Fix comment about removeRange.
[oota-llvm.git] / lib / Target / X86 / X86ISelLowering.cpp
2009-01-28 Mon P WangFixed lowering of v816 shuffles.
2009-01-28 Mon P WangAdd shuffle splat pattern for x86 sse shifts.
2009-01-26 Dan GohmanTake the next steps in making SDUse more consistent...
2009-01-26 Nate BegemanDe-identifying per sabre review
2009-01-26 Nate BegemanSupport pattern matching various x86 sse shifts.
2009-01-22 Bob WilsonAdd SelectionDAG::getNOT method to construct bitwise...
2009-01-19 Evan ChengMinor tweak to LowerUINT_TO_FP_i32. Bias (after scalar_...
2009-01-19 Evan ChengNow not UINT_TO_FP is legal (it's marked custom), dag...
2009-01-17 Bill WendlingExtend thi
2009-01-17 Bill WendlingTemporarily revert my last change. It is causing a...
2009-01-17 Bill WendlingImplement a special algorithm for converting uint_to_fp...
2009-01-16 Bill WendlingAdd support for non-zero __builtin_return_address value...
2009-01-15 Mon P WangExpand insert/extract of a <4 x i32> with a variable...
2009-01-15 Dan GohmanMake getWidenVectorType const.
2009-01-13 Dan GohmanBT appears to be available on all >= i386 chips.
2009-01-13 Dan GohmanDon't use a BT instruction if the AND has multiple...
2009-01-13 Devang PatelUse DebugInfo interface to lower dbg_* intrinsics.
2009-01-07 Dan GohmanX86_COND_C and X86_COND_NC are alternate mnemonics for
2009-01-05 Devang Patelsquash warnings.
2009-01-02 Evan ChengUse movaps / movd to extract vector element 0 even...
2009-01-01 Duncan SandsFix PR3274: when promoting the condition of a BRCOND...
2008-12-25 Chris LattnerAdd a simple pattern for matching 'bt'.
2008-12-24 Chris LattnertranslateX86CC can never fail. Simplify it based on...
2008-12-24 Chris Lattnerindentation
2008-12-23 Chris Lattnersimplify some control flow and reduce indentation,...
2008-12-23 Dan GohmanAdd instruction patterns and encodings for the x86...
2008-12-23 Dan GohmanClean up the atomic opcodes in SelectionDAG.
2008-12-23 Mon P WangFixed code generation for v8i16 and v16i8 splats on...
2008-12-18 Mon P WangFixed x86 code generation of multiple for v2i64. It...
2008-12-12 Bill Wendling- Use patterns instead of creating completely new instr...
2008-12-12 Mon P WangAdded support for SELECT v8i8 v4i16 for X86 (MMX)
2008-12-12 Bill WendlingRedo the arithmetic with overflow architecture. I was...
2008-12-10 Bill WendlingWhitespace changes.
2008-12-09 Bill WendlingAdd sub/mul overflow intrinsics. This currently doesn...
2008-12-05 Dale JohannesenMake LoopStrengthReduce smarter about hoisting things...
2008-12-03 Evan ChengRefactor code. No functionality change.
2008-12-03 Bill WendlingCC should only be a ConstantSDNode at this point. Just...
2008-12-02 Bill WendlingSecond stab at target-dependent lowering of everyone...
2008-12-01 Duncan SandsThere are no longer any places that require a
2008-12-01 Duncan SandsChange the interface to the type legalization method
2008-11-27 Bill WendlingComment out code that isn't entirely correct.
2008-11-26 Bill WendlingGenerate something sensible for an [SU]ADDO op when...
2008-11-24 Bill Wendling- Make lowering of "add with overflow" customizable...
2008-11-24 Mon P WangAdded missing description for -disable-mmx option.
2008-11-23 Duncan SandsRename SetCCResultContents to BooleanContents. In
2008-11-23 Mon P WangAdded -disable-mmx using a patch from Preston Gurd.
2008-11-13 Dale JohannesenExtend InlineAsm::C_Register to allow multiple specific...
2008-11-06 Mon P WangWidening cleanup
2008-11-05 Evan ChengIndentation.
2008-10-31 Dan GohmanUse MOVSSmr instead of EXTRACTPSmr in the case of extra...
2008-10-30 Mon P WangAdd initial support for vector widening. Logic is...
2008-10-28 Chris LattnerFix a nasty miscompilation of 176.gcc on linux/x86...
