- e68ea06 Fix debug_value handling. by Dale Johannesen · 15 years ago
- 1f3e445 MC: Provide MCAssembler with a TargetAsmBackend. by Daniel Dunbar · 15 years ago
- 23ac7c7 MC: Sketch some TargetAsmBackend hooks we are going to need. by Daniel Dunbar · 15 years ago
- 6c27f5e MC: Provide the target triple to AsmBackend constructors. by Daniel Dunbar · 15 years ago
- bd97376 Add strncpy libcall creator. Use it when it should be used. by Eric Christopher · 15 years ago
- 52a50e5 When outputing a non-lazy pointer for a stub, we may need to fill in the value by Bill Wendling · 15 years ago
- ae1757b Added Thumb2 LDRD/STRD pre/post variants for disassembly only. by Johnny Chen · 15 years ago
- ae47e46 Make sure HasDebugValue is initialized. This should fix by Dale Johannesen · 15 years ago
- 5265a12 Fix ARM buildbot breakage. by Bob Wilson · 15 years ago
- 2968943 add support, testcases, and dox for the new GHC calling by Chris Lattner · 15 years ago
- 4d6113e Lower small memcpys to load/stores on Thumb2. Radar 7686922. by Bob Wilson · 15 years ago
- 2cb4202 VirtRegRewriter spring cleaning. No functional change. by Jakob Stoklund Olesen · 15 years ago
- bcc5cb4 Work around a bug in the openbsd assembler on i386, by Chris Lattner · 15 years ago
- f2f64e9 fix PR6533 by updating the br(xor) code to remember the case by Chris Lattner · 15 years ago
- e3b8533 Cosmetic: lengthen names and improve comments. by Dale Johannesen · 15 years ago
- fbacc88 Fix an obvious typo in an assert. Patch by Sean Callanan. by Bob Wilson · 15 years ago
- cebae36 Add a bit along with the MCSymbols stored in the MachineModuleInfo maps that by Bill Wendling · 15 years ago
- bfdf7f3 Progress towards shepherding debug info through SelectionDAG. by Dale Johannesen · 15 years ago
- 6663670 move PR6576 here. by Chris Lattner · 15 years ago
- 97d7b10 add DESTDIR support for TOOLALIAS, for PR6557, patch by Matthias Klose! by Chris Lattner · 15 years ago
- 8f78e3c Fix another bitwidth calculation to handle vector types; based on a by Dan Gohman · 15 years ago
- 2101d9c MC/Mach-O: Use the MCAssembler symbol map instead of reconstructing. by Daniel Dunbar · 15 years ago
- 46836a7 MC: Move the backend section and symbol data maps to MCAssembler. by Daniel Dunbar · 15 years ago
- d80066e Remove unneeded declarations. by Daniel Dunbar · 15 years ago
- f49be7c Make sure the LR gets pushed in functions that use vaargs. This fixes by Jim Grosbach · 15 years ago
- 004453e comment why we use custom epilogue for t1 functions using vaargs. by Jim Grosbach · 15 years ago
- 2cd9ffe Fix another place where DEBUG_VALUE affected codegen. by Dale Johannesen · 15 years ago
- bbf81d8 Add a DominatorTree argument to isLCSSA so that it doesn't have to by Dan Gohman · 15 years ago
- 0891d75 Constant-fold GEP-of-GEP into a single GEP. by Dan Gohman · 15 years ago
- a773384 Fix whitespace. by Dan Gohman · 15 years ago
- dd0f3cf Factored out the disassembly printing of CPS option, MSR mask, and Negative Zero by Johnny Chen · 15 years ago
- cbb524e Fix make check with cmake/lit by Tobias Grosser · 15 years ago
- 72926c2 The backend now makes a reasonable job of targeting lmul / macc by Richard Osborne · 15 years ago
- 850f1cd Handle MVT::i64 type in DAG combine for ISD::ADD. Fold 64 bit by Richard Osborne · 15 years ago
- 0965200 Remove unneeded includes. by Daniel Dunbar · 15 years ago
