- 2202360 Fix an apparent copy-and-paste problem in an error message. by Bob Wilson · 16 years ago
- c680ac9 Make EXTRACT_VECTOR_ELT a bit more flexible in terms of the returned by Eli Friedman · 16 years ago
- 5dcaceb As Chris pointed out, we don't actually need to pass the context around here. by Owen Anderson · 16 years ago
- d1474d0 Thread LLVMContext through MVT and related parts of SDISel. by Owen Anderson · 16 years ago
- 2e4284d Make SelectionDAG::getVectorShuffle work properly for VECTOR_SHUFFLE by Dan Gohman · 16 years ago
- f1daf7d Use common code for both ARM and Thumb-2 instruction and register info. by David Goodwin · 16 years ago
- f2d754b Nowadays vectors are only split if they have an even by Duncan Sands · 16 years ago
- 8c899ee Remove trailing whitespace. Reorder some methods by Duncan Sands · 16 years ago
- 7f6aa2b Remove the vicmp and vfcmp instructions. Because we never had a release with by Nick Lewycky · 16 years ago
- 2b7a271 dag combine sext(setcc) -> vsetcc before legalize. To make this safe, by Chris Lattner · 16 years ago
- a4f7318 SelectionDAG::SignBitIsZero doesn't work right for vectors, by Chris Lattner · 16 years ago
- 8ea5ec6 Operand of asm("call") (the callee function) is represented by Dale Johannesen · 16 years ago
- c2c27b3 add support for legalizing an icmp where the result is illegal (4xi1) but by Chris Lattner · 16 years ago
- 0a6c2d8 random code cleanups. by Chris Lattner · 16 years ago
- 5962ed0 implement support for spliting and scalarizing vector setcc's. This by Chris Lattner · 16 years ago
- 9800e84 lower vector icmp/fcmp to ICMP/FCMP nodes with the right result by Chris Lattner · 16 years ago
- 87c411b ScalarizeVecRes_ShiftOp and ScalarizeVecRes_BinOp are the same, by Chris Lattner · 16 years ago
- 64a3fa2 add support for vector legalizing of *_EXTEND. by Chris Lattner · 16 years ago
- a9d1f2c Have scoped mutexes take referenes instead of pointers. by Owen Anderson · 16 years ago
- 6b61cd1 Add NumFixedArgs attribute to CallSDNode which indicates the number of fixed arguments in a vararg call. by Tilmann Scheller · 16 years ago
- 7e1e31f Simplify debug info intrisinc lowering. by Devang Patel · 16 years ago
- 033080c CMake build fixes, from Xerxes Ranby by Douglas Gregor · 16 years ago
- 6d8f126 Simplify. by Devang Patel · 16 years ago
- 07b0ec0 Simplify. No intentional functionality change. by Devang Patel · 16 years ago
- ceddbe8 Refactor. No functionality change. by Devang Patel · 16 years ago
- 30d7b65 llvm.dbg.declare is always used for local variable's debug info. by Devang Patel · 16 years ago
- 4784f1f Add a bit IsUndef to MachineOperand. This indicates the def / use register operand is defined by an implicit_def. That means it can def / use any register and passes (e.g. register scavenger) can feel free to ignore them. by Evan Cheng · 16 years ago
- 5b93ea1 fix a typo that GCC should have caught that causes crashes with -view-*-dags by Chris Lattner · 16 years ago
- 505e982 fix a really subtle bug in the cross section of aliases and TLS: by Chris Lattner · 16 years ago
- d212bb8 implement DOTGraphTraits<SelectionDAG*>::getNodeLabel in terms of by Chris Lattner · 16 years ago
- 7bea038 dot graph viewing is apparently not using SDNode::print_details, this is bad, by Chris Lattner · 16 years ago
- 6ec66db propagate target operand flags from dag nodes into MachineOperands. by Chris Lattner · 16 years ago