2008-10-24 Duncan SandsFix translateX86CC: if SetCCOpcode is SETULE and
2008-10-22 Dale JohannesenRemove allocation of unused stack slot.
2008-10-22 Duncan SandsGet this working with LegalizeTypes: (1) don't
2008-10-22 Dale JohannesenAdjust comments for pedantic satisfaction.
2008-10-21 Dale JohannesenAdd comments to explain uint64->f64 algorithm,
2008-10-21 Dale JohannesenAdd an SSE2 algorithm for uint64->f64 conversion.
2008-10-21 Dan GohmanDon't create TargetGlobalAddress nodes with offsets...
2008-10-21 Dan GohmanOptimized FCMP_OEQ and FCMP_UNE for x86.
2008-10-20 Duncan SandsHave X86 custom lowering for LegalizeTypes use
2008-10-18 Dan GohmanTeach DAGCombine to fold constant offsets into GlobalAd...
2008-10-17 Chris Lattneradd support for 128 bit inputs on both x86-64 and x86-32.
2008-10-17 Chris LattnerFix a bug where the x86 backend would reject 64-bit...
2008-10-16 Dan GohmanRemove an unused variable.
2008-10-15 Evan Cheng- Add target lowering hooks that specify which setcc...
2008-10-14 Dan GohmanFastISel support for exception-handling constructs.
2008-10-14 Evan ChengRename LoadX to LoadExt.
2008-10-11 Chris LattnerChange CALLSEQ_BEGIN and CALLSEQ_END to take TargetCons...
2008-10-09 Dale JohannesenAdd a "loses information" return value to APFloat:...
2008-10-06 Evan ChengFix PR2850 and PR2863. Only generate movddup for 128...
2008-10-03 Dale JohannesenMake atomic Swap work, 64-bit on x86-32.
2008-10-03 Dale JohannesenPass MemOperand through for 64-bit atomics on 32-bit,
2008-10-03 Dan GohmanSwitch the MachineOperand accessors back to the short...
2008-10-02 Dale JohannesenHandle some 64-bit atomics on x86-32, some of the time.
2008-10-01 Bill WendlingImplement the -fno-builtin option in the front-end...
2008-09-30 Bill WendlingJust don't transform this memset into "bzero" if no...
2008-09-30 Bill WendlingAdd the new `-no-builtin' flag. This flag is meant...
2008-09-29 Dale JohannesenRemove misuse of ReplaceNodeResults for atomics with
2008-09-27 Evan ChengRe-apply 56683 with fixes.
2008-09-26 Bill WendlingTemporarily reverting r56683. This is causing a failure...
2008-09-26 Dan GohmanRename ConstantSDNode's getSignExtended to getSExtValue...
2008-09-26 Evan ChengFix @llvm.frameaddress codegen. FP elimination optimiza...
2008-09-26 Dale JohannesenAdd "inreg" field to CallSDNode (doesn't increase
2008-09-25 Evan ChengPrefer movlhps over punpcklqdq, etc. in more cases.
2008-09-25 Devang Patel Large mechanical patch.
2008-09-25 Evan ChengWith sse3 and when the source is a load or has multiple...
2008-09-24 Evan ChengX86ISD::VZEXT_LOAD should produce and fold a chain.
2008-09-24 Evan ChengProperly handle 'm' inline asm constraints. If a GV...
2008-09-23 Dan GohmanArrange for FastISel code to have access to the Machine...
2008-09-22 Evan ChengSupport x86 specific inline asm modifier 'J'.
2008-09-22 Dale JohannesenMake log, log2, log10, exp, exp2 use Expand by
2008-09-22 Arnold SchwaighoferChange the calling convention used when tail call optim...
2008-09-16 Bill WendlingReverting r56249. On further investigation, this functi...
2008-09-16 Bill Wendling- Change "ExternalSymbolSDNode" to "SymbolSDNode".
2008-09-13 Dan GohmanRemove isImm(), isReg(), and friends, in favor of
2008-09-13 Dan GohmanDefine CallSDNode, an SDNode subclass for use with...
2008-09-12 Dan GohmanRename ConstantSDNode::getValue to getZExtValue, for...
2008-09-11 Arnold SchwaighoferWhen tailcallopt is enabled all fastcc calls must have...
2008-09-11 Dale JohannesenThe version of AtomicSDNode::AtomicSDNode used (only...
next