| Commit message (Expand) | Author | Age | Files | Lines |
* | + some more change information entries added | florian | 2021-03-31 | 1 | -2/+2 |
* | * prefetch instructions read only the operand | florian | 2021-03-30 | 1 | -6/+6 |
* | + tx64tryfinallynode.dogetcopy properly implemented, resolves #37305 | florian | 2021-03-14 | 1 | -10/+46 |
* | * x86: some fixes to enable 8 and 16 bit operations | florian | 2021-01-16 | 1 | -1/+2 |
* | * simplify tx8664shlshrnode.pass_generate_code, it should not mess with the... | florian | 2021-01-16 | 1 | -18/+2 |
* | + support R_X86_64_GOTPCRELX and R_X86_64_REX_GOTPCRELX | florian | 2021-01-14 | 1 | -0/+6 |
* | * do not extend 8 bit return values to 32 bit when loading them into eax, | florian | 2021-01-11 | 1 | -3/+6 |
* | + AddLea2Lea optimization | florian | 2021-01-08 | 1 | -0/+2 |
* | + patch by J. Gareth Moreton: Advanced MOVZX optimisations, resolves #38294 | florian | 2021-01-05 | 1 | -0/+4 |
* | * Ch_ information for vreduce* fixed | florian | 2020-12-23 | 1 | -4/+4 |
* | * implemented UseAVX512 properly | florian | 2020-12-23 | 1 | -2/+4 |
* | + patch by J. Gareth Moreton: some new x86 assembler optimizations, resolve... | florian | 2020-12-20 | 1 | -0/+4 |
* | * improved change information | florian | 2020-12-15 | 1 | -6/+6 |
* | * x86: apply VOpVmov2VOp optimization to vmins*/vmaxs* | florian | 2020-12-13 | 1 | -0/+4 |
* | * x86 change information updated | florian | 2020-12-06 | 1 | -18/+18 |
* | * allow an AT&T suffix for movabs | florian | 2020-11-30 | 1 | -1/+1 |
* | -- Zusammenführen von r47033 bis r47401 in ».«: | florian | 2020-11-12 | 4 | -91/+245 |
* | * use VPXORD in avx-512 mode as VPXOR does not work for the upper 16 registers | florian | 2020-11-09 | 1 | -0/+2 |
* | + completed Ch_* information for VMOVD* | florian | 2020-11-07 | 1 | -6/+6 |
* | * patch by Marģers to unify internal error numbers, resolves #37888 | florian | 2020-10-13 | 4 | -7/+7 |
* | * proper change information for avx-512 vcvt* instructions | florian | 2020-10-10 | 1 | -24/+24 |
* | + make use of avx-512 instructions vcvtuis2s* if possible | florian | 2020-10-10 | 2 | -5/+7 |
* | * mm registers 16-31 are also volatile during a call | florian | 2020-10-10 | 1 | -2/+2 |
* | * load 0.0 by (V)XORPS/D instead of (V)PXOR in mm registers | florian | 2020-10-10 | 1 | -3/+5 |
* | * fix #35841 also for break and continue | florian | 2020-10-09 | 1 | -6/+6 |
* | * patch by J. Gareth Moreton: propagate exit use in nested try...except sta... | florian | 2020-10-08 | 1 | -0/+18 |
* | + support all XSAVE instructions, resolves #37864 | florian | 2020-10-03 | 7 | -1/+85 |
* | * x86: ImulMov2Imul optimization | florian | 2020-09-25 | 1 | -0/+4 |
* | * do not throw an internal error later on in case of a recursive call, if t... | florian | 2020-07-31 | 1 | -1/+6 |
* | * mark all external assemblers using an LLVM tool using af_llvm | jonas | 2020-07-19 | 1 | -0/+45 |
* | * fix several double ; | florian | 2020-05-24 | 1 | -1/+1 |
* | * factored out OptPass1VPXor | florian | 2020-05-24 | 1 | -1/+2 |
* | + x86: track use of full ymm registers and generate vzeroupper if needed | florian | 2020-05-24 | 1 | -0/+4 |
* | * adapt optimizer to handle push/pop for stack alignment | florian | 2020-05-23 | 1 | -0/+2 |
* | * x86-64: adjust stack by push/pop if possible | florian | 2020-05-23 | 1 | -3/+8 |
* | + PXorPXor2PXor optimization | florian | 2020-05-18 | 1 | -0/+3 |
* | + introduced labelmaxlen in tasminfo and added code in ReplaceForbiddenAsmSym... | nickysn | 2020-04-25 | 1 | -0/+1 |
* | + implement compiler support for SEH on Win64 | svenbarth | 2020-04-21 | 1 | -0/+2 |
* | * split i/t_darwin from i/t_bsd, as they don't have that much in common | jonas | 2020-03-29 | 1 | -0/+3 |
* | * there is no exit stackframe needed if a subroutine never returns | florian | 2020-02-16 | 1 | -1/+8 |
* | * patch by J. Gareth Moreton: optimize MOVSXD as well, resolves #36700 | florian | 2020-02-15 | 1 | -0/+1 |
* | * patch by J. Gareth Moreton: x86 SUB and LEA optimisations, resolves #36622 | florian | 2020-01-24 | 1 | -0/+2 |
* | * patch by J. Gareth Moreton: get rid of another pass through the assembler... | florian | 2020-01-19 | 1 | -7/+6 |
* | * patch by J. Gareth Moreton: EAX -> EDX:EAX sign extension shortcuts, and ... | florian | 2020-01-12 | 1 | -0/+2 |
* | * added is_normal_fieldvarsym() helper and use it | jonas | 2019-12-24 | 1 | -2/+2 |
* | * fix paraloc def for vectors with vectorcall calling convention (was a single | jonas | 2019-12-24 | 1 | -0/+2 |
* | * renamed getintparaloc to getcgtempparaloc | jonas | 2019-12-24 | 1 | -2/+2 |
* | * common naming for fpu_none string | florian | 2019-12-24 | 1 | -1/+2 |
* | + MULX instruction | florian | 2019-12-21 | 1 | -1/+1 |
* | * factored out TX86AsmOptimizer.OptPass1Cmp | florian | 2019-11-24 | 1 | -0/+2 |