- 2ff7fd1 Refactor X86SelectConstAddr, folding it into X86SelectAddress. This by Dan Gohman · 17 years ago
- ee2e403 Add a new "fast" scheduler. This is currently basically just a by Dan Gohman · 17 years ago
- 91aac10 Add a bit to mark operands of asm's that conflict by Dale Johannesen · 17 years ago
- 5993258 Don't worry about clobbering physical register defs that aren't used. by Dan Gohman · 17 years ago
- 1cd3327 When converting a CopyFromReg to a copy instruction, use the register class of its uses to determine the right destination register class of the copy. This is important for targets where a physical register may belong to multiple register classes. by Evan Cheng · 17 years ago
- 50284d8 Change SelectionDAG::getConstantPool to always set the alignment of the by Dan Gohman · 17 years ago
- 056292f Reverting r56249. On further investigation, this functionality isn't needed. by Bill Wendling · 17 years ago
- aed48bf Include the alignment value when displaying ConstantPoolSDNodes. by Dan Gohman · 17 years ago
- 9468a9b - Change "ExternalSymbolSDNode" to "SymbolSDNode". by Bill Wendling · 17 years ago
- 1937e2f Don't take the time to CheckDAGForTailCallsAndFixThem when tail calls by Dan Gohman · 17 years ago
- 5eb0cec Re-enable SelectionDAG CSE for calls. It matters in the case of by Dan Gohman · 17 years ago
- 095cc29 Define CallSDNode, an SDNode subclass for use with ISD::CALL. by Dan Gohman · 17 years ago
- 4fbd796 Change ConstantSDNode and ConstantFPSDNode to use ConstantInt* and by Dan Gohman · 17 years ago
- 913d3df Pass "earlyclobber" bit through to machine by Dale Johannesen · 17 years ago
- f5aeb1a Rename ConstantSDNode::getValue to getZExtValue, for consistency by Dan Gohman · 17 years ago
- e2f2083 The sequence for ppcf128 compares was not IEEE safe in the presence of NaNs. by Dale Johannesen · 17 years ago
- 74321ab FastISel support for i1 PHI nodes. by Dan Gohman · 17 years ago
- 8211648 FastISel support for i1 constants. by Dan Gohman · 17 years ago
- 0586d91 Add X86FastISel support for static allocas, and refences by Dan Gohman · 17 years ago
- 3ee25dc Add a break statement that I accidentally deleted when by Dan Gohman · 17 years ago
- fc2508e Remove unnecessary bit-wise AND from the limited precision work. by Bill Wendling · 17 years ago
- c0c3b9a Fix 80 col violation. by Daniel Dunbar · 17 years ago
- 277fc24 Check that both operands are f32 before attempting to lower. by Bill Wendling · 17 years ago
- aeb5c7b Implement "visitPow". This is mainly used to see if we have a pow() call of this by Bill Wendling · 17 years ago
- 9d24ac5 A few more places where FPOW is being ignored. by Evan Cheng · 17 years ago
- 4344a5d Change -fast-isel-no-abort to -fast-isel-abort, which now defaults by Dan Gohman · 17 years ago
- 4b88702 Legalizer was missing code that expand fpow to a libcall. by Evan Cheng · 17 years ago
- e10c814 Adding 6-, 12-, and 18-bit limited-precision floating-point support for exp2 by Bill Wendling · 17 years ago
- b4ec283 Add support for 6-, 12-, and 18-bit limited precision calculations of exp for by Bill Wendling · 17 years ago
- 293d5f8 Add a new option, -fast-isel-verbose, that can be used with by Dan Gohman · 17 years ago
- 0d95267 Clean this up, based on Evan's suggestions. by Owen Anderson · 17 years ago
- 3915025 - Add support for 6-, 12-, and 18-bit limited precision floating-point "log" by Bill Wendling · 17 years ago
- 19e861a Make safer variant of alias resolution routine to be default by Anton Korobeynikov · 17 years ago
- bd297bc Add limited precision floating-point conversions of log10 for 6- and 18-bit by Bill Wendling · 17 years ago
- baf37cd Check for type legality before materializing integer constants in fast isel. With this change, by Owen Anderson · 17 years ago
- cf01f7a Remove the code that protected FastISel from aborting in by Dan Gohman · 17 years ago
- 59fbc80 Fix a constant lowering bug. Now we can do load and store instructions with funky getelementptr embedded in the address operand. by Evan Cheng · 17 years ago
- 3eb5940 Add support for floating-point calculations of log2 with limited precisions of 6 by Bill Wendling · 17 years ago
- a0e8a1e Reapply 55904: Unbreak and fix indentation by Anton Korobeynikov · 17 years ago
- a8657e3 Fix a few I's that were meant to be renamed to BI's. by Dan Gohman · 17 years ago
- 07cabb9 Redo the 3 existing low-precision expansions to by Dale Johannesen · 17 years ago
- a425e00 Reverting r55898 to r55909. One of these patches was causing an ICE during the full bootstrap on Darwin: by Bill Wendling · 17 years ago
- 942ca7f In visitUREM, arrange for the temporary UDIV node to be by Dan Gohman · 17 years ago
- 1e0e04c Add VISIBILITY_HIDDEN on SDISelAsmOperandInfo by Daniel Dunbar · 17 years ago
- a47916d Fix the string for ISD::UDIVREM. by Dan Gohman · 17 years ago
- 9f11850 Avoid redefinition and nnbreak windows build. by Evan Cheng · 17 years ago
- d4fde0b Unbreak and fix indentation by Anton Korobeynikov · 17 years ago
- 5960e4e Add fast isel physical register definition support. by Evan Cheng · 17 years ago
