Add a test that would have found the bug in r219829.
[oota-llvm.git] / lib /
2014-10-17 Akira HatanakaARM: Fix a bug which was causing convergence failure...
2014-10-17 Rafael EspindolaRevert commit r219835 and r219829.
2014-10-17 Peter Zotov[LLVM-C] Add LLVMInstructionClone.
2014-10-17 Matt ArsenaultR600/SI: Simplify debug printing
2014-10-16 Matt ArsenaultR600/SI: Remove another VALU pattern
2014-10-16 Peter CollingbourneIntroduce LLVMParseCommandLineOptions C API function.
2014-10-16 Juergen RibutzkaReduce code duplication between patchpoint and non...
2014-10-16 Chandler Carruth[SROA] Switch the common variable name for the 'AllocaS...
2014-10-16 Chandler Carruth[SROA] More range-based cleanups to SROA, these brought...
2014-10-16 Chandler Carruth[SROA] Switch a couple of overly complex iterator acces...
2014-10-16 Robin MorissetErase fence insertion from SelectionDAGBuilder.cpp...
2014-10-16 Matt ArsenaultR600/SI: Remove unnecessary VALU patterns
2014-10-16 Chandler Carruth[SROA] Start more deeply moving SROA to use ranges...
2014-10-16 Matt ArsenaultR600: Fix nonsensical implementation of computeKnownBit...
2014-10-16 Rafael EspindolaDelete -std-compile-opts.
2014-10-16 Bjorn SteinbrinkAllow call-slop optzn for destinations with a suitable...
2014-10-16 Sanjay Patelfold: sqrt(x * x * y) -> fabs(x) * sqrt(y)
2014-10-16 Juergen Ributzka[AArch64] Fix miscompile of sdiv-by-power-of-2.
2014-10-16 Vasileios Kalintiris[mips] Account for endianess when expanding BuildPairF6...
2014-10-16 Vasileios Kalintiris[mips] Marked the DI/EI instruction aliases as MIPS32r2
2014-10-16 Vasileios KalintirisTest commit access: remove extra new line at the end...
2014-10-16 Akira HatanakaReapply r219832 - InstCombine: Narrow switch instructio...
2014-10-16 Saleem AbdulrasoolTRE: make TRE a bit more aggressive
2014-10-16 Akira HatanakaRevert r219832.
2014-10-16 Hal Finkel[LVI] Add some additional comments about caching and...
2014-10-16 Matt ArsenaultR600: Remove dead function
2014-10-15 Sanjoy DasRevert "r219834 - Teach ScalarEvolution to sharpen...
2014-10-15 Hal FinkelPreserve non-byval pointer alignment attributes using...
2014-10-15 Hal FinkelAdd CreateAlignmentAssumption to IRBuilder
2014-10-15 Adam Nemet[AVX512] Add DQ subvector inserts
2014-10-15 Adam Nemet[AVX512] Two new attributes in X86VectorVTInfo for...
2014-10-15 Adam Nemet[AVX512] Rename arg from Opcode32/64 to Opcode128/256...
2014-10-15 Matt ArsenaultR600: Remove unnecessary part of computeKnownBitsForTar...
2014-10-15 Matt ArsenaultMove variable down to use
2014-10-15 Alexander PotapenkoAdd MachOObjectFile::getUuid()
2014-10-15 Chris BienemanFixing the build failure due to compiler warnings and...
2014-10-15 Chris BienemanDefining a new API for debug options that doesn't rely...
2014-10-15 Tom StellardR600/SI: Fix bug where immediates were being used in...
2014-10-15 Eric ChristopherAvoid caching the MachineFunction, we don't use it...
2014-10-15 Sid ManningWrong attribute. LLVM_ATTRIBUTE_UNUSED not LLVM_ATTRIB...
2014-10-15 Duncan P. N. Exon... IR: Move NumOperands from User to Value, NFC
2014-10-15 Duncan P. N. Exon... IR: Cleanup comments for Value, User, and MDNode
2014-10-15 Sid ManningWrong attribute. LLVM_ATTRIBUTE_USED not LLVM_ATTRIBUTE...
2014-10-15 Rafael EspindolaAllow forward references to section symbols.
2014-10-15 Sanjoy DasTeach ScalarEvolution to sharpen range information.
2014-10-15 Sid ManningAdd LLVM_ATTRIBUTE_UNUSED to function currently just...
2014-10-15 Akira HatanakaInstCombine: Narrow switch instructions using known...
2014-10-15 Juergen RibutzkaReapply "[FastISel][AArch64] Add custom lowering for...
2014-10-15 Juergen Ributzka[FastISel][AArch64] Factor out add with immediate emiss...
