- 0261d79 Allow more than one stub to be being generated at the same time. by Jeffrey Yasskin · 16 years ago
- f81bf15 Massive refactoring of NEON instructions. Separate opcode from data size specifier suffix, move \t up stream to instruction format, and fix more 80 column violations. by Evan Cheng · 16 years ago
- 77b02be move fconst[sd] to UAL. <rdar://7414913> by Jim Grosbach · 16 years ago
- da1aea4 Partially revert r84730 by removing N2VDup from ARMInstrFormats.td and modifying by Johnny Chen · 16 years ago
- 15e6ef8 fold immediate of a + Const into the user as a subtract if it can fit as a negated two-part immediate. by Jim Grosbach · 16 years ago
- b16ed11 Revert r84572 by removing N3VImm from ARMInstrFormats.td now that we can specify by Johnny Chen · 16 years ago
- 7ebd32a Partially revert r89377 by removing NLdStLN class definition from by Johnny Chen · 16 years ago
- 131c4a5 Make it clear that the index bit(s) of Vector Get Lane and Vector Set Lane by Johnny Chen · 16 years ago
- 7776452 Minor itinerary fixes for FP instructions. by David Goodwin · 16 years ago
- a37aa54 Move default FrameReg val to getFrameIndexReference(). Otherwise, debug info can get bogus values. by Jim Grosbach · 16 years ago
- 50f8516 Generate more correct debug info for frame indices. by Jim Grosbach · 16 years ago
- 1722f06 Minor optimization: when doing eq/ne comparions and RHS is a constant - swap operands, this will allow us to fold imm into comparison. by Anton Korobeynikov · 16 years ago
- ef4e604 Drop unsupported imm operands by Anton Korobeynikov · 16 years ago
- 3741be3 Use 2-byte alignment for functions. 4 bytes are clear overkill here. by Anton Korobeynikov · 16 years ago
- 3c97fc9 Use semicolon as assembler comment string by Anton Korobeynikov · 16 years ago
- 21ce2e3 Revert 89562. We're being sneakier than I was giving us credit for, and this by Jim Grosbach · 16 years ago
- f3b33d0 Darwin requires a frame pointer for all non-leaf functions to support correct by Jim Grosbach · 16 years ago
- ac0869d Add predicate operand to NEON instructions. Fix lots (but not all) 80 col violations in ARMInstrNEON.td. by Evan Cheng · 16 years ago
- bef8888 We are not using DBG_STOPPOINT anymore. by Devang Patel · 16 years ago
- 054b52c Added two SubtargetFeatures::AddFeatures methods, which accept a comma-separated string or already parsed command line parameters as input, and some code re-factoring to use these new methods. by Viktor Kutuzov · 16 years ago
- d9e05cd Fix a thinko that caused spurious @GOTOFFs. by Dan Gohman · 16 years ago
- 71a4196 Update for new getBlockAddress signature. by Dan Gohman · 16 years ago
- 29cbade Target-independent support for TargetFlags on BlockAddress operands, by Dan Gohman · 16 years ago
- b9e6b34 Recommitting PALIGNR shift width fixes. by Sean Callanan · 16 years ago
- 5ca1246 Remove an incorrect overaggressive optimization (PPC specific). by Dale Johannesen · 16 years ago
- 1bbf6d1 Reverting PALIGNR fix until I figure out how this broke the Clang testsuite. by Sean Callanan · 16 years ago
- 201dfa7 Fixed PALIGNR to take 8-bit rotations in all cases. by Sean Callanan · 16 years ago
- 4aedb61 Remat VLDRD from constpool. Clean up some instruction property specifications. by Evan Cheng · 16 years ago
- a956256 The verify() call of CPEIsInRange() isn't right for the assertion check of by Jim Grosbach · 16 years ago