- 5202312 fit in 80 cols by Chris Lattner · 16 years ago
- f5a5546 add targetflags to jump tables and constant pool entries. by Chris Lattner · 16 years ago
- 2a4ed82 allow setting target operand flags on TargetGlobalAddress nodes. by Chris Lattner · 16 years ago
- 1af2231 start bringing targetoperand flags into isel, first up, ExternalSymbol. by Chris Lattner · 16 years ago
- b445908 Provide guards for this shared structure. I'm not sure this actually needs by Owen Anderson · 16 years ago
- f2e19d5 by David Greene · 16 years ago
- 8cbc94a Get rid of the global CFGOnly flag by threading a ShortNames parameters through the GraphViz rendering code. by Owen Anderson · 16 years ago
- 8539cfd Rewrite 73900 per Duncan's suggestion. by Dale Johannesen · 16 years ago
- 929a493 remove dead makefile flags. by Chris Lattner · 16 years ago
- 08bc98e Fix memcpy expansion so it won't generate invalid by Dale Johannesen · 16 years ago
- 1e86a66 mv CodeGen/DebugLoc.h Support/DebugLoc.h by Devang Patel · 16 years ago
- 2fb070b Minor cleanup; fixes review comments for a previous patch. Sorry for by Eli Friedman · 16 years ago
- 7d8d36a Fixed names of libcalls checked in r73480. by Sanjiv Gupta · 16 years ago
- 8aa207e Added required libcalls for PIC16 (mostly floating points to integer casting operations). by Sanjiv Gupta · 16 years ago
- db3c169 Add some generic expansion logic for SMULO and UMULO. Fixes UMULO by Eli Friedman · 16 years ago
- 15480bd Change this from an assert to a cerr+exit, since it's diagnosing an by Dan Gohman · 16 years ago
- 11a407f Gracefully handle imbalanced inline function begin and end markers. by Devang Patel · 16 years ago
- 5d2c01e CheckTailCallReturnConstraints is missing a check on the by Arnold Schwaighofer · 16 years ago
- f10d3a7 Tweak the expansion code for BIT_CONVERT to generate better code by Eli Friedman · 16 years ago
- 7a5e555 Slightly generalize the code that handles shuffles of consecutive loads by Eli Friedman · 16 years ago
- 5257106 Fix the expansion for CONCAT_VECTORS so that it doesn't create illegal types. by Eli Friedman · 16 years ago
- 7ef3d17 Factor out a couple of helpers. by Eli Friedman · 16 years ago
- 556929a Make SINT_TO_FP/UINT_TO_FP vector legalization queries query on the by Eli Friedman · 16 years ago
- 578efa9 Add new function attribute - noimplicitfloat by Devang Patel · 16 years ago
- abc0199 Adapt the x86 build_vector dagcombine to the current state of the legalizer. by Nate Begeman · 16 years ago
- a43a7ae Allow libcalls for i16 sdiv/udiv/rem operations. by Sanjiv Gupta · 16 years ago
- ae3a0be Split the Add, Sub, and Mul instruction opcodes into separate by Dan Gohman · 16 years ago
- 4c9369d Fix FP_TO_UINT->i32 on ppc32 -mcpu=g5. This was by Dale Johannesen · 16 years ago
- 77b81fe Don't do the X * 0.0 -> 0.0 transformation in instcombine, because by Dan Gohman · 16 years ago
- 68f32cb Fix comments. by Dan Gohman · 16 years ago
- 874ae25 Revert 72707 and 72709, for the moment. by Dale Johannesen · 16 years ago
- 4150d83 Make the implicit inputs and outputs of target-independent by Dale Johannesen · 16 years ago
- f43071b Rename CustomLowerResults to CustomLowerNode, since by Duncan Sands · 16 years ago
- 51b16f4 Untabification. by Bill Wendling · 16 years ago
- d3c76bb Do not try to create a MVT type of width 0. by Evan Cheng · 16 years ago
