- 792f334 Add a FIXME to TargetMachine to remove the function specific by Eric Christopher · 11 years ago
- eea543f Have setSubtarget take a const subtarget. by Eric Christopher · 11 years ago
- a6e0a6e Move resetTargetOptions from taking a MachineFunction to a Function by Eric Christopher · 11 years ago
- deaa9d8 R600: Avoid repeated check lines by Matt Arsenault · 11 years ago
- 584886c R600/SI: Fix emitting trailing whitespace after s_waitcnt by Matt Arsenault · 11 years ago
- 479f2f7 [AVX512] Simplify use of !con() by Adam Nemet · 11 years ago
- 08f261a [AVX512] Pull pattern for subvector extract into the instruction definition by Adam Nemet · 11 years ago
- 2f3ccfc [AVX512] Make vextract*x4/vinsert*x4 tests check for the index as well by Adam Nemet · 11 years ago
- 4007b30 [AVX512] Refactor subvector extracts by Adam Nemet · 11 years ago
- 1973ffe [AVX512] Fix typo by Adam Nemet · 11 years ago
- 2a7d068 Add SDAG TableGen definitions for BR_CC by Hal Finkel · 11 years ago
- 3011a60 R600: Fix some missing conversion testcases by Matt Arsenault · 11 years ago
- 556ae04 Remove duplicated RUN lines in middle of test by Matt Arsenault · 11 years ago
- f423025 [MachineSink+PGO] Teach MachineSink to use BlockFrequencyInfo by Bruno Cardoso Lopes · 11 years ago
- 5c1c055 Object: Add range iterators for Archive children by David Majnemer · 11 years ago
- 2fe3200 [Support] Fix Format.h to build on Windows by Nick Kledzik · 11 years ago
- e93da60 [Support] Add type-safe alternative to llvm::format() by Nick Kledzik · 11 years ago
- f85d5cf Refactoring: raw pointer -> unique_ptr by Anton Yartsev · 11 years ago
- 8361c84 ARM: Remove unneeded check for MI->hasPostISelHook() by Tom Stellard · 11 years ago
- bdaf056 SelectionDAG: Remove #if NDEBUG from check for a post-isel hook by Tom Stellard · 11 years ago
- 29d48e6 R600/SI: Add support for global atomic add by Tom Stellard · 11 years ago
- 79826e0 Lower idempotent RMWs to fence+load by Robin Morisset · 11 years ago
- 837a7c0 Since the DisasmMemoryObject only operates on const data, it now only accepts a const data pointer. This silences a -Wcast-qual warning. by Aaron Ballman · 11 years ago
- 733681d Add missing attributes !cmp.[eq,gt,gtu] instructions. by Sid Manning · 11 years ago
- 1d545d9 Add llvm_unreachables() for [ASZ]ExtUpper to X86FastISel.cpp to appease the buildbots. by Daniel Sanders · 11 years ago
- 03fe69e [mips] Add CCValAssign::[ASZ]ExtUpper and CCPromoteToUpperBitsInType and handle struct's correctly on big-endian N32/N64 return values. by Daniel Sanders · 11 years ago
- 6765c34 Add aliases for VAND imm to VBIC ~imm by Renato Golin · 11 years ago
- 4b667ee [x86] Teach the new vector shuffle lowering to use AVX2 instructions for by Chandler Carruth · 11 years ago
- b10671b Update my previous commit to fit 80 cols... by Sylvestre Ledru · 11 years ago
- cabd41e Details that -debug-only is not available when LLVM is built with --enable-optimized by Sylvestre Ledru · 11 years ago
- 05901d8 [x86] Teach the new vector shuffle lowering a fancier way to lower by Chandler Carruth · 11 years ago
- f220c53 [Thumb2] BXJ should be undefined for v7M, v8A by Oliver Stannard · 11 years ago
- 2e8d2c7 [x86] Fix an oversight in the v8i32 path of the new vector shuffle by Chandler Carruth · 11 years ago
- e5fb4ad [x86] Rearrange the code for v16i16 lowering a bit for clarity and to by Chandler Carruth · 11 years ago
