Fix handling of 'g'. Closes 883
[oota-llvm.git] / lib / Target / X86 / X86ISelLowering.cpp
2006-08-17 Andrew LenharthFix handling of 'g'. Closes 883
2006-08-17 Andrew LenharthAdd the 'c' constraint as needed by the linux kernel
2006-08-17 Andrew LenharthAdd support for S and D constraints, as needed to compi...
2006-08-11 Chris LattnerEliminate use of getNode that takes a vector.
2006-08-11 Evan ChengMatch tablegen changes.
2006-08-11 Evan ChengConvert more calls of getNode() that takes a vector...
2006-08-08 Chris LattnerStart eliminating temporary vectors used to create...
2006-07-31 Chris LattnerFix PR850 and CodeGen/X86/2006-07-31-SingleRegClass.ll.
2006-07-26 Chris LattnerRename RelocModel::PIC to PIC_, to avoid conflicts...
2006-07-21 Evan ChengThis opt is now handled in DAG combine.
2006-07-20 Evan ChengA splat of a vector constant of all zero or all one...
2006-07-12 Chris LattnerAdd information preventing several register class const...
2006-07-11 Chris LattnerImplement the inline asm 'A' constraint. This implemen...
2006-07-10 Evan ChengFixed stack objects do not specify alignments, but...
2006-07-10 Chris LattnerMark internal function static
2006-07-07 Evan ChengX86 target specific DAG combine: turn build_vector...
2006-07-05 Evan ChengReorg. No functionality change.
2006-06-24 Evan ChengSimplify X86CompilationCallback: always align to 16...
2006-06-15 Evan ChengType of vector extract / insert index operand should...
2006-06-14 Evan ChengAdd argument registers to the end of call operand list...
2006-06-09 Evan ChengMinor compilation speed improvement.
2006-06-06 Evan ChengAdded X86FunctionInfo subclass of MachineFunction to...
2006-06-01 Evan ChengTypos
2006-06-01 Evan ChengRemove a warning
2006-05-31 Evan ChengRemove dead code.
2006-05-26 Evan ChengChange RET node to include signness information of...
2006-05-26 Evan ChengVector argument must be passed in memory location align...
2006-05-26 Evan ChengMac OS X ABI document lied. The first four XMM register...
2006-05-26 Evan ChengMinor update to make the code more clear
2006-05-26 Evan ChengUpdate more comments.
2006-05-26 Evan ChengFix some comments.
2006-05-26 Evan ChengNo need to handle illegal types.
2006-05-25 Evan ChengConsistency
2006-05-25 Evan ChengSome clean up.
2006-05-25 Evan ChengRemove some dead code.
2006-05-25 Evan ChengBuild breakage.
2006-05-25 Evan ChengSwitch X86 over to a call-selection model where the...
2006-05-23 Chris LattnerFix file header comment
2006-05-23 Evan ChengBetter way to check for vararg.
2006-05-23 Evan ChengRemove PreprocessCCCArguments and PreprocessFastCCArgum...
2006-05-23 Chris LattnerImplement an annoying part of the Darwin/X86 abi: the...
2006-05-19 Chris LattnerCSRet allows varargs
2006-05-17 Evan ChengShould pass by reference.
2006-05-16 Chris LattnerImplement the custom lowering hook right, returning...
2006-05-16 Chris LattnerFix a bug I introduced yesterday, which broke functions...
2006-05-16 Evan ChengX86 integer register classes naming changes. Make them...
2006-05-16 Chris LattnerAdd a chain to FORMAL_ARGUMENTS. This is a minimal...
2006-05-12 Chris LattnerDead variable
2006-05-06 Chris LattnerTeach the X86 backend about non-i32 inline asm register...
2006-05-05 Chris LattnerTeach the code generator to use cvtss2sd as extload...
2006-05-03 Owen AndersonRefactor TargetMachine, pushing handling of TargetData...
2006-04-28 Evan ChengInitial caller side support (for CCC only, not FastCC...
2006-04-28 Evan ChengImplement four-wide shuffle with 2 shufps if no more...
