| 2010-10-04 |
Bruno Cardoso Lopes | AVX intrinsics and builtins were also added to clang
|
commit | commitdiff | tree |
| 2010-09-09 |
Bruno Cardoso Lopes | Add one more pattern to fallback movddup
|
commit | commitdiff | tree |
| 2010-09-08 |
Bruno Cardoso Lopes | Minor change. Fix comments and remove unused and redundant...
|
commit | commitdiff | tree |
| 2010-09-08 |
Bruno Cardoso Lopes | x86 vector shuffle lowering now relies only on target...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | Factor out some x86 vector shuffle rewriting and add...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | Move code around to prepare for moving some of the...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | decouple MMX check from regular splat checks. Some...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | Remove now useless check, because the code can be matched...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | Minor change. Since the checks are equivalent, use...
|
commit | commitdiff | tree |
| 2010-09-07 |
Bruno Cardoso Lopes | Remove unused target specific node
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Remove the last bit of isShuffleMaskLegal checks and...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | make explicit that we not handle several mmx shuffles
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Emit target specific nodes to handle palignr. Do not...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Emit target specific nodes to handle splats starting...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Emit target specific nodes for isPSHUFHWMask and isPSHUFLWMask
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Emit target specific nodes for isSHUFPMask
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Previous isMOVLMask matching already emits targets...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | One more check from the original isShuffleMaskLegal...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Remove a duplicated but useless check that i've inserted...
|
commit | commitdiff | tree |
| 2010-09-04 |
Bruno Cardoso Lopes | Refactor some code and remove the extra checks for...
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Remove check for unpckh mask
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Remove check for unpckl mask
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Inline isShuffleMaskLegal into LowerVECTOR_SHUFFLE...
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Reapply considered harmfull part of rr112934 and r112942.
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Reapply last harmless part of r112934, the pattern...
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Reintroduce a simple function refactoring done in r112934...
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Reapply piecies of r112942 and r112934 which don't do
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Reapply Fix comment
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | AVX doesn't support mm operations neither its instrinsics.
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Use punpckh and unpckh family of nodes instead of using...
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | Fix comment
|
commit | commitdiff | tree |
| 2010-09-03 |
Bruno Cardoso Lopes | - Use specific nodes to match unpckl masks.
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Move insertps mask decoding to header file
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Move decoding of insertps back to avoid unused warnings...
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Move x86 specific shuffle mask decoding to its own...
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Replace unpckl_undef and unpckh_undef matching with...
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Move condition out to prepare for more matching
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | Remove checking for isUNPCKL_v_undef_Mask, the specific...
|
commit | commitdiff | tree |
| 2010-09-02 |
Bruno Cardoso Lopes | become more strict about when it's safe to use X86ISD...
|
commit | commitdiff | tree |
| 2010-09-01 |
Bruno Cardoso Lopes | Revert r112689, avoid those kind of checks cause they...
|
commit | commitdiff | tree |
| 2010-09-01 |
Bruno Cardoso Lopes | Using target specific nodes for shuffle nodes makes...
|
commit | commitdiff | tree |
| 2010-09-01 |
Bruno Cardoso Lopes | Use movlps, movlpd, movss and movsd specific nodes...
|
commit | commitdiff | tree |
| 2010-09-01 |
Bruno Cardoso Lopes | minor change, simplify some logic
|
commit | commitdiff | tree |
| 2010-09-01 |
Bruno Cardoso Lopes | Move some functions around so they can be used for...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | Use x86 specific MOVSLDUP node, add more patterns to...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | Use x86 specific MOVSHDUP node and add more patterns...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | Use MOVHLPS node instead of matching using movhlps...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | Use MOVLHPS and MOVHLPS x86 nodes whenever possible...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | zap unused method. x86 is the only user and already...
|
commit | commitdiff | tree |
| 2010-08-31 |
Bruno Cardoso Lopes | Use X86ISD::MOVSS and MOVSD to represent the movl mask...
|
commit | commitdiff | tree |
| 2010-08-28 |
Bruno Cardoso Lopes | Clean up the logic of vector shuffles -> vector shifts.
|
commit | commitdiff | tree |
| 2010-08-26 |
Bruno Cardoso Lopes | zap the now unused MVT::getIntVectorWithNumElements
|
commit | commitdiff | tree |
| 2010-08-26 |
Bruno Cardoso Lopes | Fix PR7748 without using microsoft extensions
|
commit | commitdiff | tree |
| 2010-08-25 |
Bruno Cardoso Lopes | Revert this for now, PUNPCKLDQ dont operate on v4f32
|
commit | commitdiff | tree |
| 2010-08-25 |
Bruno Cardoso Lopes | PUNPCKLDQ should also be used for v4f32
|
commit | commitdiff | tree |
| 2010-08-25 |
Bruno Cardoso Lopes | teach lowering to get target specific nodes for pshufd...
|
commit | commitdiff | tree |
| 2010-08-25 |
Bruno Cardoso Lopes | Convert test to use filecheck and make it more specific
|
commit | commitdiff | tree |
| 2010-08-24 |
Bruno Cardoso Lopes | Use pshufhw and pshuflw in more cases and fix getTargetShuff...