- bbdd903 Re-commit r72514 and r72516 with a fixed version of BR_CC lowering. by Eli Friedman · 16 years ago
- cdcecc0 Incorporate patch feedbacks. by Evan Cheng · 16 years ago
- 43b4127 Temporarily revert r72514 (and dependent patch r72516). It was causing this by Bill Wendling · 16 years ago
- e727d7a Remove a couple of useless functions. by Eli Friedman · 16 years ago
- 1c39965 Remove special cases for more opcodes. by Eli Friedman · 16 years ago
- ad75460 Remove special case for SETCC opcode; add some comments explaining why by Eli Friedman · 16 years ago
- 3be2e51 Some minor cleanups. by Eli Friedman · 16 years ago
- 8b944d3 Added optimization that narrow load / op / store and the 'op' is a bit twiddling instruction and its second operand is an immediate. If bits that are touched by 'op' can be done with a narrower instruction, reduce the width of the load and store as well. This happens a lot with bitfield manipulation code. by Evan Cheng · 16 years ago
- b5da3f6 Minor cleanups; add a better explanation for the issue with BUILD_VECTOR. by Eli Friedman · 16 years ago
- 4bc8c71 Remove more special cases for opcodes. by Eli Friedman · 16 years ago
- 509150f Remove special cases for more opcodes. by Eli Friedman · 16 years ago
- f6f20a7 Removing more special cases from LegalizeDAG. by Eli Friedman · 16 years ago
- 26ea8f9 Eliminate more special cases for opcodes. by Eli Friedman · 16 years ago
- f6b23bf Remove more special cases from LegalizeDAG. by Eli Friedman · 16 years ago
- 47b41f7 Remove unused argument. by Eli Friedman · 16 years ago
- 3f727d6 Remove more opcode special cases. by Eli Friedman · 16 years ago
- 8c377c7 Start of refactoring LegalizeDAG so that we don't need specialized by Eli Friedman · 16 years ago
- 74807f2 Delete a bunch of dead code from LegalizeDAG. by Eli Friedman · 16 years ago
- c046c00 Add a comment which should hopefully make the purpose of this method a by Eli Friedman · 16 years ago
- 57f1a4b Minor improvement to FCOPYSIGN to use BIT_CONVERT in cases where the by Eli Friedman · 16 years ago
- ed2f8c5 Rewrite ISD::FCOPYSIGN lowering to never use i64. Not really ideal, but by Eli Friedman · 16 years ago
- 7d869fb Update for CMakeLists; untested, so tell me if there are issues. by Eli Friedman · 16 years ago
- 957bffa Remove checks of getTypeAction from LegalizeOp; we already assert that by Eli Friedman · 16 years ago
- 1fde9c5 Disable type legalization in LegalizeDAG. by Eli Friedman · 16 years ago
- 2a35b1c Fix a bug in the expansion of EXTRACT_SUBVECTOR in by Eli Friedman · 16 years ago
- 3d43b3f Add a proper implementation of EXTRACT_SUBVECTOR legalization that by Eli Friedman · 16 years ago
- 6bb4958 Fix PR4254. by Torok Edwin · 16 years ago
- 5c22c80 Add a new step to legalization to legalize vector math operations. This by Eli Friedman · 16 years ago
- b0f1e17 Add a new codegen pass that normalizes dwarf exception handling by Duncan Sands · 16 years ago
- e3e51c0 Use v.data() instead of &v[0] when SmallVector v might be empty. by Jay Foad · 16 years ago
- df7d5d3 Temporarily revert r72191. It was causing an assert during llvm-gcc by Bill Wendling · 16 years ago
- a343764 Introduce DebugScope which gets embedded into the machine instructions' DebugLoc. by Argyrios Kyrtzidis · 16 years ago
- 2ac8b32 Fix for PR4235: to build a floating-point value from integer parts, by Eli Friedman · 16 years ago