aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
AgeCommit message (Expand)AuthorFilesLines
2016-08-17Replace "fallthrough" comments with LLVM_FALLTHROUGHJustin Bogner1-1/+1
2016-08-08Revert "[X86] Support the "ms-hotpatch" attribute."Charles Davis1-15/+0
2016-08-08[X86] Support the "ms-hotpatch" attribute.Charles Davis1-0/+15
2016-08-04[LV, X86] Be more optimistic about vectorizing shifts.Michael Kuperstein1-21/+22
2016-08-04[X86][SSE] Add initial costs for vector CTTZ/CTLZSimon Pilgrim1-4/+41
2016-08-02[AVX512] Don't use i128 masked gather/scatter/load/store. Do more accurately ...Igor Breger1-3/+3
2016-07-20[X86][SSE] Add cost model values for CTPOP of vectorsSimon Pilgrim1-4/+27
2016-07-17Strip trailing whitespaceSimon Pilgrim1-6/+6
2016-07-11[X86] Make some cast costs more preciseMichael Kuperstein1-3/+16
2016-07-06[x86] fix cost of SINT_TO_FP for i32 --> float (PR21356, PR28434)Sanjay Patel1-1/+1
2016-07-06[X86] Sort cast cost tables. NFC.Michael Kuperstein1-124/+123
2016-06-20[X86][SSE] Add cost model for BSWAP of vectorsSimon Pilgrim1-3/+24
2016-06-11[CostModel][X86][SSE] Updated costs for vector BITREVERSE ops on SSSE3+ targetsSimon Pilgrim1-0/+30
2016-06-10[X86] Add costs for SSE zext/sext to v4i64 to TTIMichael Kuperstein1-0/+14
2016-05-25[x86] avoid code explosion from LoopVectorizer for gather loop (PR27826) Sanjay Patel1-2/+10
2016-05-24[CostModel][X86][XOP] Added XOP costmodel for BITREVERSE Simon Pilgrim1-1/+44
2016-05-09[X86][SSE] Improve cost model for i64 vector comparisons on pre-SSE42 targetsSimon Pilgrim1-3/+11
2016-04-22[X86]: Changing cost for “TRUNCATE v16i32 to v16i8” in SSE4.1 mode.Ashutosh Nema1-2/+0
2016-04-14Do not use getGlobalContext()... ever.Mehdi Amini1-5/+5
2016-04-05fix typo; NFCSanjay Patel1-1/+1
2016-03-09[x86] fix cost model inaccuracy for vector memory opsSanjay Patel1-4/+4
2016-03-06AVX512BW: Support llvm intrinsic masked vector load/store for i8/i16 element ...Igor Breger1-1/+2
2016-01-25AVX1 : Enable vector masked_load/store to AVX1.Igor Breger1-1/+1
2015-12-28Implemented cost model for masked gather and scatter operationsElena Demikhovsky1-0/+136
2015-12-21[X86][SSE] Transform truncations between vectors of integers into X86ISD::PAC...Cong Hou1-3/+3
2015-12-20[X86] Prevent constant hoisting for a couple compare immediates that the sele...Craig Topper1-1/+13
