- 790be31 InstCombine: salvage the debug info of DCE'ed add instructions. by Adrian Prantl · 8 years ago
- acf8758 [X86] Make FeatureAVX512 imply FeatureF16C. by Craig Topper · 8 years ago
- 490bc39 [X86] Make FeatureAVX512 imply FeatureFMA. by Craig Topper · 8 years ago
- 618cf29 [ValueTracking] readonly (const) is a requirement for converting sqrt to llvm.sqrt; nnan is not by Sanjay Patel · 8 years ago
- e209a1a Revert r317510 "[InstCombine] Pull shifts through a select plus binop with constant" by Hans Wennborg · 8 years ago
- 2c99ae8 Revert r316064 "Fix the incorrect detection of ICONV_LIBRARY_PATH" by Hans Wennborg · 8 years ago
- cebfaaf Fix comment /NFC by Xinliang David Li · 8 years ago
- 4cbab70 [MIRPrinter] Use %subreg.xxx syntax for subregister index operands by Bjorn Pettersson · 8 years ago
- aceaaf0 [InstCombine] Pull shifts through a select plus binop with constant by Craig Topper · 8 years ago
- 82cd943 update_mir_test_checks: Be careful about replacing entire vregs by Justin Bogner · 8 years ago
- e005ea7 Fix buildbot breakages from r317503. Add parentheses to assignment when using result as a condition. by Graham Yiu · 8 years ago
- 20a90ab Adds code to PPC ISEL lowering to recognize byte inserts from vector_shuffles, and use P9 shift and vector insert byte instructions instead of vperm. Extends tests from vector insert half-word. by Graham Yiu · 8 years ago
- 561a742 Include already promoted counts when computing SUM for VP. by Dehao Chen · 8 years ago
- 0ec29cb [cfi-verify] Added a simple check that stops division-by-zero error when no indirect CF instructions are found in the provided file. by Mitch Phillips · 8 years ago
- 416cdcc [PPC] Use xxbrd to speed up bswap64 by Guozhi Wei · 8 years ago
- 6e02997 Make MCAsmBackend and MCCodeEmiiter passed by unique_ptr rval by Mitch Phillips · 8 years ago
- bd04b64 AMDGPU: Select v_mad_u64_u32 and v_mad_i64_i32 by Matt Arsenault · 8 years ago
- 94bed7d Canonicalize spelling of long-form-options in dsymutil.rst by Adrian Prantl · 8 years ago
- 00e900a [IR] redefine 'UnsafeAlgebra' / 'reassoc' fast-math-flags and add 'trans' fast-math-flag by Sanjay Patel · 8 years ago
- ae8a000 [X86][SSE] Merge combineExtractVectorElt_SSE into combineExtractVectorElt. NFCI. by Simon Pilgrim · 8 years ago
- d27af43 [SLP] Test for PR35047, NFC. by Alexey Bataev · 8 years ago
- 6e1c5e0 [X86][SSE] Combine EXTRACT_VECTOR_ELT with combineExtractWithShuffle before XFormVExtractWithShuffleIntoLoad by Simon Pilgrim · 8 years ago
- ec1f0cc [AMDGPU] Change alloca addr space of r600 to 5 for amdgiz environment by Yaxun Liu · 8 years ago
- 044ea89 [SystemZ] implement hasDivRemOp() by Jonas Paulsson · 8 years ago
- 13a223e [AMDGPU] Fix assertion due to assuming pointer in default addr space is 32 bit by Yaxun Liu · 8 years ago
- dfaa4d2 [mips] Add movep for microMIPS32R6 and fix microMIPS32r3 version by Simon Dardis · 8 years ago
- 2b2534c [docs] Update code block for compatibility with Sphinx 1.5.1 by Jonas Devlieghere · 8 years ago
- 84066df [LV][X86] update the cost of interleaving mem. access of floats by Mohammed Agabaria · 8 years ago
- 194c54b [mips] Fix PR35140 by Simon Dardis · 8 years ago
- d64a599 Fixed dead links in WritingAnLLVMPass.rst by Raphael Isemann · 8 years ago
- d8ea264 [X86][AVX512] Improve lowering of AVX512 test intrinsics by Uriel Korach · 8 years ago
- 16b230f [X86] Replace duplicate function call with variable. NFC by Uriel Korach · 8 years ago
- 807235c X86 ISel: Basic support for variable-index vector permutations by Zvi Rackover · 8 years ago
- 150ca11 Revert "adding a pattern for broadcastm" by Jina Nahias · 8 years ago