- de8b1db Remove verifySizes() since it's not adding much value. by Jim Grosbach · 16 years ago
- 9b82425 Also CSE non-pic load from constant pools. by Evan Cheng · 16 years ago
- 9ef4835 Fix codegen of conditional move of immediates. We were not making use of the immediate forms of cmov instructions at all. by Evan Cheng · 16 years ago
- 0dde971 Update comment to reflect instruction. by Eric Christopher · 16 years ago
- 4d8e90a When placing constant islands and adjusting for alignment padding, inline by Jim Grosbach · 16 years ago
- 07ba906 Refactor cmov selection code out to a separate function. No functionality change. by Evan Cheng · 16 years ago
- fb7634f Reverting the EH table patches. by Bill Wendling · 16 years ago
- 5c376ff Added NLdStLN which is similar to NLdSt with the exception that op7_4 is not by Johnny Chen · 16 years ago
- dec6de9 fix typo by Jim Grosbach · 16 years ago
- 57474fa Fix a typo in a comment. by Dan Gohman · 16 years ago
- ed54de4 80 col violation. by Evan Cheng · 16 years ago
- 2ef9c8a More consistent thumb1 asm printing. by Evan Cheng · 16 years ago
- 5b397c9 Shrink ldr / str [sp, imm0-1024] to 16-bit instructions. by Evan Cheng · 16 years ago
- a64ce45 Eliminate more * 4 in Thumb1 asm printing for consistency sake. by Evan Cheng · 16 years ago
- 2045c47 - Add sugregister logic to handle f64=(f32,f32). by Bruno Cardoso Lopes · 16 years ago
- 0046de0 Only use small sections for non linux targets! by Bruno Cardoso Lopes · 16 years ago
- 01c6937 The "ReadOnlyWithRel" enum seems to apply more to what Darwin does with the EH by Bill Wendling · 16 years ago
- bea7df5 Add XCore support for indirectbr / blockaddress. by Richard Osborne · 16 years ago
- ec04458 Attempt #2: by Bill Wendling · 16 years ago
- e823db8 Added getDefaultSubtargetFeatures method to SubtargetFeatures class which returns a correct feature string for given triple. by Viktor Kutuzov · 16 years ago
- ebb5cb9 Add ARMv6 itineraries. by David Goodwin · 16 years ago
- a300300 Fix a few places that were missed when we converted to unified syntax. by Bob Wilson · 16 years ago
- 834b08a Add a target hook to allow changing the tail duplication limit based on the by Bob Wilson · 16 years ago
- 5eea342 The llvm-gcc front-end and the pass manager use two separate TargetData objects. by Bill Wendling · 16 years ago
- f04777b Enable arm jumpt table adjustment. by Jim Grosbach · 16 years ago
- bce3dbd Both Darwin as and GNU as violate ARM docs wrt printing of addrmode6 by Anton Korobeynikov · 16 years ago
- a9ea9ec Set Inst{15-12} (Rd/Rt) to 0b1111 (PC) for BR_JTadd, BR_JTr, and BR_JTm to by Johnny Chen · 16 years ago
- d15ac2f Re-apply 89011. It's not to be blamed. by Evan Cheng · 16 years ago
- 6db07ea Revert 89011. Buildbot thinks it might be breaking stuff. by Evan Cheng · 16 years ago
- a0a95a3 When moving a block for table jumps, make sure the prior block terminator by Jim Grosbach · 16 years ago
- 6cccc30 MOV64rm should be marked isReMaterializable. by Evan Cheng · 16 years ago
- 574186f A few more instructions that should be marked re-materializable. by Evan Cheng · 16 years ago
- 9d52e8d Set Rm bits of BX_RET to 0b1110 (R14); and set condition code bits of BRIND to by Johnny Chen · 16 years ago
- d1ba06b Make X86-64 in the Large model always emit 64-bit calls. by Jeffrey Yasskin · 16 years ago
- 600c043 - Check memoperand alignment instead of checking stack alignment. Most load / store folding instructions are not referencing spill stack slots. by Evan Cheng · 16 years ago
