. */
- lua_assert(ctype_hassize(ct->info));
+ lj_assertCTS(ctype_hassize(ct->info), "bad VLA without size");
/* Calculate actual size of VLA and check for overflow. */
xsz += (uint64_t)ct->size * nelem;
return xsz < 0x80000000u ? (CTSize)xsz : CTSIZE_INVALID;
@@ -323,7 +335,8 @@
} else {
if (!(qual & CTFP_ALIGNED)) qual |= (info & CTF_ALIGN);
qual |= (info & ~(CTF_ALIGN|CTMASK_CID));
- lua_assert(ctype_hassize(info) || ctype_isfunc(info));
+ lj_assertCTS(ctype_hassize(info) || ctype_isfunc(info),
+ "ctype without size");
*szp = ctype_isfunc(info) ? CTSIZE_INVALID : ct->size;
break;
}
@@ -332,6 +345,14 @@
return qual;
}
+/* Ditto, but follow a reference. */
+CTInfo lj_ctype_info_raw(CTState *cts, CTypeID id, CTSize *szp)
+{
+ CType *ct = ctype_get(cts, id);
+ if (ctype_isref(ct->info)) id = ctype_cid(ct->info);
+ return lj_ctype_info(cts, id, szp);
+}
+
/* Get ctype metamethod. */
cTValue *lj_ctype_meta(CTState *cts, CTypeID id, MMS mm)
{
@@ -528,7 +549,7 @@
ctype_appc(ctr, ')');
break;
default:
- lua_assert(0);
+ lj_assertG_(ctr->cts->g, 0, "bad ctype %08x", info);
break;
}
ct = ctype_get(ctr->cts, ctype_cid(info));
@@ -561,7 +582,7 @@
if (isunsigned) {
*--p = 'U';
} else if ((int64_t)n < 0) {
- n = (uint64_t)-(int64_t)n;
+ n = ~n+1u;
sign = 1;
}
do { *--p = (char)('0' + n % 10); } while (n /= 10);
@@ -582,7 +603,7 @@
lj_strfmt_putfnum(sb, STRFMT_G14, re.n);
if (!(im.u32.hi & 0x80000000u) || im.n != im.n) lj_buf_putchar(sb, '+');
lj_strfmt_putfnum(sb, STRFMT_G14, im.n);
- lj_buf_putchar(sb, sbufP(sb)[-1] >= 'a' ? 'I' : 'i');
+ lj_buf_putchar(sb, sb->w[-1] >= 'a' ? 'I' : 'i');
return lj_buf_str(L, sb);
}
@@ -622,6 +643,18 @@
return cts;
}
+/* Create special weak-keyed finalizer table. */
+void lj_ctype_initfin(lua_State *L)
+{
+ /* NOBARRIER: The table is new (marked white). */
+ GCtab *t = lj_tab_new(L, 0, 1);
+ setgcref(t->metatable, obj2gco(t));
+ setstrV(L, lj_tab_setstr(L, t, lj_str_newlit(L, "__mode")),
+ lj_str_newlit(L, "k"));
+ t->nomm = (uint8_t)(~(1u<<MM_mode));
+ setgcref(G(L)->gcroot[GCROOT_FFI_FIN], obj2gco(t));
+}
+
/* Free C type table and state. */
void lj_ctype_freestate(global_State *g)
{
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ctype.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** C type management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_CTYPE_H
@@ -177,7 +177,6 @@
MSize sizetab; /* Size of C type table. */
lua_State *L; /* Lua state (needed for errors and allocations). */
global_State *g; /* Global state. */
- GCtab *finalizer; /* Map of cdata to finalizer. */
GCtab *miscmap; /* Map of -CTypeID to metatable and cb slot to func. */
CCallback cb; /* Temporary callback state. */
CTypeID1 hash[CTHASH_SIZE]; /* Hash anchors for C type table. */
@@ -260,6 +259,12 @@
#define CT_MEMALIGN 3 /* Alignment guaranteed by memory allocator. */
+#ifdef LUA_USE_ASSERT
+#define lj_assertCTS(c, ...) (lj_assertG_(cts->g, (c), __VA_ARGS__))
+#else
+#define lj_assertCTS(c, ...) ((void)cts)
+#endif
+
/* -- Predefined types ---------------------------------------------------- */
/* Target-dependent types. */
@@ -270,6 +275,8 @@
#define CTTYDEFP(_)
#endif
+#define CTF_LONG_IF8 (CTF_LONG * (sizeof(long) == 8))
+
/* Common types. */
#define CTTYDEF(_) \
_(NONE, 0, CT_ATTRIB, CTATTRIB(CTA_BAD)) \
@@ -283,8 +290,8 @@
_(UINT16, 2, CT_NUM, CTF_UNSIGNED|CTALIGN(1)) \
_(INT32, 4, CT_NUM, CTALIGN(2)) \
_(UINT32, 4, CT_NUM, CTF_UNSIGNED|CTALIGN(2)) \
- _(INT64, 8, CT_NUM, CTF_LONG|CTALIGN(3)) \
- _(UINT64, 8, CT_NUM, CTF_UNSIGNED|CTF_LONG|CTALIGN(3)) \
+ _(INT64, 8, CT_NUM, CTF_LONG_IF8|CTALIGN(3)) \
+ _(UINT64, 8, CT_NUM, CTF_UNSIGNED|CTF_LONG_IF8|CTALIGN(3)) \
_(FLOAT, 4, CT_NUM, CTF_FP|CTALIGN(2)) \
_(DOUBLE, 8, CT_NUM, CTF_FP|CTALIGN(3)) \
_(COMPLEX_FLOAT, 8, CT_ARRAY, CTF_COMPLEX|CTALIGN(2)|CTID_FLOAT) \
@@ -292,6 +299,7 @@
_(P_VOID, CTSIZE_PTR, CT_PTR, CTALIGN_PTR|CTID_VOID) \
_(P_CVOID, CTSIZE_PTR, CT_PTR, CTALIGN_PTR|CTID_CVOID) \
_(P_CCHAR, CTSIZE_PTR, CT_PTR, CTALIGN_PTR|CTID_CCHAR) \
+ _(P_UINT8, CTSIZE_PTR, CT_PTR, CTALIGN_PTR|CTID_UINT8) \
_(A_CCHAR, -1, CT_ARRAY, CTF_CONST|CTALIGN(0)|CTID_CCHAR) \
_(CTYPEID, 4, CT_ENUM, CTALIGN(2)|CTID_INT32) \
CTTYDEFP(_) \
@@ -383,6 +391,16 @@
return cts;
}
+/* Load FFI library on-demand. */
+#define ctype_loadffi(L) \
+ do { \
+ if (!ctype_ctsG(G(L))) { \
+ ptrdiff_t oldtop = (char *)L->top - mref(L->stack, char); \
+ luaopen_ffi(L); \
+ L->top = (TValue *)(mref(L->stack, char) + oldtop); \
+ } \
+ } while (0)
+
/* Save and restore state of C type table. */
#define LJ_CTYPE_SAVE(cts) CTState savects_ = *(cts)
#define LJ_CTYPE_RESTORE(cts) \
@@ -392,7 +410,8 @@
/* Check C type ID for validity when assertions are enabled. */
static LJ_AINLINE CTypeID ctype_check(CTState *cts, CTypeID id)
{
- lua_assert(id > 0 && id < cts->top); UNUSED(cts);
+ UNUSED(cts);
+ lj_assertCTS(id > 0 && id < cts->top, "bad CTID %d", id);
return id;
}
@@ -408,8 +427,9 @@
/* Get child C type. */
static LJ_AINLINE CType *ctype_child(CTState *cts, CType *ct)
{
- lua_assert(!(ctype_isvoid(ct->info) || ctype_isstruct(ct->info) ||
- ctype_isbitfield(ct->info))); /* These don't have children. */
+ lj_assertCTS(!(ctype_isvoid(ct->info) || ctype_isstruct(ct->info) ||
+ ctype_isbitfield(ct->info)),
+ "ctype %08x has no children", ct->info);
return ctype_get(cts, ctype_cid(ct->info));
}
@@ -449,11 +469,13 @@
LJ_FUNC CTSize lj_ctype_size(CTState *cts, CTypeID id);
LJ_FUNC CTSize lj_ctype_vlsize(CTState *cts, CType *ct, CTSize nelem);
LJ_FUNC CTInfo lj_ctype_info(CTState *cts, CTypeID id, CTSize *szp);
+LJ_FUNC CTInfo lj_ctype_info_raw(CTState *cts, CTypeID id, CTSize *szp);
LJ_FUNC cTValue *lj_ctype_meta(CTState *cts, CTypeID id, MMS mm);
LJ_FUNC GCstr *lj_ctype_repr(lua_State *L, CTypeID id, GCstr *name);
LJ_FUNC GCstr *lj_ctype_repr_int64(lua_State *L, uint64_t n, int isunsigned);
LJ_FUNC GCstr *lj_ctype_repr_complex(lua_State *L, void *sp, CTSize size);
LJ_FUNC CTState *lj_ctype_init(lua_State *L);
+LJ_FUNC void lj_ctype_initfin(lua_State *L);
LJ_FUNC void lj_ctype_freestate(global_State *g);
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Debugging and introspection.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_debug_c
@@ -55,7 +55,8 @@
const BCIns *ins;
GCproto *pt;
BCPos pos;
- lua_assert(fn->c.gct == ~LJ_TFUNC || fn->c.gct == ~LJ_TTHREAD);
+ lj_assertL(fn->c.gct == ~LJ_TFUNC || fn->c.gct == ~LJ_TTHREAD,
+ "function or frame expected");
if (!isluafunc(fn)) { /* Cannot derive a PC for non-Lua functions. */
return NO_BCPOS;
} else if (nextframe == NULL) { /* Lua function on top. */
@@ -63,6 +64,7 @@
if (cf == NULL || (char *)cframe_pc(cf) == (char *)cframe_L(cf))
return NO_BCPOS;
ins = cframe_pc(cf); /* Only happens during error/hook handling. */
+ if (!ins) return NO_BCPOS;
} else {
if (frame_islua(nextframe)) {
ins = frame_pc(nextframe);
@@ -93,6 +95,7 @@
}
}
ins = cframe_pc(cf);
+ if (!ins) return NO_BCPOS;
}
}
pt = funcproto(fn);
@@ -99,9 +102,12 @@
pos = proto_bcpos(pt, ins) - 1;
#if LJ_HASJIT
if (pos > pt->sizebc) { /* Undo the effects of lj_trace_exit for JLOOP. */
- GCtrace *T = (GCtrace *)((char *)(ins-1) - offsetof(GCtrace, startins));
- lua_assert(bc_isret(bc_op(ins[-1])));
- pos = proto_bcpos(pt, mref(T->startpc, const BCIns));
+ if (bc_isret(bc_op(ins[-1]))) {
+ GCtrace *T = (GCtrace *)((char *)(ins-1) - offsetof(GCtrace, startins));
+ pos = proto_bcpos(pt, mref(T->startpc, const BCIns));
+ } else {
+ pos = NO_BCPOS; /* Punt in case of stack overflow for stitched trace. */
+ }
}
#endif
return pos;
@@ -133,7 +139,7 @@
BCPos pc = debug_framepc(L, fn, nextframe);
if (pc != NO_BCPOS) {
GCproto *pt = funcproto(fn);
- lua_assert(pc <= pt->sizebc);
+ lj_assertL(pc <= pt->sizebc, "PC out of range");
return lj_debug_line(pt, pc);
}
return -1;
@@ -214,7 +220,7 @@
const char *lj_debug_uvname(GCproto *pt, uint32_t idx)
{
const uint8_t *p = proto_uvinfo(pt);
- lua_assert(idx < pt->sizeuv);
+ lj_assertX(idx < pt->sizeuv, "bad upvalue index");
if (!p) return "";
if (idx) while (*p++ || --idx) ;
return (const char *)p;
@@ -221,7 +227,7 @@
}
/* Get name and value of upvalue. */
-const char *lj_debug_uvnamev(cTValue *o, uint32_t idx, TValue **tvp)
+const char *lj_debug_uvnamev(cTValue *o, uint32_t idx, TValue **tvp, GCobj **op)
{
if (tvisfunc(o)) {
GCfunc *fn = funcV(o);
@@ -228,12 +234,15 @@
if (isluafunc(fn)) {
GCproto *pt = funcproto(fn);
if (idx < pt->sizeuv) {
- *tvp = uvval(&gcref(fn->l.uvptr[idx])->uv);
+ GCobj *uvo = gcref(fn->l.uvptr[idx]);
+ *tvp = uvval(&uvo->uv);
+ *op = uvo;
return lj_debug_uvname(pt, idx);
}
} else {
if (idx < fn->c.nupvalues) {
*tvp = &fn->c.upvalue[idx];
+ *op = obj2gco(fn);
return "";
}
}
@@ -429,7 +438,7 @@
GCfunc *fn;
if (*what == '>') {
TValue *func = L->top - 1;
- api_check(L, tvisfunc(func));
+ if (!tvisfunc(func)) return 0;
fn = funcV(func);
L->top--;
what++;
@@ -436,13 +445,14 @@
} else {
uint32_t offset = (uint32_t)ar->i_ci & 0xffff;
uint32_t size = (uint32_t)ar->i_ci >> 16;
- lua_assert(offset != 0);
+ lj_assertL(offset != 0, "bad frame offset");
frame = tvref(L->stack) + offset;
if (size) nextframe = frame + size;
- lua_assert(frame <= tvref(L->maxstack) &&
- (!nextframe || nextframe <= tvref(L->maxstack)));
+ lj_assertL(frame <= tvref(L->maxstack) &&
+ (!nextframe || nextframe <= tvref(L->maxstack)),
+ "broken frame chain");
fn = frame_func(frame);
- lua_assert(fn->c.gct == ~LJ_TFUNC);
+ lj_assertL(fn->c.gct == ~LJ_TFUNC, "bad frame function");
}
for (; *what; what++) {
if (*what == 'S') {
@@ -642,7 +652,7 @@
level += dir;
}
if (lastlen)
- setsbufP(sb, sbufB(sb) + lastlen); /* Zap trailing separator. */
+ sb->w = sb->b + lastlen; /* Zap trailing separator. */
}
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_debug.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Debugging and introspection.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_DEBUG_H
@@ -29,7 +29,8 @@
LJ_FUNC cTValue *lj_debug_frame(lua_State *L, int level, int *size);
LJ_FUNC BCLine LJ_FASTCALL lj_debug_line(GCproto *pt, BCPos pc);
LJ_FUNC const char *lj_debug_uvname(GCproto *pt, uint32_t idx);
-LJ_FUNC const char *lj_debug_uvnamev(cTValue *o, uint32_t idx, TValue **tvp);
+LJ_FUNC const char *lj_debug_uvnamev(cTValue *o, uint32_t idx, TValue **tvp,
+ GCobj **op);
LJ_FUNC const char *lj_debug_slotname(GCproto *pt, const BCIns *pc,
BCReg slot, const char **name);
LJ_FUNC const char *lj_debug_funcname(lua_State *L, cTValue *frame,
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_def.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_def.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_def.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT common internal definitions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_DEF_H
@@ -8,8 +8,8 @@
#include "lua.h"
-#if defined(_MSC_VER)
-/* MSVC is stuck in the last century and doesn't have C99's stdint.h. */
+#if defined(_MSC_VER) && (_MSC_VER < 1700)
+/* Old MSVC is stuck in the last century and doesn't have C99's stdint.h. */
typedef __int8 int8_t;
typedef __int16 int16_t;
typedef __int32 int32_t;
@@ -69,7 +69,7 @@
#define LJ_MAX_UPVAL 249 /* Max. # of upvalues. */
#define LJ_MAX_IDXCHAIN 100 /* __index/__newindex chain limit. */
-#define LJ_STACK_EXTRA (5+2*LJ_FR2) /* Extra stack space (metamethods). */
+#define LJ_STACK_EXTRA (5+3*LJ_FR2) /* Extra stack space (metamethods). */
#define LJ_NUM_CBPAGE 1 /* Number of FFI callback pages. */
@@ -80,7 +80,6 @@
#define LJ_MIN_SBUF 32 /* Min. string buffer length. */
#define LJ_MIN_VECSZ 8 /* Min. size for growable vectors. */
#define LJ_MIN_IRSZ 32 /* Min. size for growable IR. */
-#define LJ_MIN_K64SZ 16 /* Min. size for chained K64Array. */
/* JIT compiler limits. */
#define LJ_MAX_JSLOTS 250 /* Max. # of stack slots for a trace. */
@@ -105,9 +104,10 @@
#define checku16(x) ((x) == (int32_t)(uint16_t)(x))
#define checki32(x) ((x) == (int32_t)(x))
#define checku32(x) ((x) == (uint32_t)(x))
+#define checkptr31(x) (((uint64_t)(uintptr_t)(x) >> 31) == 0)
#define checkptr32(x) ((uintptr_t)(x) == (uint32_t)(uintptr_t)(x))
#define checkptr47(x) (((uint64_t)(uintptr_t)(x) >> 47) == 0)
-#define checkptrGC(x) (LJ_GC64 ? checkptr47((x)) : LJ_64 ? checkptr32((x)) :1)
+#define checkptrGC(x) (LJ_GC64 ? checkptr47((x)) : LJ_64 ? checkptr31((x)) :1)
/* Every half-decent C compiler transforms this into a rotate instruction. */
#define lj_rol(x, n) (((x)<<(n)) | ((x)>>(-(int)(n)&(8*sizeof(x)-1))))
@@ -120,7 +120,7 @@
#define bloomset(b, x) ((b) |= bloombit((x)))
#define bloomtest(b, x) ((b) & bloombit((x)))
-#if defined(__GNUC__) || defined(__psp2__)
+#if defined(__GNUC__) || defined(__clang__) || defined(__psp2__)
#define LJ_NORET __attribute__((noreturn))
#define LJ_ALIGN(n) __attribute__((aligned(n)))
@@ -146,15 +146,9 @@
#define LJ_UNLIKELY(x) __builtin_expect(!!(x), 0)
#define lj_ffs(x) ((uint32_t)__builtin_ctz(x))
-/* Don't ask ... */
-#if defined(__INTEL_COMPILER) && (defined(__i386__) || defined(__x86_64__))
-static LJ_AINLINE uint32_t lj_fls(uint32_t x)
-{
- uint32_t r; __asm__("bsrl %1, %0" : "=r" (r) : "rm" (x) : "cc"); return r;
-}
-#else
#define lj_fls(x) ((uint32_t)(__builtin_clz(x)^31))
-#endif
+#define lj_ffs64(x) ((uint32_t)__builtin_ctzll(x))
+#define lj_fls64(x) ((uint32_t)(__builtin_clzll(x)^63))
#if defined(__arm__)
static LJ_AINLINE uint32_t lj_bswap(uint32_t x)
@@ -182,7 +176,7 @@
{
return ((uint64_t)lj_bswap((uint32_t)x)<<32) | lj_bswap((uint32_t)(x>>32));
}
-#elif (__GNUC__ > 4) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3)
+#elif (__GNUC__ > 4) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3) || __clang__
static LJ_AINLINE uint32_t lj_bswap(uint32_t x)
{
return (uint32_t)__builtin_bswap32((int32_t)x);
@@ -263,21 +257,38 @@
return _CountLeadingZeros(x) ^ 31;
}
#else
-unsigned char _BitScanForward(uint32_t *, unsigned long);
-unsigned char _BitScanReverse(uint32_t *, unsigned long);
+unsigned char _BitScanForward(unsigned long *, unsigned long);
+unsigned char _BitScanReverse(unsigned long *, unsigned long);
#pragma intrinsic(_BitScanForward)
#pragma intrinsic(_BitScanReverse)
static LJ_AINLINE uint32_t lj_ffs(uint32_t x)
{
- uint32_t r; _BitScanForward(&r, x); return r;
+ unsigned long r; _BitScanForward(&r, x); return (uint32_t)r;
}
static LJ_AINLINE uint32_t lj_fls(uint32_t x)
{
- uint32_t r; _BitScanReverse(&r, x); return r;
+ unsigned long r; _BitScanReverse(&r, x); return (uint32_t)r;
}
+
+#if defined(_M_X64) || defined(_M_ARM64)
+unsigned char _BitScanForward64(unsigned long *, uint64_t);
+unsigned char _BitScanReverse64(unsigned long *, uint64_t);
+#pragma intrinsic(_BitScanForward64)
+#pragma intrinsic(_BitScanReverse64)
+
+static LJ_AINLINE uint32_t lj_ffs64(uint64_t x)
+{
+ unsigned long r; _BitScanForward64(&r, x); return (uint32_t)r;
+}
+
+static LJ_AINLINE uint32_t lj_fls64(uint64_t x)
+{
+ unsigned long r; _BitScanReverse64(&r, x); return (uint32_t)r;
+}
#endif
+#endif
unsigned long _byteswap_ulong(unsigned long);
uint64_t _byteswap_uint64(uint64_t);
@@ -338,14 +349,28 @@
#define LJ_FUNCA_NORET LJ_FUNCA LJ_NORET
#define LJ_ASMF_NORET LJ_ASMF LJ_NORET
-/* Runtime assertions. */
-#ifdef lua_assert
-#define check_exp(c, e) (lua_assert(c), (e))
-#define api_check(l, e) lua_assert(e)
+/* Internal assertions. */
+#if defined(LUA_USE_ASSERT) || defined(LUA_USE_APICHECK)
+#define lj_assert_check(g, c, ...) \
+ ((c) ? (void)0 : \
+ (lj_assert_fail((g), __FILE__, __LINE__, __func__, __VA_ARGS__), 0))
+#define lj_checkapi(c, ...) lj_assert_check(G(L), (c), __VA_ARGS__)
#else
-#define lua_assert(c) ((void)0)
+#define lj_checkapi(c, ...) ((void)L)
+#endif
+
+#ifdef LUA_USE_ASSERT
+#define lj_assertG_(g, c, ...) lj_assert_check((g), (c), __VA_ARGS__)
+#define lj_assertG(c, ...) lj_assert_check(g, (c), __VA_ARGS__)
+#define lj_assertL(c, ...) lj_assert_check(G(L), (c), __VA_ARGS__)
+#define lj_assertX(c, ...) lj_assert_check(NULL, (c), __VA_ARGS__)
+#define check_exp(c, e) (lj_assertX((c), #c), (e))
+#else
+#define lj_assertG_(g, c, ...) ((void)0)
+#define lj_assertG(c, ...) ((void)g)
+#define lj_assertL(c, ...) ((void)L)
+#define lj_assertX(c, ...) ((void)0)
#define check_exp(c, e) (e)
-#define api_check luai_apicheck
#endif
/* Static assertions. */
@@ -359,4 +384,9 @@
extern void LJ_ASSERT_NAME(__LINE__)(int STATIC_ASSERTION_FAILED[(cond)?1:-1])
#endif
+/* PRNG state. Need this here, details in lj_prng.h. */
+typedef struct PRNGState {
+ uint64_t u[4];
+} PRNGState;
+
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Instruction dispatch handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_dispatch_c
@@ -68,6 +68,8 @@
/* The JIT engine is off by default. luaopen_jit() turns it on. */
disp[BC_FORL] = disp[BC_IFORL];
disp[BC_ITERL] = disp[BC_IITERL];
+ /* Workaround for stable v2.1 bytecode. TODO: Replace with BC_IITERN. */
+ disp[BC_ITERN] = &lj_vm_IITERN;
disp[BC_LOOP] = disp[BC_ILOOP];
disp[BC_FUNCF] = disp[BC_IFUNCF];
disp[BC_FUNCV] = disp[BC_IFUNCV];
@@ -118,7 +120,7 @@
mode |= (g->hookmask & LUA_MASKRET) ? DISPMODE_RET : 0;
if (oldmode != mode) { /* Mode changed? */
ASMFunction *disp = G2GG(g)->dispatch;
- ASMFunction f_forl, f_iterl, f_loop, f_funcf, f_funcv;
+ ASMFunction f_forl, f_iterl, f_itern, f_loop, f_funcf, f_funcv;
g->dispatchmode = mode;
/* Hotcount if JIT is on, but not while recording. */
@@ -125,6 +127,7 @@
if ((mode & (DISPMODE_JIT|DISPMODE_REC)) == DISPMODE_JIT) {
f_forl = makeasmfunc(lj_bc_ofs[BC_FORL]);
f_iterl = makeasmfunc(lj_bc_ofs[BC_ITERL]);
+ f_itern = makeasmfunc(lj_bc_ofs[BC_ITERN]);
f_loop = makeasmfunc(lj_bc_ofs[BC_LOOP]);
f_funcf = makeasmfunc(lj_bc_ofs[BC_FUNCF]);
f_funcv = makeasmfunc(lj_bc_ofs[BC_FUNCV]);
@@ -131,6 +134,7 @@
} else { /* Otherwise use the non-hotcounting instructions. */
f_forl = disp[GG_LEN_DDISP+BC_IFORL];
f_iterl = disp[GG_LEN_DDISP+BC_IITERL];
+ f_itern = &lj_vm_IITERN;
f_loop = disp[GG_LEN_DDISP+BC_ILOOP];
f_funcf = makeasmfunc(lj_bc_ofs[BC_IFUNCF]);
f_funcv = makeasmfunc(lj_bc_ofs[BC_IFUNCV]);
@@ -138,6 +142,7 @@
/* Init static counting instruction dispatch first (may be copied below). */
disp[GG_LEN_DDISP+BC_FORL] = f_forl;
disp[GG_LEN_DDISP+BC_ITERL] = f_iterl;
+ disp[GG_LEN_DDISP+BC_ITERN] = f_itern;
disp[GG_LEN_DDISP+BC_LOOP] = f_loop;
/* Set dynamic instruction dispatch. */
@@ -165,6 +170,7 @@
/* Otherwise set dynamic counting ins. */
disp[BC_FORL] = f_forl;
disp[BC_ITERL] = f_iterl;
+ disp[BC_ITERN] = f_itern;
disp[BC_LOOP] = f_loop;
/* Set dynamic return dispatch. */
if ((mode & DISPMODE_RET)) {
@@ -252,15 +258,8 @@
} else {
if (!(mode & LUAJIT_MODE_ON))
G2J(g)->flags &= ~(uint32_t)JIT_F_ON;
-#if LJ_TARGET_X86ORX64
- else if ((G2J(g)->flags & JIT_F_SSE2))
- G2J(g)->flags |= (uint32_t)JIT_F_ON;
else
- return 0; /* Don't turn on JIT compiler without SSE2 support. */
-#else
- else
G2J(g)->flags |= (uint32_t)JIT_F_ON;
-#endif
lj_dispatch_update(g);
}
break;
@@ -302,15 +301,15 @@
if (idx != 0) {
cTValue *tv = idx > 0 ? L->base + (idx-1) : L->top + idx;
if (tvislightud(tv))
- g->wrapf = (lua_CFunction)lightudV(tv);
+ g->wrapf = (lua_CFunction)lightudV(g, tv);
else
return 0; /* Failed. */
} else {
return 0; /* Failed. */
}
- g->bc_cfunc_ext = BCINS_AD(BC_FUNCCW, 0, 0);
+ setbc_op(&g->bc_cfunc_ext, BC_FUNCCW);
} else {
- g->bc_cfunc_ext = BCINS_AD(BC_FUNCC, 0, 0);
+ setbc_op(&g->bc_cfunc_ext, BC_FUNCC);
}
break;
default:
@@ -374,7 +373,7 @@
hook_enter(g);
#endif
hookf(L, &ar);
- lua_assert(hook_active(g));
+ lj_assertG(hook_active(g), "active hook flag removed");
setgcref(g->cur_L, obj2gco(L));
#if LJ_HASPROFILE && !LJ_PROFILE_SIGPROF
lj_profile_hook_leave(g);
@@ -422,7 +421,8 @@
#endif
J->L = L;
lj_trace_ins(J, pc-1); /* The interpreter bytecode PC is offset by 1. */
- lua_assert(L->top - L->base == delta);
+ lj_assertG(L->top - L->base == delta,
+ "unbalanced stack after tracing of instruction");
}
}
#endif
@@ -453,7 +453,7 @@
int numparams = pt->numparams;
int gotparams = (int)(L->top - L->base);
int need = pt->framesize;
- if ((pt->flags & PROTO_VARARG)) need += 1+gotparams;
+ if ((pt->flags & PROTO_VARARG)) need += 1+LJ_FR2+gotparams;
lj_state_checkstack(L, (MSize)need);
numparams -= gotparams;
return numparams >= 0 ? numparams : 0;
@@ -482,7 +482,8 @@
#endif
pc = (const BCIns *)((uintptr_t)pc & ~(uintptr_t)1);
lj_trace_hot(J, pc);
- lua_assert(L->top - L->base == delta);
+ lj_assertG(L->top - L->base == delta,
+ "unbalanced stack after hot call");
goto out;
} else if (J->state != LJ_TRACE_IDLE &&
!(g->hookmask & (HOOK_GC|HOOK_VMEVENT))) {
@@ -491,7 +492,8 @@
#endif
/* Record the FUNC* bytecodes, too. */
lj_trace_ins(J, pc-1); /* The interpreter bytecode PC is offset by 1. */
- lua_assert(L->top - L->base == delta);
+ lj_assertG(L->top - L->base == delta,
+ "unbalanced stack after hot instruction");
}
#endif
if ((g->hookmask & LUA_MASKCALL)) {
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_dispatch.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Instruction dispatch handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_DISPATCH_H
@@ -31,7 +31,7 @@
#define SFGOTDEF(_)
#endif
#if LJ_HASJIT
-#define JITGOTDEF(_) _(lj_trace_exit) _(lj_trace_hot)
+#define JITGOTDEF(_) _(lj_err_trace) _(lj_trace_exit) _(lj_trace_hot)
#else
#define JITGOTDEF(_)
#endif
@@ -89,6 +89,10 @@
typedef struct GG_State {
lua_State L; /* Main thread. */
global_State g; /* Global state. */
+#if LJ_TARGET_ARM && !LJ_TARGET_NX
+ /* Make g reachable via K12 encoded DISPATCH-relative addressing. */
+ uint8_t align1[(16-sizeof(global_State))&15];
+#endif
#if LJ_TARGET_MIPS
ASMFunction got[LJ_GOT__MAX]; /* Global offset table. */
#endif
@@ -95,7 +99,11 @@
#if LJ_HASJIT
jit_State J; /* JIT state. */
HotCount hotcount[HOTCOUNT_SIZE]; /* Hot counters. */
+#if LJ_TARGET_ARM && !LJ_TARGET_NX
+ /* Ditto for J. */
+ uint8_t align2[(16-sizeof(jit_State)-sizeof(HotCount)*HOTCOUNT_SIZE)&15];
#endif
+#endif
ASMFunction dispatch[GG_LEN_DISP]; /* Instruction dispatch tables. */
BCIns bcff[GG_NUM_ASMFF]; /* Bytecode for ASM fast functions. */
} GG_State;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** ARM instruction emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Constant encoding --------------------------------------------------- */
@@ -81,7 +81,8 @@
static void emit_lsox(ASMState *as, ARMIns ai, Reg rd, Reg rn, int32_t ofs)
{
- lua_assert(ofs >= -255 && ofs <= 255);
+ lj_assertA(ofs >= -255 && ofs <= 255,
+ "load/store offset %d out of range", ofs);
if (ofs < 0) ofs = -ofs; else ai |= ARMI_LS_U;
*--as->mcp = ai | ARMI_LS_P | ARMI_LSX_I | ARMF_D(rd) | ARMF_N(rn) |
((ofs & 0xf0) << 4) | (ofs & 0x0f);
@@ -89,7 +90,8 @@
static void emit_lso(ASMState *as, ARMIns ai, Reg rd, Reg rn, int32_t ofs)
{
- lua_assert(ofs >= -4095 && ofs <= 4095);
+ lj_assertA(ofs >= -4095 && ofs <= 4095,
+ "load/store offset %d out of range", ofs);
/* Combine LDR/STR pairs to LDRD/STRD. */
if (*as->mcp == (ai|ARMI_LS_P|ARMI_LS_U|ARMF_D(rd^1)|ARMF_N(rn)|(ofs^4)) &&
(ai & ~(ARMI_LDR^ARMI_STR)) == ARMI_STR && rd != rn &&
@@ -106,7 +108,8 @@
#if !LJ_SOFTFP
static void emit_vlso(ASMState *as, ARMIns ai, Reg rd, Reg rn, int32_t ofs)
{
- lua_assert(ofs >= -1020 && ofs <= 1020 && (ofs&3) == 0);
+ lj_assertA(ofs >= -1020 && ofs <= 1020 && (ofs&3) == 0,
+ "load/store offset %d out of range", ofs);
if (ofs < 0) ofs = -ofs; else ai |= ARMI_LS_U;
*--as->mcp = ai | ARMI_LS_P | ARMF_D(rd & 15) | ARMF_N(rn) | (ofs >> 2);
}
@@ -124,7 +127,7 @@
while (work) {
Reg r = rset_picktop(work);
IRRef ref = regcost_ref(as->cost[r]);
- lua_assert(r != d);
+ lj_assertA(r != d, "dest reg not free");
if (emit_canremat(ref)) {
int32_t delta = i - (ra_iskref(ref) ? ra_krefk(as, ref) : IR(ref)->i);
uint32_t k = emit_isk12(ARMI_ADD, delta);
@@ -142,25 +145,25 @@
}
/* Try to find a two step delta relative to another constant. */
-static int emit_kdelta2(ASMState *as, Reg d, int32_t i)
+static int emit_kdelta2(ASMState *as, Reg rd, int32_t i)
{
RegSet work = ~as->freeset & RSET_GPR;
while (work) {
Reg r = rset_picktop(work);
IRRef ref = regcost_ref(as->cost[r]);
- lua_assert(r != d);
+ lj_assertA(r != rd, "dest reg %d not free", rd);
if (emit_canremat(ref)) {
int32_t other = ra_iskref(ref) ? ra_krefk(as, ref) : IR(ref)->i;
if (other) {
int32_t delta = i - other;
uint32_t sh, inv = 0, k2, k;
- if (delta < 0) { delta = -delta; inv = ARMI_ADD^ARMI_SUB; }
+ if (delta < 0) { delta = (int32_t)(~(uint32_t)delta+1u); inv = ARMI_ADD^ARMI_SUB; }
sh = lj_ffs(delta) & ~1;
k2 = emit_isk12(0, delta & (255 << sh));
k = emit_isk12(0, delta & ~(255 << sh));
if (k) {
- emit_dn(as, ARMI_ADD^k2^inv, d, d);
- emit_dn(as, ARMI_ADD^k^inv, d, r);
+ emit_dn(as, ARMI_ADD^k2^inv, rd, rd);
+ emit_dn(as, ARMI_ADD^k^inv, rd, r);
return 1;
}
}
@@ -171,23 +174,24 @@
}
/* Load a 32 bit constant into a GPR. */
-static void emit_loadi(ASMState *as, Reg r, int32_t i)
+static void emit_loadi(ASMState *as, Reg rd, int32_t i)
{
uint32_t k = emit_isk12(ARMI_MOV, i);
- lua_assert(rset_test(as->freeset, r) || r == RID_TMP);
+ lj_assertA(rset_test(as->freeset, rd) || rd == RID_TMP,
+ "dest reg %d not free", rd);
if (k) {
/* Standard K12 constant. */
- emit_d(as, ARMI_MOV^k, r);
+ emit_d(as, ARMI_MOV^k, rd);
} else if ((as->flags & JIT_F_ARMV6T2) && (uint32_t)i < 0x00010000u) {
/* 16 bit loword constant for ARMv6T2. */
- emit_d(as, ARMI_MOVW|(i & 0x0fff)|((i & 0xf000)<<4), r);
- } else if (emit_kdelta1(as, r, i)) {
+ emit_d(as, ARMI_MOVW|(i & 0x0fff)|((i & 0xf000)<<4), rd);
+ } else if (emit_kdelta1(as, rd, i)) {
/* One step delta relative to another constant. */
} else if ((as->flags & JIT_F_ARMV6T2)) {
/* 32 bit hiword/loword constant for ARMv6T2. */
- emit_d(as, ARMI_MOVT|((i>>16) & 0x0fff)|(((i>>16) & 0xf000)<<4), r);
- emit_d(as, ARMI_MOVW|(i & 0x0fff)|((i & 0xf000)<<4), r);
- } else if (emit_kdelta2(as, r, i)) {
+ emit_d(as, ARMI_MOVT|((i>>16) & 0x0fff)|(((i>>16) & 0xf000)<<4), rd);
+ emit_d(as, ARMI_MOVW|(i & 0x0fff)|((i & 0xf000)<<4), rd);
+ } else if (emit_kdelta2(as, rd, i)) {
/* Two step delta relative to another constant. */
} else {
/* Otherwise construct the constant with up to 4 instructions. */
@@ -197,15 +201,15 @@
int32_t m = i & (255 << sh);
i &= ~(255 << sh);
if (i == 0) {
- emit_d(as, ARMI_MOV ^ emit_isk12(0, m), r);
+ emit_d(as, ARMI_MOV ^ emit_isk12(0, m), rd);
break;
}
- emit_dn(as, ARMI_ORR ^ emit_isk12(0, m), r, r);
+ emit_dn(as, ARMI_ORR ^ emit_isk12(0, m), rd, rd);
}
}
}
-#define emit_loada(as, r, addr) emit_loadi(as, (r), i32ptr((addr)))
+#define emit_loada(as, rd, addr) emit_loadi(as, (rd), i32ptr((addr)))
static Reg ra_allock(ASMState *as, intptr_t k, RegSet allow);
@@ -261,7 +265,7 @@
{
MCode *p = as->mcp;
ptrdiff_t delta = (target - p) - 1;
- lua_assert(((delta + 0x00800000) >> 24) == 0);
+ lj_assertA(((delta + 0x00800000) >> 24) == 0, "branch target out of range");
*--p = ai | ((uint32_t)delta & 0x00ffffffu);
as->mcp = p;
}
@@ -289,7 +293,7 @@
static void emit_movrr(ASMState *as, IRIns *ir, Reg dst, Reg src)
{
#if LJ_SOFTFP
- lua_assert(!irt_isnum(ir->t)); UNUSED(ir);
+ lj_assertA(!irt_isnum(ir->t), "unexpected FP op"); UNUSED(ir);
#else
if (dst >= RID_MAX_GPR) {
emit_dm(as, irt_isnum(ir->t) ? ARMI_VMOV_D : ARMI_VMOV_S,
@@ -313,7 +317,7 @@
static void emit_loadofs(ASMState *as, IRIns *ir, Reg r, Reg base, int32_t ofs)
{
#if LJ_SOFTFP
- lua_assert(!irt_isnum(ir->t)); UNUSED(ir);
+ lj_assertA(!irt_isnum(ir->t), "unexpected FP op"); UNUSED(ir);
#else
if (r >= RID_MAX_GPR)
emit_vlso(as, irt_isnum(ir->t) ? ARMI_VLDR_D : ARMI_VLDR_S, r, base, ofs);
@@ -326,7 +330,7 @@
static void emit_storeofs(ASMState *as, IRIns *ir, Reg r, Reg base, int32_t ofs)
{
#if LJ_SOFTFP
- lua_assert(!irt_isnum(ir->t)); UNUSED(ir);
+ lj_assertA(!irt_isnum(ir->t), "unexpected FP op"); UNUSED(ir);
#else
if (r >= RID_MAX_GPR)
emit_vlso(as, irt_isnum(ir->t) ? ARMI_VSTR_D : ARMI_VSTR_S, r, base, ofs);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm64.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm64.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_arm64.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** ARM64 instruction emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Contributed by Djordje Kovacevic and Stefan Pejic from RT-RK.com.
** Sponsored by Cisco Systems, Inc.
@@ -8,8 +8,9 @@
/* -- Constant encoding --------------------------------------------------- */
-static uint64_t get_k64val(IRIns *ir)
+static uint64_t get_k64val(ASMState *as, IRRef ref)
{
+ IRIns *ir = IR(ref);
if (ir->o == IR_KINT64) {
return ir_kint64(ir)->u64;
} else if (ir->o == IR_KGC) {
@@ -17,8 +18,9 @@
} else if (ir->o == IR_KPTR || ir->o == IR_KKPTR) {
return (uint64_t)ir_kptr(ir);
} else {
- lua_assert(ir->o == IR_KINT || ir->o == IR_KNULL);
- return ir->i; /* Sign-extended. */
+ lj_assertA(ir->o == IR_KINT || ir->o == IR_KNULL,
+ "bad 64 bit const IR op %d", ir->o);
+ return (uint32_t)ir->i; /* Zero-extended. */
}
}
@@ -25,42 +27,34 @@
/* Encode constant in K12 format for data processing instructions. */
static uint32_t emit_isk12(int64_t n)
{
- uint64_t k = (n < 0) ? -n : n;
- uint32_t m = (n < 0) ? 0x40000000 : 0;
+ uint64_t k = n < 0 ? ~(uint64_t)n+1u : (uint64_t)n;
+ uint32_t m = n < 0 ? 0x40000000 : 0;
if (k < 0x1000) {
- return A64I_K12|m|A64F_U12(k);
+ return (uint32_t)(A64I_K12|m|A64F_U12(k));
} else if ((k & 0xfff000) == k) {
- return A64I_K12|m|0x400000|A64F_U12(k>>12);
+ return (uint32_t)(A64I_K12|m|0x400000|A64F_U12(k>>12));
}
return 0;
}
-#define emit_clz64(n) __builtin_clzll(n)
-#define emit_ctz64(n) __builtin_ctzll(n)
+#define emit_clz64(n) (lj_fls64(n)^63)
+#define emit_ctz64(n) lj_ffs64(n)
/* Encode constant in K13 format for logical data processing instructions. */
static uint32_t emit_isk13(uint64_t n, int is64)
{
- int inv = 0, w = 128, lz, tz;
- if (n & 1) { n = ~n; w = 64; inv = 1; } /* Avoid wrap-around of ones. */
- if (!n) return 0; /* Neither all-zero nor all-ones are allowed. */
- do { /* Find the repeat width. */
- if (is64 && (uint32_t)(n^(n>>32))) break;
- n = (uint32_t)n;
- if (!n) return 0; /* Ditto when passing n=0xffffffff and is64=0. */
- w = 32; if ((n^(n>>16)) & 0xffff) break;
- n = n & 0xffff; w = 16; if ((n^(n>>8)) & 0xff) break;
- n = n & 0xff; w = 8; if ((n^(n>>4)) & 0xf) break;
- n = n & 0xf; w = 4; if ((n^(n>>2)) & 0x3) break;
- n = n & 0x3; w = 2;
- } while (0);
- lz = emit_clz64(n);
- tz = emit_ctz64(n);
- if ((int64_t)(n << lz) >> (lz+tz) != -1ll) return 0; /* Non-contiguous? */
- if (inv)
- return A64I_K13 | (((lz-w) & 127) << 16) | (((lz+tz-w-1) & 63) << 10);
- else
- return A64I_K13 | ((w-tz) << 16) | (((63-lz-tz-w-w) & 63) << 10);
+ /* Thanks to: https://dougallj.wordpress.com/2021/10/30/ */
+ int rot, ones, size, immr, imms;
+ if (!is64) n = ((uint64_t)n << 32) | (uint32_t)n;
+ if ((n+1u) <= 1u) return 0; /* Neither all-zero nor all-ones are allowed. */
+ rot = (n & (n+1u)) ? emit_ctz64(n & (n+1u)) : 64;
+ n = lj_ror(n, rot & 63);
+ ones = emit_ctz64(~n);
+ size = emit_clz64(n) + ones;
+ if (lj_ror(n, size & 63) != n) return 0; /* Non-repeating? */
+ immr = -rot & (size - 1);
+ imms = (-(size << 1) | (ones - 1)) & 63;
+ return A64I_K13 | A64F_IMMR(immr | (size & 64)) | A64F_IMMS(imms);
}
static uint32_t emit_isfpk64(uint64_t n)
@@ -72,6 +66,17 @@
return ~0u;
}
+static uint32_t emit_isfpmovi(uint64_t n)
+{
+ /* Is every byte either 0x00 or 0xff? */
+ if ((n & U64x(01010101,01010101)) * 0xff != n) return 0;
+ /* Form 8-bit value by taking one bit from each byte. */
+ n &= U64x(80402010,08040201);
+ n = (n * U64x(01010101,01010101)) >> 56;
+ /* Split into the format expected by movi. */
+ return ((n & 0xe0) << 6) | 0x700 | (n & 0x1f);
+}
+
/* -- Emit basic instructions --------------------------------------------- */
static void emit_dnma(ASMState *as, A64Ins ai, Reg rd, Reg rn, Reg rm, Reg ra)
@@ -104,6 +109,11 @@
*--as->mcp = ai | A64F_D(rd);
}
+static void emit_dl(ASMState *as, A64Ins ai, Reg rd, uint32_t l)
+{
+ *--as->mcp = ai | A64F_D(rd) | A64F_S19(l >> 2);
+}
+
static void emit_n(ASMState *as, A64Ins ai, Reg rn)
{
*--as->mcp = ai | A64F_N(rn);
@@ -119,10 +129,21 @@
}
}
-static void emit_lso(ASMState *as, A64Ins ai, Reg rd, Reg rn, int64_t ofs)
+static LJ_AINLINE uint32_t emit_lso_pair_candidate(A64Ins ai, int ofs, int sc)
{
- int ot = emit_checkofs(ai, ofs), sc = (ai >> 30) & 3;
- lua_assert(ot);
+ if (ofs >= 0) {
+ return ai | A64F_U12(ofs>>sc); /* Subsequent lj_ror checks ofs. */
+ } else if (ofs >= -256) {
+ return (ai^A64I_LS_U) | A64F_S9(ofs & 0x1ff);
+ } else {
+ return A64F_D(31); /* Will mismatch prev. */
+ }
+}
+
+static void emit_lso(ASMState *as, A64Ins ai, Reg rd, Reg rn, int64_t ofs64)
+{
+ int ot = emit_checkofs(ai, ofs64), sc = (ai >> 30) & 3, ofs = (int)ofs64;
+ lj_assertA(ot, "load/store offset %d out of range", ofs);
/* Combine LDR/STR pairs to LDP/STP. */
if ((sc == 2 || sc == 3) &&
(!(ai & 0x400000) || rd != rn) &&
@@ -130,18 +151,16 @@
uint32_t prev = *as->mcp & ~A64F_D(31);
int ofsm = ofs - (1<<sc), ofsp = ofs + (1<<sc);
A64Ins aip;
- if (prev == (ai | A64F_N(rn) | A64F_U12(ofsm>>sc)) ||
- prev == ((ai^A64I_LS_U) | A64F_N(rn) | A64F_S9(ofsm&0x1ff))) {
+ if (prev == emit_lso_pair_candidate(ai | A64F_N(rn), ofsm, sc)) {
aip = (A64F_A(rd) | A64F_D(*as->mcp & 31));
- } else if (prev == (ai | A64F_N(rn) | A64F_U12(ofsp>>sc)) ||
- prev == ((ai^A64I_LS_U) | A64F_N(rn) | A64F_S9(ofsp&0x1ff))) {
+ } else if (prev == emit_lso_pair_candidate(ai | A64F_N(rn), ofsp, sc)) {
aip = (A64F_D(rd) | A64F_A(*as->mcp & 31));
ofsm = ofs;
} else {
goto nopair;
}
- if (ofsm >= (int)((unsigned int)-64<<sc) && ofsm <= (63<<sc)) {
- *as->mcp = aip | A64F_N(rn) | ((ofsm >> sc) << 15) |
+ if (lj_ror((unsigned int)ofsm + (64u<<sc), sc) <= 127u) {
+ *as->mcp = aip | A64F_N(rn) | (((ofsm >> sc) & 0x7f) << 15) |
(ai ^ ((ai == A64I_LDRx || ai == A64I_STRx) ? 0x50000000 : 0x90000000));
return;
}
@@ -156,28 +175,28 @@
/* -- Emit loads/stores --------------------------------------------------- */
/* Prefer rematerialization of BASE/L from global_State over spills. */
-#define emit_canremat(ref) ((ref) <= ASMREF_L)
+#define emit_canremat(ref) ((ref) <= REF_BASE)
-/* Try to find an N-step delta relative to other consts with N < lim. */
-static int emit_kdelta(ASMState *as, Reg rd, uint64_t k, int lim)
+/* Try to find a one-step delta relative to other consts. */
+static int emit_kdelta(ASMState *as, Reg rd, uint64_t k, int is64)
{
- RegSet work = ~as->freeset & RSET_GPR;
- if (lim <= 1) return 0; /* Can't beat that. */
+ RegSet work = (~as->freeset & RSET_GPR) | RID2RSET(RID_GL);
while (work) {
Reg r = rset_picktop(work);
IRRef ref = regcost_ref(as->cost[r]);
- lua_assert(r != rd);
+ lj_assertA(r != rd, "dest reg %d not free", rd);
if (ref < REF_TRUE) {
uint64_t kx = ra_iskref(ref) ? (uint64_t)ra_krefk(as, ref) :
- get_k64val(IR(ref));
+ get_k64val(as, ref);
int64_t delta = (int64_t)(k - kx);
+ if (!is64) delta = (int64_t)(int32_t)delta; /* Sign-extend. */
if (delta == 0) {
- emit_dm(as, A64I_MOVx, rd, r);
+ emit_dm(as, is64|A64I_MOVw, rd, r);
return 1;
} else {
- uint32_t k12 = emit_isk12(delta < 0 ? -delta : delta);
+ uint32_t k12 = emit_isk12(delta < 0 ? (int64_t)(~(uint64_t)delta+1u) : delta);
if (k12) {
- emit_dn(as, (delta < 0 ? A64I_SUBx : A64I_ADDx)^k12, rd, r);
+ emit_dn(as, (delta < 0 ? A64I_SUBw : A64I_ADDw)^is64^k12, rd, r);
return 1;
}
/* Do other ops or multi-step deltas pay off? Probably not.