- 708a132 [test] Add test files that were missed from SVN r317459 by Martin Storsjo · 8 years ago
- 8738ed4 [ObjectYAML] Map relocation types for COFF ARMNT and ARM64 by Martin Storsjo · 8 years ago
- 028edca [x86][AVX512] Lowering Broadcastm intrinsics to LLVM IR by Jina Nahias · 8 years ago
- 2513e7c adding a pattern for broadcastm by Jina Nahias · 8 years ago
- f9f3d36 [X86] Use EVEX encoded intrinsics for legacy FMA intrinsics when possible. by Craig Topper · 8 years ago
- af14030 [X86] Add scalar FMA ISD nodes without rounding mode. NFC by Craig Topper · 8 years ago
- 9b9ae41 [X86] Add avx512vl command line to fma-instrinsics-x86.ll by Craig Topper · 8 years ago
- 1053713 [X86] Simplify command lines on the fma-instrinsics-x86.ll test and add -show-mc-encoding. by Craig Topper · 8 years ago
- 67a8012 [X86] Use EVEX encoded instructions for legacy scalar sqrt intrinsics. by Craig Topper · 8 years ago
- c8200b7 [PassManager, SimplifyCFG] Revert r316908 and r316869. by David L. Jones · 8 years ago
- d864a8d [X86] Add missing predicate to a pattern. NFC by Craig Topper · 8 years ago
- 3594a87 [X86] Remove some more RCP and RSQRT patterns from InstrAVX512.td that I missed in r317413. by Craig Topper · 8 years ago
- f99fcce [X86] Fix outdated comment. NFC by Craig Topper · 8 years ago
- 5606961 [X86][SSE] Tests for integer min/max horizontal reductions by Simon Pilgrim · 8 years ago
- ee5e318 by Dorit Nuzman · 8 years ago
- ee9947c [SLPVectorizer] minimize tests and auto-generate full checks; NFC by Sanjay Patel · 8 years ago
- 08eb02a [REVERT][LV][X86] update the cost of interleaving mem. access of floats by Mohammed Agabaria · 8 years ago
- 1511d2e [LV][X86] update the cost of interleaving mem. access of floats by Mohammed Agabaria · 8 years ago
- d346941 [CGP] Fix the bug found by asan. by Serguei Katkov · 8 years ago
- 6479e73 [CGP] Extends the scope of optimizeMemoryInst optimization. NFC by Serguei Katkov · 8 years ago
- 30669bb [CGP] Extends the scope of optimizeMemoryInst optimization by Serguei Katkov · 8 years ago
- c80a0ef [X86][AVX] Regenerate test. NFCI. by Simon Pilgrim · 8 years ago
- 73fca57 Use code voice for DIBuilder in LLVM C API by Harlan Haskins · 8 years ago
- 60528d6 Move the llvm-tblgen project into the Tablegenning folder on IDEs like Visual Studio rather than leave it in the root directory. NFC. by Aaron Ballman · 8 years ago
- 3639f55 Move the srpm, ocaml_make_directory, llvm_vcsrevision_h, and llvm-headers projects into the Misc folder on IDEs like Visual Studio rather than leave them in the root directory. NFC. by Aaron Ballman · 8 years ago
- 33cfb89 Move the LLVMCFIVerify project into the Libraries folder on IDEs like Visual Studio rather than leave it in the root directory. NFC. by Aaron Ballman · 8 years ago
- 010bb80 Move these CMake projects into the Tests folder on IDEs like Visual Studio rather than leave it in the root directory. NFC. by Aaron Ballman · 8 years ago
- 1b23a75 [X86] Don't use RCP14 and RSQRT14 for reciprocal estimations or for legacy SSE rcp/rsqrt intrinsics when AVX512 features are enabled. by Craig Topper · 8 years ago
- cf0e1f6 [X86] Regenerate a couple more tests that I missed in r317410. by Craig Topper · 8 years ago
- e775c34 [X86] Teach EVEX->VEX pass to turn SHUFI32X4/SHUFF32X4/SHUFI64X/SHUFF64X2 into VPERM2F128/VPERM2I128. by Craig Topper · 8 years ago
- 35934f8 [AMDGPU] Remove hardcoded address space value from AMDGPULibFunc by Yaxun Liu · 8 years ago
- dcf1ffe [LTO][ThinLTO] Use the linker resolutions to mark global values as dso_local. by Sean Fertile · 8 years ago
- ce4da27 llvm/test/lit.cfg.py: Don't set the feature "llvm-64-bits" if -m32 is specified. by NAKAMURA Takumi · 8 years ago
