summaryrefslogtreecommitdiff
path: root/lib/Target/X86
Commit message (Expand)AuthorAge
* PR11033: Make sure we don't generate PCMPGTQ and PCMPEQQ if the target CPU do...Eli Friedman2011-09-28
* Rename SSEDomainFix -> lib/CodeGen/ExecutionDepsFix.Jakob Stoklund Olesen2011-09-28
* Remove X86-dependent stuff from SSEDomainFix.Jakob Stoklund Olesen2011-09-27
* Promote the X86 Get/SetSSEDomain functions to TargetInstrInfo.Jakob Stoklund Olesen2011-09-27
* Fix VEX decoding in i386 mode. Fixes PR11008.Craig Topper2011-09-26
* Only run MF.verify() with EXPENSIVE_CHECKS=1.Jakob Stoklund Olesen2011-09-24
* Implement Chris's suggestion of legalizing the various SSE and AVXDuncan Sands2011-09-23
* PR10991: make fast-isel correctly check whether accessing a global through an...Eli Friedman2011-09-22
* Add support for GR32 <-> FR32 cross class copies.Jakob Stoklund Olesen2011-09-22
* Synthesize SSE3/AVX 128 bit horizontal add/sub instructions fromDuncan Sands2011-09-22
* Fix register printing in disassembling of push/pop of segment registers and i...Craig Topper2011-09-22
* The SSE version differences for fmin/fmax are more involved than I thought.Benjamin Kramer2011-09-22
* X86: Don't form min/max nodes if the target is missing SSE.Benjamin Kramer2011-09-22
* X86Disassembler: if verbose logging is going to nulls(), disable logging comp...Benjamin Kramer2011-09-21
* fix commentNadav Rotem2011-09-21
* Insert a sanity check on the combining of x86 truncing-store nodes. This come...Nadav Rotem2011-09-21
* Change:Richard Trieu2011-09-21
* In the disassembler C API, be careful not to confuse the comment streamer tha...Owen Anderson2011-09-21
* Revert r140097, working on a better approachBruno Cardoso Lopes2011-09-20
* Simplify max/minp[s|d] dagcombine matchingBruno Cardoso Lopes2011-09-20
* Tidy up a bit more, fix tab and remove trailing whitespacesBruno Cardoso Lopes2011-09-20
* The wrong relocation was being emitted for several SSSE3 instructions.Bruno Cardoso Lopes2011-09-20
* Tidy up code!Bruno Cardoso Lopes2011-09-20
* Extend changes from r139986 to produce 256-bit AVX minps/minpd/maxps/maxpd.Craig Topper2011-09-20
* Fix PR10949. Fix the encoding of VMOVPQIto64rr.Bruno Cardoso Lopes2011-09-19
* Based on the small opt Zvi's patch was trying to achieve, eliminateBruno Cardoso Lopes2011-09-19
* Match X86ISD::FSETCCsd and X86ISD::FSETCCss while in AVX mode. This fixBruno Cardoso Lopes2011-09-19
* Fix typos in my prev commit, found by Tobi.Nadav Rotem2011-09-18
* setOperationAction should be done on the return value of the type, not the op...Nadav Rotem2011-09-18
* When promoting integer vectors we often create ext-loads. This patch adds aNadav Rotem2011-09-18
* Fix typo by changing Lower256IntVETCC to Lower256IntVSETCC.Craig Topper2011-09-18
* Synthesize x86 max/min instructions also for vectors (i.e. produceDuncan Sands2011-09-17
* Describe more AVX 128-bit convert instructions without patterns to haveBruno Cardoso Lopes2011-09-16
* Add mayLoad attribute to AVX convert instructions, since non of themBruno Cardoso Lopes2011-09-16
* Fix PR10884.Bruno Cardoso Lopes2011-09-16
* Don't attach annotations to MCInst's. Instead, have the disassembler return,...Owen Anderson2011-09-15
* Add a fixme note!Bruno Cardoso Lopes2011-09-15
* Add the remaining AVX versions of instructions to X86InstrInfo, thisBruno Cardoso Lopes2011-09-15
* Factor out partial register update checks for some SSE instructions.Bruno Cardoso Lopes2011-09-15
* Add support for stored annotations to MCInst, and provide facilities for MC-b...Owen Anderson2011-09-15
* Change all checks regarding the presence of any SSE level to alwaysBruno Cardoso Lopes2011-09-15
* Enable SSEDomainFix pass for AVX mode.Bruno Cardoso Lopes2011-09-15
* Fix the code creating VZEXT_LOAD so that it creates the right memoperand. Is...Eli Friedman2011-09-14
* Fix mem type for VEX.128 form of VROUNDP*. Remove filter preventing VROUND fr...Craig Topper2011-09-14
* Make disassembling of VBLEND* print immediate as a XMM/YMM register name. Fix...Craig Topper2011-09-14
* Teach the foldable tables about 128-bit AVX instructions and make theBruno Cardoso Lopes2011-09-14
* Vector shuffle mask <i32 4, i32 5, i32 2, i32 3> should yield "movsd", not "m...Bruno Cardoso Lopes2011-09-14
* swap vselect operand order - pr10907Nadav Rotem2011-09-13
* Add versions 256-bit versions of alignedstore and alignedload, to beBruno Cardoso Lopes2011-09-13
* Revert the remaining part of r139528. According to PR10907 the bug seemsBruno Cardoso Lopes2011-09-13