|
commit | commitdiff | tree |
| 2010-08-23 |
Bruno Cardoso Lopes | Start using target speficic nodes for shuffles: pshufhw...
|
commit | commitdiff | tree |
| 2010-08-21 |
Bruno Cardoso Lopes | Prepare LowerVECTOR_SHUFFLEv8i16 to use x86 target...
|
commit | commitdiff | tree |
| 2010-08-20 |
Bruno Cardoso Lopes | This is the first step towards refactoring the x86...
|
commit | commitdiff | tree |
| 2010-08-13 |
Bruno Cardoso Lopes | Add comments to some pattern fragments in x86
|
commit | commitdiff | tree |
| 2010-08-13 |
Bruno Cardoso Lopes | Fix comment to reflect code, and remove an unused argument
|
commit | commitdiff | tree |
| 2010-08-13 |
Bruno Cardoso Lopes | Improve comment to make explicit why not to touch this...
|
commit | commitdiff | tree |
| 2010-08-12 |
Bruno Cardoso Lopes | Some small clean-up: use of pseudo instructions
|
commit | commitdiff | tree |
| 2010-08-12 |
Bruno Cardoso Lopes | - Teach SSEDomainFix to switch between different levels...
|
commit | commitdiff | tree |
| 2010-08-12 |
Bruno Cardoso Lopes | Define AVX 128-bit pattern versions of SET0PS/PD.
|
commit | commitdiff | tree |
| 2010-08-12 |
Bruno Cardoso Lopes | Fix comment order
|
commit | commitdiff | tree |
| 2010-08-12 |
Bruno Cardoso Lopes | Begin to support some vector operations for AVX 256...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Add testcases for all AVX 256-bit intrinsics added...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Remove rsqrt/sqrt_nr intrinsics since there are no...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Reapply r109881 using a more strict command line for...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Remove AVX 256-bit cast intrinsics now that clang is...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Remove AVX 256-bit unpack and interleave intrinsics...
|
commit | commitdiff | tree |
| 2010-08-11 |
Bruno Cardoso Lopes | Remove AVX 256-bit shuffle intrinsics now that clang...
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Add AVX matching patterns to Packed Bit Test intrinsics.
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Add AVX movnt{pd,ps,dq} 256-bit intrinsics
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Add AVX movmsk 256-bit intrinsics
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Remove replicate intrinsics, clang will generate shufflevect...
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Support AVX 256-bit load and store intrinsics
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Use i32 instead of i8 for dot product intrinsic
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Patterns to match AVX cmp instructions
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Add matching patterns for vblend AVX intrinsics
|
commit | commitdiff | tree |
| 2010-08-10 |
Bruno Cardoso Lopes | Fix the last argument type of AVX vblend intrinsics
|
commit | commitdiff | tree |
| 2010-08-09 |
Bruno Cardoso Lopes | Add VCVTPD2PS, VCVTPS2DQ, VCVTPS2PDY, VCVTTPD2DQY,...
|
commit | commitdiff | tree |
| 2010-08-09 |
Bruno Cardoso Lopes | Add patterns to AVX conversions instructions. Do that...
|
commit | commitdiff | tree |
| 2010-08-09 |
Bruno Cardoso Lopes | Memory version of vcvtdq2pd intrinsic
|
commit | commitdiff | tree |
| 2010-08-09 |
Bruno Cardoso Lopes | Patterns to match vinsert, vbroadcast, vmovmask and...
|
commit | commitdiff | tree |
| 2010-08-06 |
Bruno Cardoso Lopes | Patterns to match AVX 256-bit vzero intrinsics
|
commit | commitdiff | tree |
| 2010-08-06 |
Bruno Cardoso Lopes | Patterns to match AVX 256-bit permutation intrinsics
|
commit | commitdiff | tree |
| 2010-08-06 |
Bruno Cardoso Lopes | Patterns to match AVX 256-bit horizontal arithmetic...
|
commit | commitdiff | tree |
| 2010-08-06 |
Bruno Cardoso Lopes | Patterns to match AVX 256-bit arithmetic intrinsics
|
commit | commitdiff | tree |
| 2010-08-06 |
Bruno Cardoso Lopes | Remove unused AVX intrinsics
|
commit | commitdiff | tree |
| 2010-08-05 |
Bruno Cardoso Lopes | Support very basic (doesn't include ABI support in...
|
commit | commitdiff | tree |
| 2010-08-04 |
Bruno Cardoso Lopes | Fix a comment typo and add more 256-bit intrinsics
|
commit | commitdiff | tree |
| 2010-08-03 |
Bruno Cardoso Lopes | Support x86 AVX 256-bit instruction intrinsics. Right...
|
commit | commitdiff | tree |
| 2010-07-30 |
Bruno Cardoso Lopes | A *bunch* of tests for AVX intrinsics
|
commit | commitdiff | tree |
| 2010-07-30 |
Bruno Cardoso Lopes | Support all 128-bit AVX vector intrinsics. Most part...
|
commit | commitdiff | tree |
| 2010-07-30 |
Bruno Cardoso Lopes | Fix typo!
|
commit | commitdiff | tree |
| 2010-07-26 |
Bruno Cardoso Lopes | Temporary hack to let codegen assert or generate poor...
|
commit | commitdiff | tree |
| next |
projects / oota-llvm.git / search