| Commit message (Expand) | Author | Age | Files | Lines |
* | Fix PR7748 without using microsoft extensions | Bruno Cardoso Lopes | 2010-08-26 | 2 | -13/+13 |
* | we should pattern match the SSE complex arithmetic ops. | Chris Lattner | 2010-08-25 | 1 | -0/+26 |
* | Revert this for now, PUNPCKLDQ dont operate on v4f32 | Bruno Cardoso Lopes | 2010-08-25 | 1 | -1/+1 |
* | X86: Fix misencode of RI64mi8. This fixes OpenSSL / x86_64-apple-darwin10 / c... | Daniel Dunbar | 2010-08-25 | 1 | -4/+3 |
* | Remove dead recursive function. Yay for clang -Wunused-function. | Benjamin Kramer | 2010-08-25 | 1 | -7/+0 |
* | Fix nasty mingw32 bug, which e.g. prevented llvm-gcc bootstrap there. | Anton Korobeynikov | 2010-08-25 | 3 | -6/+10 |
* | PUNPCKLDQ should also be used for v4f32 | Bruno Cardoso Lopes | 2010-08-25 | 1 | -1/+1 |
* | teach lowering to get target specific nodes for pshufd, emulating the same is... | Bruno Cardoso Lopes | 2010-08-25 | 1 | -5/+51 |
* | MC/X86: Tweak imul recognition, previous hack only applies for the imul form | Daniel Dunbar | 2010-08-24 | 1 | -1/+3 |
* | MC/X86: Add custom hack for recognizing "imul $12, %eax" and friends. | Daniel Dunbar | 2010-08-24 | 1 | -0/+8 |
* | MC/X86: Warn on scale factors > 1 without index register, instead of erroring, | Daniel Dunbar | 2010-08-24 | 1 | -3/+4 |
* | Fix X86's isLegalAddressingMode to recognize that static addresses | Dan Gohman | 2010-08-24 | 1 | -1/+3 |
* | Use pshufhw and pshuflw in more cases and fix getTargetShuffleNode number of ... | Bruno Cardoso Lopes | 2010-08-24 | 2 | -2/+18 |
* | Start using target speficic nodes for shuffles: pshufhw and pshuflw | Bruno Cardoso Lopes | 2010-08-23 | 1 | -1/+19 |
* | tyops | Gabor Greif | 2010-08-23 | 1 | -1/+1 |
* | Add a new llvm.x86.int intrinsic, allowing access to the | Chris Lattner | 2010-08-23 | 1 | -3/+7 |
* | random improvement for variable shift codegen. | Chris Lattner | 2010-08-23 | 1 | -2/+14 |
* | Revert invalid r111792. Jump tables are not broken on x86-64 / coff, | Anton Korobeynikov | 2010-08-23 | 1 | -6/+0 |
* | Workaround broken jump tables on x86-64 COFF. | Michael J. Spencer | 2010-08-23 | 1 | -0/+6 |
* | Use rip-rel addressing on win64 by default. For this we just | Anton Korobeynikov | 2010-08-21 | 2 | -15/+19 |
* | MC: Add partial x86-64 support to COFF. | Michael J. Spencer | 2010-08-21 | 2 | -4/+13 |
* | Fix x86 fast-isel's cmp+branch folding to avoid folding when the | Dan Gohman | 2010-08-21 | 1 | -2/+4 |
* | Prepare LowerVECTOR_SHUFFLEv8i16 to use x86 target specific nodes directly | Bruno Cardoso Lopes | 2010-08-21 | 2 | -7/+12 |
* | This is the first step towards refactoring the x86 vector shuffle code. The | Bruno Cardoso Lopes | 2010-08-20 | 4 | -0/+493 |
* | fix PR7465, mishandling of lcall and ljmp: intersegment long | Chris Lattner | 2010-08-19 | 4 | -12/+28 |
* | minor progress towards fixing PR7465 | Chris Lattner | 2010-08-19 | 1 | -2/+2 |
* | Marked with ATTRIBUTE_USED so that clang doesn't complain. | Bill Wendling | 2010-08-18 | 1 | -1/+1 |
* | remove some code that is dead now that lea's are modeled with segment registers. | Chris Lattner | 2010-08-18 | 1 | -14/+0 |
* | Revert part of one of the prev. patches - tailjmp will follow later. | Anton Korobeynikov | 2010-08-17 | 1 | -1/+0 |
* | More fixes for win64: | Anton Korobeynikov | 2010-08-17 | 3 | -6/+10 |
* | Enable more win64 calls folding opportunities. | Anton Korobeynikov | 2010-08-17 | 1 | -0/+2 |
* | Comment out some broken/unused/useless instructions which mess up disassembly. | Eli Friedman | 2010-08-16 | 1 | -0/+6 |
* | Don't attempt to SimplifyShortMoveForm in 64-bit mode. | Eli Friedman | 2010-08-16 | 1 | -9/+13 |
* | Hookup ELF support for X86. | Matt Fleming | 2010-08-16 | 2 | -6/+14 |
* | Partially revert r111155. It looks like MSVC is calling an operator<() that | Jakob Stoklund Olesen | 2010-08-16 | 1 | -0/+3 |
* | Remove unused functions. | Jakob Stoklund Olesen | 2010-08-16 | 1 | -8/+0 |
* | Revert r111082. No warnings for this common pattern. | Argyrios Kyrtzidis | 2010-08-15 | 1 | -3/+2 |
* | Rework how the non-sse2 memory barrier is lowered so that the | Eric Christopher | 2010-08-14 | 2 | -11/+22 |
* | Add ATTRIBUTE_UNUSED to methods that are not supposed to be used. | Argyrios Kyrtzidis | 2010-08-14 | 1 | -2/+3 |
* | improve indentation | Chris Lattner | 2010-08-14 | 1 | -19/+18 |
* | Add comments to some pattern fragments in x86 | Bruno Cardoso Lopes | 2010-08-13 | 1 | -4/+10 |
* | Revert 110491. While not wrong, it was based on a | Dale Johannesen | 2010-08-13 | 1 | -13/+8 |
* | Fix comment to reflect code, and remove an unused argument | Bruno Cardoso Lopes | 2010-08-13 | 1 | -4/+3 |
* | Improve comment to make explicit why not to touch this could before JIT goes MC | Bruno Cardoso Lopes | 2010-08-13 | 1 | -1/+6 |
* | Revert last patch and r110954 as I meant to. | Eric Christopher | 2010-08-13 | 2 | -24/+24 |
* | Revert r110954 for now, pseudo instructions can't make it through to the JIT. | Eric Christopher | 2010-08-13 | 1 | -0/+1 |
* | Some small clean-up: use of pseudo instructions | Bruno Cardoso Lopes | 2010-08-12 | 1 | -24/+23 |
* | - Teach SSEDomainFix to switch between different levels of AVX instructions. ... | Bruno Cardoso Lopes | 2010-08-12 | 2 | -27/+89 |
* | Define AVX 128-bit pattern versions of SET0PS/PD. | Bruno Cardoso Lopes | 2010-08-12 | 2 | -6/+18 |
* | Fix comment order | Bruno Cardoso Lopes | 2010-08-12 | 1 | -1/+1 |