 florian
|
b776d48367
+ x86: optimziation MovMov2MovMov 2
|
5 years ago |
|
florian
|
84a1f614d9
* fix several double ;
|
5 years ago |
|
florian
|
afb68dbcf3
* factored out OptPass1VPXor
|
5 years ago |
|
florian
|
0bce44e342
* compilation on i386 fixed
|
5 years ago |
|
florian
|
809fb83334
* adapt optimizer to handle push/pop for stack alignment
|
5 years ago |
|
florian
|
9b54588d75
+ PXorPXor2PXor optimization
|
5 years ago |
|
florian
|
b239365479
* improved LeaLea2Lea
|
5 years ago |
|
florian
|
05091cf0dd
+ MovAndTest2Test optimization
|
5 years ago |
 yury
|
c1d124d497
* x86: Fixed peephole optimization introduced in r44233. tb0219 and tb0548 tests have failed with -O2 since then.
|
5 years ago |
|
florian
|
a084c8829f
* patch by J. Gareth Moreton: refactor RemoveCurrentP
|
5 years ago |
|
florian
|
951f02d188
* correctly set register size
|
5 years ago |
|
florian
|
670e778643
+ assembler optimization MovxMov2Movx
|
5 years ago |
|
florian
|
6cac8cc77b
* fix Peephole Optimization: Cmpcc2Testcc - condition AE/NB/NC/NO --> Always optimization
|
5 years ago |
|
florian
|
8f0376c888
+ JccAdd2SetccAdd optimization
|
5 years ago |
|
florian
|
cea7723080
* more use of RemoveCurrentP(p)
|
5 years ago |
|
florian
|
f42b0bf6f5
* fixed JccAdd/Inc/Dec2*Adc/Sbb optimizations
|
5 years ago |
|
florian
|
34be9e4643
* CallRet2* optimization may not misalign the stack, reported by C Western on fpc-devel
|
5 years ago |
|
florian
|
9bdfbc2473
+ GetNextInstructionUsingRegTrackingUse
|
5 years ago |
|
florian
|
a357359971
* LeaCallLeaRet2Jmp is done in TX86AsmOptimizer.PostPeepholeOptLea
|
5 years ago |
|
florian
|
2acc6337d8
* debug statement removed
|
5 years ago |
|
florian
|
4dff373fef
* (modified) patch by J. Gareth Moreton: processor-aware MOVZX optimisation cleanup, resolves #36687
|
5 years ago |
|
florian
|
49f789b92a
* reverted accidently committed changes again
|
5 years ago |
|
florian
|
95e8992e2f
+ AArch64: profiler support for aarch64-linux
|
5 years ago |
|
florian
|
3099e10f41
* revert accidently committed code
|
5 years ago |
|
florian
|
e50c4f6373
+ AAarch64: optimize divisions by constant
|
5 years ago |
|
florian
|
88836b467e
* (V)MOVA*(V)MOVS*2(V)MOVS* 1 optimization
|
5 years ago |
|
florian
|
42324235bd
+ CallRet2Call optimization
|
5 years ago |
|
florian
|
3279e63351
* patch by J. Gareth Moreton: cleanup TX86AsmOptimizer.OptPass2MOV after DeepMov changes, second part of #36670
|
5 years ago |
|
florian
|
3a5eb0ec59
* patch by J. Gareth Moreton: cleanup TX86AsmOptimizer.OptPass1MOV after DeepMov changes, first part of #36670
|
5 years ago |
|
florian
|
f632e2e7be
* simplified TX86AsmOptimizer.OptPass1Movx
|
5 years ago |
