- 7f1cce5 If run with -debug give more information about Cyclic Probability. by Jakub Staszak · 13 years ago
- 7241caf Heuristics are in descending priority now. If we use one of them, skip the rest. by Jakub Staszak · 13 years ago
- f6c6900 Handle REG_SEQUENCE with implicitly defined operands. by Jakob Stoklund Olesen · 13 years ago
- 43afb6f Remove obsolete FIXME reference in comment. by Jim Grosbach · 13 years ago
- 293a2ee ARM assembly parsing and encoding for BFC and BFI. by Jim Grosbach · 13 years ago
- fa44725 Add InEdges (edges from header to the loop) in Loop Branch Heuristics, so by Jakub Staszak · 13 years ago
- 1ddaa9c Use ArrayRef instead of requiring an std::vector. by Bill Wendling · 13 years ago
- 7379b66 The personality function should be a Function* and not just a Value*. by Bill Wendling · 13 years ago
- 6bd4842 Fix Clang attribute reader tblgen output for a corresponding fix on the Clang side by Douglas Gregor · 13 years ago
- cc07e04 Reverse order of RS_Split live ranges under -compact-regions. by Jakob Stoklund Olesen · 13 years ago
- b490450 Initial code to convert ResumeInsts into calls to _Unwind_Resume. by Bill Wendling · 13 years ago
- af7a46d Fix stupid mistake from commit 136381. by Jakub Staszak · 13 years ago
- 2f9427c Speed up BlockFrequencyInfo a little bit. by Jakub Staszak · 13 years ago
- c29769b Add fixme. by Jim Grosbach · 13 years ago
- ad759c5 Use version 402 for the GCDA files when compiling for Apple. by Bill Wendling · 13 years ago
- 354712c Update comments. by Owen Anderson · 13 years ago
- 8313b48 Fill in some encoding information for STRD instructions. by Owen Anderson · 13 years ago
- aa3402e Revert r136295. It broke nightly testers because some parts of codegen weren't aware of the changes to operand ordering. I hope to revive this sometime in the future, but it's not strictly necessary for now. by Owen Anderson · 13 years ago
- 70a0915 ARM parsing and encoding for ADR. by Jim Grosbach · 13 years ago
- 19b9d69 Update ARM tests for parsing and encoding of WFE, WFI and YIELD. by Jim Grosbach · 13 years ago
- 2ff4bc6 CR fix: The ANY_EXTEND can be removed because the input and putput type must be by Nadav Rotem · 13 years ago
- 5edfbeb Use unsigned rather than uint16_t in case anyone feels like testing by Duncan Sands · 13 years ago
- 6f7eec1 Check an additional property specific to the way LLVM iterates over SCC's. by Duncan Sands · 13 years ago
- 8537e8a Add a unittest for the simply connected components (SCC) iterator class. by Duncan Sands · 13 years ago
- 5a079eb Due to changes coming from the new LLVM type system, you now get by Duncan Sands · 13 years ago
- 4ed9600 This file was moved from Support to ADT. Correct a comment. by Duncan Sands · 13 years ago
- 9580641 Some minor cleanups. No functionalitical change. by Bill Wendling · 13 years ago
- 9a2dde3 Leverage some of the code that John wrote to manage the landing pads. by Bill Wendling · 13 years ago
- 7196184 Don't add in the asked for size so that we don't copy too much from the old to new vectors. by Bill Wendling · 13 years ago
- 297738f In DenseMapInfo<pair<T, U>> tombstone key, use the tombstone for T and U instead by Nick Lewycky · 13 years ago
- e8ce880 Automatically merge the landingpad clauses when we come across a callee's by Bill Wendling · 13 years ago
- 4ebc591 Explicitly declare a library dependency of LLVM*Desc to LLVM*AsmPrinter. by Oscar Fuentes · 13 years ago
- 3f01f67 Updated cmake library dependencies. by Oscar Fuentes · 13 years ago