@@ -190,77 +209,101 @@
return 0; /* Failed. */
}
-static void emit_loadk(ASMState *as, Reg rd, uint64_t u64, int is64)
+#define glofs(as, k) \
+ ((intptr_t)((uintptr_t)(k) - (uintptr_t)&J2GG(as->J)->g))
+#define mcpofs(as, k) \
+ ((intptr_t)((uintptr_t)(k) - (uintptr_t)(as->mcp - 1)))
+#define checkmcpofs(as, k) \
+ (A64F_S_OK(mcpofs(as, k)>>2, 19))
+
+/* Try to form a const as ADR or ADRP or ADRP + ADD. */
+static int emit_kadrp(ASMState *as, Reg rd, uint64_t k)
{
- uint32_t k13 = emit_isk13(u64, is64);
- if (k13) { /* Can the constant be represented as a bitmask immediate? */
- emit_dn(as, (is64|A64I_ORRw)^k13, rd, RID_ZERO);
- } else {
- int i, zeros = 0, ones = 0, neg;
- if (!is64) u64 = (int64_t)(int32_t)u64; /* Sign-extend. */
- /* Count homogeneous 16 bit fragments. */
- for (i = 0; i < 4; i++) {
- uint64_t frag = (u64 >> i*16) & 0xffff;
- zeros += (frag == 0);
- ones += (frag == 0xffff);
+ A64Ins ai = A64I_ADR;
+ int64_t ofs = mcpofs(as, k);
+ if (!A64F_S_OK((uint64_t)ofs, 21)) {
+ uint64_t kpage = k & ~0xfffull;
+ MCode *adrp = as->mcp - 1 - (k != kpage);
+ ofs = (int64_t)(kpage - ((uint64_t)adrp & ~0xfffull)) >> 12;
+ if (!A64F_S_OK(ofs, 21))
+ return 0; /* Failed. */
+ if (k != kpage)
+ emit_dn(as, (A64I_ADDx^A64I_K12)|A64F_U12(k - kpage), rd, rd);
+ ai = A64I_ADRP;
+ }
+ emit_dl(as, ai|(((uint32_t)ofs&3)<<29), rd, ofs);
+ return 1;
+}
+
+static void emit_loadk(ASMState *as, Reg rd, uint64_t u64)
+{
+ int zeros = 0, ones = 0, neg, lshift = 0;
+ int is64 = (u64 >> 32) ? A64I_X : 0, i = is64 ? 4 : 2;
+ /* Count non-homogeneous 16 bit fragments. */
+ while (--i >= 0) {
+ uint32_t frag = (u64 >> i*16) & 0xffff;
+ zeros += (frag != 0);
+ ones += (frag != 0xffff);
+ }
+ neg = ones < zeros; /* Use MOVN if it pays off. */
+ if ((neg ? ones : zeros) > 1) { /* Need 2+ ins. Try 1 ins encodings. */
+ uint32_t k13 = emit_isk13(u64, is64);
+ if (k13) {
+ emit_dn(as, (is64|A64I_ORRw)^k13, rd, RID_ZERO);
+ return;
}
- neg = ones > zeros; /* Use MOVN if it pays off. */
- if (!emit_kdelta(as, rd, u64, 4 - (neg ? ones : zeros))) {
- int shift = 0, lshift = 0;
- uint64_t n64 = neg ? ~u64 : u64;
- if (n64 != 0) {
- /* Find first/last fragment to be filled. */
- shift = (63-emit_clz64(n64)) & ~15;
- lshift = emit_ctz64(n64) & ~15;
- }
- /* MOVK requires the original value (u64). */
- while (shift > lshift) {
- uint32_t u16 = (u64 >> shift) & 0xffff;
- /* Skip fragments that are correctly filled by MOVN/MOVZ. */
- if (u16 != (neg ? 0xffff : 0))
- emit_d(as, is64 | A64I_MOVKw | A64F_U16(u16) | A64F_LSL16(shift), rd);
- shift -= 16;
- }
- /* But MOVN needs an inverted value (n64). */
- emit_d(as, (neg ? A64I_MOVNx : A64I_MOVZx) |
- A64F_U16((n64 >> lshift) & 0xffff) | A64F_LSL16(lshift), rd);
+ if (emit_kdelta(as, rd, u64, is64)) {
+ return;
}
+ if (emit_kadrp(as, rd, u64)) { /* Either 1 or 2 ins. */
+ return;
+ }
}
+ if (neg) {
+ u64 = ~u64;
+ if (!is64) u64 = (uint32_t)u64;
+ }
+ if (u64) {
+ /* Find first/last fragment to be filled. */
+ int shift = (63-emit_clz64(u64)) & ~15;
+ lshift = emit_ctz64(u64) & ~15;
+ for (; shift > lshift; shift -= 16) {
+ uint32_t frag = (u64 >> shift) & 0xffff;
+ if (frag == 0) continue; /* Will be correctly filled by MOVN/MOVZ. */
+ if (neg) frag ^= 0xffff; /* MOVK requires the original value. */
+ emit_d(as, is64 | A64I_MOVKw | A64F_U16(frag) | A64F_LSL16(shift), rd);
+ }
+ }
+ /* But MOVN needs an inverted value. */
+ emit_d(as, is64 | (neg ? A64I_MOVNw : A64I_MOVZw) |
+ A64F_U16((u64 >> lshift) & 0xffff) | A64F_LSL16(lshift), rd);
}
/* Load a 32 bit constant into a GPR. */
-#define emit_loadi(as, rd, i) emit_loadk(as, rd, i, 0)
+#define emit_loadi(as, rd, i) emit_loadk(as, rd, (uint32_t)i)
/* Load a 64 bit constant into a GPR. */
-#define emit_loadu64(as, rd, i) emit_loadk(as, rd, i, A64I_X)
+#define emit_loadu64(as, rd, i) emit_loadk(as, rd, i)
-#define emit_loada(as, r, addr) emit_loadu64(as, (r), (uintptr_t)(addr))
-
-#define glofs(as, k) \
- ((intptr_t)((uintptr_t)(k) - (uintptr_t)&J2GG(as->J)->g))
-#define mcpofs(as, k) \
- ((intptr_t)((uintptr_t)(k) - (uintptr_t)(as->mcp - 1)))
-#define checkmcpofs(as, k) \
- ((((mcpofs(as, k)>>2) + 0x00040000) >> 19) == 0)
-
static Reg ra_allock(ASMState *as, intptr_t k, RegSet allow);
/* Get/set from constant pointer. */
static void emit_lsptr(ASMState *as, A64Ins ai, Reg r, void *p)
{
- /* First, check if ip + offset is in range. */
- if ((ai & 0x00400000) && checkmcpofs(as, p)) {
- emit_d(as, A64I_LDRLx | A64F_S19(mcpofs(as, p)>>2), r);
- } else {
- Reg base = RID_GL; /* Next, try GL + offset. */
- int64_t ofs = glofs(as, p);
- if (!emit_checkofs(ai, ofs)) { /* Else split up into base reg + offset. */
- int64_t i64 = i64ptr(p);
- base = ra_allock(as, (i64 & ~0x7fffull), rset_exclude(RSET_GPR, r));
- ofs = i64 & 0x7fffull;
- }
- emit_lso(as, ai, r, base, ofs);
+ Reg base = RID_GL;
+ int64_t ofs = glofs(as, p);
+ if (emit_checkofs(ai, ofs)) {
+ /* GL + offset, might subsequently fuse to LDP/STP. */
+ } else if (ai == A64I_LDRx && checkmcpofs(as, p)) {
+ /* IP + offset is cheaper than allock, but address must be in range. */
+ emit_dl(as, A64I_LDRLx, r, mcpofs(as, p));
+ return;
+ } else { /* Split up into base reg + offset. */
+ int64_t i64 = i64ptr(p);
+ base = ra_allock(as, (i64 & ~0x7fffull), rset_exclude(RSET_GPR, r));
+ ofs = i64 & 0x7fffull;
}
+ emit_lso(as, ai, r, base, ofs);
}
/* Load 64 bit IR constant into register. */
@@ -273,6 +316,9 @@
if (fpk != ~0u) {
emit_d(as, A64I_FMOV_DI | A64F_FP8(fpk), (r & 31));
return;
+ } else if ((fpk = emit_isfpmovi(*k))) {
+ emit_d(as, A64I_MOVI_DI | (fpk << 5), (r & 31));
+ return;
}
}
ofs = glofs(as, k);
@@ -279,15 +325,15 @@
if (emit_checkofs(A64I_LDRx, ofs)) {
emit_lso(as, r >= RID_MAX_GPR ? A64I_LDRd : A64I_LDRx,
(r & 31), RID_GL, ofs);
+ } else if (checkmcpofs(as, k)) {
+ emit_dl(as, r >= RID_MAX_GPR ? A64I_LDRLd : A64I_LDRLx,
+ (r & 31), mcpofs(as, k));
} else {
if (r >= RID_MAX_GPR) {
emit_dn(as, A64I_FMOV_D_R, (r & 31), RID_TMP);
r = RID_TMP;
}
- if (checkmcpofs(as, k))
- emit_d(as, A64I_LDRLx | A64F_S19(mcpofs(as, k)>>2), r);
- else
- emit_loadu64(as, r, *k);
+ emit_loadu64(as, r, *k);
}
}
@@ -312,7 +358,7 @@
{
MCode *p = --as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(((delta + 0x40000) >> 19) == 0);
+ lj_assertA(A64F_S_OK(delta, 19), "branch target out of range");
*p = A64I_BCC | A64F_S19(delta) | cond;
}
@@ -320,8 +366,8 @@
{
MCode *p = --as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(((delta + 0x02000000) >> 26) == 0);
- *p = ai | ((uint32_t)delta & 0x03ffffffu);
+ lj_assertA(A64F_S_OK(delta, 26), "branch target out of range");
+ *p = ai | A64F_S26(delta);
}
static void emit_tnb(ASMState *as, A64Ins ai, Reg r, uint32_t bit, MCode *target)
@@ -328,9 +374,10 @@
{
MCode *p = --as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(bit < 63 && ((delta + 0x2000) >> 14) == 0);
+ lj_assertA(bit < 63, "bit number out of range");
+ lj_assertA(A64F_S_OK(delta, 14), "branch target out of range");
if (bit > 31) ai |= A64I_X;
- *p = ai | A64F_BIT(bit & 31) | A64F_S14((uint32_t)delta & 0x3fffu) | r;
+ *p = ai | A64F_BIT(bit & 31) | A64F_S14(delta) | r;
}
static void emit_cnb(ASMState *as, A64Ins ai, Reg r, MCode *target)
@@ -337,22 +384,28 @@
{
MCode *p = --as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(((delta + 0x40000) >> 19) == 0);
+ lj_assertA(A64F_S_OK(delta, 19), "branch target out of range");
*p = ai | A64F_S19(delta) | r;
}
#define emit_jmp(as, target) emit_branch(as, A64I_B, (target))
-static void emit_call(ASMState *as, void *target)
+static void emit_call(ASMState *as, ASMFunction target)
{
MCode *p = --as->mcp;
- ptrdiff_t delta = (char *)target - (char *)p;
- if ((((delta>>2) + 0x02000000) >> 26) == 0) {
- *p = A64I_BL | ((uint32_t)(delta>>2) & 0x03ffffffu);
+#if LJ_ABI_PAUTH
+ char *targetp = ptrauth_auth_data((char *)target,
+ ptrauth_key_function_pointer, 0);
+#else
+ char *targetp = (char *)target;
+#endif
+ ptrdiff_t delta = targetp - (char *)p;
+ if (A64F_S_OK(delta>>2, 26)) {
+ *p = A64I_BL | A64F_S26(delta>>2);
} else { /* Target out of range: need indirect call. But don't use R0-R7. */
Reg r = ra_allock(as, i64ptr(target),
RSET_RANGE(RID_X8, RID_MAX_GPR)-RSET_FIXED);
- *p = A64I_BLR | A64F_N(r);
+ *p = A64I_BLR_AUTH | A64F_N(r);
}
}
@@ -412,7 +465,8 @@
{
if (ofs)
emit_opk(as, ofs < 0 ? A64I_SUBx : A64I_ADDx, r, r,
- ofs < 0 ? -ofs : ofs, rset_exclude(RSET_GPR, r));
+ ofs < 0 ? (int32_t)(~(uint32_t)ofs+1u) : ofs,
+ rset_exclude(RSET_GPR, r));
}
#define emit_spsub(as, ofs) emit_addptr(as, RID_SP, -(ofs))
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_mips.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_mips.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_mips.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,11 +1,12 @@
/*
** MIPS instruction emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#if LJ_64
-static intptr_t get_k64val(IRIns *ir)
+static intptr_t get_k64val(ASMState *as, IRRef ref)
{
+ IRIns *ir = IR(ref);
if (ir->o == IR_KINT64) {
return (intptr_t)ir_kint64(ir)->u64;
} else if (ir->o == IR_KGC) {
@@ -12,8 +13,11 @@
return (intptr_t)ir_kgc(ir);
} else if (ir->o == IR_KPTR || ir->o == IR_KKPTR) {
return (intptr_t)ir_kptr(ir);
+ } else if (LJ_SOFTFP && ir->o == IR_KNUM) {
+ return (intptr_t)ir_knum(ir)->u64;
} else {
- lua_assert(ir->o == IR_KINT || ir->o == IR_KNULL);
+ lj_assertA(ir->o == IR_KINT || ir->o == IR_KNULL,
+ "bad 64 bit const IR op %d", ir->o);
return ir->i; /* Sign-extended. */
}
}
@@ -20,9 +24,9 @@
#endif
#if LJ_64
-#define get_kval(ir) get_k64val(ir)
+#define get_kval(as, ref) get_k64val(as, ref)
#else
-#define get_kval(ir) ((ir)->i)
+#define get_kval(as, ref) (IR((ref))->i)
#endif
/* -- Emit basic instructions --------------------------------------------- */
@@ -66,7 +70,7 @@
}
}
-#if LJ_64
+#if LJ_64 || LJ_HASBUFFER
static void emit_tsml(ASMState *as, MIPSIns mi, Reg rt, Reg rs, uint32_t msb,
uint32_t lsb)
{
@@ -80,18 +84,18 @@
#define emit_canremat(ref) ((ref) <= REF_BASE)
/* Try to find a one step delta relative to another constant. */
-static int emit_kdelta1(ASMState *as, Reg t, intptr_t i)
+static int emit_kdelta1(ASMState *as, Reg rd, intptr_t i)
{
RegSet work = ~as->freeset & RSET_GPR;
while (work) {
Reg r = rset_picktop(work);
IRRef ref = regcost_ref(as->cost[r]);
- lua_assert(r != t);
+ lj_assertA(r != rd, "dest reg %d not free", rd);
if (ref < ASMREF_L) {
intptr_t delta = (intptr_t)((uintptr_t)i -
- (uintptr_t)(ra_iskref(ref) ? ra_krefk(as, ref) : get_kval(IR(ref))));
+ (uintptr_t)(ra_iskref(ref) ? ra_krefk(as, ref) : get_kval(as, ref)));
if (checki16(delta)) {
- emit_tsi(as, MIPSI_AADDIU, t, r, delta);
+ emit_tsi(as, MIPSI_AADDIU, rd, r, delta);
return 1;
}
}
@@ -136,6 +140,7 @@
} else if (emit_kdelta1(as, r, (intptr_t)u64)) {
return;
} else {
+ /* TODO MIPSR6: Use DAHI & DATI. Caveat: sign-extension. */
if ((u64 & 0xffff)) {
emit_tsi(as, MIPSI_ORI, r, r, u64 & 0xffff);
}
@@ -220,7 +225,7 @@
{
MCode *p = as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(((delta + 0x8000) >> 16) == 0);
+ lj_assertA(((delta + 0x8000) >> 16) == 0, "branch target out of range");
*--p = mi | MIPSF_S(rs) | MIPSF_T(rt) | ((uint32_t)delta & 0xffffu);
as->mcp = p;
}
@@ -234,10 +239,22 @@
static void emit_call(ASMState *as, void *target, int needcfa)
{
MCode *p = as->mcp;
- *--p = MIPSI_NOP;
+#if LJ_TARGET_MIPSR6
+ ptrdiff_t delta = (char *)target - (char *)p;
+ if ((((delta>>2) + 0x02000000) >> 26) == 0) { /* Try compact call first. */
+ *--p = MIPSI_BALC | (((uintptr_t)delta >>2) & 0x03ffffffu);
+ as->mcp = p;
+ return;
+ }
+#endif
+ *--p = MIPSI_NOP; /* Delay slot. */
if ((((uintptr_t)target ^ (uintptr_t)p) >> 28) == 0) {
+#if !LJ_TARGET_MIPSR6
*--p = (((uintptr_t)target & 1) ? MIPSI_JALX : MIPSI_JAL) |
(((uintptr_t)target >>2) & 0x03ffffffu);
+#else
+ *--p = MIPSI_JAL | (((uintptr_t)target >>2) & 0x03ffffffu);
+#endif
} else { /* Target out of range: need indirect call. */
*--p = MIPSI_JALR | MIPSF_S(RID_CFUNCADDR);
needcfa = 1;
@@ -284,7 +301,7 @@
static void emit_addptr(ASMState *as, Reg r, int32_t ofs)
{
if (ofs) {
- lua_assert(checki16(ofs));
+ lj_assertA(checki16(ofs), "offset %d out of range", ofs);
emit_tsi(as, MIPSI_AADDIU, r, r, ofs);
}
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_ppc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_ppc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_ppc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** PPC instruction emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Emit basic instructions --------------------------------------------- */
@@ -41,13 +41,13 @@
static void emit_slwi(ASMState *as, Reg ra, Reg rs, int32_t n)
{
- lua_assert(n >= 0 && n < 32);
+ lj_assertA(n >= 0 && n < 32, "shift out or range");
emit_rot(as, PPCI_RLWINM, ra, rs, n, 0, 31-n);
}
static void emit_rotlwi(ASMState *as, Reg ra, Reg rs, int32_t n)
{
- lua_assert(n >= 0 && n < 32);
+ lj_assertA(n >= 0 && n < 32, "shift out or range");
emit_rot(as, PPCI_RLWINM, ra, rs, n, 0, 31);
}
@@ -57,17 +57,17 @@
#define emit_canremat(ref) ((ref) <= REF_BASE)
/* Try to find a one step delta relative to another constant. */
-static int emit_kdelta1(ASMState *as, Reg t, int32_t i)
+static int emit_kdelta1(ASMState *as, Reg rd, int32_t i)
{
RegSet work = ~as->freeset & RSET_GPR;
while (work) {
Reg r = rset_picktop(work);
IRRef ref = regcost_ref(as->cost[r]);
- lua_assert(r != t);
+ lj_assertA(r != rd, "dest reg %d not free", rd);
if (ref < ASMREF_L) {
int32_t delta = i - (ra_iskref(ref) ? ra_krefk(as, ref) : IR(ref)->i);
if (checki16(delta)) {
- emit_tai(as, PPCI_ADDI, t, r, delta);
+ emit_tai(as, PPCI_ADDI, rd, r, delta);
return 1;
}
}
@@ -144,7 +144,7 @@
{
MCode *p = --as->mcp;
ptrdiff_t delta = (char *)target - (char *)p;
- lua_assert(((delta + 0x8000) >> 16) == 0);
+ lj_assertA(((delta + 0x8000) >> 16) == 0, "branch target out of range");
pi ^= (delta & 0x8000) * (PPCF_Y/0x8000);
*p = pi | PPCF_CC(cc) | ((uint32_t)delta & 0xffffu);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_x86.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_x86.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_emit_x86.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** x86/x64 instruction emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* -- Emit basic instructions --------------------------------------------- */
@@ -45,7 +45,7 @@
*(uint32_t *)(p+delta-5) = (uint32_t)xo;
return p+delta-5;
}
-#if defined(__GNUC__)
+#if defined(__GNUC__) || defined(__clang__)
if (__builtin_constant_p(xo) && n == -2)
p[delta-2] = (MCode)(xo >> 24);
else if (__builtin_constant_p(xo) && n == -3)
@@ -92,7 +92,7 @@
/* [addr] is sign-extended in x64 and must be in lower 2G (not 4G). */
static int32_t ptr2addr(const void *p)
{
- lua_assert((uintptr_t)p < (uintptr_t)0x80000000);
+ lj_assertX((uintptr_t)p < (uintptr_t)0x80000000, "pointer outside 2G range");
return i32ptr(p);
}
#else
@@ -208,7 +208,7 @@
rb = RID_ESP;
#endif
} else if (LJ_GC64 && rb == RID_RIP) {
- lua_assert(as->mrm.idx == RID_NONE);
+ lj_assertA(as->mrm.idx == RID_NONE, "RIP-rel mrm cannot have index");
mode = XM_OFS0;
p -= 4;
*(int32_t *)p = as->mrm.ofs;
@@ -274,10 +274,12 @@
/* mov r, i / xor r, r */
static void emit_loadi(ASMState *as, Reg r, int32_t i)
{
- /* XOR r,r is shorter, but modifies the flags. This is bad for HIOP. */
+ /* XOR r,r is shorter, but modifies the flags. This is bad for HIOP/jcc. */
if (i == 0 && !(LJ_32 && (IR(as->curins)->o == IR_HIOP ||
(as->curins+1 < as->T->nins &&
- IR(as->curins+1)->o == IR_HIOP)))) {
+ IR(as->curins+1)->o == IR_HIOP))) &&
+ !((*as->mcp == 0x0f && (as->mcp[1] & 0xf0) == XI_JCCn) ||
+ (*as->mcp & 0xf0) == XI_JCCs)) {
emit_rr(as, XO_ARITH(XOg_XOR), r, r);
} else {
MCode *p = as->mcp;
@@ -343,9 +345,27 @@
emit_rmro(as, xo, rr, RID_DISPATCH, (int32_t)dispofs(as, addr));
} else if (checki32(mcpofs(as, addr)) && checki32(mctopofs(as, addr))) {
emit_rmro(as, xo, rr, RID_RIP, (int32_t)mcpofs(as, addr));
- } else if (!checki32((intptr_t)addr) && (xo == XO_MOV || xo == XO_MOVSD)) {
- emit_rmro(as, xo, rr, rr, 0);
- emit_loadu64(as, rr, (uintptr_t)addr);
+ } else if (!checki32((intptr_t)addr)) {
+ Reg ra = (rr & 15);
+ if (xo != XO_MOV) {
+ /* We can't allocate a register here. Use and restore DISPATCH. Ugly. */
+ uint64_t dispaddr = (uintptr_t)J2GG(as->J)->dispatch;
+ uint8_t i8 = xo == XO_GROUP3b ? *as->mcp++ : 0;
+ ra = RID_DISPATCH;
+ if (checku32(dispaddr)) {
+ emit_loadi(as, ra, (int32_t)dispaddr);
+ } else { /* Full-size 64 bit load. */
+ MCode *p = as->mcp;
+ *(uint64_t *)(p-8) = dispaddr;
+ p[-9] = (MCode)(XI_MOVri+(ra&7));
+ p[-10] = 0x48 + ((ra>>3)&1);
+ p -= 10;
+ as->mcp = p;
+ }
+ if (xo == XO_GROUP3b) emit_i8(as, i8);
+ }
+ emit_rmro(as, xo, rr, ra, 0);
+ emit_loadu64(as, ra, (uintptr_t)addr);
} else
#endif
{
@@ -381,7 +401,8 @@
emit_rma(as, xo, r64, k);
} else {
if (ir->i) {
- lua_assert(*k == *(uint64_t*)(as->mctop - ir->i));
+ lj_assertA(*k == *(uint64_t*)(as->mctop - ir->i),
+ "bad interned 64 bit constant");
} else if (as->curins <= as->stopins && rset_test(RSET_GPR, r)) {
emit_loadu64(as, r, *k);
return;
@@ -413,7 +434,7 @@
{
MCode *p = as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(delta == (int8_t)delta);
+ lj_assertA(delta == (int8_t)delta, "short jump target out of range");
p[-1] = (MCode)(int8_t)delta;
p[-2] = XI_JMPs;
as->mcp = p - 2;
@@ -425,7 +446,7 @@
{
MCode *p = as->mcp;
ptrdiff_t delta = target - p;
- lua_assert(delta == (int8_t)delta);
+ lj_assertA(delta == (int8_t)delta, "short jump target out of range");
p[-1] = (MCode)(int8_t)delta;
p[-2] = (MCode)(XI_JCCs+(cc&15));
as->mcp = p - 2;
@@ -451,10 +472,11 @@
#define emit_label(as) ((as)->mcp)
/* Compute relative 32 bit offset for jump and call instructions. */
-static LJ_AINLINE int32_t jmprel(MCode *p, MCode *target)
+static LJ_AINLINE int32_t jmprel(jit_State *J, MCode *p, MCode *target)
{
ptrdiff_t delta = target - p;
- lua_assert(delta == (int32_t)delta);
+ UNUSED(J);
+ lj_assertJ(delta == (int32_t)delta, "jump target out of range");
return (int32_t)delta;
}
@@ -462,7 +484,7 @@
static void emit_jcc(ASMState *as, int cc, MCode *target)
{
MCode *p = as->mcp;
- *(int32_t *)(p-4) = jmprel(p, target);
+ *(int32_t *)(p-4) = jmprel(as->J, p, target);
p[-5] = (MCode)(XI_JCCn+(cc&15));
p[-6] = 0x0f;
as->mcp = p - 6;
@@ -472,7 +494,7 @@
static void emit_jmp(ASMState *as, MCode *target)
{
MCode *p = as->mcp;
- *(int32_t *)(p-4) = jmprel(p, target);
+ *(int32_t *)(p-4) = jmprel(as->J, p, target);
p[-5] = XI_JMP;
as->mcp = p - 5;
}
@@ -489,7 +511,7 @@
return;
}
#endif
- *(int32_t *)(p-4) = jmprel(p, target);
+ *(int32_t *)(p-4) = jmprel(as->J, p, target);
p[-5] = XI_CALL;
as->mcp = p - 5;
}
@@ -539,10 +561,7 @@
static void emit_addptr(ASMState *as, Reg r, int32_t ofs)
{
if (ofs) {
- if ((as->flags & JIT_F_LEA_AGU))
- emit_rmro(as, XO_LEA, r|REX_GC64, r, ofs);
- else
- emit_gri(as, XG_ARITHi(XOg_ADD), r|REX_GC64, ofs);
+ emit_gri(as, XG_ARITHi(XOg_ADD), r|REX_GC64, ofs);
}
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Error handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_err_c
@@ -29,12 +29,18 @@
** Pros and Cons:
**
** - EXT requires unwind tables for *all* functions on the C stack between
-** the pcall/catch and the error/throw. This is the default on x64,
-** but needs to be manually enabled on x86/PPC for non-C++ code.
+** the pcall/catch and the error/throw. C modules used by Lua code can
+** throw errors, so these need to have unwind tables, too. Transitively
+** this applies to all system libraries used by C modules -- at least
+** when they have callbacks which may throw an error.
**
-** - INT is faster when actually throwing errors (but this happens rarely).
+** - INT is faster when actually throwing errors, but this happens rarely.
** Setting up error handlers is zero-cost in any case.
**
+** - INT needs to save *all* callee-saved registers when entering the
+** interpreter. EXT only needs to save those actually used inside the
+** interpreter. JIT-compiled code may need to save some more.
+**
** - EXT provides full interoperability with C++ exceptions. You can throw
** Lua errors or C++ exceptions through a mix of Lua frames and C++ frames.
** C++ destructors are called as needed. C++ exceptions caught by pcall
@@ -46,27 +52,38 @@
** the wrapper function feature. Lua errors thrown through C++ frames
** cannot be caught by C++ code and C++ destructors are not run.
**
-** EXT is the default on x64 systems and on Windows, INT is the default on all
-** other systems.
+** - EXT can handle errors from internal helper functions that are called
+** from JIT-compiled code (except for Windows/x86 and 32 bit ARM).
+** INT has no choice but to call the panic handler, if this happens.
+** Note: this is mainly relevant for out-of-memory errors.
**
-** EXT can be manually enabled on POSIX systems using GCC and DWARF2 stack
-** unwinding with -DLUAJIT_UNWIND_EXTERNAL. *All* C code must be compiled
-** with -funwind-tables (or -fexceptions). This includes LuaJIT itself (set
-** TARGET_CFLAGS), all of your C/Lua binding code, all loadable C modules
-** and all C libraries that have callbacks which may be used to call back
-** into Lua. C++ code must *not* be compiled with -fno-exceptions.
+** EXT is the default on all systems where the toolchain produces unwind
+** tables by default (*). This is hard-coded and/or detected in src/Makefile.
+** You can thwart the detection with: TARGET_XCFLAGS=-DLUAJIT_UNWIND_INTERNAL
**
-** EXT is mandatory on WIN64 since the calling convention has an abundance
-** of callee-saved registers (rbx, rbp, rsi, rdi, r12-r15, xmm6-xmm15).
-** The POSIX/x64 interpreter only saves r12/r13 for INT (e.g. PS4).
+** INT is the default on all other systems.
+**
+** EXT can be manually enabled for toolchains that are able to produce
+** conforming unwind tables:
+** "TARGET_XCFLAGS=-funwind-tables -DLUAJIT_UNWIND_EXTERNAL"
+** As explained above, *all* C code used directly or indirectly by LuaJIT
+** must be compiled with -funwind-tables (or -fexceptions). C++ code must
+** *not* be compiled with -fno-exceptions.
+**
+** If you're unsure whether error handling inside the VM works correctly,
+** try running this and check whether it prints "OK":
+**
+** luajit -e "print(select(2, load('OK')):match('OK'))"
+**
+** (*) Originally, toolchains only generated unwind tables for C++ code. For
+** interoperability reasons, this can be manually enabled for plain C code,
+** too (with -funwind-tables). With the introduction of the x64 architecture,
+** the corresponding POSIX and Windows ABIs mandated unwind tables for all
+** code. Over the following years most desktop and server platforms have
+** enabled unwind tables by default on all architectures. OTOH mobile and
+** embedded platforms do not consistently mandate unwind tables.
*/
-#if defined(__GNUC__) && (LJ_TARGET_X64 || defined(LUAJIT_UNWIND_EXTERNAL)) && !LJ_NO_UNWIND
-#define LJ_UNWIND_EXT 1
-#elif LJ_TARGET_WINDOWS
-#define LJ_UNWIND_EXT 1
-#endif
-
/* -- Error messages ------------------------------------------------------ */
/* Error message strings. */
@@ -150,6 +167,7 @@
case FRAME_CONT: /* Continuation frame. */
if (frame_iscont_fficb(frame))
goto unwind_c;
+ /* fallthrough */
case FRAME_VARG: /* Vararg frame. */
frame = frame_prevd(frame);
break;
@@ -156,12 +174,15 @@
case FRAME_PCALL: /* FF pcall() frame. */
case FRAME_PCALLH: /* FF pcall() frame inside hook. */
if (errcode) {
+ global_State *g;
if (errcode == LUA_YIELD) {
frame = frame_prevd(frame);
break;
}
+ g = G(L);
+ setgcref(g->cur_L, obj2gco(L));
if (frame_typep(frame) == FRAME_PCALL)
- hook_leave(G(L));
+ hook_leave(g);
L->base = frame_prevd(frame) + 1;
L->cframe = cf;
unwindstack(L, L->base);
@@ -183,9 +204,194 @@
/* -- External frame unwinding -------------------------------------------- */
-#if defined(__GNUC__) && !LJ_NO_UNWIND && !LJ_ABI_WIN
+#if LJ_ABI_WIN
/*
+** Someone in Redmond owes me several days of my life. A lot of this is
+** undocumented or just plain wrong on MSDN. Some of it can be gathered
+** from 3rd party docs or must be found by trial-and-error. They really
+** don't want you to write your own language-specific exception handler
+** or to interact gracefully with MSVC. :-(
+*/
+
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+
+#if LJ_TARGET_X86
+typedef void *UndocumentedDispatcherContext; /* Unused on x86. */
+#else
+/* Taken from: http://www.nynaeve.net/?p=99 */
+typedef struct UndocumentedDispatcherContext {
+ ULONG64 ControlPc;
+ ULONG64 ImageBase;
+ PRUNTIME_FUNCTION FunctionEntry;
+ ULONG64 EstablisherFrame;
+ ULONG64 TargetIp;
+ PCONTEXT ContextRecord;
+ void (*LanguageHandler)(void);
+ PVOID HandlerData;
+ PUNWIND_HISTORY_TABLE HistoryTable;
+ ULONG ScopeIndex;
+ ULONG Fill0;
+} UndocumentedDispatcherContext;
+#endif
+
+/* Another wild guess. */
+extern void __DestructExceptionObject(EXCEPTION_RECORD *rec, int nothrow);
+
+#define LJ_MSVC_EXCODE ((DWORD)0xe06d7363)
+#define LJ_GCC_EXCODE ((DWORD)0x20474343)
+
+#define LJ_EXCODE ((DWORD)0xe24c4a00)
+#define LJ_EXCODE_MAKE(c) (LJ_EXCODE | (DWORD)(c))
+#define LJ_EXCODE_CHECK(cl) (((cl) ^ LJ_EXCODE) <= 0xff)
+#define LJ_EXCODE_ERRCODE(cl) ((int)((cl) & 0xff))
+
+/* Windows exception handler for interpreter frame. */
+LJ_FUNCA int lj_err_unwind_win(EXCEPTION_RECORD *rec,
+ void *f, CONTEXT *ctx, UndocumentedDispatcherContext *dispatch)
+{
+#if LJ_TARGET_X86
+ void *cf = (char *)f - CFRAME_OFS_SEH;
+#elif LJ_TARGET_ARM64
+ void *cf = (char *)f - CFRAME_SIZE;
+#else
+ void *cf = f;
+#endif
+ lua_State *L = cframe_L(cf);
+ int errcode = LJ_EXCODE_CHECK(rec->ExceptionCode) ?
+ LJ_EXCODE_ERRCODE(rec->ExceptionCode) : LUA_ERRRUN;
+ if ((rec->ExceptionFlags & 6)) { /* EH_UNWINDING|EH_EXIT_UNWIND */
+ if (rec->ExceptionCode == STATUS_LONGJUMP &&
+ rec->ExceptionRecord &&
+ LJ_EXCODE_CHECK(rec->ExceptionRecord->ExceptionCode)) {
+ errcode = LJ_EXCODE_ERRCODE(rec->ExceptionRecord->ExceptionCode);
+ if ((rec->ExceptionFlags & 0x20)) { /* EH_TARGET_UNWIND */
+ /* Unwinding is about to finish; revert the ExceptionCode so that
+ ** RtlRestoreContext does not try to restore from a _JUMP_BUFFER.
+ */
+ rec->ExceptionCode = 0;
+ }
+ }
+ /* Unwind internal frames. */
+ err_unwind(L, cf, errcode);
+ } else {
+ void *cf2 = err_unwind(L, cf, 0);
+ if (cf2) { /* We catch it, so start unwinding the upper frames. */
+#if !LJ_TARGET_X86
+ EXCEPTION_RECORD rec2;
+#endif
+ if (rec->ExceptionCode == LJ_MSVC_EXCODE ||
+ rec->ExceptionCode == LJ_GCC_EXCODE) {
+#if !LJ_TARGET_CYGWIN
+ __DestructExceptionObject(rec, 1);
+#endif
+ setstrV(L, L->top++, lj_err_str(L, LJ_ERR_ERRCPP));
+ } else if (!LJ_EXCODE_CHECK(rec->ExceptionCode)) {
+ /* Don't catch access violations etc. */
+ return 1; /* ExceptionContinueSearch */
+ }
+#if LJ_TARGET_X86
+ UNUSED(ctx);
+ UNUSED(dispatch);
+ /* Call all handlers for all lower C frames (including ourselves) again
+ ** with EH_UNWINDING set. Then call the specified function, passing cf
+ ** and errcode.
+ */
+ lj_vm_rtlunwind(cf, (void *)rec,
+ (cframe_unwind_ff(cf2) && errcode != LUA_YIELD) ?
+ (void *)lj_vm_unwind_ff : (void *)lj_vm_unwind_c, errcode);
+ /* lj_vm_rtlunwind does not return. */
+#else
+ if (LJ_EXCODE_CHECK(rec->ExceptionCode)) {
+ /* For unwind purposes, wrap the EXCEPTION_RECORD in something that
+ ** looks like a longjmp, so that MSVC will execute C++ destructors in
+ ** the frames we unwind over. ExceptionInformation[0] should really
+ ** contain a _JUMP_BUFFER*, but hopefully nobody is looking too closely
+ ** at this point.
+ */
+ rec2.ExceptionCode = STATUS_LONGJUMP;
+ rec2.ExceptionRecord = rec;
+ rec2.ExceptionAddress = 0;
+ rec2.NumberParameters = 1;
+ rec2.ExceptionInformation[0] = (ULONG_PTR)ctx;
+ rec = &rec2;
+ }
+ /* Unwind the stack and call all handlers for all lower C frames
+ ** (including ourselves) again with EH_UNWINDING set. Then set
+ ** stack pointer = f, result = errcode and jump to the specified target.
+ */
+ RtlUnwindEx(f, (void *)((cframe_unwind_ff(cf2) && errcode != LUA_YIELD) ?
+ lj_vm_unwind_ff_eh :
+ lj_vm_unwind_c_eh),
+ rec, (void *)(uintptr_t)errcode, dispatch->ContextRecord,
+ dispatch->HistoryTable);
+ /* RtlUnwindEx should never return. */
+#endif
+ }
+ }
+ return 1; /* ExceptionContinueSearch */
+}
+
+#if LJ_UNWIND_JIT
+
+#if LJ_TARGET_X64
+#define CONTEXT_REG_PC Rip
+#elif LJ_TARGET_ARM64
+#define CONTEXT_REG_PC Pc
+#else
+#error "NYI: Windows arch-specific unwinder for JIT-compiled code"
+#endif
+
+/* Windows unwinder for JIT-compiled code. */
+static void err_unwind_win_jit(global_State *g, int errcode)
+{
+ CONTEXT ctx;
+ UNWIND_HISTORY_TABLE hist;
+
+ memset(&hist, 0, sizeof(hist));
+ RtlCaptureContext(&ctx);
+ while (1) {
+ DWORD64 frame, base, addr = ctx.CONTEXT_REG_PC;
+ void *hdata;
+ PRUNTIME_FUNCTION func = RtlLookupFunctionEntry(addr, &base, &hist);
+ if (!func) { /* Found frame without .pdata: must be JIT-compiled code. */
+ ExitNo exitno;
+ uintptr_t stub = lj_trace_unwind(G2J(g), (uintptr_t)(addr - sizeof(MCode)), &exitno);
+ if (stub) { /* Jump to side exit to unwind the trace. */
+ ctx.CONTEXT_REG_PC = stub;
+ G2J(g)->exitcode = errcode;
+ RtlRestoreContext(&ctx, NULL); /* Does not return. */
+ }
+ break;
+ }
+ RtlVirtualUnwind(UNW_FLAG_NHANDLER, base, addr, func,
+ &ctx, &hdata, &frame, NULL);
+ if (!addr) break;
+ }
+ /* Unwinding failed, if we end up here. */
+}
+#endif
+
+/* Raise Windows exception. */
+static void err_raise_ext(global_State *g, int errcode)
+{
+#if LJ_UNWIND_JIT
+ if (tvref(g->jit_base)) {
+ err_unwind_win_jit(g, errcode);
+ return; /* Unwinding failed. */
+ }
+#elif LJ_HASJIT
+ /* Cannot catch on-trace errors for Windows/x86 SEH. Unwind to interpreter. */
+ setmref(g->jit_base, NULL);
+#endif
+ UNUSED(g);
+ RaiseException(LJ_EXCODE_MAKE(errcode), 1 /* EH_NONCONTINUABLE */, 0, NULL);
+}
+
+#elif !LJ_NO_UNWIND && (defined(__GNUC__) || defined(__clang__))
+
+/*
** We have to use our own definitions instead of the mandatory (!) unwind.h,
** since various OS, distros and compilers mess up the header installation.
*/
@@ -193,6 +399,7 @@
typedef struct _Unwind_Context _Unwind_Context;
#define _URC_OK 0
+#define _URC_FATAL_PHASE2_ERROR 2
#define _URC_FATAL_PHASE1_ERROR 3
#define _URC_HANDLER_FOUND 6
#define _URC_INSTALL_CONTEXT 7
@@ -212,9 +419,11 @@
void (*excleanup)(int, struct _Unwind_Exception *);
uintptr_t p1, p2;
} __attribute__((__aligned__)) _Unwind_Exception;
+#define UNWIND_EXCEPTION_TYPE _Unwind_Exception
extern uintptr_t _Unwind_GetCFA(_Unwind_Context *);
extern void _Unwind_SetGR(_Unwind_Context *, int, uintptr_t);
+extern uintptr_t _Unwind_GetIP(_Unwind_Context *);
extern void _Unwind_SetIP(_Unwind_Context *, uintptr_t);
extern void _Unwind_DeleteException(_Unwind_Exception *);
extern int _Unwind_RaiseException(_Unwind_Exception *);
@@ -232,7 +441,6 @@
lua_State *L;
if (version != 1)
return _URC_FATAL_PHASE1_ERROR;
- UNUSED(uexclass);
cf = (void *)_Unwind_GetCFA(ctx);
L = cframe_L(cf);
if ((actions & _UA_SEARCH_PHASE)) {
@@ -259,10 +467,10 @@
if ((actions & _UA_FORCE_UNWIND)) {
return _URC_CONTINUE_UNWIND;
} else if (cf) {
+ ASMFunction ip;
_Unwind_SetGR(ctx, LJ_TARGET_EHRETREG, errcode);
- _Unwind_SetIP(ctx, (uintptr_t)(cframe_unwind_ff(cf) ?
- lj_vm_unwind_ff_eh :
- lj_vm_unwind_c_eh));
+ ip = cframe_unwind_ff(cf) ? lj_vm_unwind_ff_eh : lj_vm_unwind_c_eh;
+ _Unwind_SetIP(ctx, (uintptr_t)lj_ptr_strip(ip));
return _URC_INSTALL_CONTEXT;
}
#if LJ_TARGET_X86ORX64
@@ -280,26 +488,151 @@
** it on non-x64 because the interpreter restores all callee-saved regs.
*/
lj_err_throw(L, errcode);
+#if LJ_TARGET_X64
+#error "Broken build system -- only use the provided Makefiles!"
#endif
+#endif
}
return _URC_CONTINUE_UNWIND;
}
-#if LJ_UNWIND_EXT
-#if LJ_TARGET_OSX || defined(__OpenBSD__)
-/* Sorry, no thread safety for OSX. Complain to Apple, not me. */
-static _Unwind_Exception static_uex;
+#if LJ_UNWIND_EXT && defined(LUA_USE_ASSERT)
+struct dwarf_eh_bases { void *tbase, *dbase, *func; };
+extern const void *_Unwind_Find_FDE(void *pc, struct dwarf_eh_bases *bases);
+
+/* Verify that external error handling actually has a chance to work. */
+void lj_err_verify(void)
+{
+#if !LJ_TARGET_OSX
+ /* Check disabled on MacOS due to brilliant software engineering at Apple. */
+ struct dwarf_eh_bases ehb;
+ lj_assertX(_Unwind_Find_FDE((void *)lj_err_throw, &ehb), "broken build: external frame unwinding enabled, but missing -funwind-tables");
+#endif
+ /* Check disabled, because of broken Fedora/ARM64. See #722.
+ lj_assertX(_Unwind_Find_FDE((void *)_Unwind_RaiseException, &ehb), "broken build: external frame unwinding enabled, but system libraries have no unwind tables");
+ */
+}
+#endif
+
+#if LJ_UNWIND_JIT
+/* DWARF2 personality handler for JIT-compiled code. */
+static int err_unwind_jit(int version, int actions,
+ uint64_t uexclass, _Unwind_Exception *uex, _Unwind_Context *ctx)
+{
+ /* NYI: FFI C++ exception interoperability. */
+ if (version != 1 || !LJ_UEXCLASS_CHECK(uexclass))
+ return _URC_FATAL_PHASE1_ERROR;
+ if ((actions & _UA_SEARCH_PHASE)) {
+ return _URC_HANDLER_FOUND;
+ }
+ if ((actions & _UA_CLEANUP_PHASE)) {
+ global_State *g = *(global_State **)(uex+1);
+ ExitNo exitno;
+ uintptr_t addr = _Unwind_GetIP(ctx); /* Return address _after_ call. */
+ uintptr_t stub = lj_trace_unwind(G2J(g), addr - sizeof(MCode), &exitno);
+ lj_assertG(tvref(g->jit_base), "unexpected throw across mcode frame");
+ if (stub) { /* Jump to side exit to unwind the trace. */
+ G2J(g)->exitcode = LJ_UEXCLASS_ERRCODE(uexclass);
+#ifdef LJ_TARGET_MIPS
+ _Unwind_SetGR(ctx, 4, stub);
+ _Unwind_SetGR(ctx, 5, exitno);
+ _Unwind_SetIP(ctx, (uintptr_t)(void *)lj_vm_unwind_stub);
#else
-static __thread _Unwind_Exception static_uex;
+ _Unwind_SetIP(ctx, stub);
#endif
+ return _URC_INSTALL_CONTEXT;
+ }
+ return _URC_FATAL_PHASE2_ERROR;
+ }
+ return _URC_FATAL_PHASE1_ERROR;
+}
-/* Raise DWARF2 exception. */
-static void err_raise_ext(int errcode)
+/* DWARF2 template frame info for JIT-compiled code.
+**
+** After copying the template to the start of the mcode segment,
+** the frame handler function and the code size is patched.
+** The frame handler always installs a new context to jump to the exit,
+** so don't bother to add any unwind opcodes.
+*/
+static const uint8_t err_frame_jit_template[] = {
+#if LJ_BE
+ 0,0,0,
+#endif
+ LJ_64 ? 0x1c : 0x14, /* CIE length. */
+#if LJ_LE
+ 0,0,0,
+#endif
+ 0,0,0,0, 1, 'z','P','R',0, /* CIE mark, CIE version, augmentation. */
+ 1, LJ_64 ? 0x78 : 0x7c, LJ_TARGET_EHRAREG, /* Code/data align, RA. */
+#if LJ_64
+ 10, 0, 0,0,0,0,0,0,0,0, 0x1b, /* Aug. data ABS handler, PCREL|SDATA4 code. */
+ 0,0,0,0,0, /* Alignment. */
+#else
+ 6, 0, 0,0,0,0, 0x1b, /* Aug. data ABS handler, PCREL|SDATA4 code. */
+ 0, /* Alignment. */
+#endif
+#if LJ_BE
+ 0,0,0,
+#endif
+ LJ_64 ? 0x14 : 0x10, /* FDE length. */
+ 0,0,0,
+ LJ_64 ? 0x24 : 0x1c, /* CIE offset. */
+ 0,0,0,
+ LJ_64 ? 0x14 : 0x10, /* Code offset. After Final FDE. */
+#if LJ_LE
+ 0,0,0,
+#endif
+ 0,0,0,0, 0, 0,0,0, /* Code size, augmentation length, alignment. */
+#if LJ_64
+ 0,0,0,0, /* Alignment. */
+#endif
+ 0,0,0,0 /* Final FDE. */
+};
+
+#define ERR_FRAME_JIT_OFS_HANDLER 0x12
+#define ERR_FRAME_JIT_OFS_FDE (LJ_64 ? 0x20 : 0x18)
+#define ERR_FRAME_JIT_OFS_CODE_SIZE (LJ_64 ? 0x2c : 0x24)
+#if LJ_TARGET_OSX
+#define ERR_FRAME_JIT_OFS_REGISTER ERR_FRAME_JIT_OFS_FDE
+#else
+#define ERR_FRAME_JIT_OFS_REGISTER 0
+#endif
+
+extern void __register_frame(const void *);
+extern void __deregister_frame(const void *);
+
+uint8_t *lj_err_register_mcode(void *base, size_t sz, uint8_t *info)
{
- static_uex.exclass = LJ_UEXCLASS_MAKE(errcode);
- static_uex.excleanup = NULL;
- _Unwind_RaiseException(&static_uex);
+ ASMFunction handler = (ASMFunction)err_unwind_jit;
+ memcpy(info, err_frame_jit_template, sizeof(err_frame_jit_template));
+#if LJ_ABI_PAUTH
+#if LJ_TARGET_ARM64
+ handler = ptrauth_auth_and_resign(handler,
+ ptrauth_key_function_pointer, 0,
+ ptrauth_key_process_independent_code, info + ERR_FRAME_JIT_OFS_HANDLER);
+#else
+#error "missing pointer authentication support for this architecture"
+#endif
+#endif
+ memcpy(info + ERR_FRAME_JIT_OFS_HANDLER, &handler, sizeof(handler));
+ *(uint32_t *)(info + ERR_FRAME_JIT_OFS_CODE_SIZE) =
+ (uint32_t)(sz - sizeof(err_frame_jit_template) - (info - (uint8_t *)base));
+ __register_frame(info + ERR_FRAME_JIT_OFS_REGISTER);
+#ifdef LUA_USE_ASSERT
+ {
+ struct dwarf_eh_bases ehb;
+ lj_assertX(_Unwind_Find_FDE(info + sizeof(err_frame_jit_template)+1, &ehb),
+ "bad JIT unwind table registration");
+ }
+#endif
+ return info + sizeof(err_frame_jit_template);
}
+
+void lj_err_deregister_mcode(void *base, size_t sz, uint8_t *info)
+{
+ UNUSED(base); UNUSED(sz);
+ __deregister_frame(info + ERR_FRAME_JIT_OFS_REGISTER);
+}
#endif
#else /* LJ_TARGET_ARM */
@@ -310,6 +643,7 @@
#define _US_FORCE_UNWIND 8
typedef struct _Unwind_Control_Block _Unwind_Control_Block;
+#define UNWIND_EXCEPTION_TYPE _Unwind_Control_Block
struct _Unwind_Control_Block {
uint64_t exclass;
@@ -368,136 +702,63 @@
}
if (__gnu_unwind_frame(ucb, ctx) != _URC_OK)
return _URC_FAILURE;
+#ifdef LUA_USE_ASSERT
+ /* We should never get here unless this is a forced unwind aka backtrace. */
+ if (_Unwind_GetGR(ctx, 0) == 0xff33aa77) {
+ _Unwind_SetGR(ctx, 0, 0xff33aa88);
+ }
+#endif
return _URC_CONTINUE_UNWIND;
}
-#if LJ_UNWIND_EXT
-static __thread _Unwind_Control_Block static_uex;
+#if LJ_UNWIND_EXT && defined(LUA_USE_ASSERT)
+typedef int (*_Unwind_Trace_Fn)(_Unwind_Context *, void *);
+extern int _Unwind_Backtrace(_Unwind_Trace_Fn, void *);
-static void err_raise_ext(int errcode)
+static int err_verify_bt(_Unwind_Context *ctx, int *got)
{
- memset(&static_uex, 0, sizeof(static_uex));
- static_uex.exclass = LJ_UEXCLASS_MAKE(errcode);
- _Unwind_RaiseException(&static_uex);
+ if (_Unwind_GetGR(ctx, 0) == 0xff33aa88) { *got = 2; }
+ else if (*got == 0) { *got = 1; _Unwind_SetGR(ctx, 0, 0xff33aa77); }
+ return _URC_OK;
}
+
+/* Verify that external error handling actually has a chance to work. */
+void lj_err_verify(void)
+{
+ int got = 0;
+ _Unwind_Backtrace((_Unwind_Trace_Fn)err_verify_bt, &got);
+ lj_assertX(got == 2, "broken build: external frame unwinding enabled, but missing -funwind-tables");
+}
#endif
-#endif /* LJ_TARGET_ARM */
-
-#elif LJ_ABI_WIN
-
/*
-** Someone in Redmond owes me several days of my life. A lot of this is
-** undocumented or just plain wrong on MSDN. Some of it can be gathered
-** from 3rd party docs or must be found by trial-and-error. They really
-** don't want you to write your own language-specific exception handler
-** or to interact gracefully with MSVC. :-(
+** Note: LJ_UNWIND_JIT is not implemented for 32 bit ARM.
