Cleanup vector logical ops in AVX and add use int versions for simple
[oota-llvm.git] / lib / Target / X86 /
2011-08-18 Bruno Cardoso LopesCleanup vector logical ops in AVX and add use int versi...
2011-08-17 Bruno Cardoso LopesFix PR10688. Add support for spliting 256-bit vector...
2011-08-17 Owen AndersonAllow the MCDisassembler to return a "soft fail" status...
2011-08-17 Bruno Cardoso LopesIntroduce matching patterns for vbroadcast AVX instruct...
2011-08-17 Bruno Cardoso LopesUpdate comments about vector splat handling in x86
2011-08-17 Bruno Cardoso LopesNow that we have a canonical way to handle 256-bit...
2011-08-16 Bruno Cardoso LopesInstead of always leaving the work to the generic legal...
2011-08-15 Bruno Cardoso LopesWhile I'm here, remove the "_alt" hacks to a series...
2011-08-15 Bruno Cardoso LopesReorder declarations of vmovmskp* and also put the...
2011-08-15 Jim GrosbachMCTargetAsmParser target match predicate support.
2011-08-15 Bruno Cardoso LopesFix PR10656. It's only profitable to use 128-bit insert...
2011-08-12 Bruno Cardoso LopesFix comment!
2011-08-12 Bruno Cardoso LopesThe VPERM2F128 is a AVX instruction which permutes...
2011-08-12 Bruno Cardoso LopesMove code around and add comments
2011-08-12 Duncan SandsSilence a bunch (but not all) "variable written but...
2011-08-12 Andrew TrickfindDeadCallerSavedReg fix: Missing NULL terminator...
2011-08-11 Bruno Cardoso LopesAdd a dag combine to xform 256-bit shuffles into simple...
2011-08-11 Bruno Cardoso LopesFix PR10492 by teaching MOVHLPS and MOVLPS mask matchin...
2011-08-11 Nadav RotemAdd a comment, per Bruno's CR.
2011-08-11 Nadav Rotem[AVX] If the data which is going to be saved is already...
2011-08-11 Bruno Cardoso LopesCleanup: Remove Int_ CVTSS2SI* forms
2011-08-11 Bruno Cardoso LopesSplats for v8i32/v8f32 can be handled by VPERMILPSY...
2011-08-11 Bruno Cardoso LopesUse the splat index to generate the desired shuffle...
2011-08-11 Eli FriedmanFix X86TargetLowering::LowerExternalSymbol so that...
2011-08-10 Nadav RotemWhen performing a truncating store, it is sometimes...
2011-08-10 Bruno Cardoso LopesThe following X86 pattern is incorrect:
2011-08-10 Bruno Cardoso LopesFix a bug in vpermilps mask checking. Fix PR10560
2011-08-09 Bruno Cardoso LopesAdd 256-bit support for v8i32, v4i64 and v4f64 ISD...
2011-08-09 Bruno Cardoso LopesAdd v16i16 and v32i8 store patterns
2011-08-09 Bruno Cardoso LopesUse fp unpack instructions to unpack int types. Until...
2011-08-09 Eli FriedmanFix a couple ridiculous copy-paste errors. rdar:/...
2011-08-09 Bruno Cardoso LopesReapply a more appropriate solution than in r137114...
2011-08-09 Bruno Cardoso LopesRevert r137114
2011-08-09 Bruno Cardoso LopesHandle sitofp between v4f64 <- v4i32. Fix PR10559
2011-08-09 Bruno Cardoso LopesAdd support for avx vector fextend
2011-08-09 Bruno Cardoso LopesAdd AVX versions of 128-bit sitofp and fptosi
2011-08-09 Bruno Cardoso LopesAdd two patterns to match special vmovss and vmovsd...
2011-08-09 Bruno Cardoso LopesMake LowerVSETCC aware of AVX types and add patterns...
2011-08-08 Bruno Cardoso LopesAdd support for several vector shifts operations while...
2011-08-08 Jakob Stoklund OlesenHoist hasLoadFromStackSlot and hasStoreToStackSlot.
2011-08-08 Jakob Stoklund OlesenDon't clobber pending ST regs when FP regs are killed.
2011-08-05 Chandler CarruthSilence unused variable warnings in release builds.
2011-08-05 Jason W KimFix llvm.org/bugs/show_bug.cgi?id=10583\n - test for...
2011-08-04 Evan ChengFix an obvious type. Patch by Ivan Krasin.
2011-08-04 Duncan SandsAdd obviously missing "break". Noticed by Andrey Karpo...
2011-08-04 Jason W KimFix http://llvm.org/bugs/show_bug.cgi?id=10568
2011-08-04 Bill WendlingOnly access both operands of an INSERT_SUBVECTOR if...
2011-08-03 Benjamin KramerRemove unused variables.
2011-08-03 Jakob Stoklund OlesenHandle IMPLICIT_DEF instructions in X86FloatingPoint.
2011-08-02 Eli FriedmanDon't create a ridiculous EXTRACT_ELEMENT. PR10563.
