| Commit message (Expand) | Author | Age | Files | Lines |
* | Add AVX SSE4.2 instructions | Bruno Cardoso Lopes | 2010-07-07 | 1 | -114/+179 |
* | Use only one multiclass to pinsrq instructions | Bruno Cardoso Lopes | 2010-07-07 | 2 | -38/+20 |
* | Now that almost all SSE4.1 AVX instructions are added, move code around to mo... | Bruno Cardoso Lopes | 2010-07-07 | 2 | -361/+374 |
* | Add AVX SSE4.1 insertps, ptest and movntdqa instructions | Bruno Cardoso Lopes | 2010-07-07 | 1 | -18/+39 |
* | Add AVX SSE4.1 extractps and pinsr instructions | Bruno Cardoso Lopes | 2010-07-07 | 1 | -35/+67 |
* | Revert "Remove references to INSERT_SUBREG after de-SSA" r107725. | Jakob Stoklund Olesen | 2010-07-07 | 7 | -9/+136 |
* | Also use REG_SEQUENCE for VTBX instructions. | Bob Wilson | 2010-07-07 | 2 | -24/+30 |
* | Mark eh.sjlj.set/longjmp custom lowerings as Darwin-only since that's where | Jim Grosbach | 2010-07-07 | 1 | -2/+4 |
* | Add AVX SSE4.1 Extract Integer instructions | Bruno Cardoso Lopes | 2010-07-07 | 1 | -0/+11 |
* | By default, the eh.sjlj.setjmp/longjmp intrinsics should just do nothing rather | Jim Grosbach | 2010-07-06 | 2 | -2/+9 |
* | Remove references to INSERT_SUBREG after de-SSA | Jakob Stoklund Olesen | 2010-07-06 | 7 | -136/+9 |
* | Use REG_SEQUENCE nodes to make the table registers for VTBL instructions be | Bob Wilson | 2010-07-06 | 2 | -10/+61 |
* | Accept RIP-relative symbols with 'i' constraint, and | Dale Johannesen | 2010-07-06 | 2 | -2/+3 |
* | Convert INSERT_SUBREG to COPY in TwoAddressInstructionPass. | Jakob Stoklund Olesen | 2010-07-06 | 3 | -2/+35 |
* | Track defs for all aliases in NEONMoveFix. | Jakob Stoklund Olesen | 2010-07-06 | 1 | -2/+2 |
* | Add the rest of AVX SSE4.1 packed move with sign/zero extend instructions | Bruno Cardoso Lopes | 2010-07-06 | 1 | -0/+17 |
* | Add part of AVX SSE4.1 packed move with sign/zero extend instructions | Bruno Cardoso Lopes | 2010-07-06 | 1 | -0/+15 |
* | Fix comment from previous patch | Bruno Cardoso Lopes | 2010-07-06 | 1 | -1/+1 |
* | Add AVX vblendvpd, vblendvps and vpblendvb instructions | Bruno Cardoso Lopes | 2010-07-06 | 4 | -10/+61 |
* | CanLowerReturn doesn't need a SelectionDAG; it just needs an LLVMContext. | Dan Gohman | 2010-07-06 | 6 | -11/+11 |
* | Propagate debug loc. | Devang Patel | 2010-07-06 | 19 | -50/+60 |
* | Represent NEON load/store alignments in bytes, not bits. | Bob Wilson | 2010-07-06 | 3 | -7/+13 |
* | One more case assuming that subregs have live ranges. | Jakob Stoklund Olesen | 2010-07-06 | 1 | -2/+2 |
* | Fix buildbot breakage where a def is missing. | Jakob Stoklund Olesen | 2010-07-06 | 1 | -0/+2 |
* | Add fixme. | Devang Patel | 2010-07-06 | 1 | -0/+1 |
* | Be more forgiving when calculating alias interference for physreg coalescing. | Jakob Stoklund Olesen | 2010-07-06 | 3 | -105/+59 |
* | Reapply r107655 with fixes; insert the pseudo instruction into | Dan Gohman | 2010-07-06 | 14 | -212/+275 |
* | Fix to 80-col. | Eric Christopher | 2010-07-06 | 1 | -21/+21 |
* | Fix PR7545 crash. | Devang Patel | 2010-07-06 | 1 | -3/+3 |
* | Don't create neon moves in CopyRegToReg. NEONMoveFixPass will do the conversion | Rafael Espindola | 2010-07-06 | 1 | -4/+6 |
* | tighten up this code. | Chris Lattner | 2010-07-06 | 1 | -12/+7 |
* | Revert r107655. | Dan Gohman | 2010-07-06 | 13 | -269/+209 |
* | Add versions of OutputArgReg, AnalyzeReturn, and AnalyzeCallOperands | Dan Gohman | 2010-07-06 | 1 | -0/+35 |
* | Make getMinimalPhysRegClass' comment mention what makes it different | Dan Gohman | 2010-07-06 | 1 | -1/+2 |
* | Fix a major regression on COFF targets introduced by r103267: 'discardable' s... | Anton Korobeynikov | 2010-07-06 | 1 | -1/+1 |
* | Add some more TODO comments. | Dan Gohman | 2010-07-06 | 1 | -0/+6 |
* | Add a comment. | Dan Gohman | 2010-07-06 | 1 | -1/+2 |
* | Fix a bunch of custom-inserter functions to handle the case where | Dan Gohman | 2010-07-06 | 13 | -209/+269 |
* | Fix up -fstack-protector on linux to use the segment | Eric Christopher | 2010-07-06 | 3 | -2/+42 |
* | Detabify this file. | Nick Lewycky | 2010-07-06 | 1 | -7/+7 |
* | Have the X86 backend use Triple instead of a string and some enums. | Eric Christopher | 2010-07-05 | 3 | -64/+38 |
* | Remove some unused/redundant code. | Kalle Raiskila | 2010-07-05 | 2 | -20/+0 |
* | more tidying. | Chris Lattner | 2010-07-05 | 2 | -10/+5 |
* | some notes about suboptimal insertps's | Chris Lattner | 2010-07-05 | 1 | -0/+31 |
* | random tidying | Chris Lattner | 2010-07-05 | 2 | -41/+43 |
* | rip out even more sporadic v2f32 support. | Chris Lattner | 2010-07-05 | 3 | -19/+1 |
* | rip out the various v2f32 "mmx" handling logic, now that | Chris Lattner | 2010-07-05 | 1 | -6/+6 |
* | Print symbolic subreg indices on REG_SEQUENCE and INSERT_SUBREG. | Jakob Stoklund Olesen | 2010-07-04 | 1 | -0/+2 |
* | Just rip v2f32 support completely out of the X86 backend. In | Chris Lattner | 2010-07-04 | 1 | -23/+4 |
* | fix PR7518 - terrible codegen of <2 x float>, by only marking | Chris Lattner | 2010-07-04 | 1 | -6/+16 |