Use movups to lower memcpy and memset even if it's not fast (like corei7).
[oota-llvm.git] / lib / Target / X86 / X86ISelLowering.cpp
2011-01-06 Evan ChengUse movups to lower memcpy and memset even if it's...
2011-01-06 Evan ChengRe-implement r122936 with proper target hooks. Now...
2010-12-22 Benjamin KramerX86: Lower a select directly to a setcc_carry if possible.
2010-12-21 Benjamin KramerAdd some x86 specific dagcombines for conditional incre...
2010-12-21 Chris Lattnerrename MVT::Flag to MVT::Glue. "Flag" is a terrible...
2010-12-20 Nate BegemanImplement feedback from Bruno on making pblendvb an...
2010-12-20 Chris Lattnernow that addc/adde are gone, "ADDC" in the X86 backend...
2010-12-20 Chris Lattneruse for loop over types.
2010-12-20 Chris LattnerChange the X86 backend to stop using the evil ADDC...
2010-12-19 Mon P WangPrevents PerformShuffleCombine from creating a node...
2010-12-19 Chris Lattnerimprove the setcc -> setcc_carry optimization to happen...
2010-12-19 Chris Lattnersimplify some code to just reuse a setcc if we can...
2010-12-19 Chris Lattnernow that generic vector types aren't selected onto...
2010-12-19 Chris Lattnerreduce copy/paste programming with the power of for...
2010-12-19 Chris LattnerX86 supports i8/i16 overflow ops (except i8 multiplies...
2010-12-17 Nate BegemanAdd support for matching psign & plendvb to the x86...
2010-12-10 Nate BegemanFormalize the notion that AVX and SSE are non-overlappi...
2010-12-09 Eric ChristopherRewrite the darwin tlv support to use a chain and retur...
2010-12-09 Eric ChristopherStop confusing people, it's not really a chain, or...
2010-12-09 Eric ChristopherRemove extraneous copy from DAG conversion for darwin...
2010-12-05 Chris LattnerTeach X86ISelLowering that the second result of X86ISD...
2010-12-05 Chris Lattnerit turns out that when ".with.overflow" intrinsics...
2010-12-05 Chris Lattnergeneralize the previous check to handle -1 on either...
2010-12-05 Chris LattnerImprove an integer select optimization in two ways:
2010-12-04 Benjamin KramerAdd patterns for the x86 popcnt instruction.
2010-12-04 Benjamin KramerSimplify code. No functionality change.
2010-12-01 Evan ChengFix and re-enable tail call optimization of expanded...
2010-12-01 Duncan SandsI don't think it makes any sense to assert that the...
2010-12-01 Evan ChengSpeculatively disable x86 portion of r120501 to appease...
2010-11-30 Evan ChengEnable sibling call optimization of libcalls which...
2010-11-30 Eric ChristopherFix insertion point in pcmp expander.
2010-11-30 Eric ChristopherFix some cleanups from my last patch.
2010-11-30 Eric ChristopherRewrite mwait and monitor support and custom lower...
2010-11-28 Rafael EspindolaMove lowering of TLS_addr32 and TLS_addr64 to X86MCInst...
2010-11-27 Rafael EspindolaLower TLS_addr32 and TLS_addr64.
2010-11-23 Wesley PeckRenaming ISD::BIT_CONVERT to ISD::BITCAST to better...
2010-11-18 Anton KorobeynikovMove hasFP() and few related hooks to TargetFrameInfo.
2010-11-15 Chris Lattneradd targetoperand flags for jump tables, constant pool...
2010-11-14 Chris Lattnermove the pic base symbol stuff up to MachineFunction
2010-11-14 Chris Lattnersimplify getPICBaseSymbol a bit.
2010-11-13 Peter CollingbourneRecognise 32-bit ror-based bswap implementation used...
2010-11-13 Peter CollingbourneSupport ; as asm separator
2010-11-12 Dale JohannesenRemove possibly useful info from comment, per Chris.
2010-11-03 Duncan SandsSimplify uses of MVT and EVT. An MVT can be compared...
2010-10-31 Duncan SandsFactorize the duplicated logic for choosing the right...
2010-10-29 John ThompsonInline asm multiple alternative constraints development...
2010-10-27 Michael J. Spencerx86-Win32: Switch ftol2 calling convention from stdcall...
2010-10-25 Dale JohannesenAn stdcall function calling a non-stdcall function
2010-10-21 Duncan SandsAdd parentheses to pacify gcc, which warns otherwise.
2010-10-21 Michael J. SpencerX86: Add alloca probing to dynamic alloca on Windows...