- c81f0af Testcase for pr6552. I changed the code to use "ip" instead of "fp" because by Bob Wilson · 15 years ago
- ea670df Fix checking of intermediates having one use in isADDADDMUL by Richard Osborne · 15 years ago
- 886d861 Extract recognition of patterns such as add(add(mul(x,y),a),b) by Richard Osborne · 15 years ago
- 25ab35e Fix thinko. by Richard Osborne · 15 years ago
- 2dacd1e Fold add(add(mul(x,y),a),b) -> lmul(x,y,a,b) if the intermediate by Richard Osborne · 15 years ago
- 48ee0bb Remove duplicated code. No functionality change. by Benjamin Kramer · 15 years ago
- a2f767b Fix a bug in DEBUG_VALUE handling Devang ran into. I'll get this loop right yet. by Dale Johannesen · 15 years ago
- a2cc061 Prefer LMUL to MACCU as LMUL has no tied operands. by Richard Osborne · 15 years ago
- a7e7840 Custom lower (S|U)MUL_LOHI -> MACC(S|U) by Richard Osborne · 15 years ago
- 34f9d88 Fix indentation by Richard Osborne · 15 years ago
- ebc64cf Lower add (mul a, b), c into MACCU / MACCS nodes which translate by Richard Osborne · 15 years ago
- 1250ac8 Convert test to FileCheck. by Richard Osborne · 15 years ago
- 09d53fe move three lowering hooks from MAI to TLOF and make one of them by Chris Lattner · 15 years ago
- 3a62422 Fix typo. by Evan Cheng · 15 years ago
- e025908 Unbreak test on Linux. by Evan Cheng · 15 years ago
- c5cf227 This survived a bootstrap, so let's try 98104 again. by Dale Johannesen · 15 years ago
- fac4f1f Enable machine cse pass. by Evan Cheng · 15 years ago
- 974c0fb mcize the rest of EH emission, only one more directive missing by Chris Lattner · 15 years ago
- 49f618a add missing filename! by Chris Lattner · 15 years ago
- 98cdab5 set the temporary bit on MCSymbols correctly. by Chris Lattner · 15 years ago
- 2d0a91c Fix a comment. by Dan Gohman · 15 years ago
- 2938a00 Add a couple more heuristics to neuter machine cse some more. by Evan Cheng · 15 years ago
- 07a9641 MC/Mach-O: Resolve a FIXME; these relocation types are no longer semanticaly different. by Daniel Dunbar · 15 years ago
- 9f116aa Clarify the documentation for MachineFunctionPasses. by Dan Gohman · 15 years ago
- 00685bb eliminate MCContext::CreateSymbol and CreateTemporarySymbol. by Chris Lattner · 15 years ago
- d85fc6e inline away a form of IsPCRelative, eliminating the dead IsPCRel argument. by Chris Lattner · 15 years ago
- 188a87d add some fixme's for MCizing. EH still has a few things that by Chris Lattner · 15 years ago
- a015c1c MC/Mach-O: Use the SECTDIFF relocation type for (A - B + constant) where A is external. by Daniel Dunbar · 15 years ago
- 7c617b5 Clear up the last (famous last words) frame index value reuse issues for Thumb1. by Jim Grosbach · 15 years ago
- 7dda453 Speculatively revert 98104; could be what's causing crashes by Dale Johannesen · 15 years ago
- 90e4af7 mcize uses of PrintRelDirective and eliminate it. by Chris Lattner · 15 years ago
- 4aeaca4 inline the bool form of PrintRelDirective away, leaving just the unsigned form. by Chris Lattner · 15 years ago
- 233f52b eliminate EOL, adding all comments with the OutStreamer.AddComment by Chris Lattner · 15 years ago
- 7f87696 Ever more complicated DEBUG_VALUE fixes for branch folding. by Dale Johannesen · 15 years ago
- 4ecbca5 Avoid analyzing instructions in blocks not reachable from the entry block. by Dan Gohman · 15 years ago
