Recognize sign/zero-extended constant BUILD_VECTORs for VMULL operations.
[oota-llvm.git] / lib / Target / ARM / ARMISelLowering.cpp
2010-11-23 Bob WilsonRecognize sign/zero-extended constant BUILD_VECTORs...
2010-11-23 Wesley PeckRenaming ISD::BIT_CONVERT to ISD::BITCAST to better...
2010-11-19 Evan ChengThese instructions are thumb2 only.
2010-11-18 Tanya LattnerFix bug in DAGCombiner for ARM that was trying to do...
2010-11-18 Anton KorobeynikovMove hasFP() and few related hooks to TargetFrameInfo.
2010-11-18 Bob WilsonSplit up ARM LowerShift function.
2010-11-10 Nate BegemanFix an issue where we tried to turn a v2f32 build_vecto...
2010-11-09 Bob WilsonDo not use MEMBARRIER_MCR for any Thumb code.
2010-11-09 Jim GrosbachChange the ARMConstantPoolValue modifier string to...
2010-11-08 Owen AndersonAdd support for ARM's specialized vector-compare-agains...
2010-11-05 Owen AndersonDisallow the certain NEON modified-immediate forms...
2010-11-05 Owen AndersonAdd codegen and encoding support for the immediate...
2010-11-04 Evan ChengFix @llvm.prefetch isel. Selecting between pld / pldw...
2010-11-03 Owen AndersonCovert VORRIMM to be produced via early target-specific...
2010-11-03 Owen AndersonAdd support for code generation of the one register...
2010-11-03 Bob WilsonCheck for extractelement with a variable operand for...
2010-11-03 Duncan SandsSimplify uses of MVT and EVT. An MVT can be compared...
2010-11-03 Evan ChengFix preload instruction isel. Only v7 supports pli...
2010-11-03 Evan ChengAdd support to match @llvm.prefetch to pld / pldw ...
2010-11-01 Bob WilsonNEON does not support truncating vector stores. Radar...
2010-10-30 Bob WilsonOverhaul memory barriers in the ARM backend. Radar...
2010-10-29 Evan Cheng- Don't schedule nodes with only MVT::Flag and MVT...
2010-10-29 John ThompsonInline asm multiple alternative constraints development...
2010-10-27 Bob WilsonFix compiler warnings about signed/unsigned comparisons.
2010-10-27 Bob WilsonSelectionDAG shuffle nodes do not allow operands with...
2010-10-23 Evan ChengEnable ARM fastcc.
2010-10-22 Evan ChengAdd fastcc cc: pass and return VFP / NEON values in...
2010-10-20 Dale JohannesenFix crash introduced in 116852. 8573915.
2010-10-19 Jim GrosbachAdd a pre-dispatch SjLj EH hook on the unwind edge...
2010-10-19 Dale JohannesenEnable using vdup for vector constants which are splat of
2010-10-15 Jim GrosbachDon't mark argument value stores as immutable, as other...
2010-10-15 Bob WilsonRemove unused ARMISD::AND selection DAG node.
2010-09-28 Anton KorobeynikovUser proper libcall names & condcodes while compiling...
2010-09-28 Bob WilsonAdd a command line option "-arm-strict-align" to disall...
2010-09-24 Evan ChengEnable code placement optimization pass for ARM.
2010-09-22 Jim GrosbachAdd support for ELF PLT references for ARM MC asm print...
2010-09-22 Bob WilsonChange VDUPLANE DAG combiner to just return the result...
2010-09-22 Bob WilsonCombine both VMOVDRR(VMOVRRD) and VMOVRRD(VMOVDRR)...
2010-09-21 Owen AndersonEnable target-specific mul-lowering on ARM, even at...
2010-09-21 Chris Lattnerconvert a couple more places to use the new getStore()
2010-09-21 Bob WilsonDefine the TargetLowering::getTgtMemIntrinsic hook...
2010-09-21 Chris Lattnerconvert the targets off the non-MachinePointerInfo...
2010-09-21 Chris Lattnerreimplement memcpy/memmove/memset lowering to use Machi...
2010-09-17 Bob WilsonAdd target-specific DAG combiner for BUILD_VECTOR and...
2010-09-10 Eric ChristopherSplit out some of the calling convention bits so that...
2010-09-10 Evan ChengTeach if-converter to be more careful with predicating...
2010-09-08 Jim Grosbachremove trailing whitespace
2010-09-03 Bob WilsonReplace NEON vabdl, vaba, and vabal intrinsics with...