**
-** Apparently MSVC doesn't call C++ destructors for foreign exceptions
-** unless you compile your C++ code with /EHa. Unfortunately this means
-** catch (...) also catches things like access violations. The use of
-** _set_se_translator doesn't really help, because it requires /EHa, too.
+** The quirky ARM unwind API doesn't have __register_frame().
+** A potential workaround might involve _Unwind_Backtrace.
+** But most 32 bit ARM targets don't qualify for LJ_UNWIND_EXT, anyway,
+** since they are built without unwind tables by default.
*/
-#define WIN32_LEAN_AND_MEAN
-#include <windows.h>
+#endif /* LJ_TARGET_ARM */
-#if LJ_TARGET_X64
-/* Taken from: http://www.nynaeve.net/?p=99 */
-typedef struct UndocumentedDispatcherContext {
- ULONG64 ControlPc;
- ULONG64 ImageBase;
- PRUNTIME_FUNCTION FunctionEntry;
- ULONG64 EstablisherFrame;
- ULONG64 TargetIp;
- PCONTEXT ContextRecord;
- void (*LanguageHandler)(void);
- PVOID HandlerData;
- PUNWIND_HISTORY_TABLE HistoryTable;
- ULONG ScopeIndex;
- ULONG Fill0;
-} UndocumentedDispatcherContext;
-#else
-typedef void *UndocumentedDispatcherContext;
-#endif
-/* Another wild guess. */
-extern void __DestructExceptionObject(EXCEPTION_RECORD *rec, int nothrow);
+#if LJ_UNWIND_EXT
+static __thread struct {
+ UNWIND_EXCEPTION_TYPE ex;
+ global_State *g;
+} static_uex;
-#if LJ_TARGET_X64 && defined(MINGW_SDK_INIT)
-/* Workaround for broken MinGW64 declaration. */
-VOID RtlUnwindEx_FIXED(PVOID,PVOID,PVOID,PVOID,PVOID,PVOID) asm("RtlUnwindEx");
-#define RtlUnwindEx RtlUnwindEx_FIXED
-#endif
-
-#define LJ_MSVC_EXCODE ((DWORD)0xe06d7363)
-#define LJ_GCC_EXCODE ((DWORD)0x20474343)
-
-#define LJ_EXCODE ((DWORD)0xe24c4a00)
-#define LJ_EXCODE_MAKE(c) (LJ_EXCODE | (DWORD)(c))
-#define LJ_EXCODE_CHECK(cl) (((cl) ^ LJ_EXCODE) <= 0xff)
-#define LJ_EXCODE_ERRCODE(cl) ((int)((cl) & 0xff))
-
-/* Windows exception handler for interpreter frame. */
-LJ_FUNCA int lj_err_unwind_win(EXCEPTION_RECORD *rec,
- void *f, CONTEXT *ctx, UndocumentedDispatcherContext *dispatch)
+/* Raise external exception. */
+static void err_raise_ext(global_State *g, int errcode)
{
-#if LJ_TARGET_X64
- void *cf = f;
-#else
- void *cf = (char *)f - CFRAME_OFS_SEH;
-#endif
- lua_State *L = cframe_L(cf);
- int errcode = LJ_EXCODE_CHECK(rec->ExceptionCode) ?
- LJ_EXCODE_ERRCODE(rec->ExceptionCode) : LUA_ERRRUN;
- if ((rec->ExceptionFlags & 6)) { /* EH_UNWINDING|EH_EXIT_UNWIND */
- /* Unwind internal frames. */
- err_unwind(L, cf, errcode);
- } else {
- void *cf2 = err_unwind(L, cf, 0);
- if (cf2) { /* We catch it, so start unwinding the upper frames. */
- if (rec->ExceptionCode == LJ_MSVC_EXCODE ||
- rec->ExceptionCode == LJ_GCC_EXCODE) {
-#if LJ_TARGET_WINDOWS
- __DestructExceptionObject(rec, 1);
-#endif
- setstrV(L, L->top++, lj_err_str(L, LJ_ERR_ERRCPP));
- } else if (!LJ_EXCODE_CHECK(rec->ExceptionCode)) {
- /* Don't catch access violations etc. */
- return 1; /* ExceptionContinueSearch */
- }
-#if LJ_TARGET_X64
- /* Unwind the stack and call all handlers for all lower C frames
- ** (including ourselves) again with EH_UNWINDING set. Then set
- ** rsp = cf, rax = errcode and jump to the specified target.
- */
- RtlUnwindEx(cf, (void *)((cframe_unwind_ff(cf2) && errcode != LUA_YIELD) ?
- lj_vm_unwind_ff_eh :
- lj_vm_unwind_c_eh),
- rec, (void *)(uintptr_t)errcode, ctx, dispatch->HistoryTable);
- /* RtlUnwindEx should never return. */
-#else
- UNUSED(ctx);
- UNUSED(dispatch);
- /* Call all handlers for all lower C frames (including ourselves) again
- ** with EH_UNWINDING set. Then call the specified function, passing cf
- ** and errcode.
- */
- lj_vm_rtlunwind(cf, (void *)rec,
- (cframe_unwind_ff(cf2) && errcode != LUA_YIELD) ?
- (void *)lj_vm_unwind_ff : (void *)lj_vm_unwind_c, errcode);
- /* lj_vm_rtlunwind does not return. */
-#endif
- }
- }
- return 1; /* ExceptionContinueSearch */
+ memset(&static_uex, 0, sizeof(static_uex));
+ static_uex.ex.exclass = LJ_UEXCLASS_MAKE(errcode);
+ static_uex.g = g;
+ _Unwind_RaiseException(&static_uex.ex);
}
-/* Raise Windows exception. */
-static void err_raise_ext(int errcode)
-{
- RaiseException(LJ_EXCODE_MAKE(errcode), 1 /* EH_NONCONTINUABLE */, 0, NULL);
-}
+#endif
#endif
@@ -508,22 +769,23 @@
{
global_State *g = G(L);
lj_trace_abort(g);
- setmref(g->jit_base, NULL);
L->status = LUA_OK;
#if LJ_UNWIND_EXT
- err_raise_ext(errcode);
+ err_raise_ext(g, errcode);
/*
** A return from this function signals a corrupt C stack that cannot be
** unwound. We have no choice but to call the panic function and exit.
**
** Usually this is caused by a C function without unwind information.
- ** This should never happen on x64, but may happen if you've manually
- ** enabled LUAJIT_UNWIND_EXTERNAL and forgot to recompile *every*
- ** non-C++ file with -funwind-tables.
+ ** This may happen if you've manually enabled LUAJIT_UNWIND_EXTERNAL
+ ** and forgot to recompile *every* non-C++ file with -funwind-tables.
*/
if (G(L)->panic)
G(L)->panic(L);
#else
+#if LJ_HASJIT
+ setmref(g->jit_base, NULL);
+#endif
{
void *cf = err_unwind(L, NULL, errcode);
if (cframe_unwind_ff(cf))
@@ -546,6 +808,18 @@
{
if (L->status == LUA_ERRERR+1) /* Don't touch the stack during lua_open. */
lj_vm_unwind_c(L->cframe, LUA_ERRMEM);
+ if (LJ_HASJIT) {
+ TValue *base = tvref(G(L)->jit_base);
+ if (base) L->base = base;
+ }
+ if (curr_funcisL(L)) {
+ L->top = curr_topL(L);
+ if (LJ_UNLIKELY(L->top > tvref(L->maxstack))) {
+ /* The current Lua frame violates the stack. Replace it with a dummy. */
+ L->top = L->base;
+ setframe_gc(L->base - 1 - LJ_FR2, obj2gco(L), LJ_TTHREAD);
+ }
+ }
setstrV(L, L->top++, lj_err_str(L, LJ_ERR_ERRMEM));
lj_err_throw(L, LUA_ERRMEM);
}
@@ -585,6 +859,7 @@
if (cframe_canyield(cf)) return 0;
if (cframe_errfunc(cf) >= 0)
return cframe_errfunc(cf);
+ cf = cframe_prev(cf);
frame = frame_prevd(frame);
break;
case FRAME_PCALL:
@@ -593,7 +868,7 @@
return savestack(L, frame_prevd(frame)+1); /* xpcall's errorfunc. */
return 0;
default:
- lua_assert(0);
+ lj_assertL(0, "bad frame type");
return 0;
}
}
@@ -601,13 +876,15 @@
}
/* Runtime error. */
-LJ_NOINLINE void lj_err_run(lua_State *L)
+LJ_NOINLINE void LJ_FASTCALL lj_err_run(lua_State *L)
{
- ptrdiff_t ef = finderrfunc(L);
+ ptrdiff_t ef = (LJ_HASJIT && tvref(G(L)->jit_base)) ? 0 : finderrfunc(L);
if (ef) {
- TValue *errfunc = restorestack(L, ef);
- TValue *top = L->top;
+ TValue *errfunc, *top;
+ lj_state_checkstack(L, LUA_MINSTACK * 2); /* Might raise new error. */
lj_trace_abort(G(L));
+ errfunc = restorestack(L, ef);
+ top = L->top;
if (!tvisfunc(errfunc) || L->status == LUA_ERRERR) {
setstrV(L, top-1, lj_err_str(L, LJ_ERR_ERRERR));
lj_err_throw(L, LUA_ERRERR);
@@ -622,6 +899,24 @@
lj_err_throw(L, LUA_ERRRUN);
}
+/* Stack overflow error. */
+void LJ_FASTCALL lj_err_stkov(lua_State *L)
+{
+ lj_debug_addloc(L, err2msg(LJ_ERR_STKOV), L->base-1, NULL);
+ lj_err_run(L);
+}
+
+#if LJ_HASJIT
+/* Rethrow error after doing a trace exit. */
+LJ_NOINLINE void LJ_FASTCALL lj_err_trace(lua_State *L, int errcode)
+{
+ if (errcode == LUA_ERRRUN)
+ lj_err_run(L);
+ else
+ lj_err_throw(L, errcode);
+}
+#endif
+
/* Formatted runtime error message. */
LJ_NORET LJ_NOINLINE static void err_msgv(lua_State *L, ErrMsg em, ...)
{
@@ -628,6 +923,10 @@
const char *msg;
va_list argp;
va_start(argp, em);
+ if (LJ_HASJIT) {
+ TValue *base = tvref(G(L)->jit_base);
+ if (base) L->base = base;
+ }
if (curr_funcisL(L)) L->top = curr_topL(L);
msg = lj_strfmt_pushvf(L, err2msg(em), argp);
va_end(argp);
@@ -690,9 +989,9 @@
const BCIns *pc = cframe_Lpc(L);
if (((ptrdiff_t)pc & FRAME_TYPE) != FRAME_LUA) {
const char *tname = lj_typename(o);
+ setframe_gc(o, obj2gco(L), LJ_TTHREAD);
if (LJ_FR2) o++;
setframe_pc(o, pc);
- setframe_gc(o, obj2gco(L), LJ_TTHREAD);
L->top = L->base = o+1;
err_msgv(L, LJ_ERR_BADCALL, tname);
}
@@ -702,25 +1001,27 @@
/* Error in context of caller. */
LJ_NOINLINE void lj_err_callermsg(lua_State *L, const char *msg)
{
- TValue *frame = L->base-1;
- TValue *pframe = NULL;
- if (frame_islua(frame)) {
- pframe = frame_prevl(frame);
- } else if (frame_iscont(frame)) {
- if (frame_iscont_fficb(frame)) {
- pframe = frame;
- frame = NULL;
- } else {
- pframe = frame_prevd(frame);
+ TValue *frame = NULL, *pframe = NULL;
+ if (!(LJ_HASJIT && tvref(G(L)->jit_base))) {
+ frame = L->base-1;
+ if (frame_islua(frame)) {
+ pframe = frame_prevl(frame);
+ } else if (frame_iscont(frame)) {
+ if (frame_iscont_fficb(frame)) {
+ pframe = frame;
+ frame = NULL;
+ } else {
+ pframe = frame_prevd(frame);
#if LJ_HASFFI
- /* Remove frame for FFI metamethods. */
- if (frame_func(frame)->c.ffid >= FF_ffi_meta___index &&
- frame_func(frame)->c.ffid <= FF_ffi_meta___tostring) {
- L->base = pframe+1;
- L->top = frame;
- setcframe_pc(cframe_raw(L->cframe), frame_contpc(frame));
+ /* Remove frame for FFI metamethods. */
+ if (frame_func(frame)->c.ffid >= FF_ffi_meta___index &&
+ frame_func(frame)->c.ffid <= FF_ffi_meta___tostring) {
+ L->base = pframe+1;
+ L->top = frame;
+ setcframe_pc(cframe_raw(L->cframe), frame_contpc(frame));
+ }
+#endif
}
-#endif
}
}
lj_debug_addloc(L, msg, pframe, frame);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_err.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Error handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_ERR_H
@@ -23,7 +23,11 @@
LJ_FUNC GCstr *lj_err_str(lua_State *L, ErrMsg em);
LJ_FUNCA_NORET void LJ_FASTCALL lj_err_throw(lua_State *L, int errcode);
LJ_FUNC_NORET void lj_err_mem(lua_State *L);
-LJ_FUNC_NORET void lj_err_run(lua_State *L);
+LJ_FUNC_NORET void LJ_FASTCALL lj_err_stkov(lua_State *L);
+LJ_FUNC_NORET void LJ_FASTCALL lj_err_run(lua_State *L);
+#if LJ_HASJIT
+LJ_FUNCA_NORET void LJ_FASTCALL lj_err_trace(lua_State *L, int errcode);
+#endif
LJ_FUNC_NORET void lj_err_msg(lua_State *L, ErrMsg em);
LJ_FUNC_NORET void lj_err_lex(lua_State *L, GCstr *src, const char *tok,
BCLine line, ErrMsg em, va_list argp);
@@ -38,4 +42,18 @@
LJ_FUNC_NORET void lj_err_argtype(lua_State *L, int narg, const char *xname);
LJ_FUNC_NORET void lj_err_argt(lua_State *L, int narg, int tt);
+#if LJ_UNWIND_JIT && !LJ_ABI_WIN
+LJ_FUNC uint8_t *lj_err_register_mcode(void *base, size_t sz, uint8_t *info);
+LJ_FUNC void lj_err_deregister_mcode(void *base, size_t sz, uint8_t *info);
+#else
+#define lj_err_register_mcode(base, sz, info) (info)
+#define lj_err_deregister_mcode(base, sz, info) UNUSED(base)
#endif
+
+#if LJ_UNWIND_EXT && !LJ_ABI_WIN && defined(LUA_USE_ASSERT)
+LJ_FUNC void lj_err_verify(void);
+#else
+#define lj_err_verify() ((void)0)
+#endif
+
+#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_errmsg.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_errmsg.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_errmsg.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** VM error messages.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
/* This file may be included multiple times with different ERRDEF macros. */
@@ -67,6 +67,7 @@
ERRDEF(UNPACK, "too many results to unpack")
ERRDEF(RDRSTR, "reader function must return a string")
ERRDEF(PRTOSTR, LUA_QL("tostring") " must return a string to " LUA_QL("print"))
+ERRDEF(NUMRNG, "number out of range")
ERRDEF(IDXRNG, "index out of range")
ERRDEF(BASERNG, "base out of range")
ERRDEF(LVLRNG, "level out of range")
@@ -78,6 +79,7 @@
ERRDEF(CORUN, "cannot resume running coroutine")
ERRDEF(CODEAD, "cannot resume dead coroutine")
ERRDEF(COSUSP, "cannot resume non-suspended coroutine")
+ERRDEF(PRNGSD, "PRNG seeding failed")
ERRDEF(TABINS, "wrong number of arguments to " LUA_QL("insert"))
ERRDEF(TABCAT, "invalid value (%s) at index %d in table for " LUA_QL("concat"))
ERRDEF(TABSORT, "invalid order function for sorting")
@@ -101,11 +103,7 @@
ERRDEF(BADMODN, "name conflict for module " LUA_QS)
#if LJ_HASJIT
ERRDEF(JITPROT, "runtime code generation failed, restricted kernel?")
-#if LJ_TARGET_X86ORX64
-ERRDEF(NOJIT, "JIT compiler disabled, CPU does not support SSE2")
-#else
ERRDEF(NOJIT, "JIT compiler disabled")
-#endif
#elif defined(LJ_ARCH_NOJIT)
ERRDEF(NOJIT, "no JIT compiler for this architecture (yet)")
#else
@@ -183,6 +181,19 @@
ERRDEF(FFI_NYICALL, "NYI: cannot call this C function (yet)")
#endif
+#if LJ_HASBUFFER
+/* String buffer errors. */
+ERRDEF(BUFFER_SELF, "cannot put buffer into itself")
+ERRDEF(BUFFER_BADOPT, "bad options table")
+ERRDEF(BUFFER_BADENC, "cannot serialize " LUA_QS)
+ERRDEF(BUFFER_BADDEC, "cannot deserialize tag 0x%02x")
+ERRDEF(BUFFER_BADDICTX, "cannot deserialize dictionary index %d")
+ERRDEF(BUFFER_DEPTH, "too deep to serialize")
+ERRDEF(BUFFER_DUPKEY, "duplicate table key")
+ERRDEF(BUFFER_EOB, "unexpected end of buffer")
+ERRDEF(BUFFER_LEFTOV, "left-over data in buffer")
+#endif
+
#undef ERRDEF
/* Detecting unused error messages:
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ff.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ff.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ff.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Fast function IDs.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_FF_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Fast function call recorder.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_ffrecord_c
@@ -11,6 +11,7 @@
#if LJ_HASJIT
#include "lj_err.h"
+#include "lj_buf.h"
#include "lj_str.h"
#include "lj_tab.h"
#include "lj_frame.h"
@@ -28,6 +29,7 @@
#include "lj_vm.h"
#include "lj_strscan.h"
#include "lj_strfmt.h"
+#include "lj_serialize.h"
/* Some local macros to save typing. Undef'd at the end. */
#define IR(ref) (&J->cur.ir[(ref)])
@@ -96,6 +98,14 @@
return -1;
}
+static TValue *rec_stop_stitch_cp(lua_State *L, lua_CFunction dummy, void *ud)
+{
+ jit_State *J = (jit_State *)ud;
+ lj_record_stop(J, LJ_TRLINK_STITCH, 0);
+ UNUSED(L); UNUSED(dummy);
+ return NULL;
+}
+
/* Trace stitching: add continuation below frame to start a new trace. */
static void recff_stitch(jit_State *J)
{
@@ -106,6 +116,7 @@
TValue *nframe = base + 1 + LJ_FR2;
const BCIns *pc = frame_pc(base-1);
TValue *pframe = frame_prevl(base-1);
+ int errcode;
/* Move func + args up in Lua stack and insert continuation. */
memmove(&base[1], &base[-1-LJ_FR2], sizeof(TValue)*nslot);
@@ -130,7 +141,7 @@
J->baseslot += 2 + LJ_FR2;
J->framedepth++;
- lj_record_stop(J, LJ_TRLINK_STITCH, 0);
+ errcode = lj_vm_cpcall(L, NULL, J, rec_stop_stitch_cp);
/* Undo Lua stack changes. */
memmove(&base[-1-LJ_FR2], &base[1], sizeof(TValue)*nslot);
@@ -137,6 +148,14 @@
setframe_pc(base-1, pc);
L->base -= 2 + LJ_FR2;
L->top -= 2 + LJ_FR2;
+
+ if (errcode) {
+ if (errcode == LUA_ERRRUN)
+ copyTV(L, L->top-1, L->top + (1 + LJ_FR2));
+ else
+ setintV(L->top-1, (int32_t)LJ_TRERR_RECERR);
+ lj_err_throw(L, errcode); /* Propagate errors. */
+ }
}
/* Fallback handler for fast functions that are not recorded (yet). */
@@ -182,6 +201,14 @@
lj_ir_kptr(J, &J2G(J)->tmpbuf), IRBUFHDR_RESET);
}
+/* Emit TMPREF. */
+static TRef recff_tmpref(jit_State *J, TRef tr, int mode)
+{
+ if (!LJ_DUALNUM && tref_isinteger(tr))
+ tr = emitir(IRTN(IR_CONV), tr, IRCONV_NUM_INT);
+ return emitir(IRT(IR_TMPREF, IRT_PGC), tr, mode);
+}
+
/* -- Base library fast functions ----------------------------------------- */
static void LJ_FASTCALL recff_assert(jit_State *J, RecordFFData *rd)
@@ -232,7 +259,7 @@
mtref = tref_isnil(mt) ? lj_ir_knull(J, IRT_TAB) : mt;
emitir(IRT(IR_FSTORE, IRT_TAB), fref, mtref);
if (!tref_isnil(mt))
- emitir(IRT(IR_TBAR, IRT_TAB), tr, 0);
+ emitir(IRT(IR_TBAR, IRT_NIL), tr, 0);
J->base[0] = tr;
J->needsnap = 1;
} /* else: Interpreter will throw. */
@@ -281,7 +308,7 @@
if (tref_isstr(tr))
J->base[0] = emitir(IRTI(IR_FLOAD), tr, IRFL_STR_LEN);
else if (tref_istab(tr))
- J->base[0] = lj_ir_call(J, IRCALL_lj_tab_len, tr);
+ J->base[0] = emitir(IRTI(IR_ALEN), tr, TREF_NIL);
/* else: Interpreter will throw. */
UNUSED(rd);
}
@@ -296,7 +323,7 @@
} else {
TRef trptr = emitir(IRT(IR_STRREF, IRT_PGC), tr, lj_ir_kint(J, 0));
TRef trchar = emitir(IRT(IR_XLOAD, IRT_U8), trptr, IRXLOAD_READONLY);
- emitir(IRTG(IR_EQ, IRT_INT), trchar, lj_ir_kint(J, '#'));
+ emitir(IRTGI(IR_EQ), trchar, lj_ir_kint(J, '#'));
}
return 0;
} else { /* select(n, ...) */
@@ -317,9 +344,9 @@
ptrdiff_t n = (ptrdiff_t)J->maxslot;
if (start < 0) start += n;
else if (start > n) start = n;
- rd->nres = n - start;
if (start >= 1) {
ptrdiff_t i;
+ rd->nres = n - start;
for (i = 0; i < n - start; i++)
J->base[i] = J->base[start+i];
} /* else: Interpreter will throw. */
@@ -455,6 +482,7 @@
#endif
lj_record_call(J, 0, J->maxslot - 1);
rd->nres = -1; /* Pending call. */
+ J->needsnap = 1; /* Start catching on-trace errors. */
} /* else: Interpreter will throw. */
}
@@ -490,6 +518,7 @@
if (errcode)
lj_err_throw(J->L, errcode); /* Propagate errors. */
rd->nres = -1; /* Pending call. */
+ J->needsnap = 1; /* Start catching on-trace errors. */
} /* else: Interpreter will throw. */
}
@@ -505,6 +534,40 @@
recff_nyiu(J, rd);
}
+static void LJ_FASTCALL recff_next(jit_State *J, RecordFFData *rd)
+{
+#if LJ_BE
+ /* YAGNI: Disabled on big-endian due to issues with lj_vm_next,
+ ** IR_HIOP, RID_RETLO/RID_RETHI and ra_destpair.
+ */
+ recff_nyi(J, rd);
+#else
+ TRef tab = J->base[0];
+ if (tref_istab(tab)) {
+ RecordIndex ix;
+ cTValue *keyv;
+ ix.tab = tab;
+ if (tref_isnil(J->base[1])) { /* Shortcut for start of traversal. */
+ ix.key = lj_ir_kint(J, 0);
+ keyv = niltvg(J2G(J));
+ } else {
+ TRef tmp = recff_tmpref(J, J->base[1], IRTMPREF_IN1);
+ ix.key = lj_ir_call(J, IRCALL_lj_tab_keyindex, tab, tmp);
+ keyv = &rd->argv[1];
+ }
+ copyTV(J->L, &ix.tabv, &rd->argv[0]);
+ ix.keyv.u32.lo = lj_tab_keyindex(tabV(&ix.tabv), keyv);
+ /* Omit the value, if not used by the caller. */
+ ix.idxchain = (J->framedepth && frame_islua(J->L->base-1) &&
+ bc_b(frame_pc(J->L->base-1)[-1])-1 < 2);
+ ix.mobj = 0; /* We don't need the next index. */
+ rd->nres = lj_record_next(J, &ix);
+ J->base[0] = ix.key;
+ J->base[1] = ix.val;
+ } /* else: Interpreter will throw. */
+#endif
+}
+
/* -- Math library fast functions ----------------------------------------- */
static void LJ_FASTCALL recff_math_abs(jit_State *J, RecordFFData *rd)
@@ -563,7 +626,7 @@
{
TRef tr = lj_ir_tonum(J, J->base[0]);
TRef tr2 = lj_ir_tonum(J, J->base[1]);
- J->base[0] = emitir(IRTN(IR_ATAN2), tr, tr2);
+ J->base[0] = lj_ir_call(J, IRCALL_atan2, tr, tr2);
UNUSED(rd);
}
@@ -580,47 +643,16 @@
UNUSED(rd);
}
-/* Record math.asin, math.acos, math.atan. */
-static void LJ_FASTCALL recff_math_atrig(jit_State *J, RecordFFData *rd)
+static void LJ_FASTCALL recff_math_call(jit_State *J, RecordFFData *rd)
{
- TRef y = lj_ir_tonum(J, J->base[0]);
- TRef x = lj_ir_knum_one(J);
- uint32_t ffid = rd->data;
- if (ffid != FF_math_atan) {
- TRef tmp = emitir(IRTN(IR_MUL), y, y);
- tmp = emitir(IRTN(IR_SUB), x, tmp);
- tmp = emitir(IRTN(IR_FPMATH), tmp, IRFPM_SQRT);
- if (ffid == FF_math_asin) { x = tmp; } else { x = y; y = tmp; }
- }
- J->base[0] = emitir(IRTN(IR_ATAN2), y, x);
-}
-
-static void LJ_FASTCALL recff_math_htrig(jit_State *J, RecordFFData *rd)
-{
TRef tr = lj_ir_tonum(J, J->base[0]);
J->base[0] = emitir(IRTN(IR_CALLN), tr, rd->data);
}
-static void LJ_FASTCALL recff_math_modf(jit_State *J, RecordFFData *rd)
-{
- TRef tr = J->base[0];
- if (tref_isinteger(tr)) {
- J->base[0] = tr;
- J->base[1] = lj_ir_kint(J, 0);
- } else {
- TRef trt;
- tr = lj_ir_tonum(J, tr);
- trt = emitir(IRTN(IR_FPMATH), tr, IRFPM_TRUNC);
- J->base[0] = trt;
- J->base[1] = emitir(IRTN(IR_SUB), tr, trt);
- }
- rd->nres = 2;
-}
-
static void LJ_FASTCALL recff_math_pow(jit_State *J, RecordFFData *rd)
{
- J->base[0] = lj_opt_narrow_pow(J, J->base[0], J->base[1],
- &rd->argv[0], &rd->argv[1]);
+ J->base[0] = lj_opt_narrow_arith(J, J->base[0], J->base[1],
+ &rd->argv[0], &rd->argv[1], IR_POW);
UNUSED(rd);
}
@@ -647,7 +679,7 @@
GCudata *ud = udataV(&J->fn->c.upvalue[0]);
TRef tr, one;
lj_ir_kgc(J, obj2gco(ud), IRT_UDATA); /* Prevent collection. */
- tr = lj_ir_call(J, IRCALL_lj_math_random_step, lj_ir_kptr(J, uddata(ud)));
+ tr = lj_ir_call(J, IRCALL_lj_prng_u64d, lj_ir_kptr(J, uddata(ud)));
one = lj_ir_knum_one(J);
tr = emitir(IRTN(IR_SUB), tr, one);
if (J->base[0]) {
@@ -738,7 +770,7 @@
#if LJ_HASFFI
TRef hdr = recff_bufhdr(J);
TRef tr = recff_bit64_tohex(J, rd, hdr);
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
#else
recff_nyiu(J, rd); /* Don't bother working around this NYI. */
#endif
@@ -864,8 +896,10 @@
if (i > 1) { /* Concatenate the strings, if there's more than one. */
TRef hdr = recff_bufhdr(J), tr = hdr;
for (i = 0; J->base[i] != 0; i++)
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr, J->base[i]);
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr, J->base[i]);
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
+ } else if (i == 0) {
+ J->base[0] = lj_ir_kstr(J, &J2G(J)->strempty);
}
UNUSED(rd);
}
@@ -881,19 +915,19 @@
emitir(IRTGI(vrep > 1 ? IR_GT : IR_LE), rep, lj_ir_kint(J, 1));
if (vrep > 1) {
TRef hdr2 = recff_bufhdr(J);
- TRef tr2 = emitir(IRT(IR_BUFPUT, IRT_PGC), hdr2, sep);
- tr2 = emitir(IRT(IR_BUFPUT, IRT_PGC), tr2, str);
- str2 = emitir(IRT(IR_BUFSTR, IRT_STR), tr2, hdr2);
+ TRef tr2 = emitir(IRTG(IR_BUFPUT, IRT_PGC), hdr2, sep);
+ tr2 = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr2, str);
+ str2 = emitir(IRTG(IR_BUFSTR, IRT_STR), tr2, hdr2);
}
}
tr = hdr = recff_bufhdr(J);
if (str2) {
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr, str);
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr, str);
str = str2;
rep = emitir(IRTI(IR_ADD), rep, lj_ir_kint(J, -1));
}
tr = lj_ir_call(J, IRCALL_lj_buf_putstr_rep, tr, str, rep);
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
}
static void LJ_FASTCALL recff_string_op(jit_State *J, RecordFFData *rd)
@@ -901,7 +935,7 @@
TRef str = lj_ir_tostr(J, J->base[0]);
TRef hdr = recff_bufhdr(J);
TRef tr = lj_ir_call(J, rd->data, hdr, str);
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
}
static void LJ_FASTCALL recff_string_find(jit_State *J, RecordFFData *rd)
@@ -949,7 +983,8 @@
str->len-(MSize)start, pat->len)) {
TRef pos;
emitir(IRTG(IR_NE, IRT_PGC), tr, trp0);
- pos = emitir(IRTI(IR_SUB), tr, emitir(IRT(IR_STRREF, IRT_PGC), trstr, tr0));
+ /* Recompute offset. trsptr may not point into trstr after folding. */
+ pos = emitir(IRTI(IR_ADD), emitir(IRTI(IR_SUB), tr, trsptr), trstart);
J->base[0] = emitir(IRTI(IR_ADD), pos, lj_ir_kint(J, 1));
J->base[1] = emitir(IRTI(IR_ADD), pos, trplen);
rd->nres = 2;
@@ -963,34 +998,41 @@
}
}
-static void LJ_FASTCALL recff_string_format(jit_State *J, RecordFFData *rd)
+static void recff_format(jit_State *J, RecordFFData *rd, TRef hdr, int sbufx)
{
- TRef trfmt = lj_ir_tostr(J, J->base[0]);
- GCstr *fmt = argv2str(J, &rd->argv[0]);
- int arg = 1;
- TRef hdr, tr;
+ ptrdiff_t arg = sbufx;
+ TRef tr = hdr, trfmt = lj_ir_tostr(J, J->base[arg]);
+ GCstr *fmt = argv2str(J, &rd->argv[arg]);
FormatState fs;
SFormat sf;
+ int nfmt = 0;
/* Specialize to the format string. */
emitir(IRTG(IR_EQ, IRT_STR), trfmt, lj_ir_kstr(J, fmt));
- tr = hdr = recff_bufhdr(J);
lj_strfmt_init(&fs, strdata(fmt), fmt->len);
while ((sf = lj_strfmt_parse(&fs)) != STRFMT_EOF) { /* Parse format. */
- TRef tra = sf == STRFMT_LIT ? 0 : J->base[arg++];
+ TRef tra = sf == STRFMT_LIT ? 0 : J->base[++arg];
TRef trsf = lj_ir_kint(J, (int32_t)sf);
IRCallID id;
switch (STRFMT_TYPE(sf)) {
case STRFMT_LIT:
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr,
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr,
lj_ir_kstr(J, lj_str_new(J->L, fs.str, fs.len)));
break;
case STRFMT_INT:
id = IRCALL_lj_strfmt_putfnum_int;
handle_int:
- if (!tref_isinteger(tra))
+ if (!tref_isinteger(tra)) {
+#if LJ_HASFFI
+ if (tref_iscdata(tra)) {
+ tra = lj_crecord_loadiu64(J, tra, &rd->argv[arg]);
+ tr = lj_ir_call(J, IRCALL_lj_strfmt_putfxint, tr, trsf, tra);
+ break;
+ }
+#endif
goto handle_num;
+ }
if (sf == STRFMT_INT) { /* Shortcut for plain %d. */
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr,
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr,
emitir(IRT(IR_TOSTR, IRT_STR), tra, IRTOSTR_INT));
} else {
#if LJ_HASFFI
@@ -1012,15 +1054,16 @@
handle_num:
tra = lj_ir_tonum(J, tra);
tr = lj_ir_call(J, id, tr, trsf, tra);
- if (LJ_SOFTFP) lj_needsplit(J);
+ if (LJ_SOFTFP32) lj_needsplit(J);
break;
case STRFMT_STR:
if (!tref_isstr(tra)) {
recff_nyiu(J, rd); /* NYI: __tostring and non-string types for %s. */
+ /* NYI: also buffers. */
return;
}
if (sf == STRFMT_STR) /* Shortcut for plain %s. */
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr, tra);
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr, tra);
else if ((sf & STRFMT_T_QUOTED))
tr = lj_ir_call(J, IRCALL_lj_strfmt_putquoted, tr, tra);
else
@@ -1029,7 +1072,7 @@
case STRFMT_CHAR:
tra = lj_opt_narrow_toint(J, tra);
if (sf == STRFMT_CHAR) /* Shortcut for plain %c. */
- tr = emitir(IRT(IR_BUFPUT, IRT_PGC), tr,
+ tr = emitir(IRTG(IR_BUFPUT, IRT_PGC), tr,
emitir(IRT(IR_TOSTR, IRT_STR), tra, IRTOSTR_CHAR));
else
tr = lj_ir_call(J, IRCALL_lj_strfmt_putfchar, tr, trsf, tra);
@@ -1040,10 +1083,341 @@
recff_nyiu(J, rd);
return;
}
+ if (++nfmt > 100) lj_trace_err(J, LJ_TRERR_TRACEOV);
}
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ if (sbufx) {
+ emitir(IRT(IR_USE, IRT_NIL), tr, 0);
+ } else {
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
+ }
}
+static void LJ_FASTCALL recff_string_format(jit_State *J, RecordFFData *rd)
+{
+ recff_format(J, rd, recff_bufhdr(J), 0);
+}
+
+/* -- Buffer library fast functions --------------------------------------- */
+
+#if LJ_HASBUFFER
+
+static LJ_AINLINE TRef recff_sbufx_get_L(jit_State *J, TRef ud)
+{
+ return emitir(IRT(IR_FLOAD, IRT_PGC), ud, IRFL_SBUF_L);
+}
+
+static LJ_AINLINE void recff_sbufx_set_L(jit_State *J, TRef ud, TRef val)
+{
+ TRef fref = emitir(IRT(IR_FREF, IRT_PGC), ud, IRFL_SBUF_L);
+ emitir(IRT(IR_FSTORE, IRT_PGC), fref, val);
+}
+
+static LJ_AINLINE TRef recff_sbufx_get_ptr(jit_State *J, TRef ud, IRFieldID fl)
+{
+ return emitir(IRT(IR_FLOAD, IRT_PTR), ud, fl);
+}
+
+static LJ_AINLINE void recff_sbufx_set_ptr(jit_State *J, TRef ud, IRFieldID fl, TRef val)
+{
+ TRef fref = emitir(IRT(IR_FREF, IRT_PTR), ud, fl);
+ emitir(IRT(IR_FSTORE, IRT_PTR), fref, val);
+}
+
+static LJ_AINLINE TRef recff_sbufx_len(jit_State *J, TRef trr, TRef trw)
+{
+ TRef len = emitir(IRT(IR_SUB, IRT_INTP), trw, trr);
+ if (LJ_64)
+ len = emitir(IRTI(IR_CONV), len, (IRT_INT<<5)|IRT_INTP|IRCONV_NONE);
+ return len;
+}
+
+/* Emit typecheck for string buffer. */
+static TRef recff_sbufx_check(jit_State *J, RecordFFData *rd, ptrdiff_t arg)
+{
+ TRef trtype, ud = J->base[arg];
+ if (!tvisbuf(&rd->argv[arg])) lj_trace_err(J, LJ_TRERR_BADTYPE);
+ trtype = emitir(IRT(IR_FLOAD, IRT_U8), ud, IRFL_UDATA_UDTYPE);
+ emitir(IRTGI(IR_EQ), trtype, lj_ir_kint(J, UDTYPE_BUFFER));
+ J->needsnap = 1;
+ return ud;
+}
+
+/* Emit BUFHDR for write to extended string buffer. */
+static TRef recff_sbufx_write(jit_State *J, TRef ud)
+{
+ TRef trbuf = emitir(IRT(IR_ADD, IRT_PGC), ud, lj_ir_kintpgc(J, sizeof(GCudata)));
+ return emitir(IRT(IR_BUFHDR, IRT_PGC), trbuf, IRBUFHDR_WRITE);
+}
+
+/* Check for integer in range for the buffer API. */
+static TRef recff_sbufx_checkint(jit_State *J, RecordFFData *rd, ptrdiff_t arg)
+{
+ TRef tr = J->base[arg];
+ TRef trlim = lj_ir_kint(J, LJ_MAX_BUF);
+ if (tref_isinteger(tr)) {
+ emitir(IRTGI(IR_ULE), tr, trlim);
+ } else if (tref_isnum(tr)) {
+ tr = emitir(IRTI(IR_CONV), tr, IRCONV_INT_NUM|IRCONV_ANY);
+ emitir(IRTGI(IR_ULE), tr, trlim);
+#if LJ_HASFFI
+ } else if (tref_iscdata(tr)) {
+ tr = lj_crecord_loadiu64(J, tr, &rd->argv[arg]);
+ emitir(IRTG(IR_ULE, IRT_U64), tr, lj_ir_kint64(J, LJ_MAX_BUF));
+ tr = emitir(IRTI(IR_CONV), tr, (IRT_INT<<5)|IRT_I64|IRCONV_NONE);
+#else
+ UNUSED(rd);
+#endif
+ } else {
+ lj_trace_err(J, LJ_TRERR_BADTYPE);
+ }
+ return tr;
+}
+
+static void LJ_FASTCALL recff_buffer_method_reset(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ SBufExt *sbx = bufV(&rd->argv[0]);
+ int iscow = (int)sbufiscow(sbx);
+ TRef trl = recff_sbufx_get_L(J, ud);
+ TRef trcow = emitir(IRT(IR_BAND, IRT_IGC), trl, lj_ir_kintpgc(J, SBUF_FLAG_COW));
+ TRef zeropgc = lj_ir_kintpgc(J, 0);
+ emitir(IRTG(iscow ? IR_NE : IR_EQ, IRT_IGC), trcow, zeropgc);
+ if (iscow) {
+ TRef zerop = lj_ir_kintp(J, 0);
+ trl = emitir(IRT(IR_BXOR, IRT_IGC), trl, lj_ir_kintpgc(J, SBUF_FLAG_COW));
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_W, zerop);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_E, zerop);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_B, zerop);
+ recff_sbufx_set_L(J, ud, trl);
+ emitir(IRT(IR_FSTORE, IRT_PGC),
+ emitir(IRT(IR_FREF, IRT_PGC), ud, IRFL_SBUF_REF), zeropgc);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_R, zerop);
+ } else {
+ TRef trb = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_B);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_W, trb);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_R, trb);
+ }
+}
+
+static void LJ_FASTCALL recff_buffer_method_skip(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trr = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ TRef len = recff_sbufx_len(J, trr, trw);
+ TRef trn = recff_sbufx_checkint(J, rd, 1);
+ len = emitir(IRTI(IR_MIN), len, trn);
+ trr = emitir(IRT(IR_ADD, IRT_PTR), trr, len);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_R, trr);
+}
+
+static void LJ_FASTCALL recff_buffer_method_set(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef tr = J->base[1];
+ if (tref_isstr(tr)) {
+ TRef trp = emitir(IRT(IR_STRREF, IRT_PGC), tr, lj_ir_kint(J, 0));
+ TRef len = emitir(IRTI(IR_FLOAD), tr, IRFL_STR_LEN);
+ IRIns *irp = IR(tref_ref(trp));
+ /* trp must point into the anchored obj, even after folding. */
+ if (irp->o == IR_STRREF)
+ tr = irp->op1;
+ else if (!tref_isk(tr))
+ trp = emitir(IRT(IR_ADD, IRT_PGC), tr, lj_ir_kintpgc(J, sizeof(GCstr)));
+ lj_ir_call(J, IRCALL_lj_bufx_set, trbuf, trp, len, tr);
+#if LJ_HASFFI
+ } else if (tref_iscdata(tr)) {
+ TRef trp = lj_crecord_topcvoid(J, tr, &rd->argv[1]);
+ TRef len = recff_sbufx_checkint(J, rd, 2);
+ lj_ir_call(J, IRCALL_lj_bufx_set, trbuf, trp, len, tr);
+#endif
+ } /* else: Interpreter will throw. */
+}
+
+static void LJ_FASTCALL recff_buffer_method_put(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef tr;
+ ptrdiff_t arg;
+ if (!J->base[1]) return;
+ for (arg = 1; (tr = J->base[arg]); arg++) {
+ if (tref_isudata(tr)) {
+ TRef ud2 = recff_sbufx_check(J, rd, arg);
+ emitir(IRTG(IR_NE, IRT_PGC), ud, ud2);
+ }
+ }
+ for (arg = 1; (tr = J->base[arg]); arg++) {
+ if (tref_isstr(tr)) {
+ trbuf = emitir(IRTG(IR_BUFPUT, IRT_PGC), trbuf, tr);
+ } else if (tref_isnumber(tr)) {
+ trbuf = emitir(IRTG(IR_BUFPUT, IRT_PGC), trbuf,
+ emitir(IRT(IR_TOSTR, IRT_STR), tr,
+ tref_isnum(tr) ? IRTOSTR_NUM : IRTOSTR_INT));
+ } else if (tref_isudata(tr)) {
+ TRef trr = recff_sbufx_get_ptr(J, tr, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, tr, IRFL_SBUF_W);
+ TRef len = recff_sbufx_len(J, trr, trw);
+ trbuf = lj_ir_call(J, IRCALL_lj_buf_putmem, trbuf, trr, len);
+ } else {
+ recff_nyiu(J, rd);
+ }
+ }
+ emitir(IRT(IR_USE, IRT_NIL), trbuf, 0);
+}
+
+static void LJ_FASTCALL recff_buffer_method_putf(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ recff_format(J, rd, trbuf, 1);
+}
+
+static void LJ_FASTCALL recff_buffer_method_get(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trr = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ TRef tr;
+ ptrdiff_t arg;
+ if (!J->base[1]) { J->base[1] = TREF_NIL; J->base[2] = 0; }
+ for (arg = 0; (tr = J->base[arg+1]); arg++) {
+ if (!tref_isnil(tr)) {
+ J->base[arg+1] = recff_sbufx_checkint(J, rd, arg+1);
+ }
+ }
+ for (arg = 0; (tr = J->base[arg+1]); arg++) {
+ TRef len = recff_sbufx_len(J, trr, trw);
+ if (tref_isnil(tr)) {
+ J->base[arg] = emitir(IRT(IR_XSNEW, IRT_STR), trr, len);
+ trr = trw;
+ } else {
+ TRef tru;
+ len = emitir(IRTI(IR_MIN), len, tr);
+ tru = emitir(IRT(IR_ADD, IRT_PTR), trr, len);
+ J->base[arg] = emitir(IRT(IR_XSNEW, IRT_STR), trr, len);
+ trr = tru; /* Doing the ADD before the SNEW generates better code. */
+ }
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_R, trr);
+ }
+ rd->nres = arg;
+}
+
+static void LJ_FASTCALL recff_buffer_method___tostring(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trr = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ J->base[0] = emitir(IRT(IR_XSNEW, IRT_STR), trr, recff_sbufx_len(J, trr, trw));
+}
+
+static void LJ_FASTCALL recff_buffer_method___len(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trr = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ J->base[0] = recff_sbufx_len(J, trr, trw);
+}
+
+#if LJ_HASFFI
+static void LJ_FASTCALL recff_buffer_method_putcdata(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef tr = lj_crecord_topcvoid(J, J->base[1], &rd->argv[1]);
+ TRef len = recff_sbufx_checkint(J, rd, 2);
+ trbuf = lj_ir_call(J, IRCALL_lj_buf_putmem, trbuf, tr, len);
+ emitir(IRT(IR_USE, IRT_NIL), trbuf, 0);
+}
+
+static void LJ_FASTCALL recff_buffer_method_reserve(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef trsz = recff_sbufx_checkint(J, rd, 1);
+ J->base[1] = lj_ir_call(J, IRCALL_lj_bufx_more, trbuf, trsz);
+ J->base[0] = lj_crecord_topuint8(J, recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W));
+ rd->nres = 2;
+}
+
+static void LJ_FASTCALL recff_buffer_method_commit(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef len = recff_sbufx_checkint(J, rd, 1);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ TRef tre = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_E);
+ TRef left = emitir(IRT(IR_SUB, IRT_INTP), tre, trw);
+ if (LJ_64)
+ left = emitir(IRTI(IR_CONV), left, (IRT_INT<<5)|IRT_INTP|IRCONV_NONE);
+ emitir(IRTGI(IR_ULE), len, left);
+ trw = emitir(IRT(IR_ADD, IRT_PTR), trw, len);
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_W, trw);
+}
+
+static void LJ_FASTCALL recff_buffer_method_ref(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trr = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_R);
+ TRef trw = recff_sbufx_get_ptr(J, ud, IRFL_SBUF_W);
+ J->base[0] = lj_crecord_topuint8(J, trr);
+ J->base[1] = recff_sbufx_len(J, trr, trw);
+ rd->nres = 2;
+}
+#endif
+
+static void LJ_FASTCALL recff_buffer_method_encode(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef tmp = recff_tmpref(J, J->base[1], IRTMPREF_IN1);
+ lj_ir_call(J, IRCALL_lj_serialize_put, trbuf, tmp);
+ /* No IR_USE needed, since the call is a store. */
+}
+
+static void LJ_FASTCALL recff_buffer_method_decode(jit_State *J, RecordFFData *rd)
+{
+ TRef ud = recff_sbufx_check(J, rd, 0);
+ TRef trbuf = recff_sbufx_write(J, ud);
+ TRef tmp = recff_tmpref(J, TREF_NIL, IRTMPREF_OUT1);
+ TRef trr = lj_ir_call(J, IRCALL_lj_serialize_get, trbuf, tmp);
+ IRType t = (IRType)lj_serialize_peektype(bufV(&rd->argv[0]));
+ /* No IR_USE needed, since the call is a store. */
+ J->base[0] = lj_record_vload(J, tmp, 0, t);
+ /* The sbx->r store must be after the VLOAD type check, in case it fails. */
+ recff_sbufx_set_ptr(J, ud, IRFL_SBUF_R, trr);
+}
+
+static void LJ_FASTCALL recff_buffer_encode(jit_State *J, RecordFFData *rd)
+{
+ TRef tmp = recff_tmpref(J, J->base[0], IRTMPREF_IN1);
+ J->base[0] = lj_ir_call(J, IRCALL_lj_serialize_encode, tmp);
+ /* IR_USE needed for IR_CALLA, because the encoder may throw non-OOM. */
+ emitir(IRT(IR_USE, IRT_NIL), J->base[0], 0);
+ UNUSED(rd);
+}
+
+static void LJ_FASTCALL recff_buffer_decode(jit_State *J, RecordFFData *rd)
+{
+ if (tvisstr(&rd->argv[0])) {
+ GCstr *str = strV(&rd->argv[0]);
+ SBufExt sbx;
+ IRType t;
+ TRef tmp = recff_tmpref(J, TREF_NIL, IRTMPREF_OUT1);
+ TRef tr = lj_ir_call(J, IRCALL_lj_serialize_decode, tmp, J->base[0]);
+ /* IR_USE needed for IR_CALLA, because the decoder may throw non-OOM.
