- e4b94b4 Basic implementation of SSEDomainFix pass. by Jakob Stoklund Olesen · 15 years ago
- 352aa50 Add a late SSEDomainFix pass that twiddles SSE instructions to avoid domain crossings. by Jakob Stoklund Olesen · 15 years ago
- fe4b92b Revert "Add a late SSEDomainFix pass that twiddles SSE instructions to avoid domain crossings." by Jakob Stoklund Olesen · 15 years ago
- c75c5fa Add a late SSEDomainFix pass that twiddles SSE instructions to avoid domain crossings. by Jakob Stoklund Olesen · 15 years ago
- b779033 add encoder support and tests for rdtscp by Chris Lattner · 16 years ago
- a599de2 remove special cases for vmlaunch, vmresume, vmxoff, and swapgs by Chris Lattner · 16 years ago
- 835acab implement infrastructure to support fixups for rip-rel by Chris Lattner · 16 years ago
- a033119 enhance the immediate field encoding to know whether the immediate by Chris Lattner · 16 years ago
- 0d8db8e add a bunch of mod/rm encoding types for fixed mod/rm bytes. by Chris Lattner · 16 years ago
- c96f6d6 revert r95949, it turns out that adding new prefixes is not a by Chris Lattner · 16 years ago
- 239a1ed add another bit of space for new kinds of instruction prefixes. by Chris Lattner · 16 years ago
- 39a612e port X86InstrInfo::determineREX over to the new encoder. by Chris Lattner · 16 years ago
- 74a2151 move functions for decoding X86II values into the X86II namespace. by Chris Lattner · 16 years ago
- a62fe66 constant propagate a method away. by Chris Lattner · 16 years ago
- a3a0db0 change getSizeOfImm and getBaseOpcodeFor to just take by Chris Lattner · 16 years ago
- 1e80f40 enhance new encoder to support prefixes + RawFrm by Chris Lattner · 16 years ago
- 96dc115 Add two target hooks to determine whether two loads are near and should be scheduled together. by Evan Cheng · 16 years ago
- 7da9ecf Add a quick pass to optimize sign / zero extension instructions. For targets where the pre-extension values are available in the subreg of the result of the extension, replace the uses of the pre-extension value with the result + extract_subreg. by Evan Cheng · 16 years ago
- a5a81d7 Add TargetInstrInfo::isCoalescableInstr. It returns true if the specified by Evan Cheng · 16 years ago
- 656e514 Add support to 3-addressify 16-bit instructions. by Evan Cheng · 16 years ago
- 864e2ef Remove the target hook TargetInstrInfo::BlockHasNoFallThrough in favor of by Dan Gohman · 16 years ago
- 29dbf50 by David Greene · 16 years ago
- 15217e6 Remove isProfitableToDuplicateIndirectBranch target hook. It is profitable by Bob Wilson · 16 years ago
- 4a66e5b Based on the testcase for pr3120, running on my MacPro with Xeon processors, by Bob Wilson · 16 years ago
- d57cdd5 - Change TargetInstrInfo::reMaterialize to pass in TargetRegisterInfo. by Evan Cheng · 16 years ago
- dda3978 by David Greene · 16 years ago
- b87bc95 by David Greene · 16 years ago
- 0115e16 Fix MachineLICM to use the correct virtual register class when by Dan Gohman · 16 years ago
- 3731bc0 Replace X86's CanRematLoadWithDispOperand by calling the target-independent by Dan Gohman · 16 years ago
- 91e69c3 Add basic infrastructure and x86 support for preserving MachineMemOperand by Dan Gohman · 16 years ago
- e33f44c Replace TargetInstrInfo::isInvariantLoad and its target-specific by Dan Gohman · 16 years ago
- 01a76ce Remove explicit enum integer values. They don't appear to be needed, and by Dan Gohman · 16 years ago
- 9cef48e It's not legal to fold a load from a narrower stack slot into a wider instruction. If done, the instruction does a 64-bit load and that's not by Evan Cheng · 16 years ago
- 63476a8 Reference to hidden symbols do not have to go through non-lazy pointer in non-pic mode. rdar://7187172. by Evan Cheng · 16 years ago
- b4dc13c Add crc32 instruction and intrinsics. Add a new class of prefix by Eric Christopher · 16 years ago
