projects / oota-llvm.git / history
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
LSR: Don't optimize loops if an outer loop has no preheader.
[oota-llvm.git] / lib / Target / X86 /
2012-01-07 Eric ChristopherMake the 'x' constraint work for AVX registers as well. tree | commitdiff
2012-01-05 Craig TopperMark scalar FMA4 instructions as ignoring the VEX.L... tree | commitdiff
2012-01-05 Victor UmanskyPeephole optimization of ptest-conditioned branch in... tree | commitdiff
2012-01-05 Bill WendlingReplace the uint64_t -> double convertion algorithm... tree | commitdiff
2012-01-04 Benjamin KramerSilence warnings of a mysterious compiler that still... tree | commitdiff
2012-01-04 Evan ChengFor x86, canonicalize max tree | commitdiff
2012-01-03 Chad RosierFix 80-column violations. tree | commitdiff
2012-01-03 Nadav RotemRevert 147426 because it caused pr11696. tree | commitdiff
2012-01-03 Chad RosierEnhance DAGCombine for transforming 128->256 casts... tree | commitdiff
2012-01-03 Devang PatelIntel style asm variant does not need '%' prefix. tree | commitdiff
2012-01-02 Craig TopperMiscellaneous shuffle lowering cleanup. No functional... tree | commitdiff
2012-01-02 Craig TopperMake CanXFormVExtractWithShuffleIntoLoad reject loads... tree | commitdiff
2012-01-02 Nadav RotemOptimize the sequence blend(sign_extend(x)) to blend... tree | commitdiff
2012-01-01 Craig TopperAllow CRC32 instructions to be selected when AVX is... tree | commitdiff
2012-01-01 Craig TopperFix sfence, lfence, mfence, and clflush to be able... tree | commitdiff
2012-01-01 Benjamin KramerX86Disassembler: Fix undefined behavior found by GCC 4.6 tree | commitdiff
2011-12-31 Craig TopperMerge X86 SHUFPS and SHUFPD node types. tree | commitdiff
2011-12-31 Craig TopperAdd patterns for integer forms of SHUFPD/VSHUFPD with... tree | commitdiff
2011-12-31 Craig TopperFix typo in a SHUFPD and VSHUFPD pattern that prevented... tree | commitdiff
2011-12-30 Craig TopperMake FMA4 imply AVX so that YMM registers would be... tree | commitdiff
2011-12-30 Craig TopperAdd disassembler support for VPERMIL2PD and VPERMIL2PS. tree | commitdiff
2011-12-30 Craig TopperAdd FMA4 instructions to disassembler. tree | commitdiff
2011-12-30 Craig TopperSeparate the concept of having memory access in operand... tree | commitdiff
2011-12-30 Craig TopperCombine FMA4 SS/SD patterns with the instruction defini... tree | commitdiff
2011-12-30 Craig TopperCombine FMA4 PS/PD patterns with the instruction defini... tree | commitdiff
2011-12-30 Craig TopperChange FMA4 memory forms to use memopv* instead of... tree | commitdiff
2011-12-30 Craig TopperFix load size for FMA4 SS/SD instructions. They need... tree | commitdiff
2011-12-29 Craig TopperFix execution domains for PS/PD FMA3 instructions.... tree | commitdiff
2011-12-29 Craig TopperExpose FMA3 instructions to the disassembler. tree | commitdiff
2011-12-29 Craig TopperMake FMA3 imply AVX needs to be enabled. Particularly... tree | commitdiff
2011-12-29 Craig TopperChange XOP detection to use the correct CPUID bit inste... tree | commitdiff
2011-12-29 Craig TopperAdd FeaturePOPCNT to all CPU types that lost it was... tree | commitdiff
2011-12-29 Craig TopperMark non-VEX forms of PCLMUL instructions as requiring... tree | commitdiff
2011-12-29 Craig TopperMark non-VEX forms of AES instructions as requiring... tree | commitdiff
2011-12-29 Craig TopperRemove the separate explicit AES instruction patterns... tree | commitdiff
2011-12-29 Craig TopperMake SSE42 and SSE4A not imply POPCNT. POPCNT should... tree | commitdiff
2011-12-29 Craig TopperMake LowerBUILD_VECTOR keep node vector types consisten... tree | commitdiff
2011-12-29 Craig TopperRemove some elses after returns. tree | commitdiff
2011-12-29 Craig TopperRemove trailing spaces. Fix an assert to use && instead... tree | commitdiff
2011-12-28 Eli FriedmanFix type-checking for load transformation which is... tree | commitdiff
2011-12-28 Elena DemikhovskyFixed a bug in LowerVECTOR_SHUFFLE and LowerBUILD_VECTOR. tree | commitdiff
2011-12-27 Craig TopperAdd handling of x86_avx2_pmovmskb to computeMaskedBitsF... tree | commitdiff
2011-12-24 Rafael EspindolaSection relative fixups are a coff concept, not a x86... tree | commitdiff
2011-12-24 Chandler CarruthUse standard promotion for i8 CTTZ nodes and i8 CTLZ... tree | commitdiff
2011-12-24 Chandler CarruthAdd systematic testing for cttz as well, and fix the... tree | commitdiff
2011-12-24 Benjamin KramerChandler fixed this. tree | commitdiff
2011-12-24 Chandler CarruthExpand more when we have a nice 'tzcnt' instruction... tree | commitdiff
2011-12-24 Chandler CarruthSwitch the lowering of CTLZ_ZERO_UNDEF from a .td patte... tree | commitdiff
2011-12-24 Rafael EspindolaMove x86 specific bits of the COFF writer to lib/Target... tree | commitdiff
