- 7855006 Convert an improper CodeGen test to a MC test. by Evan Cheng · 12 years ago
- e29c880 instcombine: Query target library information to gate libcall simplifications by Meador Inge · 12 years ago
- 2f69102 xfail a bad test. This is a MC test but it's dependent on a codegen optimization which is now disabled. by Evan Cheng · 12 years ago
- b341fac Disable the Thumb no-return call optimization: mov lr, pc b.w _foo by Evan Cheng · 12 years ago
- 9c7ae01 Cleanup pcmp(e/i)str(m/i) instruction definitions and load folding support. by Craig Topper · 12 years ago
- 89443ff [NVPTX] Use ABI alignment for parameters when alignment is not specified. by Justin Holewinski · 12 years ago
- 722c9a7 Fix assertions in updateRegMaskSlots(). by Jakob Stoklund Olesen · 12 years ago
- c2e9ca1 tsan: switch to new memory_order constants (ABI compatible) by Dmitry Vyukov · 12 years ago
- 9f8a90b tsan: instrument all atomics (including fetch_add, exchange, cas, etc) by Dmitry Vyukov · 12 years ago
- b1bf1ee Add support for memory runtime check. When we can, we calculate array bounds. by Nadav Rotem · 12 years ago
- 562b1d8 llvm/ConstantFolding.cpp: Make ReadDataFromGlobal() and FoldReinterpretLoadFromConstPtr() Big-endian-aware. by NAKAMURA Takumi · 12 years ago
- 214fd3d Recommit modified r167540. by Amara Emerson · 12 years ago
- be02a90 Add support of RTM from TSX extension by Michael Liao · 12 years ago
- 7629de3 instcombine: Migrate strspn optimizations by Meador Inge · 12 years ago
- 806e03d Add a relocation visitor to lib object. This works via caching relocated by Eric Christopher · 12 years ago
- b031996 Only do switch-to-lookup table transformation when TargetTransformInfo by Hans Wennborg · 12 years ago
- e90a3bc [mips] Custom-lower ISD::FRAME_TO_ARGS_OFFSET node. by Akira Hatanaka · 12 years ago
- ab1b6c8 Fix bad test IR in switch_to_lookup_table.ll by Hans Wennborg · 12 years ago
- 3b87f62 misched: Heuristics based on the machine model. by Andrew Trick · 12 years ago
- a6fb97a CostModel: add another known vector trunc optimization. by Nadav Rotem · 12 years ago
- b042868 Cost Model: add tables for some avx type-conversion hacks. by Nadav Rotem · 12 years ago
- 7ae3bcc CostModel: Add tables for the common x86 compares. by Nadav Rotem · 12 years ago
- 2d1528b Code Model: Improve the accuracy of the zext/sext/trunc vector cost estimation. by Nadav Rotem · 12 years ago
- 32c1a82 Fix for PR14264 cause by commit r167237 which did not take into account a by Kevin Enderby · 12 years ago
- a4ab529 Cost Model: Normalize the insert/extract index when splitting types by Nadav Rotem · 12 years ago
- 75138f5 Cost Model: teach the cost model about expanding integers. by Nadav Rotem · 12 years ago
- 86aef0a On PowerPC64, integer return values (as well as arguments) are supposed by Ulrich Weigand · 12 years ago
- e623702 Implement the cost of abnormal x86 instruction lowering as a table. by Nadav Rotem · 12 years ago
- 827b7a0 Add support for the PowerPC-specific inline asm Z constraint and y modifier. by Hal Finkel · 12 years ago
- cfe09ed [PATCH] PowerPC: Expand load extend vector operations by Adhemerval Zanella · 12 years ago
- 5445333 Don't infer whether a value is captured in the current function from the by Richard Osborne · 12 years ago
- c5b969a Generalize the transform that boosts GEP indices to the size of a pointer to by Duncan Sands · 12 years ago
- 3c77033 [mips] Set flag neverHasSideEffects flag on floating point conversion by Akira Hatanaka · 12 years ago