2010-10-20 Dale JohannesenRemove Synthesizable from the Type system; as MMX vector
2010-10-19 Michael J. SpencerX86: Add MS-CRT libcalls.
2010-10-19 Michael J. SpencerFix Whitespace.
2010-10-12 Eric ChristopherCombine these together - should probably have some...
2010-10-12 Nick LewyckyMark variable 'NoImplicitFloatOps' used only in an...
2010-10-12 Dan GohmanInitial va_arg support for x86-64. Patch by David Meyer!
2010-10-11 Andrew TrickFixes bug 8297: i386 cmpxchg8b, missing MachineMemOperand
2010-10-11 Michael J. SpencerX86: Call ulldiv and ftol2 on Windows instead of their...
2010-10-10 Michael J. SpencerX86: MinGW should always use libgcc on Windows.
2010-10-10 Michael J. SpencerX86: Call _alldiv instead of __divdi3 on Windows (exclu...
2010-10-10 Michael J. SpencerFix Whitespace.
2010-10-08 Cameron EsfahaniRecommit 116056, now with the missing file...
2010-10-08 Andrew Trickreverting 116056: win64_params.ll may need to be condit...
2010-10-08 Cameron EsfahaniSmall patch to restore home register stack space alloca...
2010-10-07 Evan ChengCanonicalize X86ISD::MOVDDUP nodes to v2f64 to make...
2010-10-03 Anton Korobeynikovva_args support for Win64.
2010-09-30 Dale JohannesenMassive rewrite of MMX:
2010-09-27 Chris Lattnerimprove indentation
2010-09-27 Eric ChristopherThis code should never fire on non-darwin subtargets.
2010-09-24 Dale JohannesenWe can't return SSE/MMX vectors if SSE is disabled.
2010-09-22 Bob WilsonAttempt to fix llvm-gcc build. It was crashing when...
2010-09-22 Chris Lattnerreimplement elf TLS support in terms of addressing...
2010-09-22 Chris Lattnerconvert the last 4 X86ISD nodes that should have memope...
2010-09-22 Chris Lattnergive X86ISD::FNSTCW16m a memoperand, since it touches...
2010-09-22 Chris Lattnergive FP_TO_INT16_IN_MEM and friends a memoperand. ...
2010-09-22 Chris Lattnergive VZEXT_LOAD a memory operand, it now works with...
2010-09-21 Chris Lattnergive LCMPXCHG_DAG[8] a memory operand, allowing it...
2010-09-21 Owen AndersonReimplement r114460 in target-independent DAGCombine...
2010-09-21 Chris Lattnerconvert a couple more places to use the new getStore()
2010-09-21 Owen AndersonWhen adding the carry bit to another value on X86,...
2010-09-21 Chris Lattnereliminate some uses of the getStore overload.
2010-09-21 Chris Lattnerconvert the targets off the non-MachinePointerInfo...
2010-09-21 Chris Lattnerit's more elegant to put the "getConstantPool" and
2010-09-21 Chris Lattnerupdate the X86 backend to use the MachinePointerInfo...
2010-09-21 Chris Lattnerreimplement memcpy/memmove/memset lowering to use Machi...
2010-09-21 Chris Lattnerconvert targets to the new MF.getMachineMemOperand...
2010-09-13 John ThompsonAdded skeleton for inline asm multiple alternative...
2010-09-08 Bruno Cardoso LopesMinor change. Fix comments and remove unused and redund...
2010-09-08 Bruno Cardoso Lopesx86 vector shuffle lowering now relies only on target...
2010-09-07 Bruno Cardoso LopesFactor out some x86 vector shuffle rewriting and add...
2010-09-07 Bruno Cardoso LopesMove code around to prepare for moving some of the...
2010-09-07 Bill WendlingAdd an MVT::x86mmx type. It will take the place of...
2010-09-07 Bruno Cardoso Lopesdecouple MMX check from regular splat checks. Some...
2010-09-07 Bruno Cardoso LopesRemove now useless check, because the code can be match...
2010-09-07 Bruno Cardoso LopesMinor change. Since the checks are equivalent, use...
2010-09-04 Bruno Cardoso LopesRemove the last bit of isShuffleMaskLegal checks and...
2010-09-04 Bruno Cardoso Lopesmake explicit that we not handle several mmx shuffles
2010-09-04 Bruno Cardoso LopesEmit target specific nodes to handle palignr. Do not...
2010-09-04 Bruno Cardoso LopesEmit target specific nodes to handle splats starting...
2010-09-04 Bruno Cardoso LopesEmit target specific nodes for isPSHUFHWMask and isPSHU...
next