- 3784453 Let callers decide the sub-register index on the def operand of rematerialized instructions. by Evan Cheng · 16 years ago
- f9b36f0 Move load / store folding alignment require into the table(s). by Evan Cheng · 16 years ago
- 19a2011 Undo my brain cramp. by Evan Cheng · 16 years ago
- 49ddb61 CMOVxx doesn't swap operands which it's commuted. by Evan Cheng · 16 years ago
- 7478ab8 add a predicate to determine if a global var reference requires a by Chris Lattner · 16 years ago
- 3b6b36d change isGlobalStubReference to take target flags instead of a MachineOperand. by Chris Lattner · 16 years ago
- 281bada add a new predicate method that says whether a GlobalValue by Chris Lattner · 16 years ago
- 75cdf27 move reasoning about darwin $non_lazy_ptr stubs from asmprinter into isel. by Chris Lattner · 16 years ago
- 74e726e make isel decide whether to emit $stub's on darwin instead of asmprinter. by Chris Lattner · 16 years ago
- 4aa21aa move handling of dllimport linkage in isel, not in asmprinter. by Chris Lattner · 16 years ago
- b903bed Move all the TLS processing logic into isel, don't do it in asmprinter at all. by Chris Lattner · 16 years ago
- 55e7c82 start adding logic in isel to determine asm printer semantics, step N of M. by Chris Lattner · 16 years ago
- ac5e887 Use target-specific machine operand flags to eliminate a gross hack by Chris Lattner · 16 years ago
- 094fad3 Re-apply 68552. Tested by bootstrapping llvm-gcc and using that to build llvm. by Rafael Espindola · 16 years ago
- 044b534 Temporarily revert r68552. This was causing a failure in the self-hosting LLVM by Bill Wendling · 16 years ago
- 2a6411b Reduce code duplication on the TLS implementation. by Rafael Espindola · 16 years ago
- da945e3 Have only one definition of X86AddrNumOperands. by Rafael Espindola · 16 years ago
- dc54d31 Turns out AnalyzeBranch can modify the mbb being analyzed. This is a nasty by Evan Cheng · 17 years ago
- 770bcc7 Move getPointerRegClass from TargetInstrInfo to TargetRegisterInfo. by Evan Cheng · 17 years ago
- 4350eb8 Add TargetInstrInfo::isSafeToMoveRegisterClassDefs. It returns true if it's safe to move an instruction which defines a value in the register class. Replace pre-splitting specific IgnoreRegisterClassBarriers with this new hook. by Evan Cheng · 17 years ago
- 04ee5a1 Change TargetInstrInfo::isMoveInstr to return source and destination sub-register indices as well. by Evan Cheng · 17 years ago
- 653456c X86_COND_C and X86_COND_NC are alternate mnemonics for by Dan Gohman · 17 years ago
- d68a076 Tidy up #includes, deleting a bunch of unnecessary #includes. by Dan Gohman · 17 years ago
- c54baa2 Split foldMemoryOperand into public non-virtual and protected virtual by Dan Gohman · 17 years ago
- 3fafd93 Generate something sensible for an [SU]ADDO op when the overflow/carry flag is by Bill Wendling · 17 years ago
- cbad42c Add more const qualifiers. This fixes build breakage from r59540. by Dan Gohman · 17 years ago
- 2306628 For now, don't split live intervals around x87 stack register barriers. FpGET_ST0_80 must be right after a call instruction (and ADJCALLSTACKUP) so we need to find a way to prevent reload of x87 registers between them. by Evan Cheng · 17 years ago
- 279c22e Optimized FCMP_OEQ and FCMP_UNE for x86. by Dan Gohman · 17 years ago
- 244911b getX86RegNum has long been moved to X86RegisterInfo. by Evan Cheng · 17 years ago
- 8e8b8a2 Const-ify several TargetInstrInfo methods. by Dan Gohman · 17 years ago
- ef93cec Add ability to override segment (mostly for code emitter purposes). by Anton Korobeynikov · 17 years ago
- d735b80 Switch the MachineOperand accessors back to the short names like by Dan Gohman · 17 years ago
- 57c3dac Move the GlobalBaseReg field out of X86ISelDAGToDAG.cpp by Dan Gohman · 17 years ago
- 8b74696 Move the code for initializing the global base reg out of by Dan Gohman · 17 years ago
- 94a50da Backing out 55521. Not safe. by Evan Cheng · 17 years ago