- b4b04c3 X86 CostModel: Add support for a some of the common arithmetic instructions for SSE4, AVX and AVX2. by Nadav Rotem · 12 years ago
- 3c9c1ab [mips] Set flag isAsCheapAsAMove flag on instruction LUi. by Akira Hatanaka · 12 years ago
- 11a45c2 [mips] Stop reserving register AT and use register scavenger when a scratch by Akira Hatanaka · 12 years ago
- 0c31e43 Add a stub for the x86 cost model impl. Implement a basic cost rule for inserting/extracting from XMM registers. by Nadav Rotem · 12 years ago
- f149567 CostModel: add support for Vector Insert and Extract. by Nadav Rotem · 12 years ago
- 53e93e8 [mips] Fix disassembler test cases. by Akira Hatanaka · 12 years ago
- 6bed58e Add a cost model analysis that allows us to estimate the cost of IR-level instructions. by Nadav Rotem · 12 years ago
- fb60a4e [mips] Fix bug in test case. Disable machine LICM to prevent instruction from by Akira Hatanaka · 12 years ago
- 43934ae Vext Lowering was missing opportunities by Quentin Colombet · 12 years ago
- 5c87b73 [mips] Use register number instead of name to print register $AT. by Akira Hatanaka · 12 years ago
- 173192f [mips] Delete MipsFunctionInfo::EmitNOAT. Unconditionally print directive by Akira Hatanaka · 12 years ago
- 52df6c2 Add a testcase to loop-idiom to cover PR14241 when we start handling by Chandler Carruth · 12 years ago
- a864748 Revert the switch of loop-idiom to use the new dependence analysis. by Chandler Carruth · 12 years ago
- 4ac300d BBVectorize: Commit the rest of the test-case change. by Hal Finkel · 12 years ago
- 78fd353 BBVectorize: Use target costs for incoming and outgoing values instead of the depth heuristic. by Hal Finkel · 12 years ago
- 938482f Add support for generating dwarf debugging info with assembly files by Kevin Enderby · 12 years ago
- b6843c2 llvm/test/lit.cfg: Don't use mcjit to ppc32 yet, not ready. by NAKAMURA Takumi · 12 years ago
- a5f54f1 [asan] don't instrument globals that we've created ourselves (reduces the binary size a bit) by Kostya Serebryany · 12 years ago
- 9cb285d Add a test case for PR14233. by Chandler Carruth · 12 years ago
- ece6c6b Revert the series of commits starting with r166578 which introduced the by Chandler Carruth · 12 years ago
- d0dd065 [CMake] Add llvm-mcmarkup to check-llvm. by NAKAMURA Takumi · 12 years ago
- b75111f test/CodeGen/X86/fp-fast.ll: Add +avx. by NAKAMURA Takumi · 12 years ago
- 607ebde Add a few more simple fast-math constant propagations and cancellations. by Owen Anderson · 12 years ago
- 7dd4dc8 MC: Simple example parser for MC assembly markup. by Jim Grosbach · 12 years ago
- a5526a9 (For X86) Enhancement to add-carray/sub-borrow (adc/sbb) optimization. by Shuxin Yang · 12 years ago
- e57b2cb LoopVectorize: Preserve NSW, NUW and IsExact flags. by Nadav Rotem · 12 years ago
- 0dba9a9 Fix a bug in the cost calculation of vector casts. Detect situations where bitcasts cost zero. by Nadav Rotem · 12 years ago
- 497204a [mips] Set isAsCheapAsAMove flag on ADDiu and DADDiu, which enables by Akira Hatanaka · 12 years ago
- 39fa606 Test case for r167039. Check that tail-call optimization is disabled for mips16. by Akira Hatanaka · 12 years ago
- eb3e089 Remove fixme about unreachable cases from SwitchToLookupTable by Hans Wennborg · 12 years ago
- 72465ea BBVectorize: Choose pair ordering to minimize shuffles by Hal Finkel · 12 years ago
- e0f1dca instcombine: Migrate strto* optimizations by Meador Inge · 12 years ago
- e03d9e4 Do simple constant propagation in lookup table formation for switches by Hans Wennborg · 12 years ago