- 7bde297 Make the pass class name more explicit. by Jim Grosbach · 16 years ago
- 074fb02 make pass name a bit more clear by Jim Grosbach · 16 years ago
- 08cbda5 Simplify thumb2 jump table adjustments. Remove unnecessary calculation and by Jim Grosbach · 16 years ago
- 9249efe clarify comment by Jim Grosbach · 16 years ago
- b2e86bb back off for a bit. tracking down weirdness by Jim Grosbach · 16 years ago
- ca215e7 Analyze has to be before checking the condition, obviously. Properly construct an iterator for prior. by Jim Grosbach · 16 years ago
- add2076 Disable ldc1/sdc1 instructions for mips1 targets. by Bruno Cardoso Lopes · 16 years ago
- 6e0b658 - Fix a small bug while handling target constant pools (one param was missing). by Bruno Cardoso Lopes · 16 years ago
- a443217 Detect need for autoalignment of the stack earlier to catch spills more by Jim Grosbach · 16 years ago
- 6cb6788 set the def of the VLD1q64 properly by Jim Grosbach · 16 years ago
- 93f9f7a Add a complex missed optimization opportunity I came across while investigating by Nick Lewycky · 16 years ago
- 9c477f5 cleanup. by Jim Grosbach · 16 years ago
- 00a6a1f Cleanup flow, and only update the jump table we're analyzing when replacing a destination MBB. by Jim Grosbach · 16 years ago
- 13c4fab Add XCore support for arbitrary-sized aggregate returns. by Richard Osborne · 16 years ago
- b23f3aa Temporary disable the error - it seems to be too conservative. by Anton Korobeynikov · 16 years ago
- 067d024 Add llvm::sys::getHostCPUName, for detecting the LLVM name for the host CPU. by Daniel Dunbar · 16 years ago
- 9a501cf revert 88761 as it fails builds. by Sanjiv Gupta · 16 years ago
- 6fb01a7 Fix debug info crashes for PIC16. by Sanjiv Gupta · 16 years ago
- d57cdd5 - Change TargetInstrInfo::reMaterialize to pass in TargetRegisterInfo. by Evan Cheng · 16 years ago
- 0cd22dd When expanding t2STRDi8 r, r to two stores, add kill markers correctly. by Evan Cheng · 16 years ago
- 52cd548 The instruction pointer %RIP is a reserved register on x86_64. by Jakob Stoklund Olesen · 16 years ago
- 1924aab by David Greene · 16 years ago
- 87d21b9 Allow target to specify regclass for which antideps will only be broken along the critical path. by David Goodwin · 16 years ago
- a8173b9 Support fp64 immediate zero, this fixes only part of PR5445 by Bruno Cardoso Lopes · 16 years ago
- 1e60881 Adjust isConstantSplat to allow for big-endian targets. by Dale Johannesen · 16 years ago
- c1a07be Block renumbering by Jim Grosbach · 16 years ago
- f4cbc0e use lower case for readability by Jim Grosbach · 16 years ago
- dda3978 by David Greene · 16 years ago
- b9c2fd9 by David Greene · 16 years ago
- b87bc95 by David Greene · 16 years ago
- 3f2bf85 by David Greene · 16 years ago
- 05872ea Add compare_lower and equals_lower methods to StringRef. Switch all users of by Benjamin Kramer · 16 years ago
- 80697d1 Update TB[BH] layout optimization. Add support for moving the target block by Jim Grosbach · 16 years ago
- b9d2c03 Use table to separate opcode from operands. by Evan Cheng · 16 years ago
- 06b53c0 isLegalICmpImmediate should take a signed integer; code clean up. by Evan Cheng · 16 years ago
- 01dec0e Revert 86857. It's causing consumer-typeset to fail, and there's a better way to do it forthcoming anyway. by Jim Grosbach · 16 years ago