| Commit message (Expand) | Author | Age |
* | PR9848: pandn is not commutative. | Eli Friedman | 2011-05-05 |
* | The pshufw instruction came about in MMX2 when SSE was introduced. Don't place | Bill Wendling | 2010-10-04 |
* | the immediate field of pshufw is actually an 8-bit field, not a 8-bit field t... | Chris Lattner | 2010-10-03 |
* | add support for the prefetch/prefetchw instructions, move femms into | Chris Lattner | 2010-10-03 |
* | stub out a header to put 3dNow! instructions into. | Chris Lattner | 2010-10-02 |
* | fix a regression introduced in r115243, in which the instruction | Chris Lattner | 2010-10-02 |
* | Massive rewrite of MMX: | Dale Johannesen | 2010-09-30 |
* | Move remaining MMX instructions from SSE to MMX. | Dale Johannesen | 2010-09-09 |
* | Move most MMX instructions (defined as anything that | Dale Johannesen | 2010-09-09 |
* | Add intrinsic-based patterns for MMX PINSRW and PEXTRW. | Dale Johannesen | 2010-09-08 |
* | Slight cleanup, use only one form of MMXI_binop_rm_int. | Dale Johannesen | 2010-09-08 |
* | Add intrinsic forms of mmx<->sse conversions. Notes: | Dale Johannesen | 2010-09-08 |
* | Add patterns for MMX that use the new intrinsics. | Dale Johannesen | 2010-09-07 |
* | fix the encoding of MMX_MOVFR642Qrr, it starts with 0xF2 not 0xF3, | Chris Lattner | 2010-07-15 |
* | rip out even more sporadic v2f32 support. | Chris Lattner | 2010-07-05 |
* | Fix an mmx movd encoding. | Dan Gohman | 2010-05-24 |
* | Delete MMX_MOVQ64gmr. It was the same as MMX_MOVQ64mr, but it didn't | Dan Gohman | 2010-05-20 |
* | Fixed the encoding of two of the X86 movq instuctions. The Move quadword from | Kevin Enderby | 2010-05-03 |
* | Add some missing x86 patterns for movdq2q. Fixes two (LLVM-)GCC DejaGNU test... | Stuart Hastings | 2010-04-23 |
* | remove a bunch of dead patterns. | Chris Lattner | 2010-03-28 |
* | fix a few more ambiguous types. | Chris Lattner | 2010-03-15 |
* | fix some more ambiguous patterns, remove another nontemporalstore | Chris Lattner | 2010-03-08 |
* | The mayHaveSideEffects flag is no longer used. | Dan Gohman | 2010-02-27 |
* | remove a confused pattern that is trying to match an address | Chris Lattner | 2010-02-23 |
* | X86InstrInfoSSE.td declares PINSRW as having type v8i16, | Chris Lattner | 2010-02-23 |
* | TableGen fragment refactoring. | David Greene | 2010-02-09 |
* | lower the last of the MRMInitReg instructions in MCInstLower. | Chris Lattner | 2010-02-05 |
* | Improved widening loads by adding support for wider loads if | Mon P Wang | 2010-01-24 |
* | Instruction fixes, added instructions, and AsmString changes in the | Sean Callanan | 2009-12-18 |
* | Rename usesCustomDAGSchedInserter to usesCustomInserter, and update a | Dan Gohman | 2009-10-29 |
* | Add 'isCodeGenOnly' bit to Instruction .td records. | Daniel Dunbar | 2009-08-11 |
* | Whitespace, 80-column, and isTwoAddress -> Constraints = "" changes. | Eric Christopher | 2009-08-10 |
* | Remove neverHasSideEffects on MMX_MOVD64rrv164 since it has a matching pattern. | Evan Cheng | 2009-08-03 |
* | Use movd instead of movq | Rafael Espindola | 2009-08-03 |
* | Fix the instruction encoding. | Rafael Espindola | 2009-08-03 |
* | Use movq to move 64 bits in and out of mmx registers. | Rafael Espindola | 2009-08-03 |
* | Add support for MMX VSETCC. | Eli Friedman | 2009-07-22 |
* | Misc encoding fixes; reported on llvmdev. | Eli Friedman | 2009-07-09 |
* | "The MMX_MASKMOVQ and MMX_MASKMOVQ64 instructions are labeled as MRMDestMem | Bill Wendling | 2009-06-23 |
* | Get rid of some bogus patterns for X86vzmovl. Don't create VZEXT_MOVL | Eli Friedman | 2009-06-06 |
* | Get rid of a bogus pattern that interferes with optimization. | Eli Friedman | 2009-06-06 |
* | Evan says it's wrong; back out 72808. | Stuart Hastings | 2009-06-03 |
* | Recognize another euphemism for MOVDQ2Q. | Stuart Hastings | 2009-06-03 |
* | "The instructions MMX_PSADBWrm and MMX_PSADBWrr have opcode 0b11100000 (e0), but | Bill Wendling | 2009-05-28 |
* | 2nd attempt, fixing SSE4.1 issues and implementing feedback from duncan. | Nate Begeman | 2009-04-27 |
* | Revert 69952. Causes testsuite failures on linux x86-64. | Rafael Espindola | 2009-04-24 |
* | PR2957 | Nate Begeman | 2009-04-24 |
* | Only v1i16 (i.e. _m64) is returned via RAX / RDX. | Evan Cheng | 2009-02-23 |
* | Added support for SELECT v8i8 v4i16 for X86 (MMX) | Mon P Wang | 2008-12-12 |
* | Use mmx (punpckldq VR64, (mmx_v_set0)) to clear high 32-bits of a VR64 register. | Evan Cheng | 2008-12-03 |