+ ** That's why IRCALL_lj_serialize_decode needs a fake INT result.
+ */
+ emitir(IRT(IR_USE, IRT_NIL), tr, 0);
+ memset(&sbx, 0, sizeof(SBufExt));
+ lj_bufx_set_cow(J->L, &sbx, strdata(str), str->len);
+ t = (IRType)lj_serialize_peektype(&sbx);
+ J->base[0] = lj_record_vload(J, tmp, 0, t);
+ } /* else: Interpreter will throw. */
+}
+
+#endif
+
/* -- Table library fast functions ---------------------------------------- */
static void LJ_FASTCALL recff_table_insert(jit_State *J, RecordFFData *rd)
@@ -1054,7 +1428,7 @@
rd->nres = 0;
if (tref_istab(ix.tab) && ix.val) {
if (!J->base[2]) { /* Simple push: t[#t+1] = v */
- TRef trlen = lj_ir_call(J, IRCALL_lj_tab_len, ix.tab);
+ TRef trlen = emitir(IRTI(IR_ALEN), ix.tab, TREF_NIL);
GCtab *t = tabV(&rd->argv[0]);
ix.key = emitir(IRTI(IR_ADD), trlen, lj_ir_kint(J, 1));
settabV(J->L, &ix.tabv, t);
@@ -1078,11 +1452,11 @@
lj_opt_narrow_toint(J, J->base[2]) : lj_ir_kint(J, 1);
TRef tre = (J->base[1] && J->base[2] && !tref_isnil(J->base[3])) ?
lj_opt_narrow_toint(J, J->base[3]) :
- lj_ir_call(J, IRCALL_lj_tab_len, tab);
+ emitir(IRTI(IR_ALEN), tab, TREF_NIL);
TRef hdr = recff_bufhdr(J);
TRef tr = lj_ir_call(J, IRCALL_lj_buf_puttab, hdr, tab, sep, tri, tre);
emitir(IRTG(IR_NE, IRT_PTR), tr, lj_ir_kptr(J, NULL));
- J->base[0] = emitir(IRT(IR_BUFSTR, IRT_STR), tr, hdr);
+ J->base[0] = emitir(IRTG(IR_BUFSTR, IRT_STR), tr, hdr);
} /* else: Interpreter will throw. */
UNUSED(rd);
}
@@ -1091,6 +1465,15 @@
{
TRef tra = lj_opt_narrow_toint(J, J->base[0]);
TRef trh = lj_opt_narrow_toint(J, J->base[1]);
+ if (tref_isk(tra) && tref_isk(trh)) {
+ int32_t a = IR(tref_ref(tra))->i;
+ if (a < 0x7fff) {
+ uint32_t hbits = hsize2hbits(IR(tref_ref(trh))->i);
+ a = a > 0 ? a+1 : 0;
+ J->base[0] = emitir(IRTG(IR_TNEW, IRT_TAB), (uint32_t)a, hbits);
+ return;
+ }
+ }
J->base[0] = lj_ir_call(J, IRCALL_lj_tab_new_ah, tra, trh);
UNUSED(rd);
}
@@ -1114,13 +1497,7 @@
{
TRef tr, ud, fp;
if (id) { /* io.func() */
-#if LJ_GC64
- /* TODO: fix ARM32 asm_fload(), so we can use this for all archs. */
ud = lj_ir_ggfload(J, IRT_UDATA, GG_OFS(g.gcroot[id]));
-#else
- tr = lj_ir_kptr(J, &J2G(J)->gcroot[id]);
- ud = emitir(IRT(IR_XLOAD, IRT_UDATA), tr, 0);
-#endif
} else { /* fp:method() */
ud = J->base[0];
if (!tref_isudata(ud))
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ffrecord.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Fast function call recorder.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_FFRECORD_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_frame.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_frame.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_frame.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Stack frames.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_FRAME_H
@@ -46,7 +46,7 @@
#define frame_gc(f) (gcval((f)-1))
#define frame_ftsz(f) ((ptrdiff_t)(f)->ftsz)
#define frame_pc(f) ((const BCIns *)frame_ftsz(f))
-#define setframe_gc(f, p, tp) (setgcVraw((f)-1, (p), (tp)))
+#define setframe_gc(f, p, tp) (setgcVraw((f), (p), (tp)))
#define setframe_ftsz(f, sz) ((f)->ftsz = (sz))
#define setframe_pc(f, pc) ((f)->ftsz = (int64_t)(intptr_t)(pc))
#else
@@ -192,12 +192,12 @@
#endif
#define CFRAME_SHIFT_MULTRES 3
#elif LJ_TARGET_ARM64
-#define CFRAME_OFS_ERRF 196
-#define CFRAME_OFS_NRES 200
-#define CFRAME_OFS_PREV 160
-#define CFRAME_OFS_L 176
-#define CFRAME_OFS_PC 168
-#define CFRAME_OFS_MULTRES 192
+#define CFRAME_OFS_ERRF 36
+#define CFRAME_OFS_NRES 40
+#define CFRAME_OFS_PREV 0
+#define CFRAME_OFS_L 16
+#define CFRAME_OFS_PC 8
+#define CFRAME_OFS_MULTRES 32
#define CFRAME_SIZE 208
#define CFRAME_SHIFT_MULTRES 3
#elif LJ_TARGET_PPC
@@ -226,7 +226,7 @@
#define CFRAME_OFS_L 36
#define CFRAME_OFS_PC 32
#define CFRAME_OFS_MULTRES 28
-#define CFRAME_SIZE 272
+#define CFRAME_SIZE (LJ_ARCH_HASFPU ? 272 : 128)
#define CFRAME_SHIFT_MULTRES 3
#endif
#elif LJ_TARGET_MIPS32
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Function handling (prototypes, functions and upvalues).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -24,9 +24,11 @@
/* -- Upvalues ------------------------------------------------------------ */
-static void unlinkuv(GCupval *uv)
+static void unlinkuv(global_State *g, GCupval *uv)
{
- lua_assert(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv);
+ UNUSED(g);
+ lj_assertG(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv,
+ "broken upvalue chain");
setgcrefr(uvnext(uv)->prev, uv->prev);
setgcrefr(uvprev(uv)->next, uv->next);
}
@@ -40,7 +42,7 @@
GCupval *uv;
/* Search the sorted list of open upvalues. */
while (gcref(*pp) != NULL && uvval((p = gco2uv(gcref(*pp)))) >= slot) {
- lua_assert(!p->closed && uvval(p) != &p->tv);
+ lj_assertG(!p->closed && uvval(p) != &p->tv, "closed upvalue in chain");
if (uvval(p) == slot) { /* Found open upvalue pointing to same slot? */
if (isdead(g, obj2gco(p))) /* Resurrect it, if it's dead. */
flipwhite(obj2gco(p));
@@ -61,7 +63,8 @@
setgcrefr(uv->next, g->uvhead.next);
setgcref(uvnext(uv)->prev, obj2gco(uv));
setgcref(g->uvhead.next, obj2gco(uv));
- lua_assert(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv);
+ lj_assertG(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv,
+ "broken upvalue chain");
return uv;
}
@@ -84,12 +87,13 @@
while (gcref(L->openupval) != NULL &&
uvval((uv = gco2uv(gcref(L->openupval)))) >= level) {
GCobj *o = obj2gco(uv);
- lua_assert(!isblack(o) && !uv->closed && uvval(uv) != &uv->tv);
+ lj_assertG(!isblack(o), "bad black upvalue");
+ lj_assertG(!uv->closed && uvval(uv) != &uv->tv, "closed upvalue in chain");
setgcrefr(L->openupval, uv->nextgc); /* No longer in open list. */
if (isdead(g, o)) {
lj_func_freeuv(g, uv);
} else {
- unlinkuv(uv);
+ unlinkuv(g, uv);
lj_gc_closeuv(g, uv);
}
}
@@ -98,7 +102,7 @@
void LJ_FASTCALL lj_func_freeuv(global_State *g, GCupval *uv)
{
if (!uv->closed)
- unlinkuv(uv);
+ unlinkuv(g, uv);
lj_mem_freet(g, uv);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_func.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Function handling (prototypes, functions and upvalues).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_FUNC_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Garbage collector.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -25,7 +25,9 @@
#include "lj_cdata.h"
#endif
#include "lj_trace.h"
+#include "lj_dispatch.h"
#include "lj_vm.h"
+#include "lj_vmevent.h"
#define GCSTEPSIZE 1024u
#define GCSWEEPMAX 40
@@ -41,7 +43,8 @@
/* Mark a TValue (if needed). */
#define gc_marktv(g, tv) \
- { lua_assert(!tvisgcv(tv) || (~itype(tv) == gcval(tv)->gch.gct)); \
+ { lj_assertG(!tvisgcv(tv) || (~itype(tv) == gcval(tv)->gch.gct), \
+ "TValue and GC type mismatch"); \
if (tviswhite(tv)) gc_mark(g, gcV(tv)); }
/* Mark a GCobj (if needed). */
@@ -55,7 +58,8 @@
static void gc_mark(global_State *g, GCobj *o)
{
int gct = o->gch.gct;
- lua_assert(iswhite(o) && !isdead(g, o));
+ lj_assertG(iswhite(o), "mark of non-white object");
+ lj_assertG(!isdead(g, o), "mark of dead object");
white2gray(o);
if (LJ_UNLIKELY(gct == ~LJ_TUDATA)) {
GCtab *mt = tabref(gco2ud(o)->metatable);
@@ -62,6 +66,15 @@
gray2black(o); /* Userdata are never gray. */
if (mt) gc_markobj(g, mt);
gc_markobj(g, tabref(gco2ud(o)->env));
+ if (LJ_HASBUFFER && gco2ud(o)->udtype == UDTYPE_BUFFER) {
+ SBufExt *sbx = (SBufExt *)uddata(gco2ud(o));
+ if (sbufiscow(sbx) && gcref(sbx->cowref))
+ gc_markobj(g, gcref(sbx->cowref));
+ if (gcref(sbx->dict_str))
+ gc_markobj(g, gcref(sbx->dict_str));
+ if (gcref(sbx->dict_mt))
+ gc_markobj(g, gcref(sbx->dict_mt));
+ }
} else if (LJ_UNLIKELY(gct == ~LJ_TUPVAL)) {
GCupval *uv = gco2uv(o);
gc_marktv(g, uvval(uv));
@@ -68,8 +81,9 @@
if (uv->closed)
gray2black(o); /* Closed upvalues are never gray. */
} else if (gct != ~LJ_TSTR && gct != ~LJ_TCDATA) {
- lua_assert(gct == ~LJ_TFUNC || gct == ~LJ_TTAB ||
- gct == ~LJ_TTHREAD || gct == ~LJ_TPROTO || gct == ~LJ_TTRACE);
+ lj_assertG(gct == ~LJ_TFUNC || gct == ~LJ_TTAB ||
+ gct == ~LJ_TTHREAD || gct == ~LJ_TPROTO || gct == ~LJ_TTRACE,
+ "bad GC type %d", gct);
setgcrefr(o->gch.gclist, g->gc.gray);
setgcref(g->gc.gray, o);
}
@@ -102,7 +116,8 @@
{
GCupval *uv;
for (uv = uvnext(&g->uvhead); uv != &g->uvhead; uv = uvnext(uv)) {
- lua_assert(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv);
+ lj_assertG(uvprev(uvnext(uv)) == uv && uvnext(uvprev(uv)) == uv,
+ "broken upvalue chain");
if (isgray(obj2gco(uv)))
gc_marktv(g, uvval(uv));
}
@@ -172,8 +187,7 @@
}
if (weak) { /* Weak tables are cleared in the atomic phase. */
#if LJ_HASFFI
- CTState *cts = ctype_ctsG(g);
- if (cts && cts->finalizer == t) {
+ if (gcref(g->gcroot[GCROOT_FFI_FIN]) == obj2gco(t)) {
weak = (int)(~0u & ~LJ_GC_WEAKVAL);
} else
#endif
@@ -197,7 +211,7 @@
for (i = 0; i <= hmask; i++) {
Node *n = &node[i];
if (!tvisnil(&n->val)) { /* Mark non-empty slot. */
- lua_assert(!tvisnil(&n->key));
+ lj_assertG(!tvisnil(&n->key), "mark of nil key in non-empty slot");
if (!(weak & LJ_GC_WEAKKEY)) gc_marktv(g, &n->key);
if (!(weak & LJ_GC_WEAKVAL)) gc_marktv(g, &n->val);
}
@@ -212,7 +226,8 @@
gc_markobj(g, tabref(fn->c.env));
if (isluafunc(fn)) {
uint32_t i;
- lua_assert(fn->l.nupvalues <= funcproto(fn)->sizeuv);
+ lj_assertG(fn->l.nupvalues <= funcproto(fn)->sizeuv,
+ "function upvalues out of range");
gc_markobj(g, funcproto(fn));
for (i = 0; i < fn->l.nupvalues; i++) /* Mark Lua function upvalues. */
gc_markobj(g, &gcref(fn->l.uvptr[i])->uv);
@@ -228,7 +243,7 @@
static void gc_marktrace(global_State *g, TraceNo traceno)
{
GCobj *o = obj2gco(traceref(G2J(g), traceno));
- lua_assert(traceno != G2J(g)->cur.traceno);
+ lj_assertG(traceno != G2J(g)->cur.traceno, "active trace escaped");
if (iswhite(o)) {
white2gray(o);
setgcrefr(o->gch.gclist, g->gc.gray);
@@ -309,7 +324,7 @@
{
GCobj *o = gcref(g->gc.gray);
int gct = o->gch.gct;
- lua_assert(isgray(o));
+ lj_assertG(isgray(o), "propagation of non-gray object");
gray2black(o);
setgcrefr(g->gc.gray, o->gch.gclist); /* Remove from gray list. */
if (LJ_LIKELY(gct == ~LJ_TTAB)) {
@@ -341,7 +356,7 @@
return ((sizeof(GCtrace)+7)&~7) + (T->nins-T->nk)*sizeof(IRIns) +
T->nsnap*sizeof(SnapShot) + T->nsnapmap*sizeof(SnapEntry);
#else
- lua_assert(0);
+ lj_assertG(0, "bad GC type %d", gct);
return 0;
#endif
}
@@ -395,11 +410,13 @@
if (o->gch.gct == ~LJ_TTHREAD) /* Need to sweep open upvalues, too. */
gc_fullsweep(g, &gco2th(o)->openupval);
if (((o->gch.marked ^ LJ_GC_WHITES) & ow)) { /* Black or current white? */
- lua_assert(!isdead(g, o) || (o->gch.marked & LJ_GC_FIXED));
+ lj_assertG(!isdead(g, o) || (o->gch.marked & LJ_GC_FIXED),
+ "sweep of undead object");
makewhite(g, o); /* Value is alive, change to the current white. */
p = &o->gch.nextgc;
} else { /* Otherwise value is dead, free it. */
- lua_assert(isdead(g, o) || ow == LJ_GC_SFIXED);
+ lj_assertG(isdead(g, o) || ow == LJ_GC_SFIXED,
+ "sweep of unlive object");
setgcrefr(*p, o->gch.nextgc);
if (o == gcref(g->gc.root))
setgcrefr(g->gc.root, o->gch.nextgc); /* Adjust list anchor. */
@@ -409,6 +426,32 @@
return p;
}
+/* Sweep one string interning table chain. Preserves hashalg bit. */
+static void gc_sweepstr(global_State *g, GCRef *chain)
+{
+ /* Mask with other white and LJ_GC_FIXED. Or LJ_GC_SFIXED on shutdown. */
+ int ow = otherwhite(g);
+ uintptr_t u = gcrefu(*chain);
+ GCRef q;
+ GCRef *p = &q;
+ GCobj *o;
+ setgcrefp(q, (u & ~(uintptr_t)1));
+ while ((o = gcref(*p)) != NULL) {
+ if (((o->gch.marked ^ LJ_GC_WHITES) & ow)) { /* Black or current white? */
+ lj_assertG(!isdead(g, o) || (o->gch.marked & LJ_GC_FIXED),
+ "sweep of undead string");
+ makewhite(g, o); /* String is alive, change to the current white. */
+ p = &o->gch.nextgc;
+ } else { /* Otherwise string is dead, free it. */
+ lj_assertG(isdead(g, o) || ow == LJ_GC_SFIXED,
+ "sweep of unlive string");
+ setgcrefr(*p, o->gch.nextgc);
+ lj_str_free(g, gco2str(o));
+ }
+ }
+ setgcrefp(*chain, (gcrefu(q) | (u & 1)));
+}
+
/* Check whether we can clear a key or a value slot from a table. */
static int gc_mayclear(cTValue *o, int val)
{
@@ -426,11 +469,12 @@
}
/* Clear collected entries from weak tables. */
-static void gc_clearweak(GCobj *o)
+static void gc_clearweak(global_State *g, GCobj *o)
{
+ UNUSED(g);
while (o) {
GCtab *t = gco2tab(o);
- lua_assert((t->marked & LJ_GC_WEAK));
+ lj_assertG((t->marked & LJ_GC_WEAK), "clear of non-weak table");
if ((t->marked & LJ_GC_WEAKVAL)) {
MSize i, asize = t->asize;
for (i = 0; i < asize; i++) {
@@ -466,6 +510,7 @@
TValue *top;
lj_trace_abort(g);
hook_entergc(g); /* Disable hooks and new traces during __gc. */
+ if (LJ_HASPROFILE && (oldh & HOOK_PROFILE)) lj_dispatch_update(g);
g->gc.threshold = LJ_MAX_MEM; /* Prevent GC steps. */
top = L->top;
copyTV(L, top++, mo);
@@ -474,9 +519,15 @@
L->top = top+1;
errcode = lj_vm_pcall(L, top, 1+0, -1); /* Stack: |mo|o| -> | */
hook_restore(g, oldh);
+ if (LJ_HASPROFILE && (oldh & HOOK_PROFILE)) lj_dispatch_update(g);
g->gc.threshold = oldt; /* Restore GC threshold. */
- if (errcode)
- lj_err_throw(L, errcode); /* Propagate errors. */
+ if (errcode) {
+ ptrdiff_t errobj = savestack(L, L->top-1); /* Stack may be resized. */
+ lj_vmevent_send(L, ERRFIN,
+ copyTV(L, L->top++, restorestack(L, errobj));
+ );
+ L->top--;
+ }
}
/* Finalize one userdata or cdata object from the mmudata list. */
@@ -485,7 +536,7 @@
global_State *g = G(L);
GCobj *o = gcnext(gcref(g->gc.mmudata));
cTValue *mo;
- lua_assert(tvref(g->jit_base) == NULL); /* Must not be called on trace. */
+ lj_assertG(tvref(g->jit_base) == NULL, "finalizer called on trace");
/* Unchain from list of userdata to be finalized. */
if (o == gcref(g->gc.mmudata))
setgcrefnull(g->gc.mmudata);
@@ -501,9 +552,8 @@
o->gch.marked &= (uint8_t)~LJ_GC_CDATA_FIN;
/* Resolve finalizer. */
setcdataV(L, &tmp, gco2cd(o));
- tv = lj_tab_set(L, ctype_ctsG(g)->finalizer, &tmp);
+ tv = lj_tab_set(L, tabref(g->gcroot[GCROOT_FFI_FIN]), &tmp);
if (!tvisnil(tv)) {
- g->gc.nocdatafin = 0;
copyTV(L, &tmp, tv);
setnilV(tv); /* Clear entry in finalizer table. */
gc_call_finalizer(g, L, &tmp, o);
@@ -533,23 +583,20 @@
void lj_gc_finalize_cdata(lua_State *L)
{
global_State *g = G(L);
- CTState *cts = ctype_ctsG(g);
- if (cts) {
- GCtab *t = cts->finalizer;
- Node *node = noderef(t->node);
- ptrdiff_t i;
- setgcrefnull(t->metatable); /* Mark finalizer table as disabled. */
- for (i = (ptrdiff_t)t->hmask; i >= 0; i--)
- if (!tvisnil(&node[i].val) && tviscdata(&node[i].key)) {
- GCobj *o = gcV(&node[i].key);
- TValue tmp;
- makewhite(g, o);
- o->gch.marked &= (uint8_t)~LJ_GC_CDATA_FIN;
- copyTV(L, &tmp, &node[i].val);
- setnilV(&node[i].val);
- gc_call_finalizer(g, L, &tmp, o);
- }
- }
+ GCtab *t = tabref(g->gcroot[GCROOT_FFI_FIN]);
+ Node *node = noderef(t->node);
+ ptrdiff_t i;
+ setgcrefnull(t->metatable); /* Mark finalizer table as disabled. */
+ for (i = (ptrdiff_t)t->hmask; i >= 0; i--)
+ if (!tvisnil(&node[i].val) && tviscdata(&node[i].key)) {
+ GCobj *o = gcV(&node[i].key);
+ TValue tmp;
+ makewhite(g, o);
+ o->gch.marked &= (uint8_t)~LJ_GC_CDATA_FIN;
+ copyTV(L, &tmp, &node[i].val);
+ setnilV(&node[i].val);
+ gc_call_finalizer(g, L, &tmp, o);
+ }
}
#endif
@@ -556,13 +603,12 @@
/* Free all remaining GC objects. */
void lj_gc_freeall(global_State *g)
{
- MSize i, strmask;
+ MSize i;
/* Free everything, except super-fixed objects (the main thread). */
g->gc.currentwhite = LJ_GC_WHITES | LJ_GC_SFIXED;
gc_fullsweep(g, &g->gc.root);
- strmask = g->strmask;
- for (i = 0; i <= strmask; i++) /* Free all string hash chains. */
- gc_fullsweep(g, &g->strhash[i]);
+ for (i = g->str.mask; i != ~(MSize)0; i--) /* Free all string hash chains. */
+ gc_sweepstr(g, &g->str.tab[i]);
}
/* -- Collector ----------------------------------------------------------- */
@@ -577,7 +623,7 @@
setgcrefr(g->gc.gray, g->gc.weak); /* Empty the list of weak tables. */
setgcrefnull(g->gc.weak);
- lua_assert(!iswhite(obj2gco(mainthread(g))));
+ lj_assertG(!iswhite(obj2gco(mainthread(g))), "main thread turned white");
gc_markobj(g, L); /* Mark running thread. */
gc_traverse_curtrace(g); /* Traverse current trace. */
gc_mark_gcroot(g); /* Mark GC roots (again). */
@@ -592,7 +638,7 @@
udsize += gc_propagate_gray(g); /* And propagate the marks. */
/* All marking done, clear weak tables. */
- gc_clearweak(gcref(g->gc.weak));
+ gc_clearweak(g, gcref(g->gc.weak));
lj_buf_shrink(L, &g->tmpbuf); /* Shrink temp buffer. */
@@ -625,10 +671,10 @@
return 0;
case GCSsweepstring: {
GCSize old = g->gc.total;
- gc_fullsweep(g, &g->strhash[g->gc.sweepstr++]); /* Sweep one chain. */
- if (g->gc.sweepstr > g->strmask)
+ gc_sweepstr(g, &g->str.tab[g->gc.sweepstr++]); /* Sweep one chain. */
+ if (g->gc.sweepstr > g->str.mask)
g->gc.state = GCSsweep; /* All string hash chains sweeped. */
- lua_assert(old >= g->gc.total);
+ lj_assertG(old >= g->gc.total, "sweep increased memory");
g->gc.estimate -= old - g->gc.total;
return GCSWEEPCOST;
}
@@ -635,16 +681,13 @@
case GCSsweep: {
GCSize old = g->gc.total;
setmref(g->gc.sweep, gc_sweep(g, mref(g->gc.sweep, GCRef), GCSWEEPMAX));
- lua_assert(old >= g->gc.total);
+ lj_assertG(old >= g->gc.total, "sweep increased memory");
g->gc.estimate -= old - g->gc.total;
if (gcref(*mref(g->gc.sweep, GCRef)) == NULL) {
- if (g->strnum <= (g->strmask >> 2) && g->strmask > LJ_MIN_STRTAB*2-1)
- lj_str_resize(L, g->strmask >> 1); /* Shrink string table. */
+ if (g->str.num <= (g->str.mask >> 2) && g->str.mask > LJ_MIN_STRTAB*2-1)
+ lj_str_resize(L, g->str.mask >> 1); /* Shrink string table. */
if (gcref(g->gc.mmudata)) { /* Need any finalizations? */
g->gc.state = GCSfinalize;
-#if LJ_HASFFI
- g->gc.nocdatafin = 1;
-#endif
} else { /* Otherwise skip this phase to help the JIT. */
g->gc.state = GCSpause; /* End of GC cycle. */
g->gc.debt = 0;
@@ -654,21 +697,21 @@
}
case GCSfinalize:
if (gcref(g->gc.mmudata) != NULL) {
+ GCSize old = g->gc.total;
if (tvref(g->jit_base)) /* Don't call finalizers on trace. */
return LJ_MAX_MEM;
gc_finalize(L); /* Finalize one userdata object. */
+ if (old >= g->gc.total && g->gc.estimate > old - g->gc.total)
+ g->gc.estimate -= old - g->gc.total;
if (g->gc.estimate > GCFINALIZECOST)
g->gc.estimate -= GCFINALIZECOST;
return GCFINALIZECOST;
}
-#if LJ_HASFFI
- if (!g->gc.nocdatafin) lj_tab_rehash(L, ctype_ctsG(g)->finalizer);
-#endif
g->gc.state = GCSpause; /* End of GC cycle. */
g->gc.debt = 0;
return 0;
default:
- lua_assert(0);
+ lj_assertG(0, "bad GC state");
return 0;
}
}
@@ -742,7 +785,8 @@
}
while (g->gc.state == GCSsweepstring || g->gc.state == GCSsweep)
gc_onestep(L); /* Finish sweep. */
- lua_assert(g->gc.state == GCSfinalize || g->gc.state == GCSpause);
+ lj_assertG(g->gc.state == GCSfinalize || g->gc.state == GCSpause,
+ "bad GC state");
/* Now perform a full GC. */
g->gc.state = GCSpause;
do { gc_onestep(L); } while (g->gc.state != GCSpause);
@@ -755,9 +799,11 @@
/* Move the GC propagation frontier forward. */
void lj_gc_barrierf(global_State *g, GCobj *o, GCobj *v)
{
- lua_assert(isblack(o) && iswhite(v) && !isdead(g, v) && !isdead(g, o));
- lua_assert(g->gc.state != GCSfinalize && g->gc.state != GCSpause);
- lua_assert(o->gch.gct != ~LJ_TTAB);
+ lj_assertG(isblack(o) && iswhite(v) && !isdead(g, v) && !isdead(g, o),
+ "bad object states for forward barrier");
+ lj_assertG(g->gc.state != GCSfinalize && g->gc.state != GCSpause,
+ "bad GC state");
+ lj_assertG(o->gch.gct != ~LJ_TTAB, "barrier object is not a table");
/* Preserve invariant during propagation. Otherwise it doesn't matter. */
if (g->gc.state == GCSpropagate || g->gc.state == GCSatomic)
gc_mark(g, v); /* Move frontier forward. */
@@ -794,7 +840,8 @@
lj_gc_barrierf(g, o, gcV(&uv->tv));
} else {
makewhite(g, o); /* Make it white, i.e. sweep the upvalue. */
- lua_assert(g->gc.state != GCSfinalize && g->gc.state != GCSpause);
+ lj_assertG(g->gc.state != GCSfinalize && g->gc.state != GCSpause,
+ "bad GC state");
}
}
}
@@ -814,12 +861,13 @@
void *lj_mem_realloc(lua_State *L, void *p, GCSize osz, GCSize nsz)
{
global_State *g = G(L);
- lua_assert((osz == 0) == (p == NULL));
+ lj_assertG((osz == 0) == (p == NULL), "realloc API violation");
p = g->allocf(g->allocd, p, osz, nsz);
if (p == NULL && nsz > 0)
lj_err_mem(L);
- lua_assert((nsz == 0) == (p == NULL));
- lua_assert(checkptrGC(p));
+ lj_assertG((nsz == 0) == (p == NULL), "allocf API violation");
+ lj_assertG(checkptrGC(p),
+ "allocated memory address %p outside required range", p);
g->gc.total = (g->gc.total - osz) + nsz;
return p;
}
@@ -831,7 +879,8 @@
GCobj *o = (GCobj *)g->allocf(g->allocd, NULL, 0, size);
if (o == NULL)
lj_err_mem(L);
- lua_assert(checkptrGC(o));
+ lj_assertG(checkptrGC(o),
+ "allocated memory address %p outside required range", o);
g->gc.total += size;
setgcrefr(o->gch.nextgc, g->gc.root);
setgcref(g->gc.root, o);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gc.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Garbage collector.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_GC_H
@@ -81,8 +81,10 @@
static LJ_AINLINE void lj_gc_barrierback(global_State *g, GCtab *t)
{
GCobj *o = obj2gco(t);
- lua_assert(isblack(o) && !isdead(g, o));
- lua_assert(g->gc.state != GCSfinalize && g->gc.state != GCSpause);
+ lj_assertG(isblack(o) && !isdead(g, o),
+ "bad object states for backward barrier");
+ lj_assertG(g->gc.state != GCSfinalize && g->gc.state != GCSpause,
+ "bad GC state");
black2gray(o);
setgcrefr(t->gclist, g->gc.grayagain);
setgcref(g->gc.grayagain, o);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Client for the GDB JIT API.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_gdbjit_c
@@ -363,7 +363,7 @@
.eosabi = 12,
#elif defined(__DragonFly__)
.eosabi = 0,
-#elif (defined(__sun__) && defined(__svr4__))
+#elif LJ_TARGET_SOLARIS
.eosabi = 6,
#else
.eosabi = 0,
@@ -637,7 +637,7 @@
DUV(DW_AT_low_pc); DUV(DW_FORM_addr);
DUV(DW_AT_high_pc); DUV(DW_FORM_addr);
DUV(DW_AT_stmt_list); DUV(DW_FORM_data4);
- DB(0); DB(0);
+ DB(0); DB(0); DB(0);
ctx->p = p;
}
@@ -724,7 +724,7 @@
SECTALIGN(ctx->p, sizeof(uintptr_t));
gdbjit_initsect(ctx, GDBJIT_SECT_eh_frame, gdbjit_ehframe);
ctx->objsize = (size_t)((char *)ctx->p - (char *)obj);
- lua_assert(ctx->objsize < sizeof(GDBJITobj));
+ lj_assertX(ctx->objsize < sizeof(GDBJITobj), "GDBJITobj overflow");
}
#undef SECTALIGN
@@ -782,7 +782,8 @@
ctx.spadjp = CFRAME_SIZE_JIT +
(MSize)(parent ? traceref(J, parent)->spadjust : 0);
ctx.spadj = CFRAME_SIZE_JIT + T->spadjust;
- lua_assert(startpc >= proto_bc(pt) && startpc < proto_bc(pt) + pt->sizebc);
+ lj_assertJ(startpc >= proto_bc(pt) && startpc < proto_bc(pt) + pt->sizebc,
+ "start PC out of range");
ctx.lineno = lj_debug_line(pt, proto_bcpos(pt, startpc));
ctx.filename = proto_chunknamestr(pt);
if (*ctx.filename == '@' || *ctx.filename == '=')
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_gdbjit.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Client for the GDB JIT API.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_GDBJIT_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** SSA IR (Intermediate Representation) emitter.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_ir_c
@@ -30,8 +30,9 @@
#endif
#include "lj_vm.h"
#include "lj_strscan.h"
+#include "lj_serialize.h"
#include "lj_strfmt.h"
-#include "lj_lib.h"
+#include "lj_prng.h"
/* Some local macros to save typing. Undef'd at the end. */
#define IR(ref) (&J->cur.ir[(ref)])
@@ -38,7 +39,7 @@
#define fins (&J->fold.ins)
/* Pass IR on to next optimization in chain (FOLD). */
-#define emitir(ot, a, b) (lj_ir_set(J, (ot), (a), (b)), lj_opt_fold(J))
+#define emitir(ot, a, b) (lj_ir_set(J, (ot), (a), (b)), lj_opt_fold(J))
/* -- IR tables ----------------------------------------------------------- */
@@ -90,8 +91,9 @@
{
IRIns *baseir = J->irbuf + J->irbotlim;
MSize szins = J->irtoplim - J->irbotlim;
- lua_assert(szins != 0);
- lua_assert(J->cur.nk == J->irbotlim || J->cur.nk-1 == J->irbotlim);
+ lj_assertJ(szins != 0, "zero IR size");
+ lj_assertJ(J->cur.nk == J->irbotlim || J->cur.nk-1 == J->irbotlim,
+ "unexpected IR growth");
if (J->cur.nins + (szins >> 1) < J->irtoplim) {
/* More than half of the buffer is free on top: shift up by a quarter. */
MSize ofs = szins >> 2;
@@ -146,11 +148,12 @@
}
/* Load field of type t from GG_State + offset. Must be 32 bit aligned. */
-LJ_FUNC TRef lj_ir_ggfload(jit_State *J, IRType t, uintptr_t ofs)
+TRef lj_ir_ggfload(jit_State *J, IRType t, uintptr_t ofs)
{
- lua_assert((ofs & 3) == 0);
+ lj_assertJ((ofs & 3) == 0, "unaligned GG_State field offset");
ofs >>= 2;
- lua_assert(ofs >= IRFL__MAX && ofs <= 0x3ff); /* 10 bit FOLD key limit. */
+ lj_assertJ(ofs >= IRFL__MAX && ofs <= 0x3ff,
+ "GG_State field offset breaks 10 bit FOLD key limit");
lj_ir_set(J, IRT(IR_FLOAD, t), REF_NIL, ofs);
return lj_opt_fold(J);
}
@@ -181,7 +184,7 @@
static LJ_AINLINE IRRef ir_nextk64(jit_State *J)
{
IRRef ref = J->cur.nk - 2;
- lua_assert(J->state != LJ_TRACE_ASM);
+ lj_assertJ(J->state != LJ_TRACE_ASM, "bad JIT state");
if (LJ_UNLIKELY(ref < J->irbotlim)) lj_ir_growbot(J);
J->cur.nk = ref;
return ref;
@@ -277,7 +280,7 @@
{
IRIns *ir, *cir = J->cur.ir;
IRRef ref;
- lua_assert(!isdead(J2G(J), o));
+ lj_assertJ(!isdead(J2G(J), o), "interning of dead GC object");
for (ref = J->chain[IR_KGC]; ref; ref = cir[ref].prev)
if (ir_kgc(&cir[ref]) == o)
goto found;
@@ -299,7 +302,7 @@
{
IRRef ref = ir_nextkgc(J);
IRIns *ir = IR(ref);
- lua_assert(irt_toitype_(IRT_P64) == LJ_TTRACE);
+ lj_assertJ(irt_toitype_(IRT_P64) == LJ_TTRACE, "mismatched type mapping");
ir->t.irt = IRT_P64;
ir->o = LJ_GC64 ? IR_KNUM : IR_KNULL; /* Not IR_KGC yet, but same size. */
ir->op12 = 0;
@@ -313,7 +316,7 @@
IRIns *ir, *cir = J->cur.ir;
IRRef ref;
#if LJ_64 && !LJ_GC64
- lua_assert((void *)(uintptr_t)u32ptr(ptr) == ptr);
+ lj_assertJ((void *)(uintptr_t)u32ptr(ptr) == ptr, "out-of-range GC pointer");
#endif
for (ref = J->chain[op]; ref; ref = cir[ref].prev)
if (ir_kptr(&cir[ref]) == ptr)
@@ -360,7 +363,8 @@
IRRef2 op12 = IRREF2((IRRef1)key, (IRRef1)slot);
IRRef ref;
/* Const part is not touched by CSE/DCE, so 0-65535 is ok for IRMlit here. */
- lua_assert(tref_isk(key) && slot == (IRRef)(IRRef1)slot);
+ lj_assertJ(tref_isk(key) && slot == (IRRef)(IRRef1)slot,
+ "out-of-range key/slot");
for (ref = J->chain[IR_KSLOT]; ref; ref = cir[ref].prev)
if (cir[ref].op12 == op12)
goto found;
@@ -381,13 +385,15 @@
void lj_ir_kvalue(lua_State *L, TValue *tv, const IRIns *ir)
{
UNUSED(L);
- lua_assert(ir->o != IR_KSLOT); /* Common mistake. */
+ lj_assertL(ir->o != IR_KSLOT, "unexpected KSLOT"); /* Common mistake. */
switch (ir->o) {
case IR_KPRI: setpriV(tv, irt_toitype(ir->t)); break;
case IR_KINT: setintV(tv, ir->i); break;
case IR_KGC: setgcV(L, tv, ir_kgc(ir), irt_toitype(ir->t)); break;
- case IR_KPTR: case IR_KKPTR: setlightudV(tv, ir_kptr(ir)); break;
- case IR_KNULL: setlightudV(tv, NULL); break;
+ case IR_KPTR: case IR_KKPTR:
+ setnumV(tv, (lua_Number)(uintptr_t)ir_kptr(ir));
+ break;
+ case IR_KNULL: setintV(tv, 0); break;
case IR_KNUM: setnumV(tv, ir_knum(ir)->n); break;
#if LJ_HASFFI
case IR_KINT64: {
@@ -397,7 +403,7 @@
break;
}
#endif
- default: lua_assert(0); break;
+ default: lj_assertL(0, "bad IR constant op %d", ir->o); break;
}
}
@@ -457,7 +463,7 @@
case IR_UGE: return !(a < b);
case IR_ULE: return !(a > b);
case IR_UGT: return !(a <= b);
- default: lua_assert(0); return 0;
+ default: lj_assertX(0, "bad IR op %d", op); return 0;
}
}
@@ -470,7 +476,7 @@
case IR_GE: return (res >= 0);
case IR_LE: return (res <= 0);
case IR_GT: return (res > 0);
- default: lua_assert(0); return 0;
+ default: lj_assertX(0, "bad IR op %d", op); return 0;
}
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ir.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** SSA IR (Intermediate Representation) format.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_IR_H
@@ -75,10 +75,9 @@
_(NEG, N , ref, ref) \
\
_(ABS, N , ref, ref) \
- _(ATAN2, N , ref, ref) \
_(LDEXP, N , ref, ref) \
- _(MIN, C , ref, ref) \
- _(MAX, C , ref, ref) \
+ _(MIN, N , ref, ref) \
+ _(MAX, N , ref, ref) \
_(FPMATH, N , ref, lit) \
\
/* Overflow-checking arithmetic ops. */ \
@@ -96,6 +95,7 @@
_(UREFO, LW, ref, lit) \
_(UREFC, LW, ref, lit) \
_(FREF, R , ref, lit) \
+ _(TMPREF, S , ref, lit) \
_(STRREF, N , ref, ref) \
_(LREF, L , ___, ___) \
\
@@ -106,7 +106,8 @@
_(FLOAD, L , ref, lit) \
_(XLOAD, L , ref, lit) \
_(SLOAD, L , lit, lit) \
- _(VLOAD, L , ref, ___) \
+ _(VLOAD, L , ref, lit) \
+ _(ALEN, L , ref, ref) \
\
_(ASTORE, S , ref, ref) \
_(HSTORE, S , ref, ref) \
@@ -124,8 +125,8 @@
\
/* Buffer operations. */ \
_(BUFHDR, L , ref, lit) \
- _(BUFPUT, L , ref, ref) \
- _(BUFSTR, A , ref, ref) \
+ _(BUFPUT, LW, ref, ref) \
+ _(BUFSTR, AW, ref, ref) \
\
/* Barriers. */ \
_(TBAR, S , ref, ___) \
@@ -133,15 +134,15 @@
_(XBAR, S , ___, ___) \
\
/* Type conversions. */ \
- _(CONV, NW, ref, lit) \
+ _(CONV, N , ref, lit) \
_(TOBIT, N , ref, ref) \
_(TOSTR, N , ref, lit) \
_(STRTO, N , ref, ___) \
\
/* Calls. */ \
- _(CALLN, N , ref, lit) \
- _(CALLA, A , ref, lit) \
- _(CALLL, L , ref, lit) \
+ _(CALLN, NW, ref, lit) \
+ _(CALLA, AW, ref, lit) \
+ _(CALLL, LW, ref, lit) \
_(CALLS, S , ref, lit) \
_(CALLXS, S , ref, ref) \
_(CARG, N , ref, ref) \
@@ -178,8 +179,7 @@
/* FPMATH sub-functions. ORDER FPM. */
#define IRFPMDEF(_) \
_(FLOOR) _(CEIL) _(TRUNC) /* Must be first and in this order. */ \
- _(SQRT) _(EXP) _(EXP2) _(LOG) _(LOG2) _(LOG10) \
- _(SIN) _(COS) _(TAN) \
+ _(SQRT) _(LOG) _(LOG2) \
_(OTHER)
typedef enum {
@@ -205,9 +205,15 @@
_(UDATA_META, offsetof(GCudata, metatable)) \
_(UDATA_UDTYPE, offsetof(GCudata, udtype)) \
_(UDATA_FILE, sizeof(GCudata)) \
+ _(SBUF_W, sizeof(GCudata) + offsetof(SBufExt, w)) \
+ _(SBUF_E, sizeof(GCudata) + offsetof(SBufExt, e)) \
+ _(SBUF_B, sizeof(GCudata) + offsetof(SBufExt, b)) \
+ _(SBUF_L, sizeof(GCudata) + offsetof(SBufExt, L)) \
+ _(SBUF_REF, sizeof(GCudata) + offsetof(SBufExt, cowref)) \
+ _(SBUF_R, sizeof(GCudata) + offsetof(SBufExt, r)) \
_(CDATA_CTYPEID, offsetof(GCcdata, ctypeid)) \
_(CDATA_PTR, sizeof(GCcdata)) \
- _(CDATA_INT, sizeof(GCcdata)) \
+ _(CDATA_INT, sizeof(GCcdata)) \
_(CDATA_INT64, sizeof(GCcdata)) \
_(CDATA_INT64_4, sizeof(GCcdata) + 4)
@@ -218,6 +224,11 @@
IRFL__MAX
} IRFieldID;
+/* TMPREF mode bits, stored in op2. */
+#define IRTMPREF_IN1 0x01 /* First input value. */
+#define IRTMPREF_OUT1 0x02 /* First output value. */
+#define IRTMPREF_OUT2 0x04 /* Second output value. */
+
/* SLOAD mode bits, stored in op2. */
#define IRSLOAD_PARENT 0x01 /* Coalesce with parent trace. */
#define IRSLOAD_FRAME 0x02 /* Load 32 bits of ftsz. */
@@ -225,15 +236,17 @@
#define IRSLOAD_CONVERT 0x08 /* Number to integer conversion. */
#define IRSLOAD_READONLY 0x10 /* Read-only, omit slot store. */
#define IRSLOAD_INHERIT 0x20 /* Inherited by exits/side traces. */
+#define IRSLOAD_KEYINDEX 0x40 /* Table traversal key index. */
-/* XLOAD mode, stored in op2. */
-#define IRXLOAD_READONLY 1 /* Load from read-only data. */
-#define IRXLOAD_VOLATILE 2 /* Load from volatile data. */
-#define IRXLOAD_UNALIGNED 4 /* Unaligned load. */
+/* XLOAD mode bits, stored in op2. */
+#define IRXLOAD_READONLY 0x01 /* Load from read-only data. */
+#define IRXLOAD_VOLATILE 0x02 /* Load from volatile data. */
+#define IRXLOAD_UNALIGNED 0x04 /* Unaligned load. */
/* BUFHDR mode, stored in op2. */
#define IRBUFHDR_RESET 0 /* Reset buffer. */
#define IRBUFHDR_APPEND 1 /* Append to buffer. */
+#define IRBUFHDR_WRITE 2 /* Write to string buffer. */
/* CONV mode, stored in op2. */
#define IRCONV_SRCMASK 0x001f /* Source IRType. */
@@ -250,6 +263,7 @@
#define IRCONV_ANY (1<<IRCONV_CSH) /* Any FP number is ok. */
#define IRCONV_INDEX (2<<IRCONV_CSH) /* Check + special backprop rules. */
#define IRCONV_CHECK (3<<IRCONV_CSH) /* Number checked for integerness. */
+#define IRCONV_NONE IRCONV_ANY /* INT|*64 no conv, but change type. */
/* TOSTR mode, stored in op2. */
#define IRTOSTR_INT 0 /* Convert integer to string. */
@@ -369,6 +383,7 @@
#define irt_isu32(t) (irt_type(t) == IRT_U32)
#define irt_isi64(t) (irt_type(t) == IRT_I64)
#define irt_isu64(t) (irt_type(t) == IRT_U64)
+#define irt_isp32(t) (irt_type(t) == IRT_P32)
#define irt_isfp(t) (irt_isnum(t) || irt_isfloat(t))
#define irt_isinteger(t) (irt_typerange((t), IRT_I8, IRT_INT))
@@ -377,10 +392,12 @@
#define irt_isint64(t) (irt_typerange((t), IRT_I64, IRT_U64))
#if LJ_GC64