- 9a0d1ba Revert my previous change -- the subtraction of two constants was a no-op by Bill Wendling · 17 years ago
- ca460d5 Convert by Bill Wendling · 17 years ago
- f099178 Indentation. by Evan Cheng · 17 years ago
- c7fcfa0 - Doh. Pass vector by value is bad. by Evan Cheng · 17 years ago
- 853244f Next limited float precision expansion (log2 12 bits) by Dale Johannesen · 17 years ago
- 6e60745 Revert r55859. This is breaking the build in the abscence of its companion commit. by Owen Anderson · 17 years ago
- 8a11053 Move the code that inserts copies for function livein registers by Dan Gohman · 17 years ago
- 852680a Add the next limited-precision expansion. by Dale Johannesen · 17 years ago
- 638c683 FastISel support for AND and OR with type i1. by Dan Gohman · 17 years ago
- 59e577f Add hooks for other intrinsics to get low-precision expansions. by Dale Johannesen · 17 years ago
- 40b189e FastISel support for ConstantExprs. by Dan Gohman · 17 years ago
- be0034e Revert r55817. It broke PIC. FastISel will need to find a different by Dan Gohman · 17 years ago
- c89d2fe Add a variant of AnalyzeCallOperands that can be used by fast isel. by Evan Cheng · 17 years ago
- 1c11deb "Fix" PR2762. The testcase now crashes codegen by Duncan Sands · 17 years ago
- 3dd168d Fix a search+replace-o. by Dan Gohman · 17 years ago
- 601d3c0 Add -flimit-float-precision to enable some faster, by Dale Johannesen · 17 years ago
- 087c850 FastISel support for unreachable. by Dan Gohman · 17 years ago
- c7f4a8a In FastISel mode, the scheduler may be invoked multiple times by Dan Gohman · 17 years ago
- 95267a1 Add initial support for selecting constant materializations that require constant by Owen Anderson · 17 years ago
- 5b22980 Add an include of SmallSet.h. by Dan Gohman · 17 years ago
- ae73dc1 Tidy up several unbeseeming casts from pointer to intptr_t. by Dan Gohman · 17 years ago
- 6448d91 Clean up uses of TargetLowering::getTargetMachine. by Dan Gohman · 17 years ago
- 7794f2a Add intrinsics for log, log2, log10, exp, exp2. by Dale Johannesen · 17 years ago
- 104e4ce Do trivial local CSE for constants and other non-Instruction values in FastISel. by Dan Gohman · 17 years ago
- 2a7c671 Put RegsForValue in the llvm namespace to avoid warnings about by Dan Gohman · 17 years ago
- 3df24e6 Create HandlePHINodesInSuccessorBlocksFast, a version of by Dan Gohman · 17 years ago
- d5d81a4 Oops, I accidentally broke the fallback case with my last commit. by Owen Anderson · 17 years ago
- 99aaf10 Fix an issue where we were reusing materializations of constants in blocks not dominated by the materialization. This is by Owen Anderson · 17 years ago
- f0cbcd4 Split the SelectionDAG-building code, including the FunctionLoweringInfo by Dan Gohman · 17 years ago
- 94b8d7e Separate MachineInstr-emitting routines from actual scheduling by Dan Gohman · 17 years ago
- f45261f If TargetSelectInstruction returns true, move to next instruction. by Evan Cheng · 17 years ago
- 36fd941 80 col violations. by Evan Cheng · 17 years ago
- d2ff647 Ensure that HandlePHINodesInSuccessorBlocks is run for all blocks, by Dan Gohman · 17 years ago
- ed63214 Provide two overloads of AnalyzeNewNode. by Gabor Greif · 17 years ago
- 1189f3a Even though no caller actually uses the new value by Duncan Sands · 17 years ago
- 2692d59 Another situation where ROTR is cheaper than ROTL. by Bill Wendling · 17 years ago
- 353dea2 For this pattern, ROTR is the cheaper option. by Bill Wendling · 17 years ago
- c5cbda1 - Fix comment so that it describes how the code really works: by Bill Wendling · 17 years ago
- dc71563 typo by Gabor Greif · 17 years ago
- 12632d2 fix some 80-col violations by Gabor Greif · 17 years ago
- eb9f892 Transform (x << (y&31)) -> (x << y). This takes advantage of the fact x86 shift instructions 2nd operand (shift count) is limited to 0 to 31 (or 63 in the x86-64 case). by Evan Cheng · 17 years ago
- cc54e76 Fix an issue where a use might be selected before a def, and then we didn't respect the pre-chosen vreg by Owen Anderson · 17 years ago
- 4529966 Fix 80 col. violations. by Evan Cheng · 17 years ago
- 17a568b Back out 55498. It broken Apple style bootstrapping. by Evan Cheng · 17 years ago
- 99b2182 Add a target callback for FastISel. by Dan Gohman · 17 years ago
- ba36cb5 erect abstraction boundaries for accessing SDValue members, rename Val -> Node to reflect semantics by Gabor Greif · 17 years ago
- 205d925 Implement null and undef values for FastISel. by Dan Gohman · 17 years ago
- a3f8b7a Optimize DAGCombiner's worklist processing. Previously it started by Dan Gohman · 17 years ago
- 9ffd8b5 Move CaseBlock, JumpTable, and BitTestBlock to be members of by Dan Gohman · 17 years ago
- 5edd361 Fix a FastISel bug where the instructions from lowering the arguments by Dan Gohman · 17 years ago
- 91ee545 Reduce the size of the Parts vector. by Rafael Espindola · 17 years ago
- c0bb68b Hook up support for fast-isel of trunc instructions, using the newly working support for EXTRACT_SUBREG. by Owen Anderson · 17 years ago