- 19bc3f9 [X86] Teach shuffle lowering to use 256-bit SHUF128 when possible. by Craig Topper · 8 years ago
- 5473af6 CMake: Let LLVM_BUILD_32_BITS aware of large file. by NAKAMURA Takumi · 8 years ago
- ac439ba llvm/test/Object/archive-SYM64-write.test: Delete large temp files. They are 8GiB total. by NAKAMURA Takumi · 8 years ago
- f1b2e0b Revert "[LTO][ThinLTO] Use the linker resolutions to mark global values ..." by Sean Fertile · 8 years ago
- f073293 [CallSiteSplitting] clang-format my last commit. NFCI. by Davide Italiano · 8 years ago
- 1024a37 [CallSiteSplitting] Silence GCC's -Wparentheses. NFCI. by Davide Italiano · 8 years ago
- be2858c [X86] Give unary PERMI priority over SHUF128 in lowerV8I64VectorShuffle to make it possible to fold a load. by Craig Topper · 8 years ago
- 803f827 Move TargetFrameLowering.h to CodeGen where it's implemented by David Blaikie · 8 years ago
- 29c52e4 Invoke salvageDebugInfo from CodeGenPrepare's SinkCast() by Adrian Prantl · 8 years ago
- cdc5782 [LTO][ThinLTO] Use the linker resolutions to mark global values as dso_local. by Sean Fertile · 8 years ago
- ce8f24e Fix a crash in llvm-objdump when printing a bad x86_64 relocation in a Mach-O by Kevin Enderby · 8 years ago
- 9cf32a0 Revert r317046, "Object: Move some code from ELF.h into ELF.cpp." by Peter Collingbourne · 8 years ago
- 4fedc84 [SimplifyCFG] When merging conditional stores, don't count the store we're merging against the PHINodeFoldingThreshold by Craig Topper · 8 years ago
- 7711c31 GCOV: Move GCOV from IR & Support into ProfileData to fix layering by David Blaikie · 8 years ago
- 352adf2 llvm-objdump: Fix unused-lambda-capture warning by removing unused lambda capture by David Blaikie · 8 years ago
- 79eed69 [cfi-verify] Add blacklist parsing for result filtering. by Mitch Phillips · 8 years ago
- f4beb75 Recommit r317351 : Add CallSiteSplitting pass by Jun Bum Lim · 8 years ago
- ceb5b1b Modularize: Include some required headers by David Blaikie · 8 years ago
- af481e4 [llvm-ar] Support an options string that start with a dash by Martin Storsjo · 8 years ago
- 2619256 Correcting some CRLFs that snuck in with my previous commit; NFC. by Aaron Ballman · 8 years ago
- bdc30c02 Add llvm::for_each as a range-based extensions to <algorithm> and make use of it in some cases where it is a more clear alternative to std::for_each. by Aaron Ballman · 8 years ago
- b72a3a9 [cfi-verify] Add an interesting unit test where undef search length changes result. by Mitch Phillips · 8 years ago
- b24883f [X86] Promote athlon, athlon-xp, k8, and k8-sse3 to types instead of subtypes in getHostCPUName. NFCI by Craig Topper · 8 years ago
- c86c85f Revert "Add CallSiteSplitting pass" by Jun Bum Lim · 8 years ago
- 06dbf5a Reland "Add support for writing 64-bit symbol tables for archives when offsets become too large for 32-bit" by Jake Ehrlich · 8 years ago
- 1b91c5e Add CallSiteSplitting pass by Jun Bum Lim · 8 years ago
- a8631b8 [llvm-objcopy] Add support for dwarf fission by Jake Ehrlich · 8 years ago
- 8f80505 [AArch64] Fix the number of iterations for the Newton series by Evandro Menezes · 8 years ago
- aba0da1 The patch fixes PR35131 by Evgeny Stupachenko · 8 years ago
- 6a8da4f Revert "Invoke salvageDebugInfo from CodeGenPrepare's SinkCast()" by Adrian Prantl · 8 years ago
- aaf1db1 [CodeGen] Remove unnecessary semicolons to fix a warning. NFC by Craig Topper · 8 years ago
- 761cb9c [X86] Initialize Type and Subtype in getHostCPUName to 0. by Craig Topper · 8 years ago
- 604f04f Invoke salvageDebugInfo from CodeGenPrepare's SinkCast() by Adrian Prantl · 8 years ago