+/* Include IRT_NIL, so IR(ASMREF_L) (aka REF_NIL) is considered 64 bit. */
#define IRT_IS64 \
((1u<<IRT_NUM)|(1u<<IRT_I64)|(1u<<IRT_U64)|(1u<<IRT_P64)|\
(1u<<IRT_LIGHTUD)|(1u<<IRT_STR)|(1u<<IRT_THREAD)|(1u<<IRT_PROTO)|\
- (1u<<IRT_FUNC)|(1u<<IRT_CDATA)|(1u<<IRT_TAB)|(1u<<IRT_UDATA))
+ (1u<<IRT_FUNC)|(1u<<IRT_CDATA)|(1u<<IRT_TAB)|(1u<<IRT_UDATA)|\
+ (1u<<IRT_NIL))
#elif LJ_64
#define IRT_IS64 \
((1u<<IRT_NUM)|(1u<<IRT_I64)|(1u<<IRT_U64)|(1u<<IRT_P64)|(1u<<IRT_LIGHTUD))
@@ -412,11 +429,12 @@
static LJ_AINLINE uint32_t irt_toitype_(IRType t)
{
- lua_assert(!LJ_64 || LJ_GC64 || t != IRT_LIGHTUD);
+ lj_assertX(!LJ_64 || LJ_GC64 || t != IRT_LIGHTUD,
+ "no plain type tag for lightuserdata");
if (LJ_DUALNUM && t > IRT_NUM) {
return LJ_TISNUM;
} else {
- lua_assert(t <= IRT_NUM);
+ lj_assertX(t <= IRT_NUM, "no plain type tag for IR type %d", t);
return ~(uint32_t)t;
}
}
@@ -479,6 +497,7 @@
#define TREF_REFMASK 0x0000ffff
#define TREF_FRAME 0x00010000
#define TREF_CONT 0x00020000
+#define TREF_KEYINDEX 0x00100000
#define TREF(ref, t) ((TRef)((ref) + ((t)<<24)))
@@ -560,6 +579,11 @@
TValue tv; /* TValue constant (overlaps entire slot). */
} IRIns;
+#define ir_isk64(ir) \
+ ((ir)->o == IR_KNUM || (ir)->o == IR_KINT64 || \
+ (LJ_GC64 && \
+ ((ir)->o == IR_KGC || (ir)->o == IR_KPTR || (ir)->o == IR_KKPTR)))
+
#define ir_kgc(ir) check_exp((ir)->o == IR_KGC, gcref((ir)[LJ_GC64].gcr))
#define ir_kstr(ir) (gco2str(ir_kgc((ir))))
#define ir_ktab(ir) (gco2tab(ir_kgc((ir))))
@@ -567,12 +591,7 @@
#define ir_kcdata(ir) (gco2cd(ir_kgc((ir))))
#define ir_knum(ir) check_exp((ir)->o == IR_KNUM, &(ir)[1].tv)
#define ir_kint64(ir) check_exp((ir)->o == IR_KINT64, &(ir)[1].tv)
-#define ir_k64(ir) \
- check_exp((ir)->o == IR_KNUM || (ir)->o == IR_KINT64 || \
- (LJ_GC64 && \
- ((ir)->o == IR_KGC || \
- (ir)->o == IR_KPTR || (ir)->o == IR_KKPTR)), \
- &(ir)[1].tv)
+#define ir_k64(ir) check_exp(ir_isk64(ir), &(ir)[1].tv)
#define ir_kptr(ir) \
check_exp((ir)->o == IR_KPTR || (ir)->o == IR_KKPTR, \
mref((ir)[LJ_GC64].ptr, void))
@@ -585,4 +604,12 @@
LJ_STATIC_ASSERT((int)IRT_GUARD == (int)IRM_W);
+/* Replace IR instruction with NOP. */
+static LJ_AINLINE void lj_ir_nop(IRIns *ir)
+{
+ ir->ot = IRT(IR_NOP, IRT_NIL);
+ ir->op1 = ir->op2 = 0;
+ ir->prev = 0;
+}
+
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ircall.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ircall.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_ircall.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** IR CALL* instruction definitions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_IRCALL_H
@@ -21,6 +21,7 @@
#define CCI_OTSHIFT 16
#define CCI_OPTYPE(ci) ((ci)->flags >> CCI_OTSHIFT) /* Get op/type. */
+#define CCI_TYPE(ci) (((ci)->flags>>CCI_OTSHIFT) & IRT_TYPE)
#define CCI_OPSHIFT 24
#define CCI_OP(ci) ((ci)->flags >> CCI_OPSHIFT) /* Get op. */
@@ -29,10 +30,12 @@
#define CCI_CALL_L (IR_CALLL << CCI_OPSHIFT)
#define CCI_CALL_S (IR_CALLS << CCI_OPSHIFT)
#define CCI_CALL_FN (CCI_CALL_N|CCI_CC_FASTCALL)
+#define CCI_CALL_FA (CCI_CALL_A|CCI_CC_FASTCALL)
#define CCI_CALL_FL (CCI_CALL_L|CCI_CC_FASTCALL)
#define CCI_CALL_FS (CCI_CALL_S|CCI_CC_FASTCALL)
/* C call info flags. */
+#define CCI_T (IRT_GUARD << CCI_OTSHIFT) /* May throw. */
#define CCI_L 0x0100 /* Implicit L arg. */
#define CCI_CASTU64 0x0200 /* Cast u64 result to number. */
#define CCI_NOFPRCLOBBER 0x0400 /* Does not clobber any FPRs. */
@@ -51,7 +54,7 @@
#define CCI_XARGS(ci) (((ci)->flags >> CCI_XARGS_SHIFT) & 3)
#define CCI_XA (1u << CCI_XARGS_SHIFT)
-#if LJ_SOFTFP || (LJ_32 && LJ_HASFFI)
+#if LJ_SOFTFP32 || (LJ_32 && LJ_HASFFI)
#define CCI_XNARGS(ci) (CCI_NARGS((ci)) + CCI_XARGS((ci)))
#else
#define CCI_XNARGS(ci) CCI_NARGS((ci))
@@ -60,7 +63,7 @@
/* Helpers for conditional function definitions. */
#define IRCALLCOND_ANY(x) x
-#if LJ_TARGET_X86ORX64
+#if LJ_TARGET_X86ORX64 || LJ_TARGET_ARM64
#define IRCALLCOND_FPMATH(x) NULL
#else
#define IRCALLCOND_FPMATH(x) x
@@ -78,14 +81,20 @@
#define IRCALLCOND_SOFTFP_FFI(x) NULL
#endif
-#if LJ_SOFTFP && LJ_TARGET_MIPS32
+#if LJ_SOFTFP && LJ_TARGET_MIPS
#define IRCALLCOND_SOFTFP_MIPS(x) x
#else
#define IRCALLCOND_SOFTFP_MIPS(x) NULL
#endif
-#define LJ_NEED_FP64 (LJ_TARGET_ARM || LJ_TARGET_PPC || LJ_TARGET_MIPS32)
+#if LJ_SOFTFP && LJ_TARGET_MIPS64
+#define IRCALLCOND_SOFTFP_MIPS64(x) x
+#else
+#define IRCALLCOND_SOFTFP_MIPS64(x) NULL
+#endif
+#define LJ_NEED_FP64 (LJ_TARGET_ARM || LJ_TARGET_PPC || LJ_TARGET_MIPS)
+
#if LJ_HASFFI && (LJ_SOFTFP || LJ_NEED_FP64)
#define IRCALLCOND_FP64_FFI(x) x
#else
@@ -104,6 +113,18 @@
#define IRCALLCOND_FFI32(x) NULL
#endif
+#if LJ_HASBUFFER
+#define IRCALLCOND_BUFFER(x) x
+#else
+#define IRCALLCOND_BUFFER(x) NULL
+#endif
+
+#if LJ_HASBUFFER && LJ_HASFFI
+#define IRCALLCOND_BUFFFI(x) x
+#else
+#define IRCALLCOND_BUFFFI(x) NULL
+#endif
+
#if LJ_SOFTFP
#define XA_FP CCI_XA
#define XA2_FP (CCI_XA+CCI_XA)
@@ -112,6 +133,14 @@
#define XA2_FP 0
#endif
+#if LJ_SOFTFP32
+#define XA_FP32 CCI_XA
+#define XA2_FP32 (CCI_XA+CCI_XA)
+#else
+#define XA_FP32 0
+#define XA2_FP32 0
+#endif
+
#if LJ_32
#define XA_64 CCI_XA
#define XA2_64 (CCI_XA+CCI_XA)
@@ -124,40 +153,57 @@
#define IRCALLDEF(_) \
_(ANY, lj_str_cmp, 2, FN, INT, CCI_NOFPRCLOBBER) \
_(ANY, lj_str_find, 4, N, PGC, 0) \
- _(ANY, lj_str_new, 3, S, STR, CCI_L) \
+ _(ANY, lj_str_new, 3, S, STR, CCI_L|CCI_T) \
_(ANY, lj_strscan_num, 2, FN, INT, 0) \
- _(ANY, lj_strfmt_int, 2, FN, STR, CCI_L) \
- _(ANY, lj_strfmt_num, 2, FN, STR, CCI_L) \
- _(ANY, lj_strfmt_char, 2, FN, STR, CCI_L) \
- _(ANY, lj_strfmt_putint, 2, FL, PGC, 0) \
- _(ANY, lj_strfmt_putnum, 2, FL, PGC, 0) \
- _(ANY, lj_strfmt_putquoted, 2, FL, PGC, 0) \
- _(ANY, lj_strfmt_putfxint, 3, L, PGC, XA_64) \
- _(ANY, lj_strfmt_putfnum_int, 3, L, PGC, XA_FP) \
- _(ANY, lj_strfmt_putfnum_uint, 3, L, PGC, XA_FP) \
- _(ANY, lj_strfmt_putfnum, 3, L, PGC, XA_FP) \
- _(ANY, lj_strfmt_putfstr, 3, L, PGC, 0) \
- _(ANY, lj_strfmt_putfchar, 3, L, PGC, 0) \
- _(ANY, lj_buf_putmem, 3, S, PGC, 0) \
- _(ANY, lj_buf_putstr, 2, FL, PGC, 0) \
- _(ANY, lj_buf_putchar, 2, FL, PGC, 0) \
- _(ANY, lj_buf_putstr_reverse, 2, FL, PGC, 0) \
- _(ANY, lj_buf_putstr_lower, 2, FL, PGC, 0) \
- _(ANY, lj_buf_putstr_upper, 2, FL, PGC, 0) \
- _(ANY, lj_buf_putstr_rep, 3, L, PGC, 0) \
- _(ANY, lj_buf_puttab, 5, L, PGC, 0) \
- _(ANY, lj_buf_tostr, 1, FL, STR, 0) \
- _(ANY, lj_tab_new_ah, 3, A, TAB, CCI_L) \
- _(ANY, lj_tab_new1, 2, FS, TAB, CCI_L) \
- _(ANY, lj_tab_dup, 2, FS, TAB, CCI_L) \
+ _(ANY, lj_strfmt_int, 2, FN, STR, CCI_L|CCI_T) \
+ _(ANY, lj_strfmt_num, 2, FN, STR, CCI_L|CCI_T) \
+ _(ANY, lj_strfmt_char, 2, FN, STR, CCI_L|CCI_T) \
+ _(ANY, lj_strfmt_putint, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_strfmt_putnum, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_strfmt_putquoted, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_strfmt_putfxint, 3, L, PGC, XA_64|CCI_T) \
+ _(ANY, lj_strfmt_putfnum_int, 3, L, PGC, XA_FP|CCI_T) \
+ _(ANY, lj_strfmt_putfnum_uint, 3, L, PGC, XA_FP|CCI_T) \
+ _(ANY, lj_strfmt_putfnum, 3, L, PGC, XA_FP|CCI_T) \
+ _(ANY, lj_strfmt_putfstr, 3, L, PGC, CCI_T) \
+ _(ANY, lj_strfmt_putfchar, 3, L, PGC, CCI_T) \
+ _(ANY, lj_buf_putmem, 3, S, PGC, CCI_T) \
+ _(ANY, lj_buf_putstr, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_buf_putchar, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_buf_putstr_reverse, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_buf_putstr_lower, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_buf_putstr_upper, 2, FL, PGC, CCI_T) \
+ _(ANY, lj_buf_putstr_rep, 3, L, PGC, CCI_T) \
+ _(ANY, lj_buf_puttab, 5, L, PGC, CCI_T) \
+ _(BUFFER, lj_bufx_set, 4, S, NIL, 0) \
+ _(BUFFFI, lj_bufx_more, 2, FS, INT, CCI_T) \
+ _(BUFFER, lj_serialize_put, 2, FS, PGC, CCI_T) \
+ _(BUFFER, lj_serialize_get, 2, FS, PTR, CCI_T) \
+ _(BUFFER, lj_serialize_encode, 2, FA, STR, CCI_L|CCI_T) \
+ _(BUFFER, lj_serialize_decode, 3, A, INT, CCI_L|CCI_T) \
+ _(ANY, lj_buf_tostr, 1, FL, STR, CCI_T) \
+ _(ANY, lj_tab_new_ah, 3, A, TAB, CCI_L|CCI_T) \
+ _(ANY, lj_tab_new1, 2, FA, TAB, CCI_L|CCI_T) \
+ _(ANY, lj_tab_dup, 2, FA, TAB, CCI_L|CCI_T) \
_(ANY, lj_tab_clear, 1, FS, NIL, 0) \
- _(ANY, lj_tab_newkey, 3, S, PGC, CCI_L) \
+ _(ANY, lj_tab_newkey, 3, S, PGC, CCI_L|CCI_T) \
+ _(ANY, lj_tab_keyindex, 2, FL, INT, 0) \
+ _(ANY, lj_vm_next, 2, FL, PTR, 0) \
_(ANY, lj_tab_len, 1, FL, INT, 0) \
+ _(ANY, lj_tab_len_hint, 2, FL, INT, 0) \
_(ANY, lj_gc_step_jit, 2, FS, NIL, CCI_L) \
_(ANY, lj_gc_barrieruv, 2, FS, NIL, 0) \
- _(ANY, lj_mem_newgco, 2, FS, PGC, CCI_L) \
- _(ANY, lj_math_random_step, 1, FS, NUM, CCI_CASTU64) \
+ _(ANY, lj_mem_newgco, 2, FA, PGC, CCI_L|CCI_T) \
+ _(ANY, lj_prng_u64d, 1, FS, NUM, CCI_CASTU64) \
_(ANY, lj_vm_modi, 2, FN, INT, 0) \
+ _(ANY, log10, 1, N, NUM, XA_FP) \
+ _(ANY, exp, 1, N, NUM, XA_FP) \
+ _(ANY, sin, 1, N, NUM, XA_FP) \
+ _(ANY, cos, 1, N, NUM, XA_FP) \
+ _(ANY, tan, 1, N, NUM, XA_FP) \
+ _(ANY, asin, 1, N, NUM, XA_FP) \
+ _(ANY, acos, 1, N, NUM, XA_FP) \
+ _(ANY, atan, 1, N, NUM, XA_FP) \
_(ANY, sinh, 1, N, NUM, XA_FP) \
_(ANY, cosh, 1, N, NUM, XA_FP) \
_(ANY, tanh, 1, N, NUM, XA_FP) \
@@ -169,32 +215,26 @@
_(FPMATH, lj_vm_ceil, 1, N, NUM, XA_FP) \
_(FPMATH, lj_vm_trunc, 1, N, NUM, XA_FP) \
_(FPMATH, sqrt, 1, N, NUM, XA_FP) \
- _(ANY, exp, 1, N, NUM, XA_FP) \
- _(ANY, lj_vm_exp2, 1, N, NUM, XA_FP) \
_(ANY, log, 1, N, NUM, XA_FP) \
_(ANY, lj_vm_log2, 1, N, NUM, XA_FP) \
- _(ANY, log10, 1, N, NUM, XA_FP) \
- _(ANY, sin, 1, N, NUM, XA_FP) \
- _(ANY, cos, 1, N, NUM, XA_FP) \
- _(ANY, tan, 1, N, NUM, XA_FP) \
- _(ANY, lj_vm_powi, 2, N, NUM, XA_FP) \
_(ANY, pow, 2, N, NUM, XA2_FP) \
_(ANY, atan2, 2, N, NUM, XA2_FP) \
_(ANY, ldexp, 2, N, NUM, XA_FP) \
- _(SOFTFP, lj_vm_tobit, 2, N, INT, 0) \
- _(SOFTFP, softfp_add, 4, N, NUM, 0) \
- _(SOFTFP, softfp_sub, 4, N, NUM, 0) \
- _(SOFTFP, softfp_mul, 4, N, NUM, 0) \
- _(SOFTFP, softfp_div, 4, N, NUM, 0) \
- _(SOFTFP, softfp_cmp, 4, N, NIL, 0) \
+ _(SOFTFP, lj_vm_tobit, 1, N, INT, XA_FP32) \
+ _(SOFTFP, softfp_add, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP, softfp_sub, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP, softfp_mul, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP, softfp_div, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP, softfp_cmp, 2, N, NIL, XA2_FP32) \
_(SOFTFP, softfp_i2d, 1, N, NUM, 0) \
- _(SOFTFP, softfp_d2i, 2, N, INT, 0) \
- _(SOFTFP_MIPS, lj_vm_sfmin, 4, N, NUM, 0) \
- _(SOFTFP_MIPS, lj_vm_sfmax, 4, N, NUM, 0) \
+ _(SOFTFP, softfp_d2i, 1, N, INT, XA_FP32) \
+ _(SOFTFP_MIPS, lj_vm_sfmin, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP_MIPS, lj_vm_sfmax, 2, N, NUM, XA2_FP32) \
+ _(SOFTFP_MIPS64, lj_vm_tointg, 1, N, INT, 0) \
_(SOFTFP_FFI, softfp_ui2d, 1, N, NUM, 0) \
_(SOFTFP_FFI, softfp_f2d, 1, N, NUM, 0) \
- _(SOFTFP_FFI, softfp_d2ui, 2, N, INT, 0) \
- _(SOFTFP_FFI, softfp_d2f, 2, N, FLOAT, 0) \
+ _(SOFTFP_FFI, softfp_d2ui, 1, N, INT, XA_FP32) \
+ _(SOFTFP_FFI, softfp_d2f, 1, N, FLOAT, XA_FP32) \
_(SOFTFP_FFI, softfp_i2f, 1, N, FLOAT, 0) \
_(SOFTFP_FFI, softfp_ui2f, 1, N, FLOAT, 0) \
_(SOFTFP_FFI, softfp_f2i, 1, N, INT, 0) \
@@ -272,7 +312,7 @@
#define fp64_f2l __aeabi_f2lz
#define fp64_f2ul __aeabi_f2ulz
#endif
-#elif LJ_TARGET_MIPS
+#elif LJ_TARGET_MIPS || LJ_TARGET_PPC
#define softfp_add __adddf3
#define softfp_sub __subdf3
#define softfp_mul __muldf3
@@ -315,7 +355,7 @@
#endif
#if LJ_HASFFI && LJ_NEED_FP64 && !(LJ_TARGET_ARM && LJ_SOFTFP)
-#ifdef __GNUC__
+#if defined(__GNUC__) || defined(__clang__)
#define fp64_l2d __floatdidf
#define fp64_ul2d __floatundidf
#define fp64_l2f __floatdisf
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_iropt.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_iropt.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_iropt.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Common header for IR emitter and optimizations.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_IROPT_H
@@ -56,6 +56,12 @@
#define lj_ir_kintp(J, k) lj_ir_kint(J, (int32_t)(k))
#endif
+#if LJ_GC64
+#define lj_ir_kintpgc lj_ir_kintp
+#else
+#define lj_ir_kintpgc lj_ir_kint
+#endif
+
static LJ_AINLINE TRef lj_ir_knum(jit_State *J, lua_Number n)
{
TValue tv;
@@ -120,10 +126,11 @@
LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_uload(jit_State *J);
LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_fload(jit_State *J);
LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_xload(jit_State *J);
-LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_tab_len(jit_State *J);
+LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_alen(jit_State *J);
LJ_FUNC TRef LJ_FASTCALL lj_opt_fwd_hrefk(jit_State *J);
LJ_FUNC int LJ_FASTCALL lj_opt_fwd_href_nokey(jit_State *J);
LJ_FUNC int LJ_FASTCALL lj_opt_fwd_tptr(jit_State *J, IRRef lim);
+LJ_FUNC int LJ_FASTCALL lj_opt_fwd_sbuf(jit_State *J, IRRef lim);
LJ_FUNC int lj_opt_fwd_wasnonnil(jit_State *J, IROpT loadop, IRRef xref);
/* Dead-store elimination. */
@@ -144,13 +151,12 @@
TValue *vb, TValue *vc, IROp op);
LJ_FUNC TRef lj_opt_narrow_unm(jit_State *J, TRef rc, TValue *vc);
LJ_FUNC TRef lj_opt_narrow_mod(jit_State *J, TRef rb, TRef rc, TValue *vb, TValue *vc);
-LJ_FUNC TRef lj_opt_narrow_pow(jit_State *J, TRef rb, TRef rc, TValue *vb, TValue *vc);
LJ_FUNC IRType lj_opt_narrow_forl(jit_State *J, cTValue *forbase);
/* Optimization passes. */
LJ_FUNC void lj_opt_dce(jit_State *J);
LJ_FUNC int lj_opt_loop(jit_State *J);
-#if LJ_SOFTFP || (LJ_32 && LJ_HASFFI)
+#if LJ_SOFTFP32 || (LJ_32 && LJ_HASFFI)
LJ_FUNC void lj_opt_split(jit_State *J);
#else
#define lj_opt_split(J) UNUSED(J)
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_jit.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_jit.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_jit.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Common definitions for the JIT compiler.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_JIT_H
@@ -7,77 +7,91 @@
#define _LJ_JIT_H
#include "lj_obj.h"
+#if LJ_HASJIT
#include "lj_ir.h"
-/* JIT engine flags. */
+/* -- JIT engine flags ---------------------------------------------------- */
+
+/* General JIT engine flags. 4 bits. */
#define JIT_F_ON 0x00000001
-/* CPU-specific JIT engine flags. */
+/* CPU-specific JIT engine flags. 12 bits. Flags and strings must match. */
+#define JIT_F_CPU 0x00000010
+
#if LJ_TARGET_X86ORX64
-#define JIT_F_SSE2 0x00000010
-#define JIT_F_SSE3 0x00000020
-#define JIT_F_SSE4_1 0x00000040
-#define JIT_F_PREFER_IMUL 0x00000080
-#define JIT_F_LEA_AGU 0x00000100
-#define JIT_F_BMI2 0x00000200
-/* Names for the CPU-specific flags. Must match the order above. */
-#define JIT_F_CPU_FIRST JIT_F_SSE2
-#define JIT_F_CPUSTRING "\4SSE2\4SSE3\6SSE4.1\3AMD\4ATOM\4BMI2"
+#define JIT_F_SSE3 (JIT_F_CPU << 0)
+#define JIT_F_SSE4_1 (JIT_F_CPU << 1)
+#define JIT_F_BMI2 (JIT_F_CPU << 2)
+
+
+#define JIT_F_CPUSTRING "\4SSE3\6SSE4.1\4BMI2"
+
#elif LJ_TARGET_ARM
-#define JIT_F_ARMV6_ 0x00000010
-#define JIT_F_ARMV6T2_ 0x00000020
-#define JIT_F_ARMV7 0x00000040
-#define JIT_F_VFPV2 0x00000080
-#define JIT_F_VFPV3 0x00000100
-#define JIT_F_ARMV6 (JIT_F_ARMV6_|JIT_F_ARMV6T2_|JIT_F_ARMV7)
-#define JIT_F_ARMV6T2 (JIT_F_ARMV6T2_|JIT_F_ARMV7)
+#define JIT_F_ARMV6_ (JIT_F_CPU << 0)
+#define JIT_F_ARMV6T2_ (JIT_F_CPU << 1)
+#define JIT_F_ARMV7 (JIT_F_CPU << 2)
+#define JIT_F_ARMV8 (JIT_F_CPU << 3)
+#define JIT_F_VFPV2 (JIT_F_CPU << 4)
+#define JIT_F_VFPV3 (JIT_F_CPU << 5)
+
+#define JIT_F_ARMV6 (JIT_F_ARMV6_|JIT_F_ARMV6T2_|JIT_F_ARMV7|JIT_F_ARMV8)
+#define JIT_F_ARMV6T2 (JIT_F_ARMV6T2_|JIT_F_ARMV7|JIT_F_ARMV8)
#define JIT_F_VFP (JIT_F_VFPV2|JIT_F_VFPV3)
-/* Names for the CPU-specific flags. Must match the order above. */
-#define JIT_F_CPU_FIRST JIT_F_ARMV6_
-#define JIT_F_CPUSTRING "\5ARMv6\7ARMv6T2\5ARMv7\5VFPv2\5VFPv3"
+#define JIT_F_CPUSTRING "\5ARMv6\7ARMv6T2\5ARMv7\5ARMv8\5VFPv2\5VFPv3"
+
#elif LJ_TARGET_PPC
-#define JIT_F_SQRT 0x00000010
-#define JIT_F_ROUND 0x00000020
-/* Names for the CPU-specific flags. Must match the order above. */
-#define JIT_F_CPU_FIRST JIT_F_SQRT
+#define JIT_F_SQRT (JIT_F_CPU << 0)
+#define JIT_F_ROUND (JIT_F_CPU << 1)
+
#define JIT_F_CPUSTRING "\4SQRT\5ROUND"
+
#elif LJ_TARGET_MIPS
-#define JIT_F_MIPSXXR2 0x00000010
-/* Names for the CPU-specific flags. Must match the order above. */
-#define JIT_F_CPU_FIRST JIT_F_MIPSXXR2
+#define JIT_F_MIPSXXR2 (JIT_F_CPU << 0)
+
#if LJ_TARGET_MIPS32
+#if LJ_TARGET_MIPSR6
+#define JIT_F_CPUSTRING "\010MIPS32R6"
+#else
#define JIT_F_CPUSTRING "\010MIPS32R2"
+#endif
#else
+#if LJ_TARGET_MIPSR6
+#define JIT_F_CPUSTRING "\010MIPS64R6"
+#else
#define JIT_F_CPUSTRING "\010MIPS64R2"
#endif
+#endif
+
#else
-#define JIT_F_CPU_FIRST 0
+
#define JIT_F_CPUSTRING ""
+
#endif
-/* Optimization flags. */
+/* Optimization flags. 12 bits. */
+#define JIT_F_OPT 0x00010000
#define JIT_F_OPT_MASK 0x0fff0000
-#define JIT_F_OPT_FOLD 0x00010000
-#define JIT_F_OPT_CSE 0x00020000
-#define JIT_F_OPT_DCE 0x00040000
-#define JIT_F_OPT_FWD 0x00080000
-#define JIT_F_OPT_DSE 0x00100000
-#define JIT_F_OPT_NARROW 0x00200000
-#define JIT_F_OPT_LOOP 0x00400000
-#define JIT_F_OPT_ABC 0x00800000
-#define JIT_F_OPT_SINK 0x01000000
-#define JIT_F_OPT_FUSE 0x02000000
+#define JIT_F_OPT_FOLD (JIT_F_OPT << 0)
+#define JIT_F_OPT_CSE (JIT_F_OPT << 1)
+#define JIT_F_OPT_DCE (JIT_F_OPT << 2)
+#define JIT_F_OPT_FWD (JIT_F_OPT << 3)
+#define JIT_F_OPT_DSE (JIT_F_OPT << 4)
+#define JIT_F_OPT_NARROW (JIT_F_OPT << 5)
+#define JIT_F_OPT_LOOP (JIT_F_OPT << 6)
+#define JIT_F_OPT_ABC (JIT_F_OPT << 7)
+#define JIT_F_OPT_SINK (JIT_F_OPT << 8)
+#define JIT_F_OPT_FUSE (JIT_F_OPT << 9)
+#define JIT_F_OPT_FMA (JIT_F_OPT << 10)
/* Optimizations names for -O. Must match the order above. */
-#define JIT_F_OPT_FIRST JIT_F_OPT_FOLD
#define JIT_F_OPTSTRING \
- "\4fold\3cse\3dce\3fwd\3dse\6narrow\4loop\3abc\4sink\4fuse"
+ "\4fold\3cse\3dce\3fwd\3dse\6narrow\4loop\3abc\4sink\4fuse\3fma"
/* Optimization levels set a fixed combination of flags. */
#define JIT_F_OPT_0 0
@@ -86,9 +100,12 @@
#define JIT_F_OPT_3 (JIT_F_OPT_2|\
JIT_F_OPT_FWD|JIT_F_OPT_DSE|JIT_F_OPT_ABC|JIT_F_OPT_SINK|JIT_F_OPT_FUSE)
#define JIT_F_OPT_DEFAULT JIT_F_OPT_3
+/* Note: FMA is not set by default. */
+/* -- JIT engine parameters ----------------------------------------------- */
+
#if LJ_TARGET_WINDOWS || LJ_64
-/* See: http://blogs.msdn.com/oldnewthing/archive/2003/10/08/55239.aspx */
+/* See: https://devblogs.microsoft.com/oldnewthing/20031008-00/?p=42223 */
#define JIT_P_sizemcode_DEFAULT 64
#else
/* Could go as low as 4K, but the mmap() overhead would be rather high. */
@@ -129,11 +146,14 @@
#define JIT_PARAMSTR(len, name, value) #len #name
#define JIT_P_STRING JIT_PARAMDEF(JIT_PARAMSTR)
+/* -- JIT engine data structures ------------------------------------------ */
+
/* Trace compiler state. */
typedef enum {
LJ_TRACE_IDLE, /* Trace compiler idle. */
LJ_TRACE_ACTIVE = 0x10,
LJ_TRACE_RECORD, /* Bytecode recording active. */
+ LJ_TRACE_RECORD_1ST, /* Record 1st instruction, too. */
LJ_TRACE_START, /* New trace started. */
LJ_TRACE_END, /* End of trace. */
LJ_TRACE_ASM, /* Assemble trace. */
@@ -158,10 +178,17 @@
typedef uint32_t MCode;
#endif
+/* Linked list of MCode areas. */
+typedef struct MCLink {
+ MCode *next; /* Next area. */
+ size_t size; /* Size of current area. */
+} MCLink;
+
/* Stack snapshot header. */
typedef struct SnapShot {
- uint16_t mapofs; /* Offset into snapshot map. */
+ uint32_t mapofs; /* Offset into snapshot map. */
IRRef1 ref; /* First IR ref for this snapshot. */
+ uint16_t mcofs; /* Offset into machine code in MCode units. */
uint8_t nslots; /* Number of valid slots. */
uint8_t topslot; /* Maximum frame extent. */
uint8_t nent; /* Number of compressed entries. */
@@ -177,12 +204,15 @@
#define SNAP_CONT 0x020000 /* Continuation slot. */
#define SNAP_NORESTORE 0x040000 /* No need to restore slot. */
#define SNAP_SOFTFPNUM 0x080000 /* Soft-float number. */
+#define SNAP_KEYINDEX 0x100000 /* Traversal key index. */
LJ_STATIC_ASSERT(SNAP_FRAME == TREF_FRAME);
LJ_STATIC_ASSERT(SNAP_CONT == TREF_CONT);
+LJ_STATIC_ASSERT(SNAP_KEYINDEX == TREF_KEYINDEX);
#define SNAP(slot, flags, ref) (((SnapEntry)(slot) << 24) + (flags) + (ref))
#define SNAP_TR(slot, tr) \
- (((SnapEntry)(slot) << 24) + ((tr) & (TREF_CONT|TREF_FRAME|TREF_REFMASK)))
+ (((SnapEntry)(slot) << 24) + \
+ ((tr) & (TREF_KEYINDEX|TREF_CONT|TREF_FRAME|TREF_REFMASK)))
#if !LJ_FR2
#define SNAP_MKPC(pc) ((SnapEntry)u32ptr(pc))
#endif
@@ -227,8 +257,7 @@
/* Trace object. */
typedef struct GCtrace {
GCHeader;
- uint8_t topslot; /* Top stack slot already checked to be allocated. */
- uint8_t linktype; /* Type of link. */
+ uint16_t nsnap; /* Number of snapshots. */
IRRef nins; /* Next IR instruction. Biased with REF_BIAS. */
#if LJ_GC64
uint32_t unused_gc64;
@@ -236,8 +265,7 @@
GCRef gclist;
IRIns *ir; /* IR instructions/constants. Biased with REF_BIAS. */
IRRef nk; /* Lowest IR constant. Biased with REF_BIAS. */
- uint16_t nsnap; /* Number of snapshots. */
- uint16_t nsnapmap; /* Number of snapshot map elements. */
+ uint32_t nsnapmap; /* Number of snapshot map elements. */
SnapShot *snap; /* Snapshot array. */
SnapEntry *snapmap; /* Snapshot map. */
GCRef startpt; /* Starting prototype. */
@@ -245,6 +273,9 @@
BCIns startins; /* Original bytecode of starting instruction. */
MSize szmcode; /* Size of machine code. */
MCode *mcode; /* Start of machine code. */
+#if LJ_ABI_PAUTH
+ ASMFunction mcauth; /* Start of machine code, with ptr auth applied. */
+#endif
MSize mcloop; /* Offset of loop start in machine code. */
uint16_t nchild; /* Number of child traces (root trace only). */
uint16_t spadjust; /* Stack pointer adjustment (offset in bytes). */
@@ -254,6 +285,8 @@
TraceNo1 nextroot; /* Next root trace for same prototype. */
TraceNo1 nextside; /* Next side trace of same root trace. */
uint8_t sinktags; /* Trace has SINK tags. */
+ uint8_t topslot; /* Top stack slot already checked to be allocated. */
+ uint8_t linktype; /* Type of link. */
uint8_t unused1;
#ifdef LUAJIT_USE_GDBJIT
void *gdbjit_entry; /* GDB JIT entry. */
@@ -344,6 +377,7 @@
#endif
LJ_K64__MAX,
};
+#define LJ_K64__USED (LJ_TARGET_X86ORX64 || LJ_TARGET_MIPS)
enum {
#if LJ_TARGET_X86ORX64
@@ -362,6 +396,7 @@
#endif
LJ_K32__MAX
};
+#define LJ_K32__USED (LJ_TARGET_X86ORX64 || LJ_TARGET_PPC || LJ_TARGET_MIPS)
/* Get 16 byte aligned pointer to SIMD constant. */
#define LJ_KSIMD(J, n) \
@@ -368,7 +403,7 @@
((TValue *)(((intptr_t)&J->ksimd[2*(n)] + 15) & ~(intptr_t)15))
/* Set/reset flag to activate the SPLIT pass for the current trace. */
-#if LJ_SOFTFP || (LJ_32 && LJ_HASFFI)
+#if LJ_SOFTFP32 || (LJ_32 && LJ_HASFFI)
#define lj_needsplit(J) (J->needsplit = 1)
#define lj_resetsplit(J) (J->needsplit = 0)
#else
@@ -416,13 +451,17 @@
int32_t framedepth; /* Current frame depth. */
int32_t retdepth; /* Return frame depth (count of RETF). */
+#if LJ_K32__USED
+ uint32_t k32[LJ_K32__MAX]; /* Common 4 byte constants used by backends. */
+#endif
TValue ksimd[LJ_KSIMD__MAX*2+1]; /* 16 byte aligned SIMD constants. */
- TValue k64[LJ_K64__MAX]; /* Common 8 byte constants used by backends. */
- uint32_t k32[LJ_K32__MAX]; /* Ditto for 4 byte constants. */
+#if LJ_K64__USED
+ TValue k64[LJ_K64__MAX]; /* Common 8 byte constants. */
+#endif
IRIns *irbuf; /* Temp. IR instruction buffer. Biased with REF_BIAS. */
- IRRef irtoplim; /* Upper limit of instuction buffer (biased). */
- IRRef irbotlim; /* Lower limit of instuction buffer (biased). */
+ IRRef irtoplim; /* Upper limit of instruction buffer (biased). */
+ IRRef irbotlim; /* Lower limit of instruction buffer (biased). */
IRRef loopref; /* Last loop reference or ref of final LOOP (or 0). */
MSize sizesnap; /* Size of temp. snapshot buffer. */
@@ -431,7 +470,7 @@
MSize sizesnapmap; /* Size of temp. snapshot map buffer. */
PostProc postproc; /* Required post-processing after execution. */
-#if LJ_SOFTFP || (LJ_32 && LJ_HASFFI)
+#if LJ_SOFTFP32 || (LJ_32 && LJ_HASFFI)
uint8_t needsplit; /* Need SPLIT pass. */
#endif
uint8_t retryrec; /* Retry recording. */
@@ -450,7 +489,6 @@
HotPenalty penalty[PENALTY_SLOTS]; /* Penalty slots. */
uint32_t penaltyslot; /* Round-robin index into penalty slots. */
- uint32_t prngstate; /* PRNG state. */
#ifdef LUAJIT_ENABLE_TABLE_BUMP
RBCHashEntry rbchash[RBCHASH_SLOTS]; /* Reverse bytecode map. */
@@ -464,6 +502,7 @@
const BCIns *startpc; /* Bytecode PC of starting instruction. */
TraceNo parent; /* Parent of current side trace (0 for root traces). */
ExitNo exitno; /* Exit number in parent of current side trace. */
+ int exitcode; /* Exit code from unwound trace. */
BCIns *patchpc; /* PC for pending re-patch. */
BCIns patchins; /* Instruction for pending re-patch. */
@@ -482,18 +521,13 @@
BCLine prev_line; /* Previous line. */
int prof_mode; /* Profiling mode: 0, 'f', 'l'. */
#endif
-}
-#if LJ_TARGET_ARM
-LJ_ALIGN(16) /* For DISPATCH-relative addresses in assembler part. */
+} jit_State;
+
+#ifdef LUA_USE_ASSERT
+#define lj_assertJ(c, ...) lj_assertG_(J2G(J), (c), __VA_ARGS__)
+#else
+#define lj_assertJ(c, ...) ((void)J)
#endif
-jit_State;
+#endif
-/* Trivial PRNG e.g. used for penalty randomization. */
-static LJ_AINLINE uint32_t LJ_PRNG_BITS(jit_State *J, int bits)
-{
- /* Yes, this LCG is very weak, but that doesn't matter for our use case. */
- J->prngstate = J->prngstate * 1103515245 + 12345;
- return J->prngstate >> (32-bits);
-}
-
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Lexical analyzer.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -48,6 +48,12 @@
size_t sz;
const char *p = ls->rfunc(ls->L, ls->rdata, &sz);
if (p == NULL || sz == 0) return LEX_EOF;
+ if (sz >= LJ_MAX_BUF) {
+ if (sz != ~(size_t)0) lj_err_mem(ls->L);
+ sz = ~(uintptr_t)0 - (uintptr_t)p;
+ if (sz >= LJ_MAX_BUF) sz = LJ_MAX_BUF-1;
+ ls->endmark = 1;
+ }
ls->pe = p + sz;
ls->p = p + 1;
return (LexChar)(uint8_t)p[0];
@@ -76,7 +82,7 @@
static void lex_newline(LexState *ls)
{
LexChar old = ls->c;
- lua_assert(lex_iseol(ls));
+ lj_assertLS(lex_iseol(ls), "bad usage");
lex_next(ls); /* Skip "\n" or "\r". */
if (lex_iseol(ls) && ls->c != old) lex_next(ls); /* Skip "\n\r" or "\r\n". */
if (++ls->linenumber >= LJ_MAX_LINE)
@@ -90,7 +96,7 @@
{
StrScanFmt fmt;
LexChar c, xp = 'e';
- lua_assert(lj_char_isdigit(ls->c));
+ lj_assertLS(lj_char_isdigit(ls->c), "bad usage");
if ((c = ls->c) == '0' && (lex_savenext(ls) | 0x20) == 'x')
xp = 'p';
while (lj_char_isident(ls->c) || ls->c == '.' ||
@@ -99,7 +105,7 @@
lex_savenext(ls);
}
lex_save(ls, '\0');
- fmt = lj_strscan_scan((const uint8_t *)sbufB(&ls->sb), tv,
+ fmt = lj_strscan_scan((const uint8_t *)ls->sb.b, sbuflen(&ls->sb)-1, tv,
(LJ_DUALNUM ? STRSCAN_OPT_TOINT : STRSCAN_OPT_TONUM) |
(LJ_HASFFI ? (STRSCAN_OPT_LL|STRSCAN_OPT_IMAG) : 0));
if (LJ_DUALNUM && fmt == STRSCAN_INT) {
@@ -110,12 +116,9 @@
} else if (fmt != STRSCAN_ERROR) {
lua_State *L = ls->L;
GCcdata *cd;
- lua_assert(fmt == STRSCAN_I64 || fmt == STRSCAN_U64 || fmt == STRSCAN_IMAG);
- if (!ctype_ctsG(G(L))) {
- ptrdiff_t oldtop = savestack(L, L->top);
- luaopen_ffi(L); /* Load FFI library on-demand. */
- L->top = restorestack(L, oldtop);
- }
+ lj_assertLS(fmt == STRSCAN_I64 || fmt == STRSCAN_U64 || fmt == STRSCAN_IMAG,
+ "unexpected number format %d", fmt);
+ ctype_loadffi(L);
if (fmt == STRSCAN_IMAG) {
cd = lj_cdata_new_(L, CTID_COMPLEX_DOUBLE, 2*sizeof(double));
((double *)cdataptr(cd))[0] = 0;
@@ -127,7 +130,8 @@
lj_parse_keepcdata(ls, tv, cd);
#endif
} else {
- lua_assert(fmt == STRSCAN_ERROR);
+ lj_assertLS(fmt == STRSCAN_ERROR,
+ "unexpected number format %d", fmt);
lj_lex_error(ls, TK_number, LJ_ERR_XNUMBER);
}
}
@@ -137,8 +141,8 @@
{
int count = 0;
LexChar s = ls->c;
- lua_assert(s == '[' || s == ']');
- while (lex_savenext(ls) == '=')
+ lj_assertLS(s == '[' || s == ']', "bad usage");
+ while (lex_savenext(ls) == '=' && count < 0x20000000)
count++;
return (ls->c == s) ? count : (-count) - 1;
}
@@ -172,7 +176,7 @@
}
} endloop:
if (tv) {
- GCstr *str = lj_parse_keepstr(ls, sbufB(&ls->sb) + (2 + (MSize)sep),
+ GCstr *str = lj_parse_keepstr(ls, ls->sb.b + (2 + (MSize)sep),
sbuflen(&ls->sb) - 2*(2 + (MSize)sep));
setstrV(ls->L, tv, str);
}
@@ -278,7 +282,7 @@
}
lex_savenext(ls); /* Skip trailing delimiter. */
setstrV(ls->L, tv,
- lj_parse_keepstr(ls, sbufB(&ls->sb)+1, sbuflen(&ls->sb)-2));
+ lj_parse_keepstr(ls, ls->sb.b+1, sbuflen(&ls->sb)-2));
}
/* -- Main lexical scanner ------------------------------------------------ */
@@ -298,7 +302,7 @@
do {
lex_savenext(ls);
} while (lj_char_isident(ls->c));
- s = lj_parse_keepstr(ls, sbufB(&ls->sb), sbuflen(&ls->sb));
+ s = lj_parse_keepstr(ls, ls->sb.b, sbuflen(&ls->sb));
setstrV(ls->L, tv, s);
if (s->reserved > 0) /* Reserved word? */
return TK_OFS + s->reserved;
@@ -406,6 +410,8 @@
ls->lookahead = TK_eof; /* No look-ahead token. */
ls->linenumber = 1;
ls->lastline = 1;
+ ls->endmark = 0;
+ ls->fr2 = LJ_FR2; /* Generate native bytecode by default. */
lex_next(ls); /* Read-ahead first char. */
if (ls->c == 0xef && ls->p + 2 <= ls->pe && (uint8_t)ls->p[0] == 0xbb &&
(uint8_t)ls->p[1] == 0xbf) { /* Skip UTF-8 BOM (if buffered). */
@@ -462,7 +468,7 @@
/* Look ahead for the next token. */
LexToken lj_lex_lookahead(LexState *ls)
{
- lua_assert(ls->lookahead == TK_eof);
+ lj_assertLS(ls->lookahead == TK_eof, "double lookahead");
ls->lookahead = lex_scan(ls, &ls->lookaheadval);
return ls->lookahead;
}
@@ -487,7 +493,7 @@
tokstr = NULL;
} else if (tok == TK_name || tok == TK_string || tok == TK_number) {
lex_save(ls, '\0');
- tokstr = sbufB(&ls->sb);
+ tokstr = ls->sb.b;
} else {
tokstr = lj_lex_token2str(ls, tok);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lex.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Lexical analyzer.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_LEX_H
@@ -73,6 +73,8 @@
BCInsLine *bcstack; /* Stack for bytecode instructions/line numbers. */
MSize sizebcstack; /* Size of bytecode stack. */
uint32_t level; /* Syntactical nesting level. */
+ int endmark; /* Trust bytecode end marker, even if not at EOF. */
+ int fr2; /* Generate bytecode for LJ_FR2 mode. */
} LexState;
LJ_FUNC int lj_lex_setup(lua_State *L, LexState *ls);
@@ -83,4 +85,10 @@
LJ_FUNC_NORET void lj_lex_error(LexState *ls, LexToken tok, ErrMsg em, ...);
LJ_FUNC void lj_lex_init(lua_State *L);
+#ifdef LUA_USE_ASSERT
+#define lj_assertLS(c, ...) (lj_assertG_(G(ls->L), (c), __VA_ARGS__))
+#else
+#define lj_assertLS(c, ...) ((void)ls)
#endif
+
+#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Library function support.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_lib_c
@@ -16,6 +16,9 @@
#include "lj_func.h"
#include "lj_bc.h"
#include "lj_dispatch.h"
+#if LJ_HASFFI
+#include "lj_ctype.h"
+#endif
#include "lj_vm.h"
#include "lj_strscan.h"
#include "lj_strfmt.h"
@@ -59,6 +62,7 @@
ls.pe = (const char *)~(uintptr_t)0;
ls.c = -1;
ls.level = (BCDUMP_F_STRIP|(LJ_BE*BCDUMP_F_BE));
+ ls.fr2 = LJ_FR2;
ls.chunkname = name;
pt = lj_bcread_proto(&ls);
pt->firstline = ~(BCLine)0;
@@ -263,6 +267,23 @@
return funcV(o);
}
+GCproto *lj_lib_checkLproto(lua_State *L, int narg, int nolua)
+{
+ TValue *o = L->base + narg-1;
+ if (L->top > o) {
+ if (tvisproto(o)) {
+ return protoV(o);
+ } else if (tvisfunc(o)) {
+ if (isluafunc(funcV(o)))
+ return funcproto(funcV(o));
+ else if (nolua)
+ return NULL;
+ }
+ }
+ lj_err_argt(L, narg, LUA_TFUNCTION);
+ return NULL; /* unreachable */
+}
+
GCtab *lj_lib_checktab(lua_State *L, int narg)
{
TValue *o = L->base + narg-1;
@@ -301,3 +322,56 @@
return def;
}
+/* -- Strict type checks -------------------------------------------------- */
+
+/* The following type checks do not coerce between strings and numbers.
+** And they handle plain int64_t/uint64_t FFI numbers, too.
+*/
+
+#if LJ_HASBUFFER
+GCstr *lj_lib_checkstrx(lua_State *L, int narg)
+{
+ TValue *o = L->base + narg-1;
+ if (!(o < L->top && tvisstr(o))) lj_err_argt(L, narg, LUA_TSTRING);
+ return strV(o);
+}
+
+int32_t lj_lib_checkintrange(lua_State *L, int narg, int32_t a, int32_t b)
+{
+ TValue *o = L->base + narg-1;
+ lj_assertL(b >= 0, "expected range must be non-negative");
+ if (o < L->top) {
+ if (LJ_LIKELY(tvisint(o))) {
+ int32_t i = intV(o);
+ if (i >= a && i <= b) return i;
+ } else if (LJ_LIKELY(tvisnum(o))) {
+ /* For performance reasons, this doesn't check for integerness or
+ ** integer overflow. Overflow detection still works, since all FPUs
+ ** return either MININT or MAXINT, which is then out of range.