2010-09-01 Bob WilsonRemove NEON vmull, vmlal, and vmlsl intrinsics, replaci...
2010-08-29 Bill WendlingCreate an ARMISD::AND node. This node is exactly like...
2010-08-25 Daniel DunbarARM/Thumb2: Fix a misselect in getARMCmp, when attempti...
2010-08-20 Bob WilsonReplace the arm.neon.vmovls and vmovlu intrinsics with...
2010-08-18 Bob WilsonExpand ZERO_EXTEND operations for NEON vector types.
2010-08-17 Bob WilsonAllow more cases of undef shuffle indices and add tests...
2010-08-16 Bob WilsonIgnore undef shuffle indices when checking for a VTRN...
2010-08-13 Bob WilsonTemporarily disable tail calls on ARM to work around...
2010-08-11 Jim Grosbachcortex m4 has floating point support, but only single...
2010-08-11 Bill WendlingConsider this code snippet:
2010-08-11 Evan Cheng- Add subtarget feature -mattr=+db which determine...
2010-08-10 Evan ChengDelete some unused instructions.
2010-08-10 Evan ChengRe-apply r110655 with fixes. Epilogue must restore...
2010-08-10 Daniel DunbarRevert r110655, "Fix ARM hasFP() semantics. It should...
2010-08-10 Evan ChengFix ARM hasFP() semantics. It should return true whenev...
2010-08-04 Dale JohannesenRemove switch for disabling ARM tail calls. They
2010-08-04 Bob WilsonCombine NEON VABD (absolute difference) intrinsics...
2010-08-03 Nate BegemanAdd support for getting & setting the FPSCR application...
2010-07-29 Bob WilsonRefactor ARM-specific DAG combining in preparation...
2010-07-29 Dale JohannesenImplement vector constants which are splat of
2010-07-24 Anton KorobeynikovHook in GlobalMerge pass
2010-07-23 Jim GrosbachUse the appropriate register class for an i32 when...
2010-07-23 Evan Cheng- Allow target to specify when is register pressure...
2010-07-22 Chandler CarruthMark an assert-only variable as used.
2010-07-21 Evan ChengMore register pressure aware scheduling work.
2010-07-21 Eric ChristopherBaby steps towards ARM fast-isel.
2010-07-21 Rafael EspindolaFix calling convention on ARM if vfp2+ is enabled.
2010-07-21 Evan ChengTeach bottom up pre-ra scheduler to track register...
2010-07-20 Jim GrosbachRemoved un-used code.
2010-07-19 Evan ChengARM has to provide its own TargetLowering::findRepresen...
2010-07-19 Jim GrosbachSince ARM emits inline jump tables as part of the Const...
2010-07-19 Jim Grosbachrevert so I can get the right PR# in the log message.
2010-07-19 Jim GrosbachSince ARM emits inline jump tables as part of the Const...
2010-07-17 Jim GrosbachAdd combiner patterns to more effectively utilize the...
2010-07-17 Jim Grosbachadd BFI to getTargetNodeName()
2010-07-17 Jim GrosbachFix logic think-o
2010-07-16 Jim GrosbachAdd basic support to code-gen the ARM/Thumb2 bit-field...
2010-07-15 Evan ChengSplit -enable-finite-only-fp-math to two options:
2010-07-14 Bob WilsonAdd support for NEON VMVN immediate instructions.
2010-07-14 Bob WilsonAdd an ARM-specific DAG combining to avoid redundant...
2010-07-13 Bob WilsonUse a target-specific VMOVIMM DAG node instead of BUILD...
2010-07-13 Evan ChengExtend the r107852 optimization which turns some fp...
2010-07-13 Bob WilsonMove NEON "modified immediate" encode/decode into ARMAd...
2010-07-12 Bob WilsonRemove some code that doesn't appear to do anything...
2010-07-11 Rafael EspindolaFix va_arg for doubles. With this patch VAARG nodes...
2010-07-08 Evan ChengCheck for FiniteOnlyFPMath as well.
2010-07-08 Evan Chengr107852 is only safe with -enable-unsafe-fp-math to...
2010-07-08 Evan ChengOptimize some vfp comparisons to integer ones. This...
2010-07-08 Dale JohannesenChanges to ARM tail calls, mostly cosmetic.
2010-07-07 Dan GohmanSplit the SDValue out of OutputArg so that SelectionDAG...
2010-07-07 Jim GrosbachMark eh.sjlj.set/longjmp custom lowerings as Darwin...
2010-07-06 Jim GrosbachBy default, the eh.sjlj.setjmp/longjmp intrinsics shoul...
next