- 28d7352 Make sure that the landingpad instruction takes a Constant* as the clause's value. by Bill Wendling · 13 years ago
- 7f66c45 Add a couple of convenience functions: by Bill Wendling · 13 years ago
- 15d03fb Invert the subvector insertion to be more likely to be taken as a COPY by Bruno Cardoso Lopes · 13 years ago
- 93fa476 Add patterns to generate copies for extract_subvector instead of by Bruno Cardoso Lopes · 13 years ago
- 735bccd movd/movq write zeros in the high 128-bit part of the vector. Use by Bruno Cardoso Lopes · 13 years ago
- a23236c Add a few patterns to match allzeros without having to use the fp unit. by Bruno Cardoso Lopes · 13 years ago
- 2e64ae4 Add SINT_TO_FP and FP_TO_SINT support for v8i32 types. Also move by Bruno Cardoso Lopes · 13 years ago
- b2e7c82 Fix a use after free. An instruction can't be both an intrinsic call and a fence. by Benjamin Kramer · 13 years ago
- 2bf84c1 Initial stab at getting inlining working with the EH rewrite. by Bill Wendling · 13 years ago
- 8050a61 ARM parsing and encoding tests. by Jim Grosbach · 13 years ago
- 814450a Add an optional 'bool makeAbsolute' in llvm::sys::fs::unique_file function. by Argyrios Kyrtzidis · 13 years ago
- 7b29583 Refactor and improve the encodings/decodings for addrmode3 loads, and make the writeback operand always the first. by Owen Anderson · 13 years ago
- 5de728c Emit an error is asm parser parsed X86_64 only registers, e.g. %rax, %sil. by Evan Cheng · 13 years ago
- ed39846 ARM assembly parsing and encoding for USUB16 and USUB8. by Jim Grosbach · 13 years ago
- 953e2e8 ARM assembly parsing and encoding for USAX. by Jim Grosbach · 13 years ago
- c37d4bb Fix llvm-mc handing of x86 instructions that take 8-bit unsigned immediates. by Kevin Enderby · 13 years ago
- fc2eb31 Clean up tabs. by Jim Grosbach · 13 years ago
- addec77 ARM assembly parsing and encoding support for USAT and USAT16. by Jim Grosbach · 13 years ago
- 5f33d13 ARM assembly parsing and encoding tests for USAD8 and USADA8. by Jim Grosbach · 13 years ago
- 1464846 Code generation for 'fence' instruction. by Eli Friedman · 13 years ago
- 144da2c ARM assembly parsing and encoding tests for UQSUB16 and UQSUB8. by Jim Grosbach · 13 years ago
- 4143839 Fix comment copy/paste-o. by Jim Grosbach · 13 years ago
- 29e85bc ARM assembly parsing and encoding tests for UQASX and UQSAX. by Jim Grosbach · 13 years ago
- 24a541b ARM assembly parsing and encoding tests for UQADD16 and UQADD8. by Jim Grosbach · 13 years ago
- 9a24f1f Use BlockFrequency instead of uint32_t in BlockFrequencyInfo. by Jakub Staszak · 13 years ago
- 49f2ced ARM assembly parsing and encoding for UMULL. by Jim Grosbach · 13 years ago
- 0748f83 Remove outdated FIXME comment. by Devang Patel · 13 years ago
- 71725a0 ARM assembly parsing and encoding for UMLAL. by Jim Grosbach · 13 years ago
- 2adba41 ARM assembly parsing and encoding tests for UMAAL. by Jim Grosbach · 13 years ago
- 91c56a8 Refuse to inline two functions which use different personality functions. by Bill Wendling · 13 years ago
- f36b0a2 ARM assembly parsing and encoding tests for UHSUB16 and UHSUB8. by Jim Grosbach · 13 years ago
- 66c8982 ARM assembly parsing and encoding tests for UHADD16, UHADD8 and UHASX. by Jim Grosbach · 13 years ago
- fb8989e ARM parsing and encoding of SBFX and UBFX. by Jim Grosbach · 13 years ago
- c94eefb Keep enums stable. Append EH stuff to the end. by Bill Wendling · 13 years ago
- b6854ad ARM assembly parsing and encoding tests for UADD16, UADD8 and UASX. by Jim Grosbach · 13 years ago
- 8b3fd56 ARM assembly parsing and encoding tests for TST instruction. by Jim Grosbach · 13 years ago