+ */
+ int32_t i = (int32_t)numV(o);
+ if (i >= a && i <= b) return i;
+#if LJ_HASFFI
+ } else if (tviscdata(o)) {
+ GCcdata *cd = cdataV(o);
+ if (cd->ctypeid == CTID_INT64) {
+ int64_t i = *(int64_t *)cdataptr(cd);
+ if (i >= (int64_t)a && i <= (int64_t)b) return (int32_t)i;
+ } else if (cd->ctypeid == CTID_UINT64) {
+ uint64_t i = *(uint64_t *)cdataptr(cd);
+ if ((a < 0 || i >= (uint64_t)a) && i <= (uint64_t)b) return (int32_t)i;
+ } else {
+ goto badtype;
+ }
+#endif
+ } else {
+ goto badtype;
+ }
+ lj_err_arg(L, narg, LJ_ERR_NUMRNG);
+ }
+badtype:
+ lj_err_argt(L, narg, LUA_TNUMBER);
+ return 0; /* unreachable */
+}
+#endif
+
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_lib.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Library function support.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_LIB_H
@@ -42,10 +42,17 @@
LJ_FUNC int32_t lj_lib_checkint(lua_State *L, int narg);
LJ_FUNC int32_t lj_lib_optint(lua_State *L, int narg, int32_t def);
LJ_FUNC GCfunc *lj_lib_checkfunc(lua_State *L, int narg);
+LJ_FUNC GCproto *lj_lib_checkLproto(lua_State *L, int narg, int nolua);
LJ_FUNC GCtab *lj_lib_checktab(lua_State *L, int narg);
LJ_FUNC GCtab *lj_lib_checktabornil(lua_State *L, int narg);
LJ_FUNC int lj_lib_checkopt(lua_State *L, int narg, int def, const char *lst);
+#if LJ_HASBUFFER
+LJ_FUNC GCstr *lj_lib_checkstrx(lua_State *L, int narg);
+LJ_FUNC int32_t lj_lib_checkintrange(lua_State *L, int narg,
+ int32_t a, int32_t b);
+#endif
+
/* Avoid including lj_frame.h. */
#if LJ_GC64
#define lj_lib_upvalue(L, n) \
@@ -107,9 +114,4 @@
#define LIBINIT_FFID 0xfe
#define LIBINIT_END 0xff
-/* Exported library functions. */
-
-typedef struct RandomState RandomState;
-LJ_FUNC uint64_t LJ_FASTCALL lj_math_random_step(RandomState *rs);
-
#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_load.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_load.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_load.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Load and dump code.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#include <errno.h>
@@ -34,14 +34,28 @@
UNUSED(dummy);
cframe_errfunc(L->cframe) = -1; /* Inherit error function. */
bc = lj_lex_setup(L, ls);
- if (ls->mode && !strchr(ls->mode, bc ? 'b' : 't')) {
- setstrV(L, L->top++, lj_err_str(L, LJ_ERR_XMODE));
- lj_err_throw(L, LUA_ERRSYNTAX);
+ if (ls->mode) {
+ int xmode = 1;
+ const char *mode = ls->mode;
+ char c;
+ while ((c = *mode++)) {
+ if (c == (bc ? 'b' : 't')) xmode = 0;
+ if (c == (LJ_FR2 ? 'W' : 'X')) ls->fr2 = !LJ_FR2;
+ }
+ if (xmode) {
+ setstrV(L, L->top++, lj_err_str(L, LJ_ERR_XMODE));
+ lj_err_throw(L, LUA_ERRSYNTAX);
+ }
}
pt = bc ? lj_bcread(ls) : lj_parse(ls);
- fn = lj_func_newL_empty(L, pt, tabref(L->env));
- /* Don't combine above/below into one statement. */
- setfuncV(L, L->top++, fn);
+ if (ls->fr2 == LJ_FR2) {
+ fn = lj_func_newL_empty(L, pt, tabref(L->env));
+ /* Don't combine above/below into one statement. */
+ setfuncV(L, L->top++, fn);
+ } else {
+ /* Non-native generation returns a dumpable, but non-runnable prototype. */
+ setprotoV(L, L->top++, pt);
+ }
return NULL;
}
@@ -87,30 +101,31 @@
FileReaderCtx ctx;
int status;
const char *chunkname;
+ int err = 0;
if (filename) {
+ chunkname = lua_pushfstring(L, "@%s", filename);
ctx.fp = fopen(filename, "rb");
if (ctx.fp == NULL) {
+ L->top--;
lua_pushfstring(L, "cannot open %s: %s", filename, strerror(errno));
return LUA_ERRFILE;
}
- chunkname = lua_pushfstring(L, "@%s", filename);
} else {
ctx.fp = stdin;
chunkname = "=stdin";
}
status = lua_loadx(L, reader_file, &ctx, chunkname, mode);
- if (ferror(ctx.fp)) {
- L->top -= filename ? 2 : 1;
- lua_pushfstring(L, "cannot read %s: %s", chunkname+1, strerror(errno));
- if (filename)
- fclose(ctx.fp);
- return LUA_ERRFILE;
- }
+ if (ferror(ctx.fp)) err = errno;
if (filename) {
+ fclose(ctx.fp);
L->top--;
copyTV(L, L->top-1, L->top);
- fclose(ctx.fp);
}
+ if (err) {
+ L->top--;
+ lua_pushfstring(L, "cannot read %s: %s", chunkname+1, strerror(err));
+ return LUA_ERRFILE;
+ }
return status;
}
@@ -159,18 +174,19 @@
LUA_API int lua_dump(lua_State *L, lua_Writer writer, void *data)
{
cTValue *o = L->top-1;
- api_check(L, L->top > L->base);
+ uint32_t flags = LJ_FR2*BCDUMP_F_FR2; /* Default mode for legacy C API. */
+ lj_checkapi(L->top > L->base, "top slot empty");
if (tvisfunc(o) && isluafunc(funcV(o)))
- return lj_bcwrite(L, funcproto(funcV(o)), writer, data, 0);
+ return lj_bcwrite(L, funcproto(funcV(o)), writer, data, flags);
else
return 1;
}
-
-/* -- Luajittex needs this one because it's faster than make it Lua -- */
+
+/* -- Luajittex needs this one because it's faster than make it with Lua -- */
LUA_API int RESERVED_lua_dump(lua_State *L, lua_Writer writer, void *data, int strip)
{
cTValue *o = L->top-1;
- api_check(L, L->top > L->base);
+ lj_checkapi(L->top > L->base, "top slot empty");
if (tvisfunc(o) && isluafunc(funcV(o)))
return lj_bcwrite(L, funcproto(funcV(o)), writer, data, strip);
else
@@ -194,3 +210,4 @@
}
}
+
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Machine code management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_mcode_c
@@ -14,6 +14,7 @@
#include "lj_mcode.h"
#include "lj_trace.h"
#include "lj_dispatch.h"
+#include "lj_prng.h"
#endif
#if LJ_HASJIT || LJ_HASFFI
#include "lj_vm.h"
@@ -28,6 +29,11 @@
#include <valgrind/valgrind.h>
#endif
+#if LJ_TARGET_WINDOWS
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+#endif
+
#if LJ_TARGET_IOS
void sys_icache_invalidate(void *start, size_t len);
#endif
@@ -40,11 +46,13 @@
#endif
#if LJ_TARGET_X86ORX64
UNUSED(start); UNUSED(end);
+#elif LJ_TARGET_WINDOWS
+ FlushInstructionCache(GetCurrentProcess(), start, (char *)end-(char *)start);
#elif LJ_TARGET_IOS
sys_icache_invalidate(start, (char *)end-(char *)start);
#elif LJ_TARGET_PPC
lj_vm_cachesync(start, end);
-#elif defined(__GNUC__)
+#elif defined(__GNUC__) || defined(__clang__)
__clear_cache(start, end);
#else
#error "Missing builtin to flush instruction cache"
@@ -57,9 +65,6 @@
#if LJ_TARGET_WINDOWS
-#define WIN32_LEAN_AND_MEAN
-#include <windows.h>
-
#define MCPROT_RW PAGE_READWRITE
#define MCPROT_RX PAGE_EXECUTE_READ
#define MCPROT_RWX PAGE_EXECUTE_READWRITE
@@ -66,8 +71,8 @@
static void *mcode_alloc_at(jit_State *J, uintptr_t hint, size_t sz, DWORD prot)
{
- void *p = VirtualAlloc((void *)hint, sz,
- MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN, prot);
+ void *p = LJ_WIN_VALLOC((void *)hint, sz,
+ MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN, prot);
if (!p && !hint)
lj_trace_err(J, LJ_TRERR_MCODEAL);
return p;
@@ -82,7 +87,7 @@
static int mcode_setprot(void *p, size_t sz, DWORD prot)
{
DWORD oprot;
- return !VirtualProtect(p, sz, prot, &oprot);
+ return !LJ_WIN_VPROTECT(p, sz, prot, &oprot);
}
#elif LJ_TARGET_POSIX
@@ -96,10 +101,15 @@
#define MCPROT_RW (PROT_READ|PROT_WRITE)
#define MCPROT_RX (PROT_READ|PROT_EXEC)
#define MCPROT_RWX (PROT_READ|PROT_WRITE|PROT_EXEC)
+#ifdef PROT_MPROTECT
+#define MCPROT_CREATE (PROT_MPROTECT(MCPROT_RWX))
+#else
+#define MCPROT_CREATE 0
+#endif
static void *mcode_alloc_at(jit_State *J, uintptr_t hint, size_t sz, int prot)
{
- void *p = mmap((void *)hint, sz, prot, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0);
+ void *p = mmap((void *)hint, sz, prot|MCPROT_CREATE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0);
if (p == MAP_FAILED) {
if (!hint) lj_trace_err(J, LJ_TRERR_MCODEAL);
p = NULL;
@@ -118,45 +128,27 @@
return mprotect(p, sz, prot);
}
-#elif LJ_64
+#else
#error "Missing OS support for explicit placement of executable memory"
-#else
-
-/* Fallback allocator. This will fail if memory is not executable by default. */
-#define LUAJIT_UNPROTECT_MCODE
-#define MCPROT_RW 0
-#define MCPROT_RX 0
-#define MCPROT_RWX 0
-
-static void *mcode_alloc_at(jit_State *J, uintptr_t hint, size_t sz, int prot)
-{
- UNUSED(hint); UNUSED(prot);
- return lj_mem_new(J->L, sz);
-}
-
-static void mcode_free(jit_State *J, void *p, size_t sz)
-{
- lj_mem_free(J2G(J), p, sz);
-}
-
#endif
/* -- MCode area protection ----------------------------------------------- */
-/* Define this ONLY if page protection twiddling becomes a bottleneck. */
-#ifdef LUAJIT_UNPROTECT_MCODE
+#if LUAJIT_SECURITY_MCODE == 0
-/* It's generally considered to be a potential security risk to have
+/* Define this ONLY if page protection twiddling becomes a bottleneck.
+**
+** It's generally considered to be a potential security risk to have
** pages with simultaneous write *and* execute access in a process.
**
** Do not even think about using this mode for server processes or
-** apps handling untrusted external data (such as a browser).
+** apps handling untrusted external data.
**
** The security risk is not in LuaJIT itself -- but if an adversary finds
-** any *other* flaw in your C application logic, then any RWX memory page
-** simplifies writing an exploit considerably.
+** any *other* flaw in your C application logic, then any RWX memory pages
+** simplify writing an exploit considerably.
*/
#define MCPROT_GEN MCPROT_RWX
#define MCPROT_RUN MCPROT_RWX
@@ -163,7 +155,7 @@
static void mcode_protect(jit_State *J, int prot)
{
- UNUSED(J); UNUSED(prot);
+ UNUSED(J); UNUSED(prot); UNUSED(mcode_setprot);
}
#else
@@ -180,7 +172,7 @@
#define MCPROT_RUN MCPROT_RX
/* Protection twiddling failed. Probably due to kernel security. */
-static LJ_NOINLINE void mcode_protfail(jit_State *J)
+static LJ_NORET LJ_NOINLINE void mcode_protfail(jit_State *J)
{
lua_CFunction panic = J2G(J)->panic;
if (panic) {
@@ -188,6 +180,7 @@
setstrV(L, L->top++, lj_err_str(L, LJ_ERR_JITPROT));
panic(L);
}
+ exit(EXIT_FAILURE);
}
/* Change protection of MCode area. */
@@ -242,7 +235,7 @@
}
/* Next try probing 64K-aligned pseudo-random addresses. */
do {
- hint = LJ_PRNG_BITS(J, LJ_TARGET_JUMPRANGE-16) << 16;
+ hint = lj_prng_u64(&J2G(J)->prng) & ((1u<<LJ_TARGET_JUMPRANGE)-0x10000);
} while (!(hint + sz < range+range));
hint = target + hint - range;
}
@@ -255,7 +248,7 @@
/* All memory addresses are reachable by relative jumps. */
static void *mcode_alloc(jit_State *J, size_t sz)
{
-#ifdef __OpenBSD__
+#if defined(__OpenBSD__) || defined(__NetBSD__) || LJ_TARGET_UWP
/* Allow better executable memory allocation for OpenBSD W^X mode. */
void *p = mcode_alloc_at(J, 0, sz, MCPROT_RUN);
if (p && mcode_setprot(p, sz, MCPROT_GEN)) {
@@ -272,12 +265,6 @@
/* -- MCode area management ----------------------------------------------- */
-/* Linked list of MCode areas. */
-typedef struct MCLink {
- MCode *next; /* Next area. */
- size_t size; /* Size of current area. */
-} MCLink;
-
/* Allocate a new MCode area. */
static void mcode_allocarea(jit_State *J)
{
@@ -292,6 +279,7 @@
((MCLink *)J->mcarea)->next = oldarea;
((MCLink *)J->mcarea)->size = sz;
J->szallmcarea += sz;
+ J->mcbot = (MCode *)lj_err_register_mcode(J->mcarea, sz, (uint8_t *)J->mcbot);
}
/* Free all MCode areas. */
@@ -302,7 +290,9 @@
J->szallmcarea = 0;
while (mc) {
MCode *next = ((MCLink *)mc)->next;
- mcode_free(J, mc, ((MCLink *)mc)->size);
+ size_t sz = ((MCLink *)mc)->size;
+ lj_err_deregister_mcode(mc, sz, (uint8_t *)mc + sizeof(MCLink));
+ mcode_free(J, mc, sz);
mc = next;
}
}
@@ -337,35 +327,36 @@
/* Set/reset protection to allow patching of MCode areas. */
MCode *lj_mcode_patch(jit_State *J, MCode *ptr, int finish)
{
-#ifdef LUAJIT_UNPROTECT_MCODE
- UNUSED(J); UNUSED(ptr); UNUSED(finish);
- return NULL;
-#else
if (finish) {
+#if LUAJIT_SECURITY_MCODE
if (J->mcarea == ptr)
mcode_protect(J, MCPROT_RUN);
else if (LJ_UNLIKELY(mcode_setprot(ptr, ((MCLink *)ptr)->size, MCPROT_RUN)))
mcode_protfail(J);
+#endif
return NULL;
} else {
MCode *mc = J->mcarea;
/* Try current area first to use the protection cache. */
if (ptr >= mc && ptr < (MCode *)((char *)mc + J->szmcarea)) {
+#if LUAJIT_SECURITY_MCODE
mcode_protect(J, MCPROT_GEN);
+#endif
return mc;
}
/* Otherwise search through the list of MCode areas. */
for (;;) {
mc = ((MCLink *)mc)->next;
- lua_assert(mc != NULL);
+ lj_assertJ(mc != NULL, "broken MCode area chain");
if (ptr >= mc && ptr < (MCode *)((char *)mc + ((MCLink *)mc)->size)) {
+#if LUAJIT_SECURITY_MCODE
if (LJ_UNLIKELY(mcode_setprot(mc, ((MCLink *)mc)->size, MCPROT_GEN)))
mcode_protfail(J);
+#endif
return mc;
}
}
}
-#endif
}
/* Limit of MCode reservation reached. */
@@ -376,7 +367,7 @@
sizemcode = (size_t)J->param[JIT_P_sizemcode] << 10;
sizemcode = (sizemcode + LJ_PAGESIZE-1) & ~(size_t)(LJ_PAGESIZE - 1);
maxmcode = (size_t)J->param[JIT_P_maxmcode] << 10;
- if ((size_t)need > sizemcode)
+ if (need * sizeof(MCode) > sizemcode)
lj_trace_err(J, LJ_TRERR_MCODEOV); /* Too long for any area. */
if (J->szallmcarea + sizemcode > maxmcode)
lj_trace_err(J, LJ_TRERR_MCODEAL);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_mcode.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Machine code management.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_MCODE_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Metamethod handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -47,7 +47,7 @@
cTValue *lj_meta_cache(GCtab *mt, MMS mm, GCstr *name)
{
cTValue *mo = lj_tab_getstr(mt, name);
- lua_assert(mm <= MM_FAST);
+ lj_assertX(mm <= MM_FAST, "bad metamethod %d", mm);
if (!mo || tvisnil(mo)) { /* No metamethod? */
mt->nomm |= (uint8_t)(1u<<mm); /* Set negative cache flag. */
return NULL;
@@ -86,8 +86,8 @@
else
top->u32.lo = LJ_CONT_TAILCALL;
setframe_pc(top++, pc);
+ setframe_gc(top, obj2gco(L), LJ_TTHREAD); /* Dummy frame object. */
if (LJ_FR2) top++;
- setframe_gc(top, obj2gco(L), LJ_TTHREAD); /* Dummy frame object. */
setframe_ftsz(top, ((char *)(top+1) - (char *)base) + FRAME_CONT);
L->base = L->top = top+1;
/*
@@ -240,8 +240,8 @@
int fromc = 0;
if (left < 0) { left = -left; fromc = 1; }
do {
- if (!(tvisstr(top) || tvisnumber(top)) ||
- !(tvisstr(top-1) || tvisnumber(top-1))) {
+ if (!(tvisstr(top) || tvisnumber(top) || tvisbuf(top)) ||
+ !(tvisstr(top-1) || tvisnumber(top-1) || tvisbuf(top-1))) {
cTValue *mo = lj_meta_lookup(L, top-1, MM_concat);
if (tvisnil(mo)) {
mo = lj_meta_lookup(L, top, MM_concat);
@@ -277,10 +277,12 @@
** next step: [...][CAT stack ............]
*/
TValue *e, *o = top;
- uint64_t tlen = tvisstr(o) ? strV(o)->len : STRFMT_MAXBUF_NUM;
+ uint64_t tlen = tvisstr(o) ? strV(o)->len :
+ tvisbuf(o) ? sbufxlen(bufV(o)) : STRFMT_MAXBUF_NUM;
SBuf *sb;
do {
- o--; tlen += tvisstr(o) ? strV(o)->len : STRFMT_MAXBUF_NUM;
+ o--; tlen += tvisstr(o) ? strV(o)->len :
+ tvisbuf(o) ? sbufxlen(bufV(o)) : STRFMT_MAXBUF_NUM;
} while (--left > 0 && (tvisstr(o-1) || tvisnumber(o-1)));
if (tlen >= LJ_MAX_STR) lj_err_msg(L, LJ_ERR_STROV);
sb = lj_buf_tmp_(L);
@@ -290,6 +292,9 @@
GCstr *s = strV(o);
MSize len = s->len;
lj_buf_putmem(sb, strdata(s), len);
+ } else if (tvisbuf(o)) {
+ SBufExt *sbx = bufV(o);
+ lj_buf_putmem(sb, sbx->r, sbufxlen(sbx));
} else if (tvisint(o)) {
lj_strfmt_putint(sb, intV(o));
} else {
@@ -363,7 +368,7 @@
} else if (op == BC_ISEQN) {
o2 = &mref(curr_proto(L)->k, cTValue)[bc_d(ins)];
} else {
- lua_assert(op == BC_ISEQP);
+ lj_assertL(op == BC_ISEQP, "bad bytecode op %d", op);
setpriV(&tv, ~bc_d(ins));
o2 = &tv;
}
@@ -426,7 +431,7 @@
{
L->top = curr_topL(L);
ra++; tp--;
- lua_assert(LJ_DUALNUM || tp != ~LJ_TNUMX); /* ISTYPE -> ISNUM broken. */
+ lj_assertL(LJ_DUALNUM || tp != ~LJ_TNUMX, "bad type for ISTYPE");
if (LJ_DUALNUM && tp == ~LJ_TNUMX) lj_lib_checkint(L, ra);
else if (tp == ~LJ_TNUMX+1) lj_lib_checknum(L, ra);
else if (tp == ~LJ_TSTR) lj_lib_checkstr(L, ra);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_meta.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Metamethod handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_META_H
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Miscellaneous object handling.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_obj_c
@@ -34,12 +34,13 @@
}
/* Return pointer to object or its object data. */
-const void * LJ_FASTCALL lj_obj_ptr(cTValue *o)
+const void * LJ_FASTCALL lj_obj_ptr(global_State *g, cTValue *o)
{
+ UNUSED(g);
if (tvisudata(o))
return uddata(udataV(o));
else if (tvislightud(o))
- return lightudV(o);
+ return lightudV(g, o);
else if (LJ_HASFFI && tviscdata(o))
return cdataptr(cdataV(o));
else if (tvisgcv(o))
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_obj.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LuaJIT VM tags, values and objects.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -13,7 +13,7 @@
#include "lj_def.h"
#include "lj_arch.h"
-/* -- Memory references (32 bit address space) ---------------------------- */
+/* -- Memory references --------------------------------------------------- */
/* Memory and GC object sizes. */
typedef uint32_t MSize;
@@ -34,17 +34,21 @@
#if LJ_GC64
#define mref(r, t) ((t *)(void *)(r).ptr64)
+#define mrefu(r) ((r).ptr64)
#define setmref(r, p) ((r).ptr64 = (uint64_t)(void *)(p))
+#define setmrefu(r, u) ((r).ptr64 = (uint64_t)(u))
#define setmrefr(r, v) ((r).ptr64 = (v).ptr64)
#else
#define mref(r, t) ((t *)(void *)(uintptr_t)(r).ptr32)
+#define mrefu(r) ((r).ptr32)
#define setmref(r, p) ((r).ptr32 = (uint32_t)(uintptr_t)(void *)(p))
+#define setmrefu(r, u) ((r).ptr32 = (uint32_t)(u))
#define setmrefr(r, v) ((r).ptr32 = (v).ptr32)
#endif
-/* -- GC object references (32 bit address space) ------------------------- */
+/* -- GC object references ------------------------------------------------ */
/* GCobj reference */
typedef struct GCRef {
@@ -153,11 +157,9 @@
typedef void (*ASMFunction)(void);
/* Resizable string buffer. Need this here, details in lj_buf.h. */
+#define SBufHeader char *w, *e, *b; MRef L
typedef struct SBuf {
- MRef p; /* String buffer pointer. */
- MRef e; /* String buffer end pointer. */
- MRef b; /* String buffer base. */
- MRef L; /* lua_State, used for buffer resizing. */
+ SBufHeader;
} SBuf;
/* -- Tags and values ----------------------------------------------------- */
@@ -232,7 +234,7 @@
** ---MSW---.---LSW---
** primitive types | itype | |
** lightuserdata | itype | void * | (32 bit platforms)
-** lightuserdata |ffff| void * | (64 bit platforms, 47 bit pointers)
+** lightuserdata |ffff|seg| ofs | (64 bit platforms)
** GC objects | itype | GCRef |
** int (LJ_DUALNUM)| itype | int |
** number -------double------
@@ -245,7 +247,8 @@
**
** ------MSW------.------LSW------
** primitive types |1..1|itype|1..................1|
-** GC objects/lightud |1..1|itype|-------GCRef--------|
+** GC objects |1..1|itype|-------GCRef--------|
+** lightuserdata |1..1|itype|seg|------ofs-------|
** int (LJ_DUALNUM) |1..1|itype|0..0|-----int-------|
** number ------------double-------------
**
@@ -281,18 +284,31 @@
#define LJ_TISGCV (LJ_TSTR+1)
#define LJ_TISTABUD LJ_TTAB
+/* Type marker for slot holding a traversal index. Must be lightuserdata. */
+#define LJ_KEYINDEX 0xfffe7fffu
+
#if LJ_GC64
#define LJ_GCVMASK (((uint64_t)1 << 47) - 1)
#endif
+#if LJ_64
+/* To stay within 47 bits, lightuserdata is segmented. */
+#define LJ_LIGHTUD_BITS_SEG 8
+#define LJ_LIGHTUD_BITS_LO (47 - LJ_LIGHTUD_BITS_SEG)
+#endif
+
/* -- String object ------------------------------------------------------- */
+typedef uint32_t StrHash; /* String hash value. */
+typedef uint32_t StrID; /* String ID. */
+
/* String object header. String payload follows. */
typedef struct GCstr {
GCHeader;
uint8_t reserved; /* Used by lexer for fast lookup of reserved words. */
- uint8_t unused;
- MSize hash; /* Hash of string. */
+ uint8_t hashalg; /* Hash algorithm. */
+ StrID sid; /* Interned string ID. */
+ StrHash hash; /* Hash of string. */
MSize len; /* Size of string. */
} GCstr;
@@ -300,7 +316,6 @@
#define strdata(s) ((const char *)((s)+1))
#define strdatawr(s) ((char *)((s)+1))
#define strVdata(o) strdata(strV(o))
-#define sizestring(s) (sizeof(struct GCstr)+(s)->len+1)
/* -- Userdata object ----------------------------------------------------- */
@@ -320,6 +335,7 @@
UDTYPE_USERDATA, /* Regular userdata. */
UDTYPE_IO_FILE, /* I/O library FILE. */
UDTYPE_FFI_CLIB, /* FFI C library namespace. */
+ UDTYPE_BUFFER, /* String buffer. */
UDTYPE__MAX
};
@@ -397,7 +413,7 @@
#define PROTO_UV_IMMUTABLE 0x4000 /* Immutable upvalue. */
#define proto_kgc(pt, idx) \
- check_exp((uintptr_t)(intptr_t)(idx) >= (uintptr_t)-(intptr_t)(pt)->sizekgc, \
+ check_exp((uintptr_t)(intptr_t)(idx) >= ~(uintptr_t)(pt)->sizekgc+1u, \
gcref(mref((pt)->k, GCRef)[(idx)]))
#define proto_knumtv(pt, idx) \
check_exp((uintptr_t)(idx) < (pt)->sizekn, &mref((pt)->k, TValue)[(idx)])
@@ -495,7 +511,7 @@
} GCtab;
#define sizetabcolo(n) ((n)*sizeof(TValue) + sizeof(GCtab))
-#define tabref(r) (&gcref((r))->tab)
+#define tabref(r) ((GCtab *)gcref((r)))
#define noderef(r) (mref((r), Node))
#define nextnode(n) (mref((n)->next, Node))
#if LJ_GC64
@@ -563,6 +579,9 @@
GCROOT_BASEMT_NUM = GCROOT_BASEMT + ~LJ_TNUMX,
GCROOT_IO_INPUT, /* Userdata for default I/O input file. */
GCROOT_IO_OUTPUT, /* Userdata for default I/O output file. */
+#if LJ_HASFFI
+ GCROOT_FFI_FIN, /* FFI finalizer table. */
+#endif
GCROOT_MAX
} GCRootID;
@@ -570,13 +589,18 @@
#define basemt_obj(g, o) ((g)->gcroot[GCROOT_BASEMT+itypemap(o)])
#define mmname_str(g, mm) (strref((g)->gcroot[GCROOT_MMNAME+(mm)]))
+/* Garbage collector state. */
typedef struct GCState {
GCSize total; /* Memory currently allocated. */
GCSize threshold; /* Memory threshold. */
uint8_t currentwhite; /* Current white color. */
uint8_t state; /* GC state. */
- uint8_t nocdatafin; /* No cdata finalizer called. */
- uint8_t unused2;
+ uint8_t unused0;
+#if LJ_64
+ uint8_t lightudnum; /* Number of lightuserdata segments - 1. */
+#else
+ uint8_t unused1;
+#endif
MSize sweepstr; /* Sweep position in string table. */
GCRef root; /* List of all collectable objects. */
MRef sweep; /* Sweep position in root list. */
@@ -588,27 +612,41 @@
GCSize estimate; /* Estimate of memory actually in use. */
MSize stepmul; /* Incremental GC step granularity. */
MSize pause; /* Pause between successive GC cycles. */
+#if LJ_64
+ MRef lightudseg; /* Upper bits of lightuserdata segments. */
+#endif
} GCState;
+/* String interning state. */
+typedef struct StrInternState {
+ GCRef *tab; /* String hash table anchors. */
+ MSize mask; /* String hash mask (size of hash table - 1). */
+ MSize num; /* Number of strings in hash table. */
+ StrID id; /* Next string ID. */
+ uint8_t idreseed; /* String ID reseed counter. */
+ uint8_t second; /* String interning table uses secondary hashing. */
+ uint8_t unused1;
+ uint8_t unused2;
+ LJ_ALIGN(8) uint64_t seed; /* Random string seed. */
+} StrInternState;
+
/* Global state, shared by all threads of a Lua universe. */
typedef struct global_State {
- GCRef *strhash; /* String hash table (hash chain anchors). */
- MSize strmask; /* String hash mask (size of hash table - 1). */
- MSize strnum; /* Number of strings in hash table. */
lua_Alloc allocf; /* Memory allocator. */
void *allocd; /* Memory allocator data. */
GCState gc; /* Garbage collector. */
- volatile int32_t vmstate; /* VM state or current JIT code trace number. */
- SBuf tmpbuf; /* Temporary string buffer. */
GCstr strempty; /* Empty string. */
uint8_t stremptyz; /* Zero terminator of empty string. */
uint8_t hookmask; /* Hook mask. */
uint8_t dispatchmode; /* Dispatch mode. */
uint8_t vmevmask; /* VM event mask. */
+ StrInternState str; /* String interning. */
+ volatile int32_t vmstate; /* VM state or current JIT code trace number. */
GCRef mainthref; /* Link to main thread. */
- TValue registrytv; /* Anchor for registry. */
+ SBuf tmpbuf; /* Temporary string buffer. */
TValue tmptv, tmptv2; /* Temporary TValues. */
Node nilnode; /* Fallback 1-element hash part (nil key and value). */
+ TValue registrytv; /* Anchor for registry. */
GCupval uvhead; /* Head of double-linked list of all open upvalues. */
int32_t hookcount; /* Instruction hook countdown. */
int32_t hookcstart; /* Start count for instruction hook counter. */
@@ -620,6 +658,7 @@
GCRef cur_L; /* Currently executing lua_State. */
MRef jit_base; /* Current JIT code L->base or NULL. */
MRef ctype_state; /* Pointer to C type state. */
+ PRNGState prng; /* Global PRNG state. */
GCRef gcroot[GCROOT_MAX]; /* GC roots. */
} global_State;
@@ -638,7 +677,8 @@
#define HOOK_PROFILE 0x80
#define hook_active(g) ((g)->hookmask & HOOK_ACTIVE)
#define hook_enter(g) ((g)->hookmask |= HOOK_ACTIVE)
-#define hook_entergc(g) ((g)->hookmask |= (HOOK_ACTIVE|HOOK_GC))
+#define hook_entergc(g) \
+ ((g)->hookmask = ((g)->hookmask | (HOOK_ACTIVE|HOOK_GC)) & ~HOOK_PROFILE)
#define hook_vmevent(g) ((g)->hookmask |= (HOOK_ACTIVE|HOOK_VMEVENT))
#define hook_leave(g) ((g)->hookmask &= ~HOOK_ACTIVE)
#define hook_save(g) ((g)->hookmask & ~HOOK_EVENTMASK)
@@ -678,6 +718,11 @@
#define curr_topL(L) (L->base + curr_proto(L)->framesize)
#define curr_top(L) (curr_funcisL(L) ? curr_topL(L) : L->top)
+#if defined(LUA_USE_ASSERT) || defined(LUA_USE_APICHECK)
+LJ_FUNC_NORET void lj_assert_fail(global_State *g, const char *file, int line,
+ const char *func, const char *fmt, ...);
+#endif
+
/* -- GC object definition and conversions -------------------------------- */
/* GC header for generic access to common fields of GC objects. */
@@ -731,10 +776,6 @@
/* -- TValue getters/setters ---------------------------------------------- */
-#ifdef LUA_USE_ASSERT
-#include "lj_gc.h"
-#endif
-
/* Macros to test types. */
#if LJ_GC64
#define itype(o) ((uint32_t)((o)->it64 >> 47))
@@ -795,10 +836,24 @@
#endif
#define boolV(o) check_exp(tvisbool(o), (LJ_TFALSE - itype(o)))
#if LJ_64
-#define lightudV(o) \
- check_exp(tvislightud(o), (void *)((o)->u64 & U64x(00007fff,ffffffff)))
+#define lightudseg(u) \
+ (((u) >> LJ_LIGHTUD_BITS_LO) & ((1 << LJ_LIGHTUD_BITS_SEG)-1))
+#define lightudlo(u) \
+ ((u) & (((uint64_t)1 << LJ_LIGHTUD_BITS_LO) - 1))
+#define lightudup(p) \
+ ((uint32_t)(((p) >> LJ_LIGHTUD_BITS_LO) << (LJ_LIGHTUD_BITS_LO-32)))
+static LJ_AINLINE void *lightudV(global_State *g, cTValue *o)
+{
+ uint64_t u = o->u64;
+ uint64_t seg = lightudseg(u);
+ uint32_t *segmap = mref(g->gc.lightudseg, uint32_t);
+ lj_assertG(tvislightud(o), "lightuserdata expected");
+ if (seg == (1 << LJ_LIGHTUD_BITS_SEG)-1) return NULL;
+ lj_assertG(seg <= g->gc.lightudnum, "bad lightuserdata segment %d", seg);
+ return (void *)(((uint64_t)segmap[seg] << 32) | lightudlo(u));
+}
#else
-#define lightudV(o) check_exp(tvislightud(o), gcrefp((o)->gcr, void))
+#define lightudV(g, o) check_exp(tvislightud(o), gcrefp((o)->gcr, void))
#endif
#define gcV(o) check_exp(tvisgcv(o), gcval(o))
#define strV(o) check_exp(tvisstr(o), &gcval(o)->str)
@@ -824,7 +879,7 @@
#define setpriV(o, i) (setitype((o), (i)))
#endif
-static LJ_AINLINE void setlightudV(TValue *o, void *p)
+static LJ_AINLINE void setrawlightudV(TValue *o, void *p)
{
#if LJ_GC64
o->u64 = (uint64_t)p | (((uint64_t)LJ_TLIGHTUD) << 47);
@@ -835,29 +890,29 @@
#endif
}
-#if LJ_64
-#define checklightudptr(L, p) \
- (((uint64_t)(p) >> 47) ? (lj_err_msg(L, LJ_ERR_BADLU), NULL) : (p))
-#else
-#define checklightudptr(L, p) (p)
-#endif
-
-#if LJ_FR2
+#if LJ_FR2 || LJ_32
#define contptr(f) ((void *)(f))
#define setcont(o, f) ((o)->u64 = (uint64_t)(uintptr_t)contptr(f))
-#elif LJ_64
+#else
#define contptr(f) \
((void *)(uintptr_t)(uint32_t)((intptr_t)(f) - (intptr_t)lj_vm_asm_begin))
#define setcont(o, f) \
((o)->u64 = (uint64_t)(void *)(f) - (uint64_t)lj_vm_asm_begin)
-#else
-#define contptr(f) ((void *)(f))
-#define setcont(o, f) setlightudV((o), contptr(f))
#endif
-#define tvchecklive(L, o) \
- UNUSED(L), lua_assert(!tvisgcv(o) || \
- ((~itype(o) == gcval(o)->gch.gct) && !isdead(G(L), gcval(o))))
+static LJ_AINLINE void checklivetv(lua_State *L, TValue *o, const char *msg)
+{
+ UNUSED(L); UNUSED(o); UNUSED(msg);
+#if LUA_USE_ASSERT
+ if (tvisgcv(o)) {
+ lj_assertL(~itype(o) == gcval(o)->gch.gct,
+ "mismatch of TValue type %d vs GC type %d",
+ ~itype(o), gcval(o)->gch.gct);
+ /* Copy of isdead check from lj_gc.h to avoid circular include. */
+ lj_assertL(!(gcval(o)->gch.marked & (G(L)->gc.currentwhite ^ 3) & 3), msg);
+ }
+#endif
+}
static LJ_AINLINE void setgcVraw(TValue *o, GCobj *v, uint32_t itype)
{
@@ -870,11 +925,12 @@
static LJ_AINLINE void setgcV(lua_State *L, TValue *o, GCobj *v, uint32_t it)
{
- setgcVraw(o, v, it); tvchecklive(L, o);
+ setgcVraw(o, v, it);
+ checklivetv(L, o, "store to dead GC object");
}
#define define_setV(name, type, tag) \
-static LJ_AINLINE void name(lua_State *L, TValue *o, type *v) \
+static LJ_AINLINE void name(lua_State *L, TValue *o, const type *v) \
{ \
setgcV(L, o, obj2gco(v), tag); \
}
@@ -917,7 +973,8 @@
/* Copy tagged values. */
static LJ_AINLINE void copyTV(lua_State *L, TValue *o1, const TValue *o2)
{
- *o1 = *o2; tvchecklive(L, o1);
+ *o1 = *o2;
+ checklivetv(L, o1, "copy of dead GC object");
}
/* -- Number to integer conversion ---------------------------------------- */
@@ -924,7 +981,10 @@
#if LJ_SOFTFP
LJ_ASMF int32_t lj_vm_tobit(double x);
+#if LJ_TARGET_MIPS64
+LJ_ASMF int32_t lj_vm_tointg(double x);
#endif
+#endif
static LJ_AINLINE int32_t lj_num2bit(lua_Number n)
{
@@ -939,14 +999,22 @@
#define lj_num2int(n) ((int32_t)(n))
+/*
+** This must match the JIT backend behavior. In particular for archs
+** that don't have a common hardware instruction for this conversion.
+** Note that signed FP to unsigned int conversions have an undefined
+** result and should never be relied upon in portable FFI code.
+** See also: C99 or C11 standard, 6.3.1.4, footnote of (1).
+*/
static LJ_AINLINE uint64_t lj_num2u64(lua_Number n)
{
-#ifdef _MSC_VER
- if (n >= 9223372036854775808.0) /* They think it's a feature. */
- return (uint64_t)(int64_t)(n - 18446744073709551616.0);
- else
+#if LJ_TARGET_X86ORX64 || LJ_TARGET_MIPS
+ int64_t i = (int64_t)n;
+ if (i < 0) i = (int64_t)(n - 18446744073709551616.0);
+ return (uint64_t)i;
+#else
+ return (uint64_t)n;
#endif
- return (uint64_t)n;
}
static LJ_AINLINE int32_t numberVint(cTValue *o)
@@ -975,6 +1043,20 @@
/* Compare two objects without calling metamethods. */
LJ_FUNC int LJ_FASTCALL lj_obj_equal(cTValue *o1, cTValue *o2);
-LJ_FUNC const void * LJ_FASTCALL lj_obj_ptr(cTValue *o);
+LJ_FUNC const void * LJ_FASTCALL lj_obj_ptr(global_State *g, cTValue *o);
+#if LJ_ABI_PAUTH
+#if LJ_TARGET_ARM64
+#include <ptrauth.h>
+#define lj_ptr_sign(ptr, ctx) \
+ ptrauth_sign_unauthenticated((ptr), ptrauth_key_function_pointer, (ctx))
+#define lj_ptr_strip(ptr) ptrauth_strip((ptr), ptrauth_key_function_pointer)
+#else
+#error "No support for pointer authentication for this architecture"
#endif
+#else
+#define lj_ptr_sign(ptr, ctx) (ptr)
+#define lj_ptr_strip(ptr) (ptr)
+#endif
+
+#endif
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_dce.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_dce.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_dce.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** DCE: Dead Code Elimination. Pre-LOOP only -- ASM already performs DCE.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_dce_c
@@ -44,15 +44,12 @@
IRIns *ir = IR(ins);
if (irt_ismarked(ir->t)) {
irt_clearmark(ir->t);
- pchain[ir->o] = &ir->prev;
} else if (!ir_sideeff(ir)) {
*pchain[ir->o] = ir->prev; /* Reroute original instruction chain. */
- ir->t.irt = IRT_NIL;
- ir->o = IR_NOP; /* Replace instruction with NOP. */
- ir->op1 = ir->op2 = 0;
- ir->prev = 0;
+ lj_ir_nop(ir);
continue;
}
+ pchain[ir->o] = &ir->prev;
if (ir->op1 >= REF_FIRST) irt_setmark(IR(ir->op1)->t);
if (ir->op2 >= REF_FIRST) irt_setmark(IR(ir->op2)->t);
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_fold.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_fold.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_fold.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -2,7 +2,7 @@
** FOLD: Constant Folding, Algebraic Simplifications and Reassociation.
** ABCelim: Array Bounds Check Elimination.
** CSE: Common-Subexpression Elimination.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_fold_c
@@ -173,7 +173,6 @@
LJFOLD(SUB KNUM KNUM)
LJFOLD(MUL KNUM KNUM)
LJFOLD(DIV KNUM KNUM)
-LJFOLD(ATAN2 KNUM KNUM)
LJFOLD(LDEXP KNUM KNUM)
LJFOLD(MIN KNUM KNUM)
LJFOLD(MAX KNUM KNUM)
@@ -213,13 +212,34 @@
return lj_ir_knum(J, y);
}
-LJFOLD(POW KNUM KINT)
+LJFOLD(CALLN KNUM any)
+LJFOLDF(kfold_fpcall1)
+{
+ const CCallInfo *ci = &lj_ir_callinfo[fins->op2];
+ if (CCI_TYPE(ci) == IRT_NUM) {
+ double y = ((double (*)(double))ci->func)(knumleft);
+ return lj_ir_knum(J, y);
+ }
+ return NEXTFOLD;
+}
+
+LJFOLD(CALLN CARG IRCALL_atan2)
+LJFOLDF(kfold_fpcall2)
+{
+ if (irref_isk(fleft->op1) && irref_isk(fleft->op2)) {
+ const CCallInfo *ci = &lj_ir_callinfo[fins->op2];
+ double a = ir_knum(IR(fleft->op1))->n;
+ double b = ir_knum(IR(fleft->op2))->n;
+ double y = ((double (*)(double, double))ci->func)(a, b);
+ return lj_ir_knum(J, y);
+ }
+ return NEXTFOLD;
+}
+
+LJFOLD(POW KNUM KNUM)
LJFOLDF(kfold_numpow)
{
- lua_Number a = knumleft;
- lua_Number b = (lua_Number)fright->i;
- lua_Number y = lj_vm_foldarith(a, b, IR_POW - IR_ADD);
- return lj_ir_knum(J, y);
+ return lj_ir_knum(J, lj_vm_foldarith(knumleft, knumright, IR_POW - IR_ADD));
}
/* Must not use kfold_kref for numbers (could be NaN). */
@@ -247,7 +267,7 @@
case IR_SUB: k1 -= k2; break;
case IR_MUL: k1 *= k2; break;
case IR_MOD: k1 = lj_vm_modi(k1, k2); break;
- case IR_NEG: k1 = -k1; break;
+ case IR_NEG: k1 = (int32_t)(~(uint32_t)k1+1u); break;
case IR_BAND: k1 &= k2; break;
case IR_BOR: k1 |= k2; break;
case IR_BXOR: k1 ^= k2; break;
@@ -258,7 +278,7 @@
case IR_BROR: k1 = (int32_t)lj_ror((uint32_t)k1, (k2 & 31)); break;
case IR_MIN: k1 = k1 < k2 ? k1 : k2; break;
case IR_MAX: k1 = k1 > k2 ? k1 : k2; break;
- default: lua_assert(0); break;
+ default: lj_assertX(0, "bad IR op %d", op); break;
}
return k1;
}
@@ -330,7 +350,7 @@
case IR_ULE: return CONDFOLD((uint32_t)a <= (uint32_t)b);
case IR_ABC:
case IR_UGT: return CONDFOLD((uint32_t)a > (uint32_t)b);
- default: lua_assert(0); return FAILFOLD;
+ default: lj_assertJ(0, "bad IR op %d", fins->o); return FAILFOLD;
}
}
@@ -344,10 +364,12 @@
/* -- Constant folding for 64 bit integers -------------------------------- */
-static uint64_t kfold_int64arith(uint64_t k1, uint64_t k2, IROp op)
+static uint64_t kfold_int64arith(jit_State *J, uint64_t k1, uint64_t k2,
+ IROp op)
{
+ UNUSED(J);
+#if LJ_HASFFI
switch (op) {
-#if LJ_HASFFI
case IR_ADD: k1 += k2; break;
case IR_SUB: k1 -= k2; break;
case IR_MUL: k1 *= k2; break;
@@ -355,13 +377,16 @@
case IR_BOR: k1 |= k2; break;
case IR_BXOR: k1 ^= k2; break;
case IR_BSHL: k1 <<= (k2 & 63); break;
- case IR_BSHR: k1 = (int32_t)((uint32_t)k1 >> (k2 & 63)); break;
- case IR_BSAR: k1 >>= (k2 & 63); break;
- case IR_BROL: k1 = (int32_t)lj_rol((uint32_t)k1, (k2 & 63)); break;
- case IR_BROR: k1 = (int32_t)lj_ror((uint32_t)k1, (k2 & 63)); break;
+ case IR_BSHR: k1 >>= (k2 & 63); break;
+ case IR_BSAR: k1 = (uint64_t)((int64_t)k1 >> (k2 & 63)); break;
+ case IR_BROL: k1 = lj_rol(k1, (k2 & 63)); break;
+ case IR_BROR: k1 = lj_ror(k1, (k2 & 63)); break;
+ default: lj_assertJ(0, "bad IR op %d", op); break;
+ }
+#else
+ UNUSED(k2); UNUSED(op);
+ lj_assertJ(0, "FFI IR op without FFI");
#endif
- default: UNUSED(k2); lua_assert(0); break;
- }
return k1;
}
@@ -373,7 +398,7 @@
LJFOLD(BXOR KINT64 KINT64)
LJFOLDF(kfold_int64arith)
{
- return INT64FOLD(kfold_int64arith(ir_k64(fleft)->u64,
+ return INT64FOLD(kfold_int64arith(J, ir_k64(fleft)->u64,
ir_k64(fright)->u64, (IROp)fins->o));
}
@@ -395,7 +420,7 @@
}
return INT64FOLD(k1);
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -411,7 +436,7 @@
int32_t sh = (fright->i & 63);
return INT64FOLD(lj_carith_shift64(k, sh, fins->o - IR_BSHL));
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -421,7 +446,7 @@
#if LJ_HASFFI
return INT64FOLD(~ir_k64(fleft)->u64);
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -431,7 +456,7 @@
#if LJ_HASFFI
return INT64FOLD(lj_bswap64(ir_k64(fleft)->u64));
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -456,10 +481,10 @@
case IR_UGE: return CONDFOLD(a >= b);
case IR_ULE: return CONDFOLD(a <= b);
case IR_UGT: return CONDFOLD(a > b);
- default: lua_assert(0); return FAILFOLD;
+ default: lj_assertJ(0, "bad IR op %d", fins->o); return FAILFOLD;
}
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -471,7 +496,7 @@
return DROPFOLD;
return NEXTFOLD;
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -485,6 +510,7 @@
}
LJFOLD(SNEW any KINT)
+LJFOLD(XSNEW any KINT)
LJFOLDF(kfold_snew_empty)
{
if (fright->i == 0)
@@ -496,7 +522,7 @@
LJFOLDF(kfold_strref)
{
GCstr *str = ir_kstr(fleft);
- lua_assert((MSize)fright->i <= str->len);
+ lj_assertJ((MSize)fright->i <= str->len, "bad string ref");
return lj_ir_kkptr(J, (char *)strdata(str) + fright->i);
}
@@ -548,22 +574,51 @@
** The compromise is to declare them as loads, emit them like stores and
** CSE whole chains manually when the BUFSTR is to be emitted. Any chain
** fragments left over from CSE are eliminated by DCE.
+**
+** The string buffer methods emit a USE instead of a BUFSTR to keep the
+** chain alive.