- e3bb4bb [x86] Implement AVX2 support for v32i8 in the new vector shuffle lowering. by Chandler Carruth · 11 years ago
- 1d63231 [x86] More tweaks to the v32i8 test cases. by Chandler Carruth · 11 years ago
- a87d04a [x86] Re-work a bunch of the v32i8 test cases to actually involve byte by Chandler Carruth · 11 years ago
- dd8ce12 MC: Use @IMGREL instead of @IMGREL32, which we can't parse by Reid Kleckner · 11 years ago
- 6a289bb [x86] Remove the defunct X86ISD::BLENDV entry -- we use vector selects by Chandler Carruth · 11 years ago
- ef673b3 [x86] Fix the v16i16 blend logic I added in the prior commit and add the by Chandler Carruth · 11 years ago
- aacc919 llvm-cov: Combine segments that cover the same location by Justin Bogner · 11 years ago
- 0253523 [X86,AVX] Add an isel pattern for X86VBroadcast. by Akira Hatanaka · 11 years ago
- bdecfeb [x86] Implement v16i16 support with AVX2 in the new vector shuffle lowering. by Chandler Carruth · 11 years ago
- e793862 Flush out enough of llvm-objdump’s SymbolizerSymbolLookUp() for Mach-O files to by Kevin Enderby · 11 years ago
- 0e9d114 [asan] don't instrument module CTORs that may be run before asan.module_ctor. This fixes asan running together -coverage by Kostya Serebryany · 11 years ago
- d4c244f Removing empty ARM tests from failed revert by Renato Golin · 11 years ago
- ae3027c Removing empty tests from failed revert by Renato Golin · 11 years ago
- bb994f5 Revert 218406 - Refactor the RelocVisitor::visit method by Renato Golin · 11 years ago
- 450403a Revert 218407 - Add support for ARM and AArch64 BE object files by Renato Golin · 11 years ago
- 6c34b46 Revert 218408 - Report endianness in output of {dwarf, obj}dump by Renato Golin · 11 years ago
- 7fff39a Revert 218411 - XFAIL reloc test on x86/hexagon by Renato Golin · 11 years ago
- ebf023f XFAIL reloc test on x86/hexagon by Renato Golin · 11 years ago
- d3d620b Revert r218380. This was breaking Apple internal build bots. by Akira Hatanaka · 11 years ago
- 329e09f Report endianness in output of {dwarf, obj}dump by Renato Golin · 11 years ago
- bc7101b Add support for ARM and AArch64 BE object files by Renato Golin · 11 years ago
- c0104e4 Refactor the RelocVisitor::visit method by Renato Golin · 11 years ago
- f14380a pass environment when invoking llvm-config from lit.cfg by Scott Douglass · 11 years ago
- 4bb780a Adding #ifdef around TermColorMutex based on feedback from Craig Topper. by Chris Bieneman · 11 years ago
- c88ae96 [x86] Factor out the logic to generically decombose a vector shuffle by Chandler Carruth · 11 years ago
- 9917d2e Revert "Add support for ARM and AArch64 BE object files" by Kaelyn Takata · 11 years ago
- 48ac014 Revert "Report endianness in output of {dwarf, obj}dump" by Kaelyn Takata · 11 years ago
- a0d6422 Revert "Refactor the RelocVisitor::visit method" by Kaelyn Takata · 11 years ago
- c3a1f3e Report endianness in output of {dwarf, obj}dump by Renato Golin · 11 years ago
- 1513681 Add support for ARM and AArch64 BE object files by Renato Golin · 11 years ago
- faac033 Refactor the RelocVisitor::visit method by Renato Golin · 11 years ago
- cfc4296 Fix assertion in LICM doFinalization() by David Peixotto · 11 years ago
- 8c4e64a [Thumb] Make load/store optimizer less conservative. by Moritz Roth · 11 years ago
- 43c6b6b [Thumb] 32-bit encodings of 'cps' are not valid for v7M by Oliver Stannard · 11 years ago
