Unbreak the JIT with SSE
[oota-llvm.git] / lib / Target / X86 /
2006-01-27 Chris LattnerUnbreak the JIT with SSE
2006-01-27 Evan Chengx86 CPU detection and proper subtarget support
2006-01-27 Chris LattnerPHI and INLINEASM are now built-in instructions provide...
2006-01-26 Jeff CohenImprove compatibility with VC2005, patch by Morten...
2006-01-26 Chris LattnerImprove compatibility with VC2005, patch by Morten...
2006-01-26 Evan ChengAdded preliminary x86 subtarget support.
2006-01-26 Evan ChengWork around some x86 Darwin assembler bugs
2006-01-26 Evan ChengWhen trying to fold X86::SETCC into a Select, make...
2006-01-26 Evan ChengRemove the uses of STATUS flag register. Rely on node...
2006-01-25 Nate BegemanFirst part of bug 680:
2006-01-25 Evan ChengX86 prefer scheduling for reduced register pressure.
2006-01-25 Evan ChengFix a selectcc lowering bug. Make a copy of X86ISD...
2006-01-25 Chris LattnerLoosen up these checks to allow direct uses of ESP
2006-01-24 Chris Lattneruse ESP directly, not a copy of ESP into some other...
2006-01-24 Chris LattnerEmit the copies out of call return registers *after...
2006-01-24 Chris LattnerAllow jit-beta to work
2006-01-22 Chris LattnerAdd explicit #includes of <iostream>
2006-01-21 Evan ChengRename fcmovae to fcmovnb and fcmova to fcmovnbe (follo...
2006-01-20 Chris LattnerLowerReturn now doesn't have to handle f32 returns.
2006-01-20 Evan ChengStop doing that accidental commit.
2006-01-20 Evan ChengA few more SH{L|R}D peepholes.
2006-01-19 Evan ChengDidn't mean to commit the last one.
2006-01-19 Evan ChengAdded i16 SH{L|R}D patterns.
2006-01-19 Evan ChengAvoid generating a redundant setcc.
2006-01-19 Evan Chengadc and sbb need an incoming flag to ensure it reads...
2006-01-19 Evan ChengTwo peepholes:
2006-01-19 Evan ChengDidn't mean to check that in.
2006-01-19 Evan ChengA obvious typo
2006-01-18 Evan ChengSRA shift amount must be in i8
2006-01-18 Evan ChengIf a call return type is i1, insert a truncate from...
2006-01-17 Evan ChengFix lowering of calls which return f32 values.
2006-01-17 Evan ChengZero extending load from i1 to i8.
2006-01-17 Evan ChengSSE does not support i64 SINT_TO_FP (FP stack doesn...
2006-01-17 Evan ChengAdded a FIXME comment about why FST is currently flagge...
2006-01-17 Evan ChengBug fixes: fpGETRESULT should produces a flag result...
2006-01-16 Evan ChengMore typo's
2006-01-16 Evan ChengSome typo's
2006-01-16 Evan ChengFix FP_TO_INT**_IN_MEM lowering.
2006-01-16 Chris Lattnertransfer some notes from my email to somewhere useful.
2006-01-15 Evan ChengAdded patterns for 8-bit multiply
2006-01-15 Chris LattnerUse the default lowering of ISD::DYNAMIC_STACKALLOC...
2006-01-14 Chris Lattnersilence a warning
2006-01-14 Nate Begemanbswap implementation
2006-01-14 Evan ChengA typo.
2006-01-13 Evan ChengAdd truncstore i1 patterns.
2006-01-13 Chris LattnerFix a bug in my last X86 checkin, pointed out by cozmic
2006-01-13 Evan ChengLHS = X86ISD::CMOVcc LHS, RHS means LHS = RHS if cc...
2006-01-13 Chris LattnerEnable X86 support for savestack/restorestack
2006-01-13 Chris Lattnerexpand unsupported stacksave/stackrestore nodes
2006-01-13 Evan ChengMinor update.