2006-04-28 Evan ChengTargetLowering::LowerArguments should return a VBIT_CON...
2006-04-27 Evan ChengMake x86 isel lowering produce tailcall nodes. They...
2006-04-27 Evan ChengSupport for passing 128-bit vector arguments via XMM...
2006-04-27 Evan ChengOops
2006-04-27 Evan ChengBug fix: not updating NumIntRegs.
2006-04-27 Evan Cheng- Clean up formal argument lowering code. Prepare for...
2006-04-26 Evan ChengFix fastcc failures.
2006-04-26 Evan ChengSwitching over FORMAL_ARGUMENTS mechanism to lower...
2006-04-25 Evan ChengSeparate LowerOperation() into multiple functions,...
2006-04-24 Evan ChengSpecial case handling two wide build_vector(0, x).
2006-04-24 Evan ChengA little bit more build_vector enhancement for v8i16...
2006-04-23 Evan ChengMOVL shuffle (i.e. movd or movss / movsd from memory...
2006-04-22 Nate BegemanJumpTable support! What this represents is working...
2006-04-22 Evan ChengDon't do all the lowering stuff for 2-wide build_vector...
2006-04-22 Evan ChengFix a performance regression. Use {p}shuf* when there...
2006-04-21 Evan ChengRevamp build_vector lowering to take advantage of movss...
2006-04-21 Evan ChengNow generating perfect (I think) code for "vector set...
2006-04-20 Evan Cheng- Added support to turn "vector clear elements", e...
2006-04-20 Evan ChengHandle v2i64 BUILD_VECTOR custom lowering correctly...
2006-04-19 Evan ChengisSplatMask() bug: first element can be an undef.
2006-04-19 Evan Cheng- Added support to do aribitrary 4 wide shuffle with...
2006-04-19 Evan ChengCommute vector_shuffle to match more movlhps, movlp...
2006-04-17 Evan ChengUse movss to insert_vector_elt(v, s, 0).
2006-04-17 Evan ChengUse two pinsrw to insert an element into v4i32 / v4f32...
2006-04-17 Evan ChengImplement v8i16, v16i8 splat using unpckl + pshufd.
2006-04-17 Chris Lattnerimplement returns of a vector, testcase here: CodeGen...
2006-04-17 Evan ChengFP SETOLT, SETOLT, SETUGE, SETUGT conditions were imple...
2006-04-15 Evan ChengSilly bug
2006-04-15 Evan ChengDo not use movs{h|l}dup for a shuffle with a single...
2006-04-14 Evan ChengLast few SSE3 intrinsics.
2006-04-13 Evan ChengX86 SSE2 supports v8i16 multiplication
2006-04-12 Evan ChengAll "integer" logical ops (pand, por, pxor) are now...
2006-04-12 Evan ChengPromote v4i32, v8i16, v16i8 load to v2i64 load.
2006-04-11 Evan ChengAdded support for _mm_move_ss and _mm_move_sd.
2006-04-10 Evan ChengConditional move of vector types.
2006-04-07 Evan ChengCode clean up.
2006-04-06 Evan Cheng- movlp{s|d} and movhp{s|d} support.
2006-04-05 Evan ChengSupport for comi / ucomi intrinsics.
2006-04-05 Evan ChengHandle canonical form of e.g.
2006-04-05 Evan ChengBogus assert
2006-04-05 Evan ChengFallthrough to expand if a VECTOR_SHUFFLE cannot be...
2006-04-05 Evan ChengHandle v8i16 shuffle that must be broken into a pair...
2006-04-03 Evan ChengUse movlpd to: store lower f64 extracted from v2f64.
2006-04-03 Evan Cheng- More efficient extract_vector_elt with shuffle and...
2006-03-31 Evan ChengUse a X86 target specific node X86ISD::PINSRW instead...
2006-03-31 Evan ChengAdd support to use pextrw and pinsrw to extract and...
2006-03-31 Evan ChengExpand all INSERT_VECTOR_ELT (obviously bad) for now.
next