Add a late SSEDomainFix pass that twiddles SSE instructions to avoid domain crossings.
[oota-llvm.git] / lib / Target / X86 / X86Instr64bit.td
2010-03-24 Chris Lattnerremove 64-bit or_is_add parallels.
2010-03-19 Chris Lattnerremove the patterns that I commented out in r98930...
2010-03-19 Daniel DunbarMC/X86: Rename alternate spellings of {ADD64,CMP64...
2010-03-19 Chris Lattnercomment out a bunch of parallel store patterns that...
2010-03-19 Daniel DunbarX86: Fix encoding for TEST64rr.
2010-03-19 Chris LattnerNow that tblgen can handle matching implicit defs of...
2010-03-18 Chris Lattnercallq is pcrelative
2010-03-16 Chris Lattnerfix the same bug on the x86-64 side of the fence.
2010-03-14 Evan ChengFix jit encoding bugs.
2010-03-14 Evan ChengDo not force indirect tailcall through fixed registers...
2010-03-13 Daniel DunbarX86_64: Fix encoding for the rest of the 64i32 instruct...
2010-03-13 Daniel DunbarX86: Fix ADD64i32 encoding.
2010-03-13 Daniel DunbarMC/X86_64: Fix matching of leaq.
2010-03-13 Daniel DunbarMC/X86_64: Fix matching of callq.
2010-03-08 Chris LattnerCorrect immediate sizes.
2010-03-03 Chris Lattnerfactor the 'sign extended from 8 bit' patterns better so
2010-02-27 Dan GohmanThe mayHaveSideEffects flag is no longer used.
2010-02-27 Chris Lattnerfix an incorrect (overly conservative) predicate.
2010-02-23 Chris Lattnerremove a bunch of dead named arguments in input patterns,
2010-02-23 Chris Lattnerfix a type mismatch in this pattern, where we were...
2010-02-13 Chris Lattnerremove special cases for vmlaunch, vmresume, vmxoff...
2010-02-12 Daniel DunbarX86: Fix definition for RCL/RCR.*m? operations -- they...
2010-02-09 Chris Lattnermove target-independent opcodes out of TargetInstrInfo
2010-02-05 Chris Lattnerreally kill off the last MRMInitReg inst, remove logic...
2010-02-05 Chris Lattnerfactor code better in X86MCInstLower::Lower, teach...
2010-01-31 Evan ChengChange TAILJMP's to be varargs and transfer implicit...
2010-01-22 Daniel DunbarMark EH_RETURN64 as CodeGenOnly.
2010-01-12 Evan ChengEliminate or_not_add and just use AddedComplexity so...
2010-01-12 Dan GohmanReapply the MOV64r0 patch, with a fix: MOV64r0 clobbers...
2010-01-11 Evan ChengExtend r93152 to work on OR r, r. If the source set...
2010-01-11 Evan ChengRevert 93158. It's breaking quite a few x86_64 tests.
2010-01-11 Dan GohmanUse a 32-bit and with implicit zero-extension instead...
2010-01-11 Dan GohmanRe-instate MOV64r0 and MOV16r0, with adjustments to...
2010-01-11 Dan GohmanPattern top-level operators don't need to be restricted...
2010-01-11 Evan ChengSelect an OR with immediate as an ADD if the input...
2010-01-08 Evan ChengFix what looks to me obvious instruction definition...
2010-01-04 Evan ChengPerform this folding as a target specific dag combine:
2009-12-18 Sean CallananInstruction fixes, added instructions, and AsmString...
2009-12-16 Evan ChengRe-enable 91381 with fixes.
2009-12-15 Evan ChengFix an encoding bug.
2009-12-15 Evan ChengUse sbb x, x to materialize carry bit in a GPR. The...
2009-11-30 Dan GohmanFix a minor inconsistency.
2009-11-17 Evan ChengMOV64rm should be marked isReMaterializable.
2009-11-02 Anton KorobeynikovDo not infer the target type for COPY_TO_REGCLASS from...
2009-10-30 Dan GohmanInitial x86 support for BlockAddresses.
2009-10-29 Dan GohmanRename usesCustomDAGSchedInserter to usesCustomInserter...
2009-10-19 Chris Lattnerremove strings from instructions who are never asmprinted.
2009-10-19 Torok EdwinFix PR5247, "lock addq" pattern (and other atomics...
2009-09-18 Dan GohmanAdd support for using the FLAGS result of or, xor,...