- 0d9d70f eliminate a bunch of \n's that are being printed to O. Next up is to kill by Chris Lattner · 15 years ago
- fcdae49 convert the non-"ispcrel" case of EmitReference to MC, by Chris Lattner · 15 years ago
- 2e5dd9b make the NullStreamer set the section on a label when emitted so that isDefined() works. by Chris Lattner · 15 years ago
- f747747 Try to keep the cached inliner costs around for a bit longer for big functions. by Jakob Stoklund Olesen · 15 years ago
- f291be3 MC/X86: Rename alternate spellings of ADD{8,16,32} and mark as "code gen only" so they don't get selected by the asm matcher. by Daniel Dunbar · 15 years ago
- 1e8ee89 MC/X86: Rename alternate spellings of CMP{8,16,32} and mark as "code gen only" so they don't get selected by the asm matcher. by Daniel Dunbar · 15 years ago
- 315b1bb Continue propagating the GoogleTest flags until we can update our version to by Chandler Carruth · 15 years ago
- b5a158b Revert r98089, it was breaking a clang test. by Jakob Stoklund Olesen · 15 years ago
- 9e5d87d Try to keep the cached inliner costs around for a bit longer for big functions. by Jakob Stoklund Olesen · 15 years ago
- 74a6842 Permit inlining into huge functions. This heuristic is ancient, and inlining by Jakob Stoklund Olesen · 15 years ago
- dff4b4c Change the Value argument to eliminateFrameIndex to a type-tagged value. This by Jim Grosbach · 15 years ago
- a2c618a MSR (Move to Special Register from ARM core register) requires a mask to specify by Johnny Chen · 15 years ago
- f3a066f MC/Mach-O: For PCrel relocations, we need to compensate for the PCrel adjustment when determining if we need a scattered relocation. by Daniel Dunbar · 15 years ago
- aef9d7a MC/Mach-O: Also set the PCrel bit in the second half of paired relocation entries. by Daniel Dunbar · 15 years ago
- e180fa9 MC/Mach-O: Don't generate relocations for PCrel fixups to local labels. by Daniel Dunbar · 15 years ago
- f7b7304 The address of an indirect call must be in R12 on Darwin. by Dale Johannesen · 15 years ago
- f10e43e Consolidate GoogleTest make options and duplicate them to its own makefile. by Chandler Carruth · 15 years ago
- 4e50154 scavenged frame index value re-use gets confused when more than one base by Jim Grosbach · 15 years ago
- bdc38e5 The ARM EH experiment worked! by Bill Wendling · 15 years ago
- ad4f66c In cases where the carry / borrow unused converted ladd / lsub by Richard Osborne · 15 years ago
- 78717cd Canonicalize ladd constant to RHS. by Richard Osborne · 15 years ago
- 7b871b3 Add DAG combine for ladd / lsub. by Richard Osborne · 15 years ago
- 7ebc377 Attempt to fix random build failures seen when doing highly by Duncan Sands · 15 years ago
- bfc9999 Allow more cross-rc coalescing. by Evan Cheng · 15 years ago
- c608784 reapply r98035: by Chris Lattner · 15 years ago
- 6c7dfc0 add some extra checks. I'm not sure why, but this does unbreak a by Chris Lattner · 15 years ago
- 7ef1840 Revert accidental commit. by Evan Cheng · 15 years ago
- 1552ccc Don't do illegal cross-class coalescing. by Jakob Stoklund Olesen · 15 years ago
- 31f94c7 - Make the machine cse dumb coalescer (as opposed to the more awesome simple by Evan Cheng · 15 years ago
- 81d0c36 Don't try to fold V_SET0 and V_SETALLONES to loads in medium and by Dan Gohman · 15 years ago