2011-08-02 Bruno Cardoso LopesMake this kind of lowering to be supported by 256-bit...
2011-08-02 Nick LewyckyBail from FastISel when we encounter a volatile memset...
2011-08-01 Bruno Cardoso LopesAdd v4f64 -> v2f32 fp_round support. Also add a testcas...
2011-08-01 Bruno Cardoso LopesTeach PreprocessISelDAG to be aware of vector types...
2011-08-01 Bruno Cardoso LopesLower CONCAT_VECTORS to use two VINSERTF128 instruction...
2011-08-01 Bruno Cardoso LopesSince vectors with all ones can't be created with a...
2011-08-01 Douglas GregorUpdate CMake target names for tablegen-generated data...
2011-07-29 Eli FriedmanMisc optimizer+codegen work for 'cmpxchg' and 'atomicrm...
2011-07-29 Bruno Cardoso LopesFix two tests that I crashed in the previous commits...
2011-07-29 Bruno Cardoso LopesMatch VPERMIL masks more strictly and update the target...
2011-07-29 Bruno Cardoso LopesAdd DecodeShuffle shuffle support for VPERMIPD variantes
2011-07-29 Bruno Cardoso LopesAdd v8i32 and v4i64 vpermil patterns
2011-07-29 Bruno Cardoso LopesFix a bug while generating target specific VPERMIL...
2011-07-29 Bruno Cardoso LopesEnable usage of SSE4 extracts and inserts in their...
2011-07-29 Bruno Cardoso LopesCleanup PALIGNR handling and remove the old palign...
2011-07-29 Chandler CarruthRewrite the CMake build to use explicit dependencies...
2011-07-28 Oscar FuentesExplicitly declare a library dependency of LLVM*Desc to
2011-07-28 Bruno Cardoso LopesInvert the subvector insertion to be more likely to...
2011-07-28 Bruno Cardoso LopesAdd patterns to generate copies for extract_subvector...
2011-07-28 Bruno Cardoso Lopesmovd/movq write zeros in the high 128-bit part of the...
2011-07-28 Bruno Cardoso LopesAdd a few patterns to match allzeros without having...
2011-07-28 Bruno Cardoso LopesAdd SINT_TO_FP and FP_TO_SINT support for v8i32 types...
2011-07-27 Evan ChengEmit an error is asm parser parsed X86_64 only register...
2011-07-27 Kevin EnderbyFix llvm-mc handing of x86 instructions that take 8...
2011-07-27 Eli FriedmanCode generation for 'fence' instruction.
2011-07-27 Eli FriedmanX86ISD::MEMBARRIER does not require SSE2; it doesn...
2011-07-27 Jeffrey YasskinExplicitly cast narrowing conversions inside {}s that...
2011-07-27 Bruno Cardoso LopesMove some code around to open opportunity for more...
2011-07-27 Bruno Cardoso LopesThe vpermilps and vpermilpd have different behaviour...
2011-07-27 Bruno Cardoso LopesRemove more dead code!
2011-07-27 Evan ChengSupport .code32 and .code64 in X86 assembler.
2011-07-26 Benjamin KramerAdd a neat little two's complement hack for x86.
2011-07-26 Bruno Cardoso LopesRecognize unpckh* masks and match 256-bit versions...
2011-07-26 Eli FriedmanPrevent x86-specific DAGCombine from creating nodes...
2011-07-26 Bruno Cardoso LopesRemove now unused patterns. 0 insertions(+), 98 deletio...
2011-07-26 Bruno Cardoso LopesCleanup old matching for PUNPCK* variants
2011-07-26 Bill WendlingThe compact unwinding offsets are divided by 8 on 64...
2011-07-26 Bruno Cardoso LopesAdd 256-bit isel for movsldup/movshdup
2011-07-26 Bruno Cardoso LopesMore movsldup/movshdup cleanup. Rewrite the mask matchi...
2011-07-26 Bruno Cardoso LopesMore cleanup, subtarget info isn't used here.
2011-07-26 Bruno Cardoso LopesAdd 128-bit AVX versions of movshdup/mosldup
2011-07-26 Bruno Cardoso LopesCleanup movsldup/movshdup matching.
2011-07-26 Evan ChengRename createCodeEmitter to createMCCodeEmitter; create...
2011-07-26 Evan ChengRename TargetAsmParser to MCTargetAsmParser and TargetA...
2011-07-26 Chandler CarruthClean up a pile of hacks in our CMake build relating...
2011-07-25 Evan ChengRename TargetAsmBackend to MCAsmBackend; rename createA...
2011-07-25 Bruno Cardoso LopesCodegen allonesvector better while using AVX: vpcmpeqd...
2011-07-25 Bruno Cardoso LopesAdd remaining 256-bit vector bitcasts. This also fixes...
2011-07-25 Bruno Cardoso Lopes- Handle special scalar_to_vector case: splats. Using...
2011-07-25 Bruno Cardoso LopesReintroduce r135730, this is indeed the right approach...
next