2009-09-18 Sean CallananAdded RCL and RCR (rotate left and right with a
2009-09-16 Sean CallananAdded the LODS (load byte into register, usually
2009-09-16 Sean CallananAdded the LAR (load segment access rights)
2009-09-16 Sean CallananAdded an alternate form of register-register CMP
2009-09-16 Sean CallananAdded the definitions for one-bit left shifts to
2009-09-15 Sean CallananUpdated comments per Eli's suggestion.
2009-09-15 Sean CallananAdded register-to-register ADD instructions to the
2009-09-15 Sean CallananAdded a new register class for segment registers
2009-09-15 Dan GohmanOn x86-64, the 32-bit cmov doesn't actually clear the...
2009-09-12 Sean CallananAdded CMPS (string comparison) instructions for all
2009-09-12 Sean CallananAdded SCAS instructions in their 8, 16, 32, and
2009-09-11 Sean CallananAdded ADC, SUB, SBB, and OR instructions that operate
2009-09-10 Sean CallananAdded XOR instructions for rAX and immediates of
2009-09-10 Sean CallananAdded MOV instructions between rAX and memory offsets,
2009-09-10 Sean CallananAdded a variety of PUSH and POP instructions, including
2009-09-03 Dan GohmanAdd a -disable-16bit flag and associated support for...
2009-09-03 Sean CallananAdded opaque 32-, 48-, and 80-bit memory operand types...
2009-09-02 Sean CallananFixed the asmstrings for 8-bit, 16-bit, and 32-bit...
2009-09-01 Sean CallananAdded TEST %rAX, $imm instructions to the Intel tables...
2009-08-26 Dan GohmanDon't use INSERT_SUBREG to model anyext operations...
2009-08-11 Daniel DunbarX86/AsmParser: Mark MOV64GSrm, MOV64FSrm, GS_MOV32rm...
2009-08-11 Chris Lattnermove some 32-bit instrs to x86instrinfo.td
2009-08-10 Daniel Dunbarllvm-mc/AsmParser: Disambiguate i64i8imm.
2009-08-10 Daniel Dunbarllvm-mc/AsmMatcher: Change assembler parser match class...
2009-08-09 Daniel DunbarExtend comment on ParserMatchClass .td field, and add...
2009-08-07 Anton KorobeynikovDo not generate 32-bit call on win64 when imm does...
2009-08-06 Anton KorobeynikovWe need to sext global addresses in kernel code model...
2009-08-05 Dan GohmanEnable the new no-SP register classes by default. This...
2009-08-05 Dan GohmanMajor calling convention code refactoring.
2009-08-03 Anton KorobeynikovUnbreak Win64 CC. Step one: honour register save area...
2009-08-02 Dan GohmanResync lea32addr and lea64addr.
2009-07-30 Evan ChengOptimize some common usage patterns of atomic built...
2009-07-21 Bill WendlingAdd the following functions:
2009-07-16 Chris Lattneruse SUBREG_TO_REG instead of INSERT_SUBREG, this way...
2009-07-14 Chris Lattnerreapply r75408, which eliminates MOV64r0 in favor of...
2009-07-12 Bill WendlingTemporarily revert r75408. It appears to break the...
2009-07-12 Chris Lattnereliminate MOV64r0 in favor of a Pat<> pattern. This...
2009-07-11 Chris Lattnerfix x86-64 static codegen to materialize the address...
2009-07-11 Chris Lattnercomment cleanup, reduce nesting.
2009-07-11 Chris Lattnerremove some dead patterns, WrapperRIP doesn't exist...
2009-06-27 Chris LattnerReimplement rip-relative addressing in the X86-64 backe...
2009-06-20 Chris Lattnerchange TLS_ADDR lowering to lower to a real mem operand...
2009-06-20 Chris Lattnereliminate the "call" operand modifier from the asm...
2009-06-20 Chris Lattnerimplement support for lowering subregs when preparing...
2009-06-16 Evan ChengCALL64pcrel32 immediate field is 32-bit. Patch by Abhin...
2009-06-15 Bill WendlingThe Ls and Qs were mixed up. Patch by Sean.
2009-06-15 Bill Wendling"The Intel instruction tables should include the 64...
2009-06-03 Dan GohmanRevert r72734. The Darwin assembler doesn't support...
2009-06-02 Evan ChengOn Darwin x86_64 small code model doesn't guarantee...
2009-06-02 Dale JohannesenRevert 72707 and 72709, for the moment.
2009-06-01 Dale JohannesenMake the implicit inputs and outputs of target-independent
next