- 4d46d0a Swap fp comparison operands and change predicate to allow load folding. by Evan Cheng · 17 years ago
- 940f83e Make TargetInstrInfo::copyRegToReg return a bool indicating whether the copy requested by Owen Anderson · 17 years ago
- 44eb65c Convert uses of std::vector in TargetInstrInfo to SmallVector. This change had to be propoagated down into all the targets and up into all clients of this API. by Owen Anderson · 17 years ago
- 8e5f2c6 Pool-allocation for MachineInstrs, MachineBasicBlocks, and by Dan Gohman · 17 years ago
- f660c17 Make LiveVariables even more optional, by making it optional in the call to TargetInstrInfo::convertToThreeAddressInstruction by Owen Anderson · 17 years ago
- 1c4b5ea Make intel asmprinter child of generic asmprinter, not x86 shared asm printer. This leads to some code duplication, which will be resolved later. by Anton Korobeynikov · 17 years ago
- 58dcb0e Add option to commuteInstruction() which forces it to create a new (commuted) instruction. by Evan Cheng · 17 years ago
- c9f5f3f Change target-specific classes to use more precise static types. by Dan Gohman · 17 years ago
- 9f8fea3 Constify the machine instruction passed into the by Bill Wendling · 17 years ago
- 52e724a Infrastructure for getting the machine code size of a function and an instruction. X86, PowerPC and ARM are implemented by Nicolas Geoffray · 17 years ago
- ca1267c Move reMaterialize() from TargetRegisterInfo to TargetInstrInfo. by Evan Cheng · 17 years ago
- 950a4c4 Add explicit keywords. by Dan Gohman · 17 years ago
- 6634e26 Get rid of a pseudo instruction and replace it with subreg based operation on real instructions, ridding the asm printers of the hack used to do this previously. In the process, update LowerSubregs to be careful about eliminating copies that have side affects. by Christopher Lamb · 17 years ago
- 1fab4a6 Recommitting parts of r48130. These do not appear to cause the observed failures. by Christopher Lamb · 17 years ago
- 4499e49 Revert 48125, 48126, and 48130 for now to unbreak some x86-64 tests. by Evan Cheng · 17 years ago
- 3feb017 Allow insert_subreg into implicit, target-specific values. by Christopher Lamb · 17 years ago
- ea7da50 Add lock prefix support to x86. Also add the instructions necessary for the atomic ops. They are still marked pseudo, since I cannot figure out what format to use, but they are the correct opcode. by Andrew Lenharth · 17 years ago
- 6f0d024 Rename MRegisterInfo to TargetRegisterInfo. by Dan Gohman · 18 years ago
- 5fd79d0 It's not always safe to fold movsd into xorpd, etc. Check the alignment of the load address first to make sure it's 16 byte aligned. by Evan Cheng · 18 years ago
- a22edc8 Simplify the side effect stuff a bit more and make licm/sinking by Chris Lattner · 18 years ago
- 749c6f6 rename TargetInstrDescriptor -> TargetInstrDesc. by Chris Lattner · 18 years ago
- cc8cd0c remove MachineOpCode typedef. by Chris Lattner · 18 years ago
- 43dbe05 Move even more functionality from MRegisterInfo into TargetInstrInfo. by Owen Anderson · 18 years ago
- d94b6a1 Move some more functionality from MRegisterInfo to TargetInstrInfo. by Owen Anderson · 18 years ago
- 3100afa Machine LICM will check that operands are defined outside of the loop. Also by Bill Wendling · 18 years ago
- f6372aa Move some more instruction creation methods from RegisterInfo into InstrInfo. by Owen Anderson · 18 years ago
- 6410552 Fix a problem where lib/Target/TargetInstrInfo.h would include and use by Chris Lattner · 18 years ago
- d10fd97 Move copyRegToReg from MRegisterInfo to TargetInstrInfo. This is part of the by Owen Anderson · 18 years ago
- 6259d51 If we have a load of a global address that's not modified during the by Bill Wendling · 18 years ago
- 4ee451d Remove attribution from file headers, per discussion on llvmdev. by Chris Lattner · 18 years ago