2006-01-13 Evan ChengMore typo's. I need new eye glasses...
2006-01-13 Evan ChengOops. Typo.
2006-01-13 Evan ChengFix a SETCC / BRCOND folding bug.
2006-01-12 Evan ChengFix sint_to_fp (fild*) support.
2006-01-12 Evan ChengSpecify transformation from GlobalAddress to TargetGlob...
2006-01-12 Evan ChengX86ISD::SETCC (e.g. SETEr) produces a flag (so multiple...
2006-01-12 Evan Cheng* Materialize GlobalAddress and ExternalSym with MOV32r...
2006-01-11 Evan ChengAdded ROTL and ROTR.
2006-01-11 Evan ChengSupport for MEMCPY and MEMSET.
2006-01-11 Evan ChengSelect DYNAMIC_STACKALLOC
2006-01-11 Nate BegemanAdd bswap, rotl, and rotr nodes
2006-01-11 Evan Cheng* Add special entry code main() (to set x87 to 64-bit...
2006-01-11 Chris Lattnerimplement FP_REG_KILL insertion for the dag-dag instruc...
2006-01-11 Chris LattnerFit into 80 cols
2006-01-11 Evan ChengSSE cmov support.
2006-01-10 Evan Cheng* fp to sint patterns.
2006-01-10 Evan ChengFP_TO_INT*_IN_MEM and x87 FP Select support.
2006-01-09 Evan Cheng* Added undef patterns.
2006-01-09 Evan ChengMore typos
2006-01-09 Evan Chengtypo
2006-01-09 Evan ChengSupport for ADD_PARTS, SUB_PARTS, SHL_PARTS, SHR_PARTS...
2006-01-06 Evan Cheng* Added integer div / rem.
2006-01-06 Evan ChengISEL code for MULHU, MULHS, and UNDEF.
2006-01-06 Chris Lattnersilence a bogus gcc warning
2006-01-06 Evan ChengAddd (shl x, 1) ==> (shl x, x) peepholes.
2006-01-06 Evan Chengfold (shl x, 1) -> (add x, x)
2006-01-06 Evan Cheng* Fast call support.
2006-01-05 Evan ChengAdded ConstantFP patterns.
2006-01-05 Jim LaskeyHad expand logic backward.
2006-01-05 Jim LaskeyAdded initial support for DEBUG_LABEL allowing debug...
2006-01-05 Evan ChengDAG based isel call support.
2005-12-27 Chris LattnerFix a problem duraid pointed out to me compiling kc...
2005-12-26 Evan ChengAdded field noResults to Instruction.
2005-12-24 Evan ChengLet the helper functions know about X86::FR32RegClass...
2005-12-23 Evan Cheng* Removed the use of FLAG. Now use hasFlagIn and hasFla...
2005-12-23 Evan ChengMore X86 floating point patterns.
2005-12-23 Chris Lattnermake sure bit_convert's are expanded
2005-12-22 Evan ChengBye bye HACKTROCITY.
2005-12-21 Evan Cheng* Fix a GlobalAddress lowering bug.
2005-12-21 Evan ChengOops. Accidentally deleted RET pattern. It's still...
2005-12-21 Jim LaskeyDisengage DEBUG_LOC from non-PPC targets.
2005-12-21 Evan Cheng* Added support for X86 RET with an additional operand...
2005-12-21 Chris LattnerThis was meant to go in
2005-12-21 Chris LattnerRewrite FP stackifier support in the X86InstrInfo.td...
2005-12-21 Chris LattnerWrap some long lines: no functionality change
2005-12-21 Evan ChengRemove ISD::RET select code. Now tblgen'd.
2005-12-21 Evan Cheng* Added lowering hook for external weak global address...
2005-12-20 Evan ChengSSE2 floating point load / store patterns. SSE2 fp...
2005-12-20 Evan ChengAdded X86 readport patterns.
2005-12-20 Evan ChengAdded a hook to print out names of target specific...
next