- 6a07014 Silencing an "enumeral and non-enumeral type in conditional expression" warning. NFC. by Aaron Ballman · 11 years ago
- 7a27231 Replace a hand-written suffix compare with std::lexicographical_compare. by Benjamin Kramer · 11 years ago
- 10cd8098 [x86] Teach the instruction lowering to add comments describing constant by Chandler Carruth · 11 years ago
- 35fdc09 Allow BB duplication threshold to be adjusted through JumpThreading's ctor by Michael Liao · 11 years ago
- bfb2b18 Windows/Host.inc: Reformat the header to fit 80-col. by NAKAMURA Takumi · 11 years ago
- d968b05 Unix/Host.inc: Remove <cstdlib>. It has been unused for a long time. by NAKAMURA Takumi · 11 years ago
- 4921f68 Unix/Host.inc: Wrap a comment line in 80-col. by NAKAMURA Takumi · 11 years ago
- c44f94d Unix/Host.inc: Remove leading whitespace. It had been here since r56942! by NAKAMURA Takumi · 11 years ago
- 6b37197 valgrind/x86_64-pc-linux-gnu.supp: Suppress also /bin/bash. by NAKAMURA Takumi · 11 years ago
- 80b7979 valgrind/x86_64-pc-linux-gnu.supp: Tweak /bin/sed to let calloc recognized. by NAKAMURA Takumi · 11 years ago
- 1fe409e Clear PreferredExtendType for in each function-specific state FunctionLoweringInfo. by Jiangning Liu · 11 years ago
- f00b50b [x86] More refactoring of the shuffle comment emission. The previous by Chandler Carruth · 11 years ago
- db5e1da [x86] Bypass the shuffle mask comment generation when not using verbose by Chandler Carruth · 11 years ago
- 7e0f903 [x86] Hoist the logic for extracting the relevant bits of information by Chandler Carruth · 11 years ago
- 59da3f0 R600/SI: Add new helper isSGPRClassID by Matt Arsenault · 11 years ago
- 2e67962 R600/SI: Fix hardcoded and wrong operand numbers. by Matt Arsenault · 11 years ago
- 9b50273 R600/SI: Enable named operand table for SALU instructions by Matt Arsenault · 11 years ago
- 5f671ba [x86] Start refactoring the comment printing logic in the MC lowering of by Chandler Carruth · 11 years ago
- 0bb38df R600/SI: Fix weird CHECK-DAG usage by Matt Arsenault · 11 years ago
- 81c6c96 R600/SI: Enable selecting SALU inside branches by Tom Stellard · 11 years ago
- abe9b22 R600/SI: Move PHIs that define SGPRs to the VALU in most cases by Tom Stellard · 11 years ago
- 36ba796 R600/SI: Fix the FixSGPRLiveRanges pass by Tom Stellard · 11 years ago
- 90d1726 R600/SI: Mark EXEC_LO and EXEC_HI as reserved by Tom Stellard · 11 years ago
- b811241 R600/SI: Fix SIRegisterInfo::getPhysRegSubReg() by Tom Stellard · 11 years ago
- 860cabe R600/SI: Implement VGPR register spilling for compute at -O0 v3 by Tom Stellard · 11 years ago
- 6717f9d [x86] Teach the new vector shuffle lowering to lower v8i32 shuffles with by Chandler Carruth · 11 years ago
- 8415f84 [x86] Fix a really terrible bug in the repeated 128-bin-lane shuffle by Chandler Carruth · 11 years ago
- 73ce288 Fix swift-atomics testcase by Robin Morisset · 11 years ago
- 30ce74b [x86] Teach the new vector shuffle lowering to lower v4i64 vector by Chandler Carruth · 11 years ago
- 8577eaf GlobalOpt: Preserve comdats of unoptimized initializers by Reid Kleckner · 11 years ago
- bd84764 AArch64: allow constant expressions for shifted reg literals by Jim Grosbach · 11 years ago
- 798f284 [x86] Teach the rest of the 'target shuffle' machinery about blends and by Chandler Carruth · 11 years ago