- f1ae78a ARM assembly parsing and encoding tests for TEQ instruction. by Jim Grosbach · 13 years ago
- 0647031 Refactor the STRT and STRBT instructions to distinguish between the register-addend and immediate-addend versions. Temporarily XFAIL the asm parsing tests for these instructions. by Owen Anderson · 13 years ago
- 772fe17 Merge the contents from exception-handling-rewrite to the mainline. by Bill Wendling · 13 years ago
- 7e1547e ARM assembly parsing and encoding for extend instructions. by Jim Grosbach · 13 years ago
- e511186 Teach the ConstantMerge pass about alignment. Fixes PR10514! by Nick Lewycky · 13 years ago
- 84e7f7e X86ISD::MEMBARRIER does not require SSE2; it doesn't actually generate any code, and all x86 processors will honor the required semantics. by Eli Friedman · 13 years ago
- abcd715 The numbering of LLVMOpcode is supposed to be stable; revert my earlier change, and append Fence onto the end. by Eli Friedman · 13 years ago
- 0031b4f Add test cases for BlockFrequency. by Jakub Staszak · 13 years ago
- 18dceba Add a generic 'capacity_in_bytes' function to allow inspection of memory usage of various data structures. by Ted Kremenek · 13 years ago
- 766c63e ARM assembly parsing aliases for extend instructions w/o rotate. by Jim Grosbach · 13 years ago
- 0a2a306 Update document listing DIVariable elements to reflect recent changes. by Devang Patel · 13 years ago
- 7032741 ARM cleanup of remaining extend instructions. by Jim Grosbach · 13 years ago
- c5a8c86 ARM extend instructions simplification. by Jim Grosbach · 13 years ago
- ffcc2a5 Optimize 96-bit division a little bit. by Jakub Staszak · 13 years ago
- 636a02b Move static methods to the anonymous namespace. by Jakub Staszak · 13 years ago
- 9d9f765 Edge to itself is backedge as well. by Jakub Staszak · 13 years ago
- a09d514 Trim includes. by Frits van Bommel · 13 years ago
- 0da49df Update CMake build for new gtest file. by Frits van Bommel · 13 years ago
- c437bd5 Remove some code that is no longer needed now that googletest knows how by Jay Foad · 13 years ago
- b33f8e3 Merge gtest-1.6.0. by Jay Foad · 13 years ago
- a44defe Explicitly cast narrowing conversions inside {}s that will become errors in by Jeffrey Yasskin · 13 years ago
- cbdccde Revert r136156, which broke several buildbots. by Dan Gohman · 13 years ago
- 8a552bb Misc mid-level changes for new 'fence' instruction. by Eli Friedman · 13 years ago
- 751bef7 Minor simplification. by Eli Friedman · 13 years ago
- 9b4ad12 Move some code around to open opportunity for more shuffle matching by Bruno Cardoso Lopes · 13 years ago
- cea34e4 The vpermilps and vpermilpd have different behaviour regarding the by Bruno Cardoso Lopes · 13 years ago
- cd9e5ae Remove more dead code! by Bruno Cardoso Lopes · 13 years ago
- 6f3ba37 Fix AliasSetTracker so that it doesn't make any assumptions about instructions it doesn't know about (like the atomic instructions I'm adding). by Eli Friedman · 13 years ago
- bd27f5a Support .code32 and .code64 in X86 assembler. by Evan Cheng · 13 years ago
- 26a9200 It is quiet possible that inlined function body is split into multiple chunks of consequtive instructions. But, there is not any way to describe this in .debug_inline accelerator table used by gdb. However, describe non contiguous ranges of inlined function body appropriately using AT_range of DW_TAG_inlined_subroutine debug info entry. by Devang Patel · 13 years ago
- 2817619 Remove these two directories. The tests can be ported to dragonegg if by Eric Christopher · 13 years ago