2014-10-15 Rafael EspindolaCorrectly handle references to section symbols.
2014-10-15 Sid ManningEnable the instruction printer in HexagonMCTargetDesc
2014-10-15 Matt ArsenaultR600/SI: Also try to use 0 base for misaligned 8-byte...
2014-10-15 Matt ArsenaultR600: Fix miscompiles when BFE has multiple uses
2014-10-15 Sanjay Patelcorrect const-ness with auto and dyn_cast
2014-10-15 Hal Finkel[SLPVectorize] Basic ephemeral-value awareness
2014-10-15 Hal FinkelTreat the WorkSet used to find ephemeral values as...
2014-10-15 Derek Schuff[MC] Make bundle alignment mode setting idempotent...
2014-10-15 Duncan P. N. Exon... DI: Make comments "brief"-er, NFC
2014-10-15 Sanjay PatelUse 'auto' for easier reading; no functional change...
2014-10-15 Duncan P. N. Exon... DI: Cleanup comments, NFC
2014-10-15 Rafael EspindolaSimplify handling of --noexecstack by using getNonexecu...
2014-10-15 Duncan P. N. Exon... DI: Use a `DenseMap` instead of named metadata, NFC
2014-10-15 Rafael EspindolaMove getNonexecutableStackSection up to the base ELF...
2014-10-15 Matt ArsenaultR600: Use existing variable
2014-10-15 Matt ArsenaultR600: Remove outdated comment
2014-10-15 Juergen RibutzkaRevert "[FastISel][AArch64] Add custom lowering for...
2014-10-15 Jingyue Wu[MachineSink] Use the real post dominator tree
2014-10-15 Tim NorthoverARM: drop check for triple that's no longer used.
2014-10-15 Eric ChristopherRemove unused variable.
2014-10-14 Eric ChristopherNo need to cache this unused variable.
2014-10-14 Gerolf Hoflehner[AArch64] Wrong CC access in CSINC-conditional branch...
2014-10-14 Gerolf Hoflehner[AAarch64] Optimize CSINC-branch sequence
2014-10-14 Hal Finkel[LoopVectorize] Ignore @llvm.assume for cost estimates...
2014-10-14 Simon Pilgrim[X86][SSE] pslldq/psrldq shuffle mask decodes
2014-10-14 Tim NorthoverARM: remove ARM/Thumb distinction for preferred alignment.
2014-10-14 Tim NorthoverARM: allow misaligned local variables in Thumb1 mode.
2014-10-14 Juergen Ributzka[FastISel][AArch64] Add custom lowering for GEPs.
2014-10-14 Hans Wennborg[x86 asm] allow fwait alias in both At&t and Intel...
2014-10-14 Tim NorthoverARM: set preferred aggregate alignment to 32 universally.
2014-10-14 Hal Finkel[CFL-AA] CFL-AA should not assert on an va_arg instruction
2014-10-14 Sanjay PatelOptimize away fabs() calls when input is squared (known...
2014-10-14 Juergen Ributzka[FastISel][AArch64] Fix sign-/zero-extend folding when...
2014-10-14 David MajnemerInstCombine: Don't miscompile X % ((Pow2 << A) >>u B)
2014-10-14 Jan VeselyReapply "R600: Add new intrinsic to read work dimensions"
2014-10-14 Hal FinkelRevert "r216914 - Revert: [APFloat] Fixed a bug in...
2014-10-14 Rafael EspindolaRevert "R600: Add new intrinsic to read work dimensions"
2014-10-14 Rafael EspindolaRemove unused member variable.
2014-10-14 Jan VeselyR600: Add new intrinsic to read work dimensions
2014-10-14 Jan VeselyR600: FMA is VecALU only instruction
2014-10-14 Reed KotlerFinish getting Mips fast-isel to match up with AArch64...
2014-10-14 David BlaikieDebugInfo: Ensure that all debug location scope chains...
2014-10-14 Rafael EspindolaRemove method that is identical to the base class one.
2014-10-14 Matt ArsenaultR600/SI: Use DS offsets for constant addresses
2014-10-14 David BlaikieRevert "Fix stuff... again."
2014-10-14 David BlaikieRevert some parts of r196288 that were confusing and...
2014-10-14 David BlaikieFix stuff... again.
2014-10-14 Hal Finkel[LVI] Check for @llvm.assume dominating the edge branch
2014-10-14 NAKAMURA TakumiRevert r219638, (r219640 and r219676), "Removing the...
2014-10-14 Robert Khasanov[AVX512] Extended avx512_binop_rm to DQ/VL subsets.
2014-10-14 Robert Khasanov[AVX512] Extended avx512_binop_rm to BW/VL subsets.
next