2011-12-22 Chad RosierFix 80-column violations. tree | commitdiff
2011-12-21 Chad RosierFix 80-column violations. tree | commitdiff
2011-12-21 Chad RosierNo case stmt for BUILD_VECTOR in PerformDAGCombine... tree | commitdiff
2011-12-21 Rafael EspindolaMove the X86 specific bits of the ELF writer to the... tree | commitdiff
2011-12-21 Rafael EspindolaReduce the exposure of Triple::OSType in the ELF object... tree | commitdiff
2011-12-21 Craig TopperRemove mode specific disassembler classes and just... tree | commitdiff
2011-12-21 Craig TopperFix typo in a couple comments tree | commitdiff
2011-12-20 Elena DemikhovskyThis is the second fix related to VZEXT_MOVL node. tree | commitdiff
2011-12-20 Chandler CarruthBegin teaching the X86 target how to efficiently codege... tree | commitdiff
2011-12-20 Chandler CarruthFix up the CMake build for the new files added in r1469... tree | commitdiff
2011-12-20 David BlaikieUnweaken vtables as per llvm.org/docs/CodingStandards... tree | commitdiff
2011-12-19 Jakob Stoklund OlesenEmit a getMatchingSuperRegClass() implementation for... tree | commitdiff
2011-12-18 Benjamin KramerAnother variadics tweak. tree | commitdiff
2011-12-18 Benjamin KramerUse the fancy new VariadicFunction template instead... tree | commitdiff
2011-12-17 Craig TopperRemove an unused X86ISD node type. tree | commitdiff
2011-12-17 Benjamin KramerX86: Factor the bswap asm matching to be slightly less... tree | commitdiff
2011-12-17 Rafael EspindolaAdd back the MC bits of 126425. Original patch by Natha... tree | commitdiff
2011-12-17 Lang HamesMake sure that the lower bits on the VSELECT condition... tree | commitdiff
2011-12-16 Craig TopperDon't try to match 'unpackl/h v, v' for 32xi8 and 16xi1... tree | commitdiff
2011-12-15 Eli FriedmanMake sure we correctly note the existence of an i8... tree | commitdiff
2011-12-15 Chad RosierAdd missing zmovl AVX patterns which were causing crashes. tree | commitdiff
2011-12-15 Chad RosierFix assert in LowerBUILD_VECTOR for v16i16 type on... tree | commitdiff
2011-12-15 Lang HamesFix VSELECT operand order. Was previously backwards... tree | commitdiff
2011-12-15 Chad RosierUse SmallVector/assign(), rather than std::vector/push_... tree | commitdiff
2011-12-15 Chad RosierAdd support for lowering fneg when AVX is enabled. tree | commitdiff
2011-12-14 Bill WendlingThe saved registers weren't being processed in the... tree | commitdiff
2011-12-14 Evan Cheng- Add MachineInstrBundle.h and MachineInstrBundle.cpp... tree | commitdiff
2011-12-13 Chandler CarruthInitial CodeGen support for CTTZ/CTLZ where a zero... tree | commitdiff
2011-12-12 Daniel DunbarLLVMBuild: Introduce a common section which currently... tree | commitdiff
2011-12-12 Daniel DunbarLLVMBuild: Remove trailing newline, which irked me. tree | commitdiff
2011-12-12 Jan SjödinXOP instructions and encoding tests. tree | commitdiff
2011-12-12 Jan SjödinXOP encoding bits and logic. tree | commitdiff
2011-12-11 Craig TopperRemove some remants of the old palign pattern fragment... tree | commitdiff
2011-12-10 Rafael EspindolaHandle expressions of the form _GLOBAL_OFFSET_TABLE_... tree | commitdiff
2011-12-09 Benjamin KramerThis is now implemented. tree | commitdiff
2011-12-09 Benjamin KramerX86: Add patterns for the various rounding ops for... tree | commitdiff
2011-12-09 Benjamin KramerX86: Split (v)rounds[sd] into a normal and an intrinsic... tree | commitdiff
2011-12-09 Evan ChengRemove hasSSE1orAVX(). It's the same as hasXMM(). tree | commitdiff
2011-12-08 Evan ChengAdd 256-bit variant vmovss and vmovsd patterns. rdar... tree | commitdiff
2011-12-08 Evan ChengAdd various missing AVX patterns which was causing... tree | commitdiff
2011-12-08 Owen AndersonDon't explicitly marked libm rounding ops as legal... tree | commitdiff
2011-12-08 Owen AndersonTeach SelectionDAG to match more calls to libm function... tree | commitdiff
2011-12-08 Evan ChengMany of the SSE patterns should not be selected when... tree | commitdiff
2011-12-08 Jan SjödinSrc2 and src3 were accidentally swapped for the FMA4... tree | commitdiff
2011-12-07 Craig TopperFix a bunch of SSE/AVX patterns to use proper memop... tree | commitdiff
2011-12-07 Bill WendlingAdjust the stack by one pointer size for all frameless... tree | commitdiff
2011-12-07 Bill WendlingFix off-by-one error when encoding the stack size for... tree | commitdiff
2011-12-07 Evan ChengAdd bundle aware API for querying instruction propertie... tree | commitdiff
2011-12-06 Bill WendlingExplicitly check for the different SUB instructions. tree | commitdiff
2011-12-06 Bill WendlingEncode the total stack if there isn't a frame. tree | commitdiff
2011-12-06 Bill Wendling* Add a macro to remove a magic number. tree | commitdiff
next
C/C++ LLVM-based compilers that forbids OOTA behaviors