 florian
|
ea07d557fb
* check properly for the supported cpu type, resolves #40298
|
2 years ago |
 J. Gareth "Curious Kit" Moreton
|
7b2da347fd
* Fixed bug in "JccMovJmpMov2CMovCMov" optimisation where it didn't track registers in references getting changed
|
2 years ago |
|
florian
|
1d8c4cfac5
* another patch to fix #40223
|
2 years ago |
 Pierre Muller
|
5f4551446b
Do not always set DEBUG_AOPTCPU
|
2 years ago |
|
florian
|
0e4e498b3b
* do not replace index registers with esp/rsp, resolves #40223
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
9f8df68426
* x86: TEST/Jcc/TEST optimisations can now look beyond the next instruction under -O3
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
55f8be3ec6
* x86: RegModifiedByInstruction and RegInInstruction
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
060ae935f5
* x86: Fixed oversight in RegModifiedByInstruction not checking W0, W1 and WU-type flags
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
2d5ef5b5d0
* x86: Fix to MOV/CMP register deallocation positioning
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
2217591da4
* x86: The LEA/LEA optimisations can now work with a different
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
7122bdcd24
* x86: LeaLea2Lea now handles cases where the second LEA instruction
|
2 years ago |
|
florian
|
2cd126d4e8
+ added debug messages
|
2 years ago |
|
florian
|
995a8ae912
* fixed comment
|
2 years ago |
|
florian
|
30eca29f18
+ x86: AND/CMP -> CMP optimiziation
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
e09c2061d1
* x86: Some refactoring to use aoc_ForceNewIteration instead of manually advancing p
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
d0f5b6c302
x86: Some oversights fixed where another iteration of pass 1 wasn't performed when it should have been
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
b4d894529c
* x86: OptPass1_V_MOVAP now uses GetNextInstructionUsingReg and UpdateUsedRegsBetween
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
f5ba364c4f
* x86: PostPeepholeOptCmp and PostPeepholeOptTestOr
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
64223dc794
* x86: JccMovJmpMov2CMovCMov will no longer move
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
1a556f9543
* i386: Bug fix where EBP was marked as preserved when it wasn't, causing faulty optimisations
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
bab0daddb2
* x86: Fixed bad register tracking in OptPass2JMP
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
49c713133b
* x86: Converted bit-test optimisations to use
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
5e418b090b
* x86: New debug_hexstr instruction to help produce concise debug comments
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
082101b975
* x86-64: Extended the movl/movq optimisation to cover more distance
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
b82e066ab2
* x86-64: movl/movq -> movl/movl zero-extension optimisation
|
2 years ago |
|
florian
|
c0e0e06fac
* undid accidentely activated define
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
7fe962b731
* x86: CMOV optimisations now check optimizecputype rather than cputype
|
2 years ago |
|
J. Gareth "Curious Kit" Moreton
|
a1d2ad08c8
* x86: Extension to CanBeCMOV that permits a potentially unsafe reference if it appears in the previous comparison
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
efb0b75454
* x86: New CMP optimisation that mimics some of the CMOV optimisations and removes unnecessary conditions.
|
3 years ago |
|
J. Gareth "Curious Kit" Moreton
|
1d7e23ed6a
* x86: Refactored CMOV optimisations and permitted the use of constants in some cases.
|
4 years ago |
