 J. Gareth "Curious Kit" Moreton
|
b58fdc3e58
Improved ADD and SUB optimisations for LEA instructions
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
fd28cc0db0
Better handling of zeroing upper parts of registers
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
674ed4069a
Expanded MM block move to include YMM registers under AVX
|
3 years ago |
 florian
|
2c180cf101
* by default, DEBUG_AOPTCPU is only enabled if the compiler is compiled with -dEXTDEBUG
|
3 years ago |
|
florian
|
ec40db3da7
+ (V)Cvtss2CvtSd(V)Cvtsd2ss2Nop optimization, resolves #39360
|
3 years ago |
|
florian
|
c67d35b8cc
* patch by J. Gareth Moreton: x86: Minor bitwise optimisations, resolves #39299
|
4 years ago |
|
florian
|
7d6b01bfc4
+ SHXXMov2SHXX optimization
|
4 years ago |
|
florian
|
bae583118a
* patch by J. Gareth Moreton: x86 JccMovJmpMov2MovSetcc improvement, resolves #38761
|
4 years ago |
|
florian
|
5726428dcc
* patch by J. Gareth Moreton: Additional SETcc optimisations, resolves #38767
|
4 years ago |
|
florian
|
578424cf51
* patch by J. Gareth Moreton: TEST chain shortcutting, resolves #38908
|
4 years ago |
|
florian
|
4e9b42344e
+ AddLea2Lea optimization
|
4 years ago |
|
florian
|
28efcfba65
+ patch by J. Gareth Moreton: Advanced MOVZX optimisations, resolves #38294
|
4 years ago |
|
florian
|
2a990b8167
+ patch by J. Gareth Moreton: some new x86 assembler optimizations, resolves #38130
|
4 years ago |
|
florian
|
78244d89dc
* use VPXORD in avx-512 mode as VPXOR does not work for the upper 16 registers
|
4 years ago |
|
florian
|
c2684c0dc3
* use V* instructions if possible
|
4 years ago |
|
florian
|
8e62bf649e
* load 0.0 by (V)XORPS/D instead of (V)PXOR in mm registers
|
4 years ago |
|
florian
|
1fb7603f61
* x86: ImulMov2Imul optimization
|
4 years ago |
|
florian
|
afb68dbcf3
* factored out OptPass1VPXor
|
5 years ago |
|
florian
|
9b54588d75
+ PXorPXor2PXor optimization
|
5 years ago |
 yury
|
1b3a3a7983
* Removed lot of unused local vars. It is useful to turn on the notes in options. :)
|
5 years ago |
|
florian
|
993144b91b
* patch by J. Gareth Moreton: x86 SUB and LEA optimisations, resolves #36622
|
5 years ago |
|
florian
|
16152cf948
* patch by J. Gareth Moreton: get rid of another pass through the assembler list by integrating OptReferences
|
5 years ago |
|
florian
|
124c64152d
* patch by J. Gareth Moreton: EAX -> EDX:EAX sign extension shortcuts, and MOVSX shortcuts for AX register, part 2 of #36551
|
5 years ago |
 svenbarth
|
0149d26adb
* in PeepHoleOptPass1Cpu when an instruction contains a segment, don't replace the current instruction with the next, cause then we get an infinite loop when compiling on Win32 with -dTEST_WIN32_SEH -O4
|
5 years ago |
|
florian
|
9e307f5c3a
- removed code not used for years
|
5 years ago |
|
florian
|
d444f750f9
* factored out TX86AsmOptimizer.OptPass1Cmp
|
5 years ago |
|
florian
|
dea1855126
o patch by J. Gareth Moreton, resolves #36352:
|
5 years ago |
 pierre
|
c882828fd0
Minimalist change to avoid Invalid Typecast error on class when compiled with '-CriotR -O2' for i386-go32v2 target
|
5 years ago |
|
florian
|
87b3b089d6
* i386: more clean up of TCPUAsmOPtimizer.PeepHoleOptPass1Cpu
|
5 years ago |
|
florian
|
632f13c47a
* i386 does not use anymore its own assembler optimizer loops
|
5 years ago |