- 4ad3d98 LCSSA: Add a workaround for another nasty SCEV cache invalidation issue. by Benjamin Kramer · 12 years ago
- 061938b DependenceAnalysis: Don't crash if there is no constant operand. by Benjamin Kramer · 12 years ago
- 9441125 Implement ADJCALLSTACKUP and ADJCALLSTACKDOWN by Reed Kotler · 12 years ago
- 08684d1 instcombine: Migrate strpbrk optimizations by Meador Inge · 12 years ago
- 57cfd71 instcombine: Migrate strlen optimizations by Meador Inge · 12 years ago
- a0885fb instcombine: Migrate strncpy optimizations by Meador Inge · 12 years ago
- 5b2c4dc LoopVectorize: Do not vectorize loops with tiny constant trip counts. by Nadav Rotem · 12 years ago
- 42d4335 This patch addresses an ABI compatibility issue with empty aggregate by Bill Schmidt · 12 years ago
- 462d1ca Add support for loops that don't start with Zero. by Nadav Rotem · 12 years ago
- e6d781f instcombine: Migrate stpcpy optimizations by Meador Inge · 12 years ago
- fa9d137 instcombine: Split out the __stpcpy_chk simplifications from StrCpyChkOpt by Meador Inge · 12 years ago
- dfd0b9b X86 SSE: update rsqrtss and rcpss to use two source operands and by Manman Ren · 12 years ago
- 4c74a95 X86 MMX: optimize transfer from mmx to i32 by Manman Ren · 12 years ago
- 70dace3 Fix PR14212: For some strange reason I treated vectors differently from by Chandler Carruth · 12 years ago
- 2f34d75 [mips] Allow tail-call optimization for vararg functions and functions which by Akira Hatanaka · 12 years ago
- 415f869 LoopIdiom: Fix a serious missed optimization: we only turned top-level loops into memmove. by Benjamin Kramer · 12 years ago
- 8a69a95 BBVectorize: Fix a small bug introduced in r167042. by Hal Finkel · 12 years ago
- e709f5b LoopVectorize: Add support for write-only loops when the write destination is a single pointer. by Nadav Rotem · 12 years ago
- c83b5dc PowerPC: Expand FSRQT for vector types by Adhemerval Zanella · 12 years ago
- cc03331 LoopVectorize: Fix a bug in the initialization of reduction variables. AND needs to start at all-one by Nadav Rotem · 12 years ago
- c197a55 Set %defaultjit to use MCJIT for PowerPC targets. by Ulrich Weigand · 12 years ago
- 9a419f6 Change ForceSizeOpt attribute into MinSize attribute by Quentin Colombet · 12 years ago
- 92b4699 switch_to_lookup_table.ll: Remove some unnecessary lines, comments, by Hans Wennborg · 12 years ago
- 5f41fd6 PowerPC: More support for Altivec compare operations by Adhemerval Zanella · 12 years ago
- 3467b9f Enable some additional constant folding for PPCDoubleDouble. by Ulrich Weigand · 12 years ago
- 04d7d13 Use TargetTransformInfo to control switch-to-lookup table transformation by Hans Wennborg · 12 years ago
- c588e0e Remove an invalid assert in TargetTransformImpl by Hal Finkel · 12 years ago
- 8ba1474 ARM: Better disassembly for pc-relative LDR. by Jim Grosbach · 12 years ago
- c09856b Change mips16 delay slot jumps to non delay slot forms by default. by Reed Kotler · 12 years ago
- a24262a Re-commit r166971. I reverted it to quickly, when buildbots didn't have a chance by Jakub Staszak · 12 years ago
- 445ba85 Fix ARM's b.w instruction for thumb 2 and the encoding T4. The branch target by Kevin Enderby · 12 years ago
- c1ed096 Revert r166971. It causes buildbot failure. To be investigated. by Jakub Staszak · 12 years ago
- 926dd44 llvm/test/CodeGen/X86/vec_shuffle-30.ll: Try to unbreak builds - assuming +avx. by NAKAMURA Takumi · 12 years ago
- 6d31782 Allow to fold vector load if there is more than one bitcast, so in the case: by Jakub Staszak · 12 years ago