*/
-/* BUFHDR is emitted like a store, see below. */
+LJFOLD(BUFHDR any any)
+LJFOLDF(bufhdr_merge)
+{
+ return fins->op2 == IRBUFHDR_WRITE ? CSEFOLD : EMITFOLD;
+}
-LJFOLD(BUFPUT BUFHDR BUFSTR)
-LJFOLDF(bufput_append)
+LJFOLD(BUFPUT any BUFSTR)
+LJFOLDF(bufput_bufstr)
{
- /* New buffer, no other buffer op inbetween and same buffer? */
- if ((J->flags & JIT_F_OPT_FWD) &&
- !(fleft->op2 & IRBUFHDR_APPEND) &&
- fleft->prev == fright->op2 &&
- fleft->op1 == IR(fright->op2)->op1) {
- IRRef ref = fins->op1;
- IR(ref)->op2 = (fleft->op2 | IRBUFHDR_APPEND); /* Modify BUFHDR. */
- IR(ref)->op1 = fright->op1;
- return ref;
+ if ((J->flags & JIT_F_OPT_FWD)) {
+ IRRef hdr = fright->op2;
+ /* New buffer, no other buffer op inbetween and same buffer? */
+ if (fleft->o == IR_BUFHDR && fleft->op2 == IRBUFHDR_RESET &&
+ fleft->prev == hdr &&
+ fleft->op1 == IR(hdr)->op1 &&
+ !(irt_isphi(fright->t) && IR(hdr)->prev) &&
+ (!LJ_HASBUFFER || J->chain[IR_CALLA] < hdr)) {
+ IRRef ref = fins->op1;
+ IR(ref)->op2 = IRBUFHDR_APPEND; /* Modify BUFHDR. */
+ IR(ref)->op1 = fright->op1;
+ return ref;
+ }
+ /* Replay puts to global temporary buffer. */
+ if (IR(hdr)->op2 == IRBUFHDR_RESET && !irt_isphi(fright->t)) {
+ IRIns *ir = IR(fright->op1);
+ /* For now only handle single string.reverse .lower .upper .rep. */
+ if (ir->o == IR_CALLL &&
+ ir->op2 >= IRCALL_lj_buf_putstr_reverse &&
+ ir->op2 <= IRCALL_lj_buf_putstr_rep) {
+ IRIns *carg1 = IR(ir->op1);
+ if (ir->op2 == IRCALL_lj_buf_putstr_rep) {
+ IRIns *carg2 = IR(carg1->op1);
+ if (carg2->op1 == hdr) {
+ return lj_ir_call(J, ir->op2, fins->op1, carg2->op2, carg1->op2);
+ }
+ } else if (carg1->op1 == hdr) {
+ return lj_ir_call(J, ir->op2, fins->op1, carg1->op2);
+ }
+ }
+ }
}
return EMITFOLD; /* Always emit, CSE later. */
}
@@ -592,11 +647,12 @@
LJFOLD(BUFSTR any any)
LJFOLDF(bufstr_kfold_cse)
{
- lua_assert(fleft->o == IR_BUFHDR || fleft->o == IR_BUFPUT ||
- fleft->o == IR_CALLL);
+ lj_assertJ(fleft->o == IR_BUFHDR || fleft->o == IR_BUFPUT ||
+ fleft->o == IR_CALLL,
+ "bad buffer constructor IR op %d", fleft->o);
if (LJ_LIKELY(J->flags & JIT_F_OPT_FOLD)) {
if (fleft->o == IR_BUFHDR) { /* No put operations? */
- if (!(fleft->op2 & IRBUFHDR_APPEND)) /* Empty buffer? */
+ if (fleft->op2 == IRBUFHDR_RESET) /* Empty buffer? */
return lj_ir_kstr(J, &J2G(J)->strempty);
fins->op1 = fleft->op1;
fins->op2 = fleft->prev; /* Relies on checks in bufput_append. */
@@ -603,7 +659,7 @@
return CSEFOLD;
} else if (fleft->o == IR_BUFPUT) {
IRIns *irb = IR(fleft->op1);
- if (irb->o == IR_BUFHDR && !(irb->op2 & IRBUFHDR_APPEND))
+ if (irb->o == IR_BUFHDR && irb->op2 == IRBUFHDR_RESET)
return fleft->op2; /* Shortcut for a single put operation. */
}
}
@@ -613,9 +669,10 @@
while (ref) {
IRIns *irs = IR(ref), *ira = fleft, *irb = IR(irs->op1);
while (ira->o == irb->o && ira->op2 == irb->op2) {
- lua_assert(ira->o == IR_BUFHDR || ira->o == IR_BUFPUT ||
- ira->o == IR_CALLL || ira->o == IR_CARG);
- if (ira->o == IR_BUFHDR && !(ira->op2 & IRBUFHDR_APPEND))
+ lj_assertJ(ira->o == IR_BUFHDR || ira->o == IR_BUFPUT ||
+ ira->o == IR_CALLL || ira->o == IR_CARG,
+ "bad buffer constructor IR op %d", ira->o);
+ if (ira->o == IR_BUFHDR && ira->op2 == IRBUFHDR_RESET)
return ref; /* CSE succeeded. */
if (ira->o == IR_CALLL && ira->op2 == IRCALL_lj_buf_puttab)
break;
@@ -673,7 +730,7 @@
LJFOLDF(bufput_kfold_fmt)
{
IRIns *irc = IR(fleft->op1);
- lua_assert(irref_isk(irc->op2)); /* SFormat must be const. */
+ lj_assertJ(irref_isk(irc->op2), "SFormat must be const");
if (irref_isk(fleft->op2)) {
SFormat sf = (SFormat)IR(irc->op2)->i;
IRIns *ira = IR(fleft->op2);
@@ -978,8 +1035,7 @@
LJFOLD(SUB any KNUM)
LJFOLDF(simplify_numsub_k)
{
- lua_Number n = knumright;
- if (n == 0.0) /* x - (+-0) ==> x */
+ if (ir_knum(fright)->u64 == 0) /* x - (+0) ==> x */
return LEFTFOLD;
return NEXTFOLD;
}
@@ -1053,54 +1109,19 @@
return RETRYFOLD;
}
-LJFOLD(POW any KINT)
-LJFOLDF(simplify_numpow_xk)
+LJFOLD(POW any KNUM)
+LJFOLDF(simplify_numpow_k)
{
- int32_t k = fright->i;
- TRef ref = fins->op1;
- if (k == 0) /* x ^ 0 ==> 1 */
+ if (knumright == 0.0) /* x ^ 0 ==> 1 */
return lj_ir_knum_one(J); /* Result must be a number, not an int. */
- if (k == 1) /* x ^ 1 ==> x */
+ else if (knumright == 1.0) /* x ^ 1 ==> x */
return LEFTFOLD;
- if ((uint32_t)(k+65536) > 2*65536u) /* Limit code explosion. */
+ else if (knumright == 2.0) /* x ^ 2 ==> x * x */
+ return emitir(IRTN(IR_MUL), fins->op1, fins->op1);
+ else
return NEXTFOLD;
- if (k < 0) { /* x ^ (-k) ==> (1/x) ^ k. */
- ref = emitir(IRTN(IR_DIV), lj_ir_knum_one(J), ref);
- k = -k;
- }
- /* Unroll x^k for 1 <= k <= 65536. */
- for (; (k & 1) == 0; k >>= 1) /* Handle leading zeros. */
- ref = emitir(IRTN(IR_MUL), ref, ref);
- if ((k >>= 1) != 0) { /* Handle trailing bits. */
- TRef tmp = emitir(IRTN(IR_MUL), ref, ref);
- for (; k != 1; k >>= 1) {
- if (k & 1)
- ref = emitir(IRTN(IR_MUL), ref, tmp);
- tmp = emitir(IRTN(IR_MUL), tmp, tmp);
- }
- ref = emitir(IRTN(IR_MUL), ref, tmp);
- }
- return ref;
}
-LJFOLD(POW KNUM any)
-LJFOLDF(simplify_numpow_kx)
-{
- lua_Number n = knumleft;
- if (n == 2.0) { /* 2.0 ^ i ==> ldexp(1.0, tonum(i)) */
- fins->o = IR_CONV;
-#if LJ_TARGET_X86ORX64
- fins->op1 = fins->op2;
- fins->op2 = IRCONV_NUM_INT;
- fins->op2 = (IRRef1)lj_opt_fold(J);
-#endif
- fins->op1 = (IRRef1)lj_ir_knum_one(J);
- fins->o = IR_LDEXP;
- return RETRYFOLD;
- }
- return NEXTFOLD;
-}
-
/* -- Simplify conversions ------------------------------------------------ */
LJFOLD(CONV CONV IRCONV_NUM_INT) /* _NUM */
@@ -1114,7 +1135,7 @@
}
LJFOLD(CONV CONV IRCONV_INT_NUM) /* _INT */
-LJFOLD(CONV CONV IRCONV_U32_NUM) /* _U32*/
+LJFOLD(CONV CONV IRCONV_U32_NUM) /* _U32 */
LJFOLDF(simplify_conv_int_num)
{
/* Fold even across PHI to avoid expensive num->int conversions in loop. */
@@ -1149,8 +1170,10 @@
LJFOLD(CONV CONV IRCONV_INT_I64) /* _INT or _U32 */
LJFOLD(CONV CONV IRCONV_INT_U64) /* _INT or _U32 */
+LJFOLD(CONV CONV IRCONV_INT_U32) /* _INT or _U32 */
LJFOLD(CONV CONV IRCONV_U32_I64) /* _INT or _U32 */
LJFOLD(CONV CONV IRCONV_U32_U64) /* _INT or _U32 */
+LJFOLD(CONV CONV IRCONV_U32_INT) /* _INT or _U32 */
LJFOLDF(simplify_conv_int_i64)
{
int src;
@@ -1183,10 +1206,10 @@
{
/* Fold even across PHI to avoid expensive num->int conversions in loop. */
if ((fleft->op2 & IRCONV_SRCMASK) == IRT_INT) {
- lua_assert(irt_isnum(fleft->t));
+ lj_assertJ(irt_isnum(fleft->t), "expected TOBIT number arg");
return fleft->op1;
} else if ((fleft->op2 & IRCONV_SRCMASK) == IRT_U32) {
- lua_assert(irt_isnum(fleft->t));
+ lj_assertJ(irt_isnum(fleft->t), "expected TOBIT number arg");
fins->o = IR_CONV;
fins->op1 = fleft->op1;
fins->op2 = (IRT_INT<<5)|IRT_U32;
@@ -1195,14 +1218,13 @@
return NEXTFOLD;
}
-/* Shortcut floor/ceil/round + IRT_NUM <- IRT_INT/IRT_U32 conversion. */
+/* Shortcut floor/ceil/trunc + IRT_NUM <- integer conversion. */
LJFOLD(FPMATH CONV IRFPM_FLOOR)
LJFOLD(FPMATH CONV IRFPM_CEIL)
LJFOLD(FPMATH CONV IRFPM_TRUNC)
LJFOLDF(simplify_floor_conv)
{
- if ((fleft->op2 & IRCONV_SRCMASK) == IRT_INT ||
- (fleft->op2 & IRCONV_SRCMASK) == IRT_U32)
+ if ((uint32_t)(fleft->op2 & IRCONV_SRCMASK) - (uint32_t)IRT_I8 <= (uint32_t)(IRT_U64 - IRT_U8))
return LEFTFOLD;
return NEXTFOLD;
}
@@ -1226,8 +1248,8 @@
/* Use scalar evolution analysis results to strength-reduce sign-extension. */
if (ref == J->scev.idx) {
IRRef lo = J->scev.dir ? J->scev.start : J->scev.stop;
- lua_assert(irt_isint(J->scev.t));
- if (lo && IR(lo)->i + ofs >= 0) {
+ lj_assertJ(irt_isint(J->scev.t), "only int SCEV supported");
+ if (lo && IR(lo)->o == IR_KINT && IR(lo)->i + ofs >= 0) {
ok_reduce:
#if LJ_TARGET_X64
/* Eliminate widening. All 32 bit ops do an implicit zero-extension. */
@@ -1257,16 +1279,21 @@
LJFOLD(CONV MUL IRCONV_U32_U64)
LJFOLDF(simplify_conv_narrow)
{
+#if LJ_64
+ UNUSED(J);
+ return NEXTFOLD;
+#else
IROp op = (IROp)fleft->o;
IRType t = irt_type(fins->t);
IRRef op1 = fleft->op1, op2 = fleft->op2, mode = fins->op2;
PHIBARRIER(fleft);
- op1 = emitir(IRTI(IR_CONV), op1, mode);
- op2 = emitir(IRTI(IR_CONV), op2, mode);
+ op1 = emitir(IRT(IR_CONV, t), op1, mode);
+ op2 = emitir(IRT(IR_CONV, t), op2, mode);
fins->ot = IRT(op, t);
fins->op1 = op1;
fins->op2 = op2;
return RETRYFOLD;
+#endif
}
/* Special CSE rule for CONV. */
@@ -1302,7 +1329,8 @@
/* Narrowing ignores PHIs and repeating it inside the loop is not useful. */
if (J->chain[IR_LOOP])
return NEXTFOLD;
- lua_assert(fins->o != IR_CONV || (fins->op2&IRCONV_CONVMASK) != IRCONV_TOBIT);
+ lj_assertJ(fins->o != IR_CONV || (fins->op2&IRCONV_CONVMASK) != IRCONV_TOBIT,
+ "unexpected CONV TOBIT");
return lj_opt_narrow_convert(J);
}
@@ -1339,7 +1367,7 @@
if (fright->i == 0) /* i - 0 ==> i */
return LEFTFOLD;
fins->o = IR_ADD; /* i - k ==> i + (-k) */
- fins->op2 = (IRRef1)lj_ir_kint(J, -fright->i); /* Overflow for -2^31 ok. */
+ fins->op2 = (IRRef1)lj_ir_kint(J, (int32_t)(~(uint32_t)fright->i+1u)); /* Overflow for -2^31 ok. */
return RETRYFOLD;
}
@@ -1370,7 +1398,7 @@
if (k == 0) /* i - 0 ==> i */
return LEFTFOLD;
fins->o = IR_ADD; /* i - k ==> i + (-k) */
- fins->op2 = (IRRef1)lj_ir_kint64(J, (uint64_t)-(int64_t)k);
+ fins->op2 = (IRRef1)lj_ir_kint64(J, ~k+1u);
return RETRYFOLD;
}
@@ -1408,7 +1436,7 @@
return simplify_intmul_k(J, (int32_t)ir_kint64(fright)->u64);
return NEXTFOLD;
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -1416,7 +1444,7 @@
LJFOLDF(simplify_intmod_k)
{
int32_t k = fright->i;
- lua_assert(k != 0);
+ lj_assertJ(k != 0, "integer mod 0");
if (k > 0 && (k & (k-1)) == 0) { /* i % (2^k) ==> i & (2^k-1) */
fins->o = IR_BAND;
fins->op2 = lj_ir_kint(J, k-1);
@@ -1666,7 +1694,8 @@
fins->ot = IRTI(IR_BAND);
return RETRYFOLD;
} else if (irk->o == IR_KINT64) {
- uint64_t k = kfold_int64arith(ir_k64(irk)->u64, fright->i, (IROp)fins->o);
+ uint64_t k = kfold_int64arith(J, ir_k64(irk)->u64, fright->i,
+ (IROp)fins->o);
IROpT ot = fleft->ot;
fins->op1 = fleft->op1;
fins->op1 = (IRRef1)lj_opt_fold(J);
@@ -1714,8 +1743,8 @@
IRIns *irk = IR(fleft->op2);
PHIBARRIER(fleft);
if (irk->o == IR_KINT64) {
- uint64_t k = kfold_int64arith(ir_k64(irk)->u64,
- ir_k64(fright)->u64, (IROp)fins->o);
+ uint64_t k = kfold_int64arith(J, ir_k64(irk)->u64, ir_k64(fright)->u64,
+ (IROp)fins->o);
/* (i | k1) & k2 ==> i & k2, if (k1 & k2) == 0. */
/* (i & k1) | k2 ==> i | k2, if (k1 | k2) == -1. */
if (k == (fins->o == IR_BAND ? (uint64_t)0 : ~(uint64_t)0)) {
@@ -1725,7 +1754,7 @@
}
return NEXTFOLD;
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
@@ -1761,8 +1790,8 @@
#if LJ_HASFFI
IRIns *irk = IR(fleft->op2);
if (irk->o == IR_KINT64) {
- uint64_t k = kfold_int64arith(ir_k64(irk)->u64,
- ir_k64(fright)->u64, (IROp)fins->o);
+ uint64_t k = kfold_int64arith(J, ir_k64(irk)->u64, ir_k64(fright)->u64,
+ (IROp)fins->o);
PHIBARRIER(fleft);
fins->op1 = fleft->op1;
fins->op2 = (IRRef1)lj_ir_kint64(J, k);
@@ -1770,12 +1799,10 @@
}
return NEXTFOLD;
#else
- UNUSED(J); lua_assert(0); return FAILFOLD;
+ UNUSED(J); lj_assertJ(0, "FFI IR op without FFI"); return FAILFOLD;
#endif
}
-LJFOLD(MIN MIN any)
-LJFOLD(MAX MAX any)
LJFOLD(BAND BAND any)
LJFOLD(BOR BOR any)
LJFOLDF(reassoc_dup)
@@ -1785,6 +1812,15 @@
return NEXTFOLD;
}
+LJFOLD(MIN MIN any)
+LJFOLD(MAX MAX any)
+LJFOLDF(reassoc_dup_minmax)
+{
+ if (fins->op2 == fleft->op2)
+ return LEFTFOLD; /* (a o b) o b ==> a o b */
+ return NEXTFOLD;
+}
+
LJFOLD(BXOR BXOR any)
LJFOLDF(reassoc_bxor)
{
@@ -1823,23 +1859,12 @@
return NEXTFOLD;
}
-LJFOLD(MIN MIN KNUM)
-LJFOLD(MAX MAX KNUM)
LJFOLD(MIN MIN KINT)
LJFOLD(MAX MAX KINT)
LJFOLDF(reassoc_minmax_k)
{
IRIns *irk = IR(fleft->op2);
- if (irk->o == IR_KNUM) {
- lua_Number a = ir_knum(irk)->n;
- lua_Number y = lj_vm_foldarith(a, knumright, fins->o - IR_ADD);
- if (a == y) /* (x o k1) o k2 ==> x o k1, if (k1 o k2) == k1. */
- return LEFTFOLD;
- PHIBARRIER(fleft);
- fins->op1 = fleft->op1;
- fins->op2 = (IRRef1)lj_ir_knum(J, y);
- return RETRYFOLD; /* (x o k1) o k2 ==> x o (k1 o k2) */
- } else if (irk->o == IR_KINT) {
+ if (irk->o == IR_KINT) {
int32_t a = irk->i;
int32_t y = kfold_intop(a, fright->i, fins->o);
if (a == y) /* (x o k1) o k2 ==> x o k1, if (k1 o k2) == k1. */
@@ -1852,24 +1877,6 @@
return NEXTFOLD;
}
-LJFOLD(MIN MAX any)
-LJFOLD(MAX MIN any)
-LJFOLDF(reassoc_minmax_left)
-{
- if (fins->op2 == fleft->op1 || fins->op2 == fleft->op2)
- return RIGHTFOLD; /* (b o1 a) o2 b ==> b; (a o1 b) o2 b ==> b */
- return NEXTFOLD;
-}
-
-LJFOLD(MIN any MAX)
-LJFOLD(MAX any MIN)
-LJFOLDF(reassoc_minmax_right)
-{
- if (fins->op1 == fright->op1 || fins->op1 == fright->op2)
- return LEFTFOLD; /* a o2 (a o1 b) ==> a; a o2 (b o1 a) ==> a */
- return NEXTFOLD;
-}
-
/* -- Array bounds check elimination -------------------------------------- */
/* Eliminate ABC across PHIs to handle t[i-1] forwarding case.
@@ -1906,6 +1913,7 @@
LJFOLD(ABC any KINT)
LJFOLDF(abc_k)
{
+ PHIBARRIER(fleft);
if (LJ_LIKELY(J->flags & JIT_F_OPT_ABC)) {
IRRef ref = J->chain[IR_ABC];
IRRef asize = fins->op1;
@@ -1912,8 +1920,8 @@
while (ref > asize) {
IRIns *ir = IR(ref);
if (ir->op1 == asize && irref_isk(ir->op2)) {
- int32_t k = IR(ir->op2)->i;
- if (fright->i > k)
+ uint32_t k = (uint32_t)IR(ir->op2)->i;
+ if ((uint32_t)fright->i > k)
ir->op2 = fins->op2;
return DROPFOLD;
}
@@ -1928,9 +1936,10 @@
LJFOLD(ABC any any)
LJFOLDF(abc_invar)
{
- /* Invariant ABC marked as PTR. Drop if op1 is invariant, too. */
+ /* Invariant ABC marked as P32 or U32. Drop if op1 is invariant too. */
if (!irt_isint(fins->t) && fins->op1 < J->chain[IR_LOOP] &&
- !irt_isphi(IR(fins->op1)->t))
+ (irt_isu32(fins->t) ||
+ (!irref_isk(fins->op1) && !irt_isphi(IR(fins->op1)->t))))
return DROPFOLD;
return NEXTFOLD;
}
@@ -1965,7 +1974,10 @@
LJFOLDF(comm_equal)
{
/* For non-numbers only: x == x ==> drop; x ~= x ==> fail */
- if (fins->op1 == fins->op2 && !irt_isnum(fins->t))
+ if (fins->op1 == fins->op2 &&
+ (!irt_isnum(fins->t) ||
+ (fleft->o == IR_CONV && /* Converted integers cannot be NaN. */
+ (uint32_t)(fleft->op2 & IRCONV_SRCMASK) - (uint32_t)IRT_I8 <= (uint32_t)(IRT_U64 - IRT_U8))))
return CONDFOLD(fins->o == IR_EQ);
return fold_comm_swap(J);
}
@@ -1995,8 +2007,6 @@
LJFOLD(BAND any any)
LJFOLD(BOR any any)
-LJFOLD(MIN any any)
-LJFOLD(MAX any any)
LJFOLDF(comm_dup)
{
if (fins->op1 == fins->op2) /* x o x ==> x */
@@ -2004,6 +2014,15 @@
return fold_comm_swap(J);
}
+LJFOLD(MIN any any)
+LJFOLD(MAX any any)
+LJFOLDF(comm_dup_minmax)
+{
+ if (fins->op1 == fins->op2) /* x o x ==> x */
+ return LEFTFOLD;
+ return NEXTFOLD;
+}
+
LJFOLD(BXOR any any)
LJFOLDF(comm_bxor)
{
@@ -2040,7 +2059,7 @@
{
GCstr *kstr = ir_kstr(fright);
int32_t len = (int32_t)kstr->len;
- lua_assert(irt_isstr(fins->t));
+ lj_assertJ(irt_isstr(fins->t), "bad equality IR type");
#if LJ_TARGET_UNALIGNED
#define FOLD_SNEW_MAX_LEN 4 /* Handle string lengths 0, 1, 2, 3, 4. */
@@ -2104,7 +2123,7 @@
LJFOLDF(kfold_hload_kkptr)
{
UNUSED(J);
- lua_assert(ir_kptr(fleft) == niltvg(J2G(J)));
+ lj_assertJ(ir_kptr(fleft) == niltvg(J2G(J)), "expected niltv");
return TREF_NIL;
}
@@ -2114,11 +2133,29 @@
LJFOLD(ULOAD any)
LJFOLDX(lj_opt_fwd_uload)
-LJFOLD(CALLL any IRCALL_lj_tab_len)
-LJFOLDX(lj_opt_fwd_tab_len)
+LJFOLD(ALEN any any)
+LJFOLDX(lj_opt_fwd_alen)
+/* Try to merge UREFO/UREFC into referenced instruction. */
+static TRef merge_uref(jit_State *J, IRRef ref, IRIns* ir)
+{
+ if (ir->o == IR_UREFO && irt_isguard(ir->t)) {
+ /* Might be pointing to some other coroutine's stack.
+ ** And GC might shrink said stack, thereby repointing the upvalue.
+ ** GC might even collect said coroutine, thereby closing the upvalue.
+ */
+ if (gcstep_barrier(J, ref))
+ return EMITFOLD; /* So cannot merge. */
+ /* Current fins wants a check, but ir doesn't have one. */
+ if ((irt_t(fins->t) & (IRT_GUARD|IRT_TYPE)) == (IRT_GUARD|IRT_PGC) &&
+ irt_type(ir->t) == IRT_IGC)
+ ir->t.irt += IRT_PGC-IRT_IGC; /* So install a check. */
+ }
+ return ref; /* Not a TRef, but the caller doesn't care. */
+}
+
/* Upvalue refs are really loads, but there are no corresponding stores.
-** So CSE is ok for them, except for UREFO across a GC step (see below).
+** So CSE is ok for them, except for guarded UREFO across a GC step.
** If the referenced function is const, its upvalue addresses are const, too.
** This can be used to improve CSE by looking for the same address,
** even if the upvalues originate from a different function.
@@ -2136,9 +2173,7 @@
if (irref_isk(ir->op1)) {
GCfunc *fn2 = ir_kfunc(IR(ir->op1));
if (gco2uv(gcref(fn2->l.uvptr[(ir->op2 >> 8)])) == uv) {
- if (fins->o == IR_UREFO && gcstep_barrier(J, ref))
- break;
- return ref;
+ return merge_uref(J, ref, ir);
}
}
ref = ir->prev;
@@ -2147,6 +2182,24 @@
return EMITFOLD;
}
+/* Custom CSE for UREFO. */
+LJFOLD(UREFO any any)
+LJFOLDF(cse_urefo)
+{
+ if (LJ_LIKELY(J->flags & JIT_F_OPT_CSE)) {
+ IRRef ref = J->chain[IR_UREFO];
+ IRRef lim = fins->op1;
+ IRRef2 op12 = (IRRef2)fins->op1 + ((IRRef2)fins->op2 << 16);
+ while (ref > lim) {
+ IRIns *ir = IR(ref);
+ if (ir->op12 == op12)
+ return merge_uref(J, ref, ir);
+ ref = ir->prev;
+ }
+ }
+ return EMITFOLD;
+}
+
LJFOLD(HREFK any any)
LJFOLDX(lj_opt_fwd_hrefk)
@@ -2248,6 +2301,27 @@
return NEXTFOLD;
}
+LJFOLD(FLOAD any IRFL_SBUF_W)
+LJFOLD(FLOAD any IRFL_SBUF_E)
+LJFOLD(FLOAD any IRFL_SBUF_B)
+LJFOLD(FLOAD any IRFL_SBUF_L)
+LJFOLD(FLOAD any IRFL_SBUF_REF)
+LJFOLD(FLOAD any IRFL_SBUF_R)
+LJFOLDF(fload_sbuf)
+{
+ TRef tr = lj_opt_fwd_fload(J);
+ return lj_opt_fwd_sbuf(J, tref_ref(tr)) ? tr : EMITFOLD;
+}
+
+/* The fast function ID of function objects is immutable. */
+LJFOLD(FLOAD KGC IRFL_FUNC_FFID)
+LJFOLDF(fload_func_ffid_kgc)
+{
+ if (LJ_LIKELY(J->flags & JIT_F_OPT_FOLD))
+ return INTFOLD((int32_t)ir_kfunc(fleft)->c.ffid);
+ return NEXTFOLD;
+}
+
/* The C type ID of cdata objects is immutable. */
LJFOLD(FLOAD KGC IRFL_CDATA_CTYPEID)
LJFOLDF(fload_cdata_typeid_kgc)
@@ -2315,7 +2389,7 @@
TRef tr = lj_opt_cse(J);
return tref_ref(tr) < J->chain[IR_RETF] ? EMITFOLD : tr;
} else {
- lua_assert(J->slot[fins->op1] != 0);
+ lj_assertJ(J->slot[fins->op1] != 0, "uninitialized slot accessed");
return J->slot[fins->op1];
}
}
@@ -2331,18 +2405,24 @@
LJFOLD(XLOAD any any)
LJFOLDX(lj_opt_fwd_xload)
+/* -- Frame handling ------------------------------------------------------ */
+
+/* Prevent CSE of a REF_BASE operand across IR_RETF. */
+LJFOLD(SUB any BASE)
+LJFOLD(SUB BASE any)
+LJFOLD(EQ any BASE)
+LJFOLDF(fold_base)
+{
+ return lj_opt_cselim(J, J->chain[IR_RETF]);
+}
+
/* -- Write barriers ------------------------------------------------------ */
/* Write barriers are amenable to CSE, but not across any incremental
** GC steps.
-**
-** The same logic applies to open upvalue references, because a stack
-** may be resized during a GC step (not the current stack, but maybe that
-** of a coroutine).
*/
LJFOLD(TBAR any)
LJFOLD(OBAR any any)
-LJFOLD(UREFO any any)
LJFOLDF(barrier_tab)
{
TRef tr = lj_opt_cse(J);
@@ -2394,6 +2474,7 @@
LJFOLDX(lj_opt_dse_xstore)
LJFOLD(NEWREF any any) /* Treated like a store. */
+LJFOLD(TMPREF any any)
LJFOLD(CALLA any any)
LJFOLD(CALLL any any) /* Safeguard fallback. */
LJFOLD(CALLS any any)
@@ -2404,9 +2485,19 @@
LJFOLD(TDUP any)
LJFOLD(CNEW any any)
LJFOLD(XSNEW any any)
-LJFOLD(BUFHDR any any)
LJFOLDX(lj_ir_emit)
+/* -- Miscellaneous ------------------------------------------------------- */
+
+LJFOLD(CARG any any)
+LJFOLDF(cse_carg)
+{
+ TRef tr = lj_opt_cse(J);
+ if (tref_ref(tr) < J->chain[IR_LOOP]) /* CSE across loop? */
+ return EMITFOLD; /* Raw emit. Assumes fins is left intact by CSE. */
+ return tr;
+}
+
/* ------------------------------------------------------------------------ */
/* Every entry in the generated hash table is a 32 bit pattern:
@@ -2430,8 +2521,9 @@
IRRef ref;
if (LJ_UNLIKELY((J->flags & JIT_F_OPT_MASK) != JIT_F_OPT_DEFAULT)) {
- lua_assert(((JIT_F_OPT_FOLD|JIT_F_OPT_FWD|JIT_F_OPT_CSE|JIT_F_OPT_DSE) |
- JIT_F_OPT_DEFAULT) == JIT_F_OPT_DEFAULT);
+ lj_assertJ(((JIT_F_OPT_FOLD|JIT_F_OPT_FWD|JIT_F_OPT_CSE|JIT_F_OPT_DSE) |
+ JIT_F_OPT_DEFAULT) == JIT_F_OPT_DEFAULT,
+ "bad JIT_F_OPT_DEFAULT");
/* Folding disabled? Chain to CSE, but not for loads/stores/allocs. */
if (!(J->flags & JIT_F_OPT_FOLD) && irm_kind(lj_ir_mode[fins->o]) == IRM_N)
return lj_opt_cse(J);
@@ -2493,7 +2585,7 @@
return lj_ir_kint(J, fins->i);
if (ref == FAILFOLD)
lj_trace_err(J, LJ_TRERR_GFAIL);
- lua_assert(ref == DROPFOLD);
+ lj_assertJ(ref == DROPFOLD, "bad fold result");
return REF_DROP;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_loop.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_loop.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_loop.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** LOOP: Loop Optimizations.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_loop_c
@@ -223,8 +223,9 @@
}
J->guardemit.irt = 0;
/* Setup new snapshot. */
- snap->mapofs = (uint16_t)nmapofs;
+ snap->mapofs = (uint32_t)nmapofs;
snap->ref = (IRRef1)J->cur.nins;
+ snap->mcofs = 0;
snap->nslots = nslots;
snap->topslot = osnap->topslot;
snap->count = 0;
@@ -251,7 +252,7 @@
nmap += nn;
while (omap < nextmap) /* Copy PC + frame links. */
*nmap++ = *omap++;
- J->cur.nsnapmap = (uint16_t)(nmap - J->cur.snapmap);
+ J->cur.nsnapmap = (uint32_t)(nmap - J->cur.snapmap);
}
typedef struct LoopState {
@@ -299,7 +300,8 @@
loopmap = &J->cur.snapmap[loopsnap->mapofs];
/* The PC of snapshot #0 and the loop snapshot must match. */
psentinel = &loopmap[loopsnap->nent];
- lua_assert(*psentinel == J->cur.snapmap[J->cur.snap[0].nent]);
+ lj_assertJ(*psentinel == J->cur.snapmap[J->cur.snap[0].nent],
+ "mismatched PC for loop snapshot");
*psentinel = SNAP(255, 0, 0); /* Replace PC with temporary sentinel. */
/* Start substitution with snapshot #1 (#0 is empty for root traces). */
@@ -352,10 +354,12 @@
irr = IR(ref);
goto phiconv;
}
- } else if (ref != REF_DROP && irr->o == IR_CONV &&
- ref > invar && irr->op1 < invar) {
- /* May need an extra PHI for a CONV. */
- ref = irr->op1;
+ } else if (ref != REF_DROP && ref > invar &&
+ ((irr->o == IR_CONV && irr->op1 < invar) ||
+ (irr->o == IR_ALEN && irr->op2 < invar &&
+ irr->op2 != REF_NIL))) {
+ /* May need an extra PHI for a CONV or ALEN hint. */
+ ref = irr->o == IR_CONV ? irr->op1 : irr->op2;
irr = IR(ref);
phiconv:
if (ref < invar && !irref_isk(ref) && !irt_isphi(irr->t)) {
@@ -369,8 +373,8 @@
}
}
if (!irt_isguard(J->guardemit)) /* Drop redundant snapshot. */
- J->cur.nsnapmap = (uint16_t)J->cur.snap[--J->cur.nsnap].mapofs;
- lua_assert(J->cur.nsnapmap <= J->sizesnapmap);
+ J->cur.nsnapmap = (uint32_t)J->cur.snap[--J->cur.nsnap].mapofs;
+ lj_assertJ(J->cur.nsnapmap <= J->sizesnapmap, "bad snapshot map index");
*psentinel = J->cur.snapmap[J->cur.snap[0].nent]; /* Restore PC. */
loop_emit_phi(J, subst, phi, nphi, onsnap);
@@ -383,7 +387,7 @@
SnapShot *snap = &J->cur.snap[nsnap-1];
SnapEntry *map = J->cur.snapmap;
map[snap->mapofs + snap->nent] = map[J->cur.snap[0].nent]; /* Restore PC. */
- J->cur.nsnapmap = (uint16_t)nsnapmap;
+ J->cur.nsnapmap = (uint32_t)nsnapmap;
J->cur.nsnap = nsnap;
J->guardemit.irt = 0;
lj_ir_rollback(J, ins);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_mem.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_mem.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_mem.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -3,7 +3,7 @@
** AA: Alias Analysis using high-level semantic disambiguation.
** FWD: Load Forwarding (L2L) + Store Forwarding (S2L).
** DSE: Dead-Store Elimination.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_mem_c
@@ -18,6 +18,7 @@
#include "lj_jit.h"
#include "lj_iropt.h"
#include "lj_ircall.h"
+#include "lj_dispatch.h"
/* Some local macros to save typing. Undef'd at the end. */
#define IR(ref) (&J->cur.ir[(ref)])
@@ -56,8 +57,8 @@
{
IRIns *taba = IR(ta), *tabb = IR(tb);
int newa, newb;
- lua_assert(ta != tb);
- lua_assert(irt_istab(taba->t) && irt_istab(tabb->t));
+ lj_assertJ(ta != tb, "bad usage");
+ lj_assertJ(irt_istab(taba->t) && irt_istab(tabb->t), "bad usage");
/* Disambiguate new allocations. */
newa = (taba->o == IR_TNEW || taba->o == IR_TDUP);
newb = (tabb->o == IR_TNEW || tabb->o == IR_TDUP);
@@ -71,6 +72,34 @@
return aa_escape(J, taba, tabb);
}
+/* Check whether there's no aliasing table.clear. */
+static int fwd_aa_tab_clear(jit_State *J, IRRef lim, IRRef ta)
+{
+ IRRef ref = J->chain[IR_CALLS];
+ while (ref > lim) {
+ IRIns *calls = IR(ref);
+ if (calls->op2 == IRCALL_lj_tab_clear &&
+ (ta == calls->op1 || aa_table(J, ta, calls->op1) != ALIAS_NO))
+ return 0; /* Conflict. */
+ ref = calls->prev;
+ }
+ return 1; /* No conflict. Can safely FOLD/CSE. */
+}
+
+/* Check whether there's no aliasing NEWREF/table.clear for the left operand. */
+int LJ_FASTCALL lj_opt_fwd_tptr(jit_State *J, IRRef lim)
+{
+ IRRef ta = fins->op1;
+ IRRef ref = J->chain[IR_NEWREF];
+ while (ref > lim) {
+ IRIns *newref = IR(ref);
+ if (ta == newref->op1 || aa_table(J, ta, newref->op1) != ALIAS_NO)
+ return 0; /* Conflict. */
+ ref = newref->prev;
+ }
+ return fwd_aa_tab_clear(J, lim, ta);
+}
+
/* Alias analysis for array and hash access using key-based disambiguation. */
static AliasRet aa_ahref(jit_State *J, IRIns *refa, IRIns *refb)
{
@@ -99,7 +128,7 @@
/* Disambiguate array references based on index arithmetic. */
int32_t ofsa = 0, ofsb = 0;
IRRef basea = ka, baseb = kb;
- lua_assert(refb->o == IR_AREF);
+ lj_assertJ(refb->o == IR_AREF, "expected AREF");
/* Gather base and offset from t[base] or t[base+-ofs]. */
if (keya->o == IR_ADD && irref_isk(keya->op2)) {
basea = keya->op1;
@@ -117,8 +146,9 @@
return ALIAS_NO; /* t[base+-o1] vs. t[base+-o2] and o1 != o2. */
} else {
/* Disambiguate hash references based on the type of their keys. */
- lua_assert((refa->o==IR_HREF || refa->o==IR_HREFK || refa->o==IR_NEWREF) &&
- (refb->o==IR_HREF || refb->o==IR_HREFK || refb->o==IR_NEWREF));
+ lj_assertJ((refa->o==IR_HREF || refa->o==IR_HREFK || refa->o==IR_NEWREF) &&
+ (refb->o==IR_HREF || refb->o==IR_HREFK || refb->o==IR_NEWREF),
+ "bad xREF IR op %d or %d", refa->o, refb->o);
if (!irt_sametype(keya->t, keyb->t))
return ALIAS_NO; /* Different key types. */
}
@@ -152,9 +182,11 @@
IRIns *ir = (xr->o == IR_HREFK || xr->o == IR_AREF) ? IR(xr->op1) : xr;
IRRef tab = ir->op1;
ir = IR(tab);
- if (ir->o == IR_TNEW || (ir->o == IR_TDUP && irref_isk(xr->op2))) {
+ if ((ir->o == IR_TNEW || (ir->o == IR_TDUP && irref_isk(xr->op2))) &&
+ fwd_aa_tab_clear(J, tab, tab)) {
/* A NEWREF with a number key may end up pointing to the array part.
** But it's referenced from HSTORE and not found in the ASTORE chain.
+ ** Or a NEWREF may rehash the table and move unrelated number keys.
** For now simply consider this a conflict without forwarding anything.
*/
if (xr->o == IR_AREF) {
@@ -165,6 +197,11 @@
goto cselim;
ref2 = newref->prev;
}
+ } else {
+ IRIns *key = IR(xr->op2);
+ if (key->o == IR_KSLOT) key = IR(key->op1);
+ if (irt_isnum(key->t) && J->chain[IR_NEWREF] > tab)
+ goto cselim;
}
/* NEWREF inhibits CSE for HREF, and dependent FLOADs from HREFK/AREF.
** But the above search for conflicting stores was limited by xref.
@@ -180,11 +217,10 @@
}
ref = store->prev;
}
- lua_assert(ir->o != IR_TNEW || irt_isnil(fins->t));
- if (irt_ispri(fins->t)) {
- return TREF_PRI(irt_type(fins->t));
- } else if (irt_isnum(fins->t) || (LJ_DUALNUM && irt_isint(fins->t)) ||
- irt_isstr(fins->t)) {
+ /* Simplified here: let loop_unroll() figure out any type instability. */
+ if (ir->o == IR_TNEW) {
+ return TREF_NIL;
+ } else {
TValue keyv;
cTValue *tv;
IRIns *key = IR(xr->op2);
@@ -191,12 +227,13 @@
if (key->o == IR_KSLOT) key = IR(key->op1);
lj_ir_kvalue(J->L, &keyv, key);
tv = lj_tab_get(J->L, ir_ktab(IR(ir->op1)), &keyv);
- lua_assert(itype2irt(tv) == irt_type(fins->t));
- if (irt_isnum(fins->t))
+ if (tvispri(tv))
+ return TREF_PRI(itype2irt(tv));
+ else if (tvisnum(tv))
return lj_ir_knum_u64(J, tv->u64);
- else if (LJ_DUALNUM && irt_isint(fins->t))
+ else if (tvisint(tv))
return lj_ir_kint(J, intV(tv));
- else
+ else if (tvisgcv(tv))
return lj_ir_kstr(J, strV(tv));
}
/* Othwerwise: don't intern as a constant. */
@@ -265,7 +302,7 @@
while (ref > tab) {
IRIns *newref = IR(ref);
if (tab == newref->op1) {
- if (fright->op1 == newref->op2)
+ if (fright->op1 == newref->op2 && fwd_aa_tab_clear(J, ref, tab))
return ref; /* Forward from NEWREF. */
else
goto docse;
@@ -275,7 +312,7 @@
ref = newref->prev;
}
/* No conflicting NEWREF: key location unchanged for HREFK of TDUP. */
- if (IR(tab)->o == IR_TDUP)
+ if (IR(tab)->o == IR_TDUP && fwd_aa_tab_clear(J, tab, tab))
fins->t.irt &= ~IRT_GUARD; /* Drop HREFK guard. */
docse:
return CSEFOLD;
@@ -309,34 +346,6 @@
return 1; /* No conflict. Can fold to niltv. */
}
-/* Check whether there's no aliasing table.clear. */
-static int fwd_aa_tab_clear(jit_State *J, IRRef lim, IRRef ta)
-{
- IRRef ref = J->chain[IR_CALLS];
- while (ref > lim) {
- IRIns *calls = IR(ref);
- if (calls->op2 == IRCALL_lj_tab_clear &&
- (ta == calls->op1 || aa_table(J, ta, calls->op1) != ALIAS_NO))
- return 0; /* Conflict. */
- ref = calls->prev;
- }
- return 1; /* No conflict. Can safely FOLD/CSE. */
-}
-
-/* Check whether there's no aliasing NEWREF/table.clear for the left operand. */
-int LJ_FASTCALL lj_opt_fwd_tptr(jit_State *J, IRRef lim)
-{
- IRRef ta = fins->op1;
- IRRef ref = J->chain[IR_NEWREF];
- while (ref > lim) {
- IRIns *newref = IR(ref);
- if (ta == newref->op1 || aa_table(J, ta, newref->op1) != ALIAS_NO)
- return 0; /* Conflict. */
- ref = newref->prev;
- }
- return fwd_aa_tab_clear(J, lim, ta);
-}
-
/* ASTORE/HSTORE elimination. */
TRef LJ_FASTCALL lj_opt_dse_ahstore(jit_State *J)
{
@@ -360,16 +369,16 @@
/* Different value: try to eliminate the redundant store. */
if (ref > J->chain[IR_LOOP]) { /* Quick check to avoid crossing LOOP. */
IRIns *ir;
- /* Check for any intervening guards (includes conflicting loads). */
+ /* Check for any intervening guards (includes conflicting loads).
+ ** Note that lj_tab_keyindex and lj_vm_next don't need guards,
+ ** since they are followed by at least one guarded VLOAD.
+ */
for (ir = IR(J->cur.nins-1); ir > store; ir--)
- if (irt_isguard(ir->t) || ir->o == IR_CALLL)
+ if (irt_isguard(ir->t) || ir->o == IR_ALEN)
goto doemit; /* No elimination possible. */
/* Remove redundant store from chain and replace with NOP. */
*refp = store->prev;
- store->o = IR_NOP;
- store->t.irt = IRT_NIL;
- store->op1 = store->op2 = 0;
- store->prev = 0;
+ lj_ir_nop(store);
/* Now emit the new store instead. */
}
goto doemit;
@@ -380,6 +389,67 @@
return EMITFOLD; /* Otherwise we have a conflict or simply no match. */
}
+/* ALEN forwarding. */
+TRef LJ_FASTCALL lj_opt_fwd_alen(jit_State *J)
+{
+ IRRef tab = fins->op1; /* Table reference. */
+ IRRef lim = tab; /* Search limit. */
+ IRRef ref;
+
+ /* Search for conflicting HSTORE with numeric key. */
+ ref = J->chain[IR_HSTORE];
+ while (ref > lim) {
+ IRIns *store = IR(ref);
+ IRIns *href = IR(store->op1);
+ IRIns *key = IR(href->op2);
+ if (irt_isnum(key->o == IR_KSLOT ? IR(key->op1)->t : key->t)) {
+ lim = ref; /* Conflicting store found, limits search for ALEN. */
+ break;
+ }
+ ref = store->prev;
+ }
+
+ /* Try to find a matching ALEN. */
+ ref = J->chain[IR_ALEN];
+ while (ref > lim) {
+ /* CSE for ALEN only depends on the table, not the hint. */
+ if (IR(ref)->op1 == tab) {
+ IRRef sref;
+
+ /* Search for aliasing table.clear. */
+ if (!fwd_aa_tab_clear(J, ref, tab))
+ break;
+
+ /* Search for hint-forwarding or conflicting store. */
+ sref = J->chain[IR_ASTORE];
+ while (sref > ref) {
+ IRIns *store = IR(sref);
+ IRIns *aref = IR(store->op1);
+ IRIns *fref = IR(aref->op1);
+ if (tab == fref->op1) { /* ASTORE to the same table. */
+ /* Detect t[#t+1] = x idiom for push. */
+ IRIns *idx = IR(aref->op2);
+ if (!irt_isnil(store->t) &&
+ idx->o == IR_ADD && idx->op1 == ref &&
+ IR(idx->op2)->o == IR_KINT && IR(idx->op2)->i == 1) {
+ /* Note: this requires an extra PHI check in loop unroll. */
+ fins->op2 = aref->op2; /* Set ALEN hint. */
+ }
+ goto doemit; /* Conflicting store, possibly giving a hint. */
+ } else if (aa_table(J, tab, fref->op1) != ALIAS_NO) {
+ goto doemit; /* Conflicting store. */
+ }
+ sref = store->prev;
+ }
+
+ return ref; /* Plain ALEN forwarding. */
+ }
+ ref = IR(ref)->prev;
+ }
+doemit:
+ return EMITFOLD;
+}
+
/* -- ULOAD forwarding ---------------------------------------------------- */
/* The current alias analysis for upvalues is very simplistic. It only
@@ -392,8 +462,6 @@
*/
static AliasRet aa_uref(IRIns *refa, IRIns *refb)
{
- if (refa->o != refb->o)
- return ALIAS_NO; /* Different UREFx type. */
if (refa->op1 == refb->op1) { /* Same function. */
if (refa->op2 == refb->op2)
return ALIAS_MUST; /* Same function, same upvalue idx. */
@@ -400,10 +468,17 @@
else
return ALIAS_NO; /* Same function, different upvalue idx. */
} else { /* Different functions, check disambiguation hash values. */
- if (((refa->op2 ^ refb->op2) & 0xff))
+ if (((refa->op2 ^ refb->op2) & 0xff)) {
return ALIAS_NO; /* Upvalues with different hash values cannot alias. */
- else
- return ALIAS_MAY; /* No conclusion can be drawn for same hash value. */
+ } else if (refa->o != refb->o) {
+ /* Different UREFx type, but need to confirm the UREFO really is open. */
+ if (irt_type(refa->t) == IRT_IGC) refa->t.irt += IRT_PGC-IRT_IGC;
+ else if (irt_type(refb->t) == IRT_IGC) refb->t.irt += IRT_PGC-IRT_IGC;
+ return ALIAS_NO;
+ } else {
+ /* No conclusion can be drawn for same hash value and same UREFx type. */
+ return ALIAS_MAY;
+ }
}
}
@@ -429,7 +504,6 @@
cselim:
/* Try to find a matching load. Below the conflicting store, if any. */
-
ref = J->chain[IR_ULOAD];
while (ref > lim) {
IRIns *ir = IR(ref);
@@ -470,10 +544,7 @@
goto doemit; /* No elimination possible. */
/* Remove redundant store from chain and replace with NOP. */
*refp = store->prev;
- store->o = IR_NOP;
- store->t.irt = IRT_NIL;
- store->op1 = store->op2 = 0;
- store->prev = 0;
+ lj_ir_nop(store);
if (ref+1 < J->cur.nins &&
store[1].o == IR_OBAR && store[1].op1 == xref) {
IRRef1 *bp = &J->chain[IR_OBAR];
@@ -482,10 +553,7 @@
bp = &obar->prev;
/* Remove OBAR, too. */
*bp = obar->prev;
- obar->o = IR_NOP;
- obar->t.irt = IRT_NIL;
- obar->op1 = obar->op2 = 0;
- obar->prev = 0;
+ lj_ir_nop(obar);
}
/* Now emit the new store instead. */
}
@@ -565,8 +633,9 @@
goto doemit;
break; /* Otherwise continue searching. */
case ALIAS_MUST:
- if (store->op2 == val) /* Same value: drop the new store. */
- return DROPFOLD;
+ if (store->op2 == val &&
+ !(xr->op2 >= IRFL_SBUF_W && xr->op2 <= IRFL_SBUF_R))
+ return DROPFOLD; /* Same value: drop the new store. */
/* Different value: try to eliminate the redundant store. */
if (ref > J->chain[IR_LOOP]) { /* Quick check to avoid crossing LOOP. */
IRIns *ir;
@@ -576,10 +645,7 @@
goto doemit; /* No elimination possible. */
/* Remove redundant store from chain and replace with NOP. */
*refp = store->prev;
- store->o = IR_NOP;
- store->t.irt = IRT_NIL;
- store->op1 = store->op2 = 0;
- store->prev = 0;
+ lj_ir_nop(store);
/* Now emit the new store instead. */
}
goto doemit;
@@ -590,6 +656,29 @@
return EMITFOLD; /* Otherwise we have a conflict or simply no match. */
}
+/* Check whether there's no aliasing buffer op between IRFL_SBUF_*. */
+int LJ_FASTCALL lj_opt_fwd_sbuf(jit_State *J, IRRef lim)
+{
+ IRRef ref;
+ if (J->chain[IR_BUFPUT] > lim)
+ return 0; /* Conflict. */
+ ref = J->chain[IR_CALLS];
+ while (ref > lim) {
+ IRIns *ir = IR(ref);
+ if (ir->op2 >= IRCALL_lj_strfmt_putint && ir->op2 < IRCALL_lj_buf_tostr)
+ return 0; /* Conflict. */
+ ref = ir->prev;
+ }
+ ref = J->chain[IR_CALLL];
+ while (ref > lim) {
+ IRIns *ir = IR(ref);
+ if (ir->op2 >= IRCALL_lj_strfmt_putint && ir->op2 < IRCALL_lj_buf_tostr)
+ return 0; /* Conflict. */
+ ref = ir->prev;
+ }
+ return 1; /* No conflict. Can safely FOLD/CSE. */
+}
+
/* -- XLOAD forwarding and XSTORE elimination ----------------------------- */
/* Find cdata allocation for a reference (if any). */
@@ -830,10 +919,7 @@
goto doemit; /* No elimination possible. */
/* Remove redundant store from chain and replace with NOP. */
*refp = store->prev;
- store->o = IR_NOP;
- store->t.irt = IRT_NIL;
- store->op1 = store->op2 = 0;
- store->prev = 0;
+ lj_ir_nop(store);
/* Now emit the new store instead. */
}
goto doemit;
@@ -844,39 +930,6 @@
return EMITFOLD; /* Otherwise we have a conflict or simply no match. */
}
-/* -- Forwarding of lj_tab_len -------------------------------------------- */
-
-/* This is rather simplistic right now, but better than nothing. */
-TRef LJ_FASTCALL lj_opt_fwd_tab_len(jit_State *J)
-{
- IRRef tab = fins->op1; /* Table reference. */
- IRRef lim = tab; /* Search limit. */
- IRRef ref;
-
- /* Any ASTORE is a conflict and limits the search. */
- if (J->chain[IR_ASTORE] > lim) lim = J->chain[IR_ASTORE];
-
- /* Search for conflicting HSTORE with numeric key. */
- ref = J->chain[IR_HSTORE];
- while (ref > lim) {
- IRIns *store = IR(ref);
- IRIns *href = IR(store->op1);
- IRIns *key = IR(href->op2);
- if (irt_isnum(key->o == IR_KSLOT ? IR(key->op1)->t : key->t)) {
- lim = ref; /* Conflicting store found, limits search for TLEN. */
- break;
- }
- ref = store->prev;
- }
-
- /* Search for aliasing table.clear. */
- if (!fwd_aa_tab_clear(J, lim, tab))
- return lj_ir_emit(J);
-
- /* Try to find a matching load. Below the conflicting store, if any. */
- return lj_opt_cselim(J, lim);
-}
-
/* -- ASTORE/HSTORE previous type analysis -------------------------------- */
/* Check whether the previous value for a table store is non-nil.
@@ -907,6 +960,8 @@
if (skref == xkref || !irref_isk(skref) || !irref_isk(xkref))
return 0; /* A nil store with same const key or var key MAY alias. */
/* Different const keys CANNOT alias. */
+ } else if (irt_isp32(IR(skref)->t) != irt_isp32(IR(xkref)->t)) {
+ return 0; /* HREF and HREFK MAY alias. */
} /* Different key types CANNOT alias. */
} /* Other non-nil stores MAY alias. */
ref = store->prev;
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_narrow.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_narrow.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_narrow.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,7 +1,7 @@
/*
** NARROW: Narrowing of numbers to integers (double to int32_t).
** STRIPOV: Stripping of overflow checks.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_narrow_c
@@ -341,7 +341,8 @@
NarrowIns *savesp = nc->sp;
int count = narrow_conv_backprop(nc, ir->op1, depth);
count += narrow_conv_backprop(nc, ir->op2, depth);
- if (count <= 1) { /* Limit total number of conversions. */
+ /* Limit total number of conversions. */
+ if (count <= 1 && nc->sp < nc->maxsp) {
*nc->sp++ = NARROWINS(IRT(ir->o, nc->t), ref);
return count;
}
@@ -372,17 +373,17 @@
} else if (op == NARROW_CONV) {
*sp++ = emitir_raw(convot, ref, convop2); /* Raw emit avoids a loop. */
} else if (op == NARROW_SEXT) {
- lua_assert(sp >= nc->stack+1);
+ lj_assertJ(sp >= nc->stack+1, "stack underflow");
sp[-1] = emitir(IRT(IR_CONV, IRT_I64), sp[-1],
(IRT_I64<<5)|IRT_INT|IRCONV_SEXT);
} else if (op == NARROW_INT) {
- lua_assert(next < last);
+ lj_assertJ(next < last, "missing arg to NARROW_INT");
*sp++ = nc->t == IRT_I64 ?
lj_ir_kint64(J, (int64_t)(int32_t)*next++) :
lj_ir_kint(J, *next++);
} else { /* Regular IROpT. Pops two operands and pushes one result. */
IRRef mode = nc->mode;
- lua_assert(sp >= nc->stack+2);
+ lj_assertJ(sp >= nc->stack+2, "stack underflow");
sp--;
/* Omit some overflow checks for array indexing. See comments above. */
if ((mode & IRCONV_CONVMASK) == IRCONV_INDEX) {
@@ -398,7 +399,7 @@
narrow_bpc_set(J, narrow_ref(ref), narrow_ref(sp[-1]), mode);
}
}
- lua_assert(sp == nc->stack+1);
+ lj_assertJ(sp == nc->stack+1, "stack misalignment");
return nc->stack[0];
}
@@ -452,7 +453,7 @@
TRef LJ_FASTCALL lj_opt_narrow_index(jit_State *J, TRef tr)
{
IRIns *ir;
- lua_assert(tref_isnumber(tr));
+ lj_assertJ(tref_isnumber(tr), "expected number type");
if (tref_isnum(tr)) /* Conversion may be narrowed, too. See above. */
return emitir(IRTGI(IR_CONV), tr, IRCONV_INT_NUM|IRCONV_INDEX);
/* Omit some overflow checks for array indexing. See comments above. */
@@ -499,7 +500,7 @@
/* Narrow C array index (overflow undefined). */
TRef LJ_FASTCALL lj_opt_narrow_cindex(jit_State *J, TRef tr)
{
- lua_assert(tref_isnumber(tr));
+ lj_assertJ(tref_isnumber(tr), "expected number type");
if (tref_isnum(tr))
return emitir(IRT(IR_CONV, IRT_INTP), tr, (IRT_INTP<<5)|IRT_NUM|IRCONV_ANY);
/* Undefined overflow semantics allow stripping of ADDOV, SUBOV and MULOV. */
@@ -551,8 +552,13 @@
{
rc = conv_str_tonum(J, rc, vc);
if (tref_isinteger(rc)) {
- if ((uint32_t)numberVint(vc) != 0x80000000u)
- return emitir(IRTGI(IR_SUBOV), lj_ir_kint(J, 0), rc);
+ uint32_t k = (uint32_t)numberVint(vc);
+ if ((LJ_DUALNUM || k != 0) && k != 0x80000000u) {
+ TRef zero = lj_ir_kint(J, 0);
+ if (!LJ_DUALNUM)
+ emitir(IRTGI(IR_NE), rc, zero);
+ return emitir(IRTGI(IR_SUBOV), zero, rc);
+ }
rc = emitir(IRTN(IR_CONV), rc, IRCONV_NUM_INT);
}
return emitir(IRTN(IR_NEG), rc, lj_ir_ksimd(J, LJ_KSIMD_NEG));
@@ -579,44 +585,6 @@
return emitir(IRTN(IR_SUB), rb, tmp);
}
-/* Narrowing of power operator or math.pow. */
-TRef lj_opt_narrow_pow(jit_State *J, TRef rb, TRef rc, TValue *vb, TValue *vc)
-{
- rb = conv_str_tonum(J, rb, vb);
- rb = lj_ir_tonum(J, rb); /* Left arg is always treated as an FP number. */
- rc = conv_str_tonum(J, rc, vc);
- /* Narrowing must be unconditional to preserve (-x)^i semantics. */
- if (tvisint(vc) || numisint(numV(vc))) {
- int checkrange = 0;
- /* Split pow is faster for bigger exponents. But do this only for (+k)^i. */
- if (tref_isk(rb) && (int32_t)ir_knum(IR(tref_ref(rb)))->u32.hi >= 0) {
- int32_t k = numberVint(vc);
- if (!(k >= -65536 && k <= 65536)) goto split_pow;
- checkrange = 1;
- }
- if (!tref_isinteger(rc)) {
- /* Guarded conversion to integer! */
- rc = emitir(IRTGI(IR_CONV), rc, IRCONV_INT_NUM|IRCONV_CHECK);
- }
- if (checkrange && !tref_isk(rc)) { /* Range guard: -65536 <= i <= 65536 */
- TRef tmp = emitir(IRTI(IR_ADD), rc, lj_ir_kint(J, 65536));
- emitir(IRTGI(IR_ULE), tmp, lj_ir_kint(J, 2*65536));
- }
- return emitir(IRTN(IR_POW), rb, rc);
- }
-split_pow:
- /* FOLD covers most cases, but some are easier to do here. */
- if (tref_isk(rb) && tvispone(ir_knum(IR(tref_ref(rb)))))
- return rb; /* 1 ^ x ==> 1 */
- rc = lj_ir_tonum(J, rc);
- if (tref_isk(rc) && ir_knum(IR(tref_ref(rc)))->n == 0.5)
- return emitir(IRTN(IR_FPMATH), rb, IRFPM_SQRT); /* x ^ 0.5 ==> sqrt(x) */
- /* Split up b^c into exp2(c*log2(b)). Assembler may rejoin later. */
- rb = emitir(IRTN(IR_FPMATH), rb, IRFPM_LOG2);
- rc = emitir(IRTN(IR_MUL), rb, rc);
- return emitir(IRTN(IR_FPMATH), rc, IRFPM_EXP2);
-}
-
/* -- Predictive narrowing of induction variables ------------------------- */
/* Narrow a single runtime value. */
@@ -630,9 +598,10 @@
/* Narrow the FORL index type by looking at the runtime values. */
IRType lj_opt_narrow_forl(jit_State *J, cTValue *tv)
{
- lua_assert(tvisnumber(&tv[FORL_IDX]) &&
+ lj_assertJ(tvisnumber(&tv[FORL_IDX]) &&
tvisnumber(&tv[FORL_STOP]) &&
- tvisnumber(&tv[FORL_STEP]));
+ tvisnumber(&tv[FORL_STEP]),
+ "expected number types");
/* Narrow only if the runtime values of start/stop/step are all integers. */
if (narrow_forl(J, &tv[FORL_IDX]) &&
narrow_forl(J, &tv[FORL_STOP]) &&
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_sink.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_sink.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_sink.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** SINK: Allocation Sinking and Store Sinking.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_sink_c
@@ -36,12 +36,14 @@
}
/* Recursively check whether a value depends on a PHI. */
-static int sink_phidep(jit_State *J, IRRef ref)
+static int sink_phidep(jit_State *J, IRRef ref, int *workp)
{
IRIns *ir = IR(ref);
+ if (!*workp) return 1; /* Give up and pretend it does. */
+ (*workp)--;
if (irt_isphi(ir->t)) return 1;
- if (ir->op1 >= REF_FIRST && sink_phidep(J, ir->op1)) return 1;
- if (ir->op2 >= REF_FIRST && sink_phidep(J, ir->op2)) return 1;
+ if (ir->op1 >= REF_FIRST && sink_phidep(J, ir->op1, workp)) return 1;
+ if (ir->op2 >= REF_FIRST && sink_phidep(J, ir->op2, workp)) return 1;
return 0;
}
@@ -56,7 +58,13 @@
return 1; /* Sinkable PHI. */
}
/* Otherwise the value must be loop-invariant. */
- return ref < J->loopref && !sink_phidep(J, ref);
+ if (ref < J->loopref) {
+ /* Check for PHI dependencies, but give up after reasonable effort. */
+ int work = 64;
+ return !sink_phidep(J, ref, &work);
+ } else {
+ return 0; /* Loop-variant. */
+ }
}
return 1; /* Constant (non-PHI). */
}
@@ -78,8 +86,7 @@
switch (ir->o) {
case IR_BASE:
return; /* Finished. */
- case IR_CALLL: /* IRCALL_lj_tab_len */
- case IR_ALOAD: case IR_HLOAD: case IR_XLOAD: case IR_TBAR:
+ case IR_ALOAD: case IR_HLOAD: case IR_XLOAD: case IR_TBAR: case IR_ALEN:
irt_setmark(IR(ir->op1)->t); /* Mark ref for remaining loads. */
break;
case IR_FLOAD:
@@ -100,8 +107,8 @@
(LJ_32 && ir+1 < irlast && (ir+1)->o == IR_HIOP &&
!sink_checkphi(J, ir, (ir+1)->op2))))
irt_setmark(ir->t); /* Mark ineligible allocation. */
+#endif
/* fallthrough */
-#endif
case IR_USTORE:
irt_setmark(IR(ir->op2)->t); /* Mark stored value. */
break;
@@ -219,6 +226,7 @@
for (ir = IR(J->cur.nk); ir < irbase; ir++) {
irt_clearmark(ir->t);
ir->prev = REGSP_INIT;
+ /* The false-positive of irt_is64() for ASMREF_L (REF_NIL) is OK here. */
if (irt_is64(ir->t) && ir->o != IR_KNULL)
ir++;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_split.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_split.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_opt_split.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** SPLIT: Split 64 bit IR instructions into 32 bit IR instructions.
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#define lj_opt_split_c
@@ -8,7 +8,7 @@
#include "lj_obj.h"
-#if LJ_HASJIT && (LJ_SOFTFP || (LJ_32 && LJ_HASFFI))
+#if LJ_HASJIT && (LJ_SOFTFP32 || (LJ_32 && LJ_HASFFI))
#include "lj_err.h"
#include "lj_buf.h"
@@ -235,7 +235,7 @@
return split_emit(J, IRTI(IR_BOR), t1, t2);
} else {
IRRef t1 = ir->prev, t2;
- lua_assert(op == IR_BSHR || op == IR_BSAR);
+ lj_assertJ(op == IR_BSHR || op == IR_BSAR, "bad usage");
nir->o = IR_BSHR;
t2 = split_emit(J, IRTI(IR_BSHL), hi, lj_ir_kint(J, (-k&31)));
ir->prev = split_emit(J, IRTI(IR_BOR), t1, t2);
@@ -250,7 +250,7 @@
ir->prev = lj_ir_kint(J, 0);
return lo;
} else {
- lua_assert(op == IR_BSHR || op == IR_BSAR);
+ lj_assertJ(op == IR_BSHR || op == IR_BSAR, "bad usage");
if (k == 32) {
J->cur.nins--;
ir->prev = hi;
@@ -400,35 +400,11 @@
hi = split_call_ll(J, hisubst, oir, ir, IRCALL_softfp_div);
break;
case IR_POW:
- hi = split_call_li(J, hisubst, oir, ir, IRCALL_lj_vm_powi);
+ hi = split_call_ll(J, hisubst, oir, ir, IRCALL_pow);
break;
case IR_FPMATH:
- /* Try to rejoin pow from EXP2, MUL and LOG2. */
- if (nir->op2 == IRFPM_EXP2 && nir->op1 > J->loopref) {
- IRIns *irp = IR(nir->op1);
- if (irp->o == IR_CALLN && irp->op2 == IRCALL_softfp_mul) {
- IRIns *irm4 = IR(irp->op1);
- IRIns *irm3 = IR(irm4->op1);
- IRIns *irm12 = IR(irm3->op1);
- IRIns *irl1 = IR(irm12->op1);
- if (irm12->op1 > J->loopref && irl1->o == IR_CALLN &&
- irl1->op2 == IRCALL_lj_vm_log2) {
- IRRef tmp = irl1->op1; /* Recycle first two args from LOG2. */
- IRRef arg3 = irm3->op2, arg4 = irm4->op2;
- J->cur.nins--;
- tmp = split_emit(J, IRT(IR_CARG, IRT_NIL), tmp, arg3);
- tmp = split_emit(J, IRT(IR_CARG, IRT_NIL), tmp, arg4);
- ir->prev = tmp = split_emit(J, IRTI(IR_CALLN), tmp, IRCALL_pow);
- hi = split_emit(J, IRT(IR_HIOP, IRT_SOFTFP), tmp, tmp);
- break;
- }
- }
- }
hi = split_call_l(J, hisubst, oir, ir, IRCALL_lj_vm_floor + ir->op2);
break;
- case IR_ATAN2:
- hi = split_call_ll(J, hisubst, oir, ir, IRCALL_atan2);
- break;
case IR_LDEXP:
hi = split_call_li(J, hisubst, oir, ir, IRCALL_ldexp);
break;
@@ -453,7 +429,7 @@
hi = split_emit(J, IRT(IR_HIOP, IRT_SOFTFP), nref, nref);
break;
case IR_FLOAD:
- lua_assert(ir->op1 == REF_NIL);
+ lj_assertJ(ir->op1 == REF_NIL, "expected FLOAD from GG_State");
hi = lj_ir_kint(J, *(int32_t*)((char*)J2GG(J) + ir->op2 + LJ_LE*4));
nir->op2 += LJ_BE*4;
break;
@@ -489,8 +465,9 @@
break;
}
#endif
- lua_assert(st == IRT_INT ||
- (LJ_32 && LJ_HASFFI && (st == IRT_U32 || st == IRT_FLOAT)));
+ lj_assertJ(st == IRT_INT ||
+ (LJ_32 && LJ_HASFFI && (st == IRT_U32 || st == IRT_FLOAT)),
+ "bad source type for CONV");
nir->o = IR_CALLN;
#if LJ_32 && LJ_HASFFI
nir->op2 = st == IRT_INT ? IRCALL_softfp_i2d :
@@ -520,7 +497,8 @@
hi = nir->op2;
break;
default:
- lua_assert(ir->o <= IR_NE || ir->o == IR_MIN || ir->o == IR_MAX);
+ lj_assertJ(ir->o <= IR_NE || ir->o == IR_MIN || ir->o == IR_MAX,
+ "bad IR op %d", ir->o);
hi = split_emit(J, IRTG(IR_HIOP, IRT_SOFTFP),
hisubst[ir->op1], hisubst[ir->op2]);
break;
@@ -577,7 +555,7 @@
hi = split_bitshift(J, hisubst, oir, nir, ir);
break;
case IR_FLOAD:
- lua_assert(ir->op2 == IRFL_CDATA_INT64);
+ lj_assertJ(ir->op2 == IRFL_CDATA_INT64, "only INT64 supported");
hi = split_emit(J, IRTI(IR_FLOAD), nir->op1, IRFL_CDATA_INT64_4);
#if LJ_BE
ir->prev = hi; hi = nref;
@@ -643,7 +621,7 @@
hi = nir->op2;
break;
default:
- lua_assert(ir->o <= IR_NE); /* Comparisons. */
+ lj_assertJ(ir->o <= IR_NE, "bad IR op %d", ir->o); /* Comparisons. */
split_emit(J, IRTGI(IR_HIOP), hiref, hisubst[ir->op2]);
break;
}
@@ -667,7 +645,7 @@
tmp = split_emit(J, IRT(IR_CARG, IRT_NIL), hisubst[op1], oir[op1].prev);
#endif
ir->prev = split_emit(J, IRTI(IR_CALLN), tmp, IRCALL_lj_vm_tobit);
- } else if (ir->o == IR_TOSTR) {
+ } else if (ir->o == IR_TOSTR || ir->o == IR_TMPREF) {
if (hisubst[ir->op1]) {
if (irref_isk(ir->op1))
nir->op1 = ir->op1;
@@ -721,7 +699,7 @@
#if LJ_SOFTFP
if (st == IRT_NUM || (LJ_32 && LJ_HASFFI && st == IRT_FLOAT)) {
if (irt_isguard(ir->t)) {
- lua_assert(st == IRT_NUM && irt_isint(ir->t));
+ lj_assertJ(st == IRT_NUM && irt_isint(ir->t), "bad CONV types");
J->cur.nins--;
ir->prev = split_num2int(J, nir->op1, hisubst[ir->op1], 1);
} else {
@@ -852,7 +830,7 @@
if (!J->needsplit)
J->needsplit = split_needsplit(J);
#else
- lua_assert(J->needsplit >= split_needsplit(J)); /* Verify flag. */
+ lj_assertJ(J->needsplit >= split_needsplit(J), "bad SPLIT state");
#endif
if (J->needsplit) {
int errcode = lj_vm_cpcall(J->L, NULL, J, cpsplit);
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.c 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Lua parser (source code -> bytecode).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
**
** Major portions taken verbatim or adapted from the Lua interpreter.
** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
@@ -163,6 +163,12 @@
LJ_STATIC_ASSERT((int)BC_DIVVV-(int)BC_ADDVV == (int)OPR_DIV-(int)OPR_ADD);
LJ_STATIC_ASSERT((int)BC_MODVV-(int)BC_ADDVV == (int)OPR_MOD-(int)OPR_ADD);
+#ifdef LUA_USE_ASSERT
+#define lj_assertFS(c, ...) (lj_assertG_(G(fs->L), (c), __VA_ARGS__))
+#else
+#define lj_assertFS(c, ...) ((void)fs)
+#endif
+
/* -- Error handling ------------------------------------------------------ */
LJ_NORET LJ_NOINLINE static void err_syntax(LexState *ls, ErrMsg em)
@@ -200,7 +206,7 @@
{
lua_State *L = fs->L;
TValue *o;
- lua_assert(expr_isnumk(e));
+ lj_assertFS(expr_isnumk(e), "bad usage");
o = lj_tab_set(L, fs->kt, &e->u.nval);
if (tvhaskslot(o))
return tvkslot(o);
@@ -225,7 +231,7 @@
/* Add a string constant. */
static BCReg const_str(FuncState *fs, ExpDesc *e)
{
- lua_assert(expr_isstrk(e) || e->k == VGLOBAL);
+ lj_assertFS(expr_isstrk(e) || e->k == VGLOBAL, "bad usage");
return const_gc(fs, obj2gco(e->u.sval), LJ_TSTR);
}
@@ -313,7 +319,7 @@
{
BCIns *jmp = &fs->bcbase[pc].ins;
BCPos offset = dest-(pc+1)+BCBIAS_J;
- lua_assert(dest != NO_JMP);
+ lj_assertFS(dest != NO_JMP, "uninitialized jump target");
if (offset > BCMAX_D)
err_syntax(fs->ls, LJ_ERR_XJUMP);
setbc_d(jmp, offset);
@@ -362,7 +368,7 @@
if (target == fs->pc) {
jmp_tohere(fs, list);
} else {
- lua_assert(target < fs->pc);
+ lj_assertFS(target < fs->pc, "bad jump target");
jmp_patchval(fs, list, target, NO_REG, target);
}
}
@@ -392,7 +398,7 @@
{
if (reg >= fs->nactvar) {
fs->freereg--;
- lua_assert(reg == fs->freereg);
+ lj_assertFS(reg == fs->freereg, "bad regfree");
}
}
@@ -542,7 +548,7 @@
} else if (e->k <= VKTRUE) {
ins = BCINS_AD(BC_KPRI, reg, const_pri(e));
} else {
- lua_assert(e->k == VVOID || e->k == VJMP);
+ lj_assertFS(e->k == VVOID || e->k == VJMP, "bad expr type %d", e->k);
return;
}
bcemit_INS(fs, ins);
@@ -637,7 +643,7 @@
ins = BCINS_AD(BC_GSET, ra, const_str(fs, var));
} else {
BCReg ra, rc;
- lua_assert(var->k == VINDEXED);
+ lj_assertFS(var->k == VINDEXED, "bad expr type %d", var->k);
ra = expr_toanyreg(fs, e);
rc = var->u.s.aux;
if ((int32_t)rc < 0) {
@@ -645,10 +651,12 @@
} else if (rc > BCMAX_C) {
ins = BCINS_ABC(BC_TSETB, ra, var->u.s.info, rc-(BCMAX_C+1));
} else {
+#ifdef LUA_USE_ASSERT
/* Free late alloced key reg to avoid assert on free of value reg. */
/* This can only happen when called from expr_table(). */
- lua_assert(e->k != VNONRELOC || ra < fs->nactvar ||
- rc < ra || (bcreg_free(fs, rc),1));
+ if (e->k == VNONRELOC && ra >= fs->nactvar && rc >= ra)
+ bcreg_free(fs, rc);
+#endif
ins = BCINS_ABC(BC_TSETV, ra, var->u.s.info, rc);
}
}
@@ -659,19 +667,20 @@
/* Emit method lookup expression. */
static void bcemit_method(FuncState *fs, ExpDesc *e, ExpDesc *key)
{
- BCReg idx, func, obj = expr_toanyreg(fs, e);
+ BCReg idx, func, fr2, obj = expr_toanyreg(fs, e);
expr_free(fs, e);
func = fs->freereg;
- bcemit_AD(fs, BC_MOV, func+1+LJ_FR2, obj); /* Copy object to 1st argument. */
- lua_assert(expr_isstrk(key));
+ fr2 = fs->ls->fr2;
+ bcemit_AD(fs, BC_MOV, func+1+fr2, obj); /* Copy object to 1st argument. */
+ lj_assertFS(expr_isstrk(key), "bad usage");
idx = const_str(fs, key);
if (idx <= BCMAX_C) {
- bcreg_reserve(fs, 2+LJ_FR2);
+ bcreg_reserve(fs, 2+fr2);
bcemit_ABC(fs, BC_TGETS, func, obj, idx);
} else {
- bcreg_reserve(fs, 3+LJ_FR2);
- bcemit_AD(fs, BC_KSTR, func+2+LJ_FR2, idx);
- bcemit_ABC(fs, BC_TGETV, func, obj, func+2+LJ_FR2);
+ bcreg_reserve(fs, 3+fr2);
+ bcemit_AD(fs, BC_KSTR, func+2+fr2, idx);
+ bcemit_ABC(fs, BC_TGETV, func, obj, func+2+fr2);
fs->freereg--;
}
e->u.s.info = func;
@@ -803,7 +812,8 @@
else
rc = expr_toanyreg(fs, e2);
/* 1st operand discharged by bcemit_binop_left, but need KNUM/KSHORT. */
- lua_assert(expr_isnumk(e1) || e1->k == VNONRELOC);
+ lj_assertFS(expr_isnumk(e1) || e1->k == VNONRELOC,
+ "bad expr type %d", e1->k);
expr_toval(fs, e1);
/* Avoid two consts to satisfy bytecode constraints. */
if (expr_isnumk(e1) && !expr_isnumk(e2) &&
@@ -853,9 +863,12 @@
e1 = e2; e2 = eret; /* Swap operands. */
op = ((op-BC_ISLT)^3)+BC_ISLT;
expr_toval(fs, e1);
+ ra = expr_toanyreg(fs, e1);
+ rd = expr_toanyreg(fs, e2);
+ } else {
+ rd = expr_toanyreg(fs, e2);
+ ra = expr_toanyreg(fs, e1);
}
- rd = expr_toanyreg(fs, e2);
- ra = expr_toanyreg(fs, e1);
ins = BCINS_AD(op, ra, rd);
}
/* Using expr_free might cause asserts if the order is wrong. */
@@ -888,12 +901,12 @@
if (op <= OPR_POW) {
bcemit_arith(fs, op, e1, e2);
} else if (op == OPR_AND) {
- lua_assert(e1->t == NO_JMP); /* List must be closed. */
+ lj_assertFS(e1->t == NO_JMP, "jump list not closed");
expr_discharge(fs, e2);
jmp_append(fs, &e2->f, e1->f);
*e1 = *e2;
} else if (op == OPR_OR) {
- lua_assert(e1->f == NO_JMP); /* List must be closed. */
+ lj_assertFS(e1->f == NO_JMP, "jump list not closed");
expr_discharge(fs, e2);
jmp_append(fs, &e2->t, e1->t);
*e1 = *e2;
@@ -900,7 +913,8 @@
} else if (op == OPR_CONCAT) {
expr_toval(fs, e2);
if (e2->k == VRELOCABLE && bc_op(*bcptr(fs, e2)) == BC_CAT) {
- lua_assert(e1->u.s.info == bc_b(*bcptr(fs, e2))-1);
+ lj_assertFS(e1->u.s.info == bc_b(*bcptr(fs, e2))-1,
+ "bad CAT stack layout");
expr_free(fs, e1);
setbc_b(bcptr(fs, e2), e1->u.s.info);
e1->u.s.info = e2->u.s.info;
@@ -912,8 +926,9 @@
}
e1->k = VRELOCABLE;
} else {
- lua_assert(op == OPR_NE || op == OPR_EQ ||
- op == OPR_LT || op == OPR_GE || op == OPR_LE || op == OPR_GT);
+ lj_assertFS(op == OPR_NE || op == OPR_EQ ||
+ op == OPR_LT || op == OPR_GE || op == OPR_LE || op == OPR_GT,
+ "bad binop %d", op);
bcemit_comp(fs, op, e1, e2);
}
}
@@ -942,19 +957,19 @@
e->u.s.info = fs->freereg-1;
e->k = VNONRELOC;
} else {
- lua_assert(e->k == VNONRELOC);
+ lj_assertFS(e->k == VNONRELOC, "bad expr type %d", e->k);
}
} else {
- lua_assert(op == BC_UNM || op == BC_LEN);
+ lj_assertFS(op == BC_UNM || op == BC_LEN, "bad unop %d", op);
if (op == BC_UNM && !expr_hasjump(e)) { /* Constant-fold negations. */
#if LJ_HASFFI
if (e->k == VKCDATA) { /* Fold in-place since cdata is not interned. */
GCcdata *cd = cdataV(&e->u.nval);
- int64_t *p = (int64_t *)cdataptr(cd);
+ uint64_t *p = (uint64_t *)cdataptr(cd);
if (cd->ctypeid == CTID_COMPLEX_DOUBLE)
- p[1] ^= (int64_t)U64x(80000000,00000000);
+ p[1] ^= U64x(80000000,00000000);
else
- *p = -*p;
+ *p = ~*p+1u;
return;
} else
#endif
@@ -961,11 +976,11 @@
if (expr_isnumk(e) && !expr_numiszero(e)) { /* Avoid folding to -0. */
TValue *o = expr_numtv(e);
if (tvisint(o)) {
- int32_t k = intV(o);
- if (k == -k)
+ int32_t k = intV(o), negk = (int32_t)(~(uint32_t)k+1u);
+ if (k == negk)
setnumV(o, -(lua_Number)k);
else
- setintV(o, -k);
+ setintV(o, negk);
return;
} else {
o->u64 ^= U64x(80000000,00000000);
@@ -1040,8 +1055,9 @@
lj_lex_error(ls, 0, LJ_ERR_XLIMC, LJ_MAX_VSTACK);
lj_mem_growvec(ls->L, ls->vstack, ls->sizevstack, LJ_MAX_VSTACK, VarInfo);
}
- lua_assert((uintptr_t)name < VARNAME__MAX ||
- lj_tab_getstr(fs->kt, name) != NULL);
+ lj_assertFS((uintptr_t)name < VARNAME__MAX ||
+ lj_tab_getstr(fs->kt, name) != NULL,
+ "unanchored variable name");
/* NOBARRIER: name is anchored in fs->kt and ls->vstack is not a GCobj. */
setgcref(ls->vstack[vtop].name, obj2gco(name));
fs->varmap[fs->nactvar+n] = (uint16_t)vtop;
@@ -1096,7 +1112,7 @@
return i; /* Already exists. */
/* Otherwise create a new one. */
checklimit(fs, fs->nuv, LJ_MAX_UPVAL, "upvalues");
- lua_assert(e->k == VLOCAL || e->k == VUPVAL);
+ lj_assertFS(e->k == VLOCAL || e->k == VUPVAL, "bad expr type %d", e->k);
fs->uvmap[n] = (uint16_t)vidx;
fs->uvtmp[n] = (uint16_t)(e->k == VLOCAL ? vidx : LJ_MAX_VSTACK+e->u.s.info);
fs->nuv = n+1;
@@ -1147,7 +1163,8 @@
lj_lex_error(ls, 0, LJ_ERR_XLIMC, LJ_MAX_VSTACK);
lj_mem_growvec(ls->L, ls->vstack, ls->sizevstack, LJ_MAX_VSTACK, VarInfo);
}
- lua_assert(name == NAME_BREAK || lj_tab_getstr(fs->kt, name) != NULL);
+ lj_assertFS(name == NAME_BREAK || lj_tab_getstr(fs->kt, name) != NULL,
+ "unanchored label name");
/* NOBARRIER: name is anchored in fs->kt and ls->vstack is not a GCobj. */
setgcref(ls->vstack[vtop].name, obj2gco(name));
ls->vstack[vtop].startpc = pc;
@@ -1177,8 +1194,9 @@
FuncState *fs = ls->fs;
BCPos pc = vg->startpc;
BCIns *ip = &fs->bcbase[pc].ins;
- lua_assert(gola_isgoto(vg));
- lua_assert(bc_op(*ip) == BC_JMP || bc_op(*ip) == BC_UCLO);
+ lj_assertFS(gola_isgoto(vg), "expected goto");
+ lj_assertFS(bc_op(*ip) == BC_JMP || bc_op(*ip) == BC_UCLO,
+ "bad bytecode op %d", bc_op(*ip));
setbc_a(ip, vg->slot);
if (bc_op(*ip) == BC_JMP) {
BCPos next = jmp_next(fs, pc);
@@ -1197,9 +1215,9 @@
if (gcrefeq(vg->name, vl->name) && gola_isgoto(vg)) {
if (vg->slot < vl->slot) {
GCstr *name = strref(var_get(ls, ls->fs, vg->slot).name);
- lua_assert((uintptr_t)name >= VARNAME__MAX);
+ lj_assertLS((uintptr_t)name >= VARNAME__MAX, "expected goto name");
ls->linenumber = ls->fs->bcbase[vg->startpc].line;
- lua_assert(strref(vg->name) != NAME_BREAK);
+ lj_assertLS(strref(vg->name) != NAME_BREAK, "unexpected break");
lj_lex_error(ls, 0, LJ_ERR_XGSCOPE,
strdata(strref(vg->name)), strdata(name));
}
@@ -1263,7 +1281,7 @@
bl->vstart = fs->ls->vtop;
bl->prev = fs->bl;
fs->bl = bl;
- lua_assert(fs->freereg == fs->nactvar);
+ lj_assertFS(fs->freereg == fs->nactvar, "bad regalloc");
}
/* End a scope. */
@@ -1274,7 +1292,7 @@
fs->bl = bl->prev;
var_remove(ls, bl->nactvar);
fs->freereg = fs->nactvar;
- lua_assert(bl->nactvar == fs->nactvar);
+ lj_assertFS(bl->nactvar == fs->nactvar, "bad regalloc");
if ((bl->flags & (FSCOPE_UPVAL|FSCOPE_NOCLOSE)) == FSCOPE_UPVAL)
bcemit_AJ(fs, BC_UCLO, bl->nactvar, 0);
if ((bl->flags & FSCOPE_BREAK)) {
@@ -1309,9 +1327,12 @@
{
BCInsLine *base = fs->bcbase;
MSize i;
+ BCIns op;
pt->sizebc = n;
- bc[0] = BCINS_AD((fs->flags & PROTO_VARARG) ? BC_FUNCV : BC_FUNCF,
- fs->framesize, 0);
+ if (fs->ls->fr2 != LJ_FR2) op = BC_NOT; /* Mark non-native prototype. */
+ else if ((fs->flags & PROTO_VARARG)) op = BC_FUNCV;
+ else op = BC_FUNCF;
+ bc[0] = BCINS_AD(op, fs->framesize, 0);
for (i = 1; i < n; i++)
bc[i] = base[i].ins;
}
@@ -1361,13 +1382,13 @@
Node *n = &node[i];
if (tvhaskslot(&n->val)) {
ptrdiff_t kidx = (ptrdiff_t)tvkslot(&n->val);
- lua_assert(!tvisint(&n->key));
+ lj_assertFS(!tvisint(&n->key), "unexpected integer key");
if (tvisnum(&n->key)) {
TValue *tv = &((TValue *)kptr)[kidx];
if (LJ_DUALNUM) {
lua_Number nn = numV(&n->key);
int32_t k = lj_num2int(nn);
- lua_assert(!tvismzero(&n->key));
+ lj_assertFS(!tvismzero(&n->key), "unexpected -0 key");
if ((lua_Number)k == nn)
setintV(tv, k);
else
@@ -1415,7 +1436,7 @@
uint8_t *li = (uint8_t *)lineinfo;
do {
BCLine delta = base[i].line - first;
- lua_assert(delta >= 0 && delta < 256);
+ lj_assertFS(delta >= 0 && delta < 256, "bad line delta");
li[i] = (uint8_t)delta;
} while (++i < n);
} else if (LJ_LIKELY(numline < 65536)) {
@@ -1422,7 +1443,7 @@
uint16_t *li = (uint16_t *)lineinfo;
do {
BCLine delta = base[i].line - first;
- lua_assert(delta >= 0 && delta < 65536);
+ lj_assertFS(delta >= 0 && delta < 65536, "bad line delta");
li[i] = (uint16_t)delta;
} while (++i < n);
} else {
@@ -1429,7 +1450,7 @@
uint32_t *li = (uint32_t *)lineinfo;
do {
BCLine delta = base[i].line - first;
- lua_assert(delta >= 0);
+ lj_assertFS(delta >= 0, "bad line delta");
li[i] = (uint32_t)delta;
} while (++i < n);
}
@@ -1448,7 +1469,7 @@
MSize len = s->len+1;
char *p = lj_buf_more(&ls->sb, len);
p = lj_buf_wmem(p, strdata(s), len);
- setsbufP(&ls->sb, p);
+ ls->sb.w = p;
}
*ofsvar = sbuflen(&ls->sb);
lastpc = 0;
@@ -1469,7 +1490,7 @@
startpc = vs->startpc;
p = lj_strfmt_wuleb128(p, startpc-lastpc);
p = lj_strfmt_wuleb128(p, vs->endpc-startpc);
- setsbufP(&ls->sb, p);
+ ls->sb.w = p;
lastpc = startpc;
}
}
@@ -1482,7 +1503,7 @@
{
setmref(pt->uvinfo, p);
setmref(pt->varinfo, (char *)p + ofsvar);
- memcpy(p, sbufB(&ls->sb), sbuflen(&ls->sb)); /* Copy from temp. buffer. */
+ memcpy(p, ls->sb.b, sbuflen(&ls->sb)); /* Copy from temp. buffer. */
}
#else
@@ -1519,7 +1540,7 @@
}
fs->bl->flags |= FSCOPE_NOCLOSE; /* Handled above. */
fscope_end(fs);
- lua_assert(fs->bl == NULL);
+ lj_assertFS(fs->bl == NULL, "bad scope nesting");
/* May need to fixup returns encoded before first function was created. */
if (fs->flags & PROTO_FIXUP_RETURN) {
BCPos pc;
@@ -1537,7 +1558,7 @@
/* Replace with UCLO plus branch. */
fs->bcbase[pc].ins = BCINS_AD(BC_UCLO, 0, offset);
break;
- case BC_UCLO:
+ case BC_FNEW:
return; /* We're done. */
default:
break;
@@ -1591,7 +1612,7 @@
L->top--; /* Pop table of constants. */
ls->vtop = fs->vbase; /* Reset variable stack. */
ls->fs = fs->prev;
- lua_assert(ls->fs != NULL || ls->tok == TK_eof);
+ lj_assertL(ls->fs != NULL || ls->tok == TK_eof, "bad parser state");
return pt;
}
@@ -1685,14 +1706,15 @@
}
/* Get value of constant expression. */
-static void expr_kvalue(TValue *v, ExpDesc *e)
+static void expr_kvalue(FuncState *fs, TValue *v, ExpDesc *e)
{
+ UNUSED(fs);
if (e->k <= VKTRUE) {
setpriV(v, ~(uint32_t)e->k);
} else if (e->k == VKSTR) {
setgcVraw(v, obj2gco(e->u.sval), LJ_TSTR);
} else {
- lua_assert(tvisnumber(expr_numtv(e)));
+ lj_assertFS(tvisnumber(expr_numtv(e)), "bad number constant");
*v = *expr_numtv(e);
}
}
@@ -1742,11 +1764,11 @@
fs->bcbase[pc].ins = BCINS_AD(BC_TDUP, freg-1, kidx);
}
vcall = 0;
- expr_kvalue(&k, &key);
+ expr_kvalue(fs, &k, &key);
v = lj_tab_set(fs->L, t, &k);
lj_gc_anybarriert(fs->L, t);
if (expr_isk_nojump(&val)) { /* Add const key/value to template table. */
- expr_kvalue(v, &val);
+ expr_kvalue(fs, v, &val);
} else { /* Otherwise create dummy string key (avoids lj_tab_newkey). */
settabV(fs->L, v, t); /* Preserve key with table itself as value. */
fixt = 1; /* Fix this later, after all resizes. */
@@ -1765,8 +1787,9 @@
if (vcall) {
BCInsLine *ilp = &fs->bcbase[fs->pc-1];
ExpDesc en;
- lua_assert(bc_a(ilp->ins) == freg &&
- bc_op(ilp->ins) == (narr > 256 ? BC_TSETV : BC_TSETB));
+ lj_assertFS(bc_a(ilp->ins) == freg &&
+ bc_op(ilp->ins) == (narr > 256 ? BC_TSETV : BC_TSETB),
+ "bad CALL code generation");
expr_init(&en, VKNUM, 0);
en.u.nval.u32.lo = narr-1;
en.u.nval.u32.hi = 0x43300000; /* Biased integer to avoid denormals. */
@@ -1796,7 +1819,7 @@
for (i = 0; i <= hmask; i++) {
Node *n = &node[i];
if (tvistab(&n->val)) {
- lua_assert(tabV(&n->val) == t);
+ lj_assertFS(tabV(&n->val) == t, "bad dummy key in template table");
setnilV(&n->val); /* Turn value into nil. */
}
}
@@ -1827,7 +1850,7 @@
} while (lex_opt(ls, ','));
}
var_add(ls, nparams);
- lua_assert(fs->nactvar == nparams);
+ lj_assertFS(fs->nactvar == nparams, "bad regalloc");
bcreg_reserve(fs, nparams);
lex_check(ls, ')');
return nparams;
@@ -1914,14 +1937,14 @@
err_syntax(ls, LJ_ERR_XFUNARG);
return; /* Silence compiler. */
}
- lua_assert(e->k == VNONRELOC);
+ lj_assertFS(e->k == VNONRELOC, "bad expr type %d", e->k);
base = e->u.s.info; /* Base register for call. */
if (args.k == VCALL) {
- ins = BCINS_ABC(BC_CALLM, base, 2, args.u.s.aux - base - 1 - LJ_FR2);
+ ins = BCINS_ABC(BC_CALLM, base, 2, args.u.s.aux - base - 1 - ls->fr2);
} else {
if (args.k != VVOID)
expr_tonextreg(fs, &args);
- ins = BCINS_ABC(BC_CALL, base, 2, fs->freereg - base - LJ_FR2);
+ ins = BCINS_ABC(BC_CALL, base, 2, fs->freereg - base - ls->fr2);
}
expr_init(e, VCALL, bcemit_INS(fs, ins));
e->u.s.aux = base;
@@ -1961,7 +1984,7 @@
parse_args(ls, v);
} else if (ls->tok == '(' || ls->tok == TK_string || ls->tok == '{') {
expr_tonextreg(fs, v);
- if (LJ_FR2) bcreg_reserve(fs, 1);
+ if (ls->fr2) bcreg_reserve(fs, 1);
parse_args(ls, v);
} else {
break;
@@ -2316,11 +2339,15 @@
BCReg nret = expr_list(ls, &e);
if (nret == 1) { /* Return one result. */
if (e.k == VCALL) { /* Check for tail call. */
+#ifdef LUAJIT_DISABLE_TAILCALL
+ goto notailcall;
+#else
BCIns *ip = bcptr(fs, &e);
/* It doesn't pay off to add BC_VARGT just for 'return ...'. */
if (bc_op(*ip) == BC_VARG) goto notailcall;
fs->pc--;
ins = BCINS_AD(bc_op(*ip)-BC_CALL+BC_CALLT, bc_a(*ip), bc_c(*ip));
+#endif
} else { /* Can return the result from any register. */
ins = BCINS_AD(BC_RET1, expr_toanyreg(fs, &e), 2);
}
@@ -2499,6 +2526,7 @@
cTValue *o;
switch (bc_op(ins)) {
case BC_MOV:
+ if (bc_d(ins) >= fs->nactvar) return 0;
name = gco2str(gcref(var_get(ls, fs, bc_d(ins)).name));
break;
case BC_UGET:
@@ -2543,8 +2571,8 @@
line = ls->linenumber;
assign_adjust(ls, 3, expr_list(ls, &e), &e);
/* The iterator needs another 3 [4] slots (func [pc] | state ctl). */
- bcreg_bump(fs, 3+LJ_FR2);
- isnext = (nvars <= 5 && predict_next(ls, fs, exprpc));
+ bcreg_bump(fs, 3+ls->fr2);
+ isnext = (nvars <= 5 && fs->pc > exprpc && predict_next(ls, fs, exprpc));
var_add(ls, 3); /* Hidden control variables. */
lex_check(ls, TK_do);
loop = bcemit_AJ(fs, isnext ? BC_ISNEXT : BC_JMP, base, NO_JMP);
@@ -2667,7 +2695,8 @@
lj_lex_next(ls);
parse_goto(ls);
break;
- } /* else: fallthrough */
+ }
+ /* fallthrough */
default:
parse_call_assign(ls);
break;
@@ -2683,8 +2712,9 @@
while (!islast && !parse_isend(ls->tok)) {
islast = parse_stmt(ls);
lex_opt(ls, ';');
- lua_assert(ls->fs->framesize >= ls->fs->freereg &&
- ls->fs->freereg >= ls->fs->nactvar);
+ lj_assertLS(ls->fs->framesize >= ls->fs->freereg &&
+ ls->fs->freereg >= ls->fs->nactvar,
+ "bad regalloc");
ls->fs->freereg = ls->fs->nactvar; /* Free registers after each stmt. */
}
synlevel_end(ls);
@@ -2719,9 +2749,8 @@
err_token(ls, TK_eof);
pt = fs_finish(ls, ls->linenumber);
L->top--; /* Drop chunkname. */
- lua_assert(fs.prev == NULL);
- lua_assert(ls->fs == NULL);
- lua_assert(pt->sizeuv == 0);
+ lj_assertL(fs.prev == NULL && ls->fs == NULL, "mismatched frame nesting");
+ lj_assertL(pt->sizeuv == 0, "toplevel proto has upvalues");
return pt;
}
Modified: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.h
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.h 2025-02-02 16:41:28 UTC (rev 73694)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_parse.h 2025-02-02 17:39:28 UTC (rev 73695)
@@ -1,6 +1,6 @@
/*
** Lua parser (source code -> bytecode).
-** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
*/
#ifndef _LJ_PARSE_H
Added: trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_prng.c
===================================================================
--- trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_prng.c (rev 0)
+++ trunk/Build/source/libs/luajit/LuaJIT-src/src/lj_prng.c 2025-02-02 17:39:28 UTC (rev 73695)
@@ -0,0 +1,263 @@
+/*
+** Pseudo-random number generation.
+** Copyright (C) 2005-2025 Mike Pall. See Copyright Notice in luajit.h
+*/
+
+#define lj_prng_c
+#define LUA_CORE
+
+/* To get the syscall prototype. */
+#if defined(__linux__) && !defined(_GNU_SOURCE)
+#define _GNU_SOURCE
+#endif
+
+#include "lj_def.h"
+#include "lj_arch.h"
+#include "lj_prng.h"
+
+/* -- PRNG step function -------------------------------------------------- */
+
+/* This implements a Tausworthe PRNG with period 2^223. Based on:
+** Tables of maximally-equidistributed combined LFSR generators,
+** Pierre L'Ecuyer, 1991, table 3, 1st entry.
+** Full-period ME-CF generator with L=64, J=4, k=223, N1=49.
+**
+** Important note: This PRNG is NOT suitable for cryptographic use!
+**
+** But it works fine for math.random(), which has an API that's not
+** suitable for cryptography, anyway.
+**
+** When used as a securely seeded global PRNG, it substantially raises
+** the difficulty for various attacks on the VM.
+*/
+
+/* Update generator i and compute a running xor of all states. */
+#define TW223_GEN(rs, z, r, i, k, q, s) \
+ z = rs->u[i]; \
+ z = (((z<<q)^z) >> (k-s)) ^ ((z&((uint64_t)(int64_t)-1 << (64-k)))<<s); \
+ r ^= z; rs->u[i] = z;
+
+#define TW223_STEP(rs, z, r) \
+ TW223_GEN(rs, z, r, 0, 63, 31, 18) \
+ TW223_GEN(rs, z, r, 1, 58, 19, 28) \
+ TW223_GEN(rs, z, r, 2, 55, 24, 7) \
+ TW223_GEN(rs, z, r, 3, 47, 21, 8)
+
+/* PRNG step function with uint64_t result. */
+LJ_NOINLINE uint64_t LJ_FASTCALL lj_prng_u64(PRNGState *rs)
+{
+ uint64_t z, r = 0;
+ TW223_STEP(rs, z, r)
+ return r;
+}
+
+/* PRNG step function with double in uint64_t result. */
+LJ_NOINLINE uint64_t LJ_FASTCALL lj_prng_u64d(PRNGState *rs)
+{
+ uint64_t z, r = 0;
+ TW223_STEP(rs, z, r)
+ /* Returns a double bit pattern in the range 1.0 <= d < 2.0. */
+ return (r & U64x(000fffff,ffffffff)) | U64x(3ff00000,00000000);
+}
+
+/* Condition seed: ensure k[i] MSB of u[i] are non-zero. */
+static LJ_AINLINE void lj_prng_condition(PRNGState *rs)
+{
+ if (rs->u[0] < (1u << 1)) rs->u[0] += (1u << 1);
+ if (rs->u[1] < (1u << 6)) rs->u[1] += (1u << 6);
+ if (rs->u[2] < (1u << 9)) rs->u[2] += (1u << 9);
+ if (rs->u[3] < (1u << 17)) rs->u[3] += (1u << 17);
+}
+
+/* -- PRNG seeding from OS ------------------------------------------------ */
+
+#if LUAJIT_SECURITY_PRNG == 0
+
+/* Nothing to define. */
+
+#elif LJ_TARGET_XBOX360
+
+extern int XNetRandom(void *buf, unsigned int len);
+
+#elif LJ_TARGET_PS3
+
+extern int sys_get_random_number(void *buf, uint64_t len);
+
+#elif LJ_TARGET_PS4 || LJ_TARGET_PS5 || LJ_TARGET_PSVITA
+
+extern int sceRandomGetRandomNumber(void *buf, size_t len);
+
+#elif LJ_TARGET_NX
+
+#include <unistd.h>
+
+#elif LJ_TARGET_WINDOWS || LJ_TARGET_XBOXONE
+
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+
+#if LJ_TARGET_UWP || LJ_TARGET_XBOXONE
+/* Must use BCryptGenRandom. */
+#include <bcrypt.h>
+#pragma comment(lib, "bcrypt.lib")
+#else
+/* If you wonder about this mess, then search online for RtlGenRandom. */
+typedef BOOLEAN (WINAPI *PRGR)(void *buf, ULONG len);
+static PRGR libfunc_rgr;
+#endif
+
+#elif LJ_TARGET_POSIX
+
+#if LJ_TARGET_LINUX
+/* Avoid a dependency on glibc 2.25+ and use the getrandom syscall instead. */
+#include <sys/syscall.h>
+#else
+
+#if LJ_TARGET_OSX && !LJ_TARGET_IOS
+/*
+** In their infinite wisdom Apple decided to disallow getentropy() in the
+** iOS App Store. Even though the call is common to all BSD-ish OS, it's
+** recommended by Apple in their own security-related docs, and, to top
+** off the foolery, /dev/urandom is handled by the same kernel code,
+** yet accessing it is actually permitted (but less efficient).
+*/
+#include <Availability.h>
+#if __MAC_OS_X_VERSION_MIN_REQUIRED >= 101200
+#define LJ_TARGET_HAS_GETENTROPY 1
+#endif
+#elif (LJ_TARGET_BSD && !defined(__NetBSD__)) || LJ_TARGET_SOLARIS || LJ_TARGET_CYGWIN || LJ_TARGET_QNX
+#define LJ_TARGET_HAS_GETENTROPY 1
+#endif
+
+#if LJ_TARGET_HAS_GETENTROPY
+extern int getentropy(void *buf, size_t len)
+#ifdef __ELF__
+ __attribute__((weak))
+#endif
+;
+#endif
+
+#endif
+
+/* For the /dev/urandom fallback. */
+#include <fcntl.h>
+#include <unistd.h>
+
+#endif
+
+#if LUAJIT_SECURITY_PRNG == 0
+
+/* If you really don't care about security, then define
+** LUAJIT_SECURITY_PRNG=0. This yields a predictable seed
+** and provides NO SECURITY against various attacks on the VM.
+**
+** BTW: This is NOT the way to get predictable table iteration,
+** predictable trace generation, predictable bytecode generation, etc.
+*/
@@ Diff output truncated at 1234567 characters. @@
More information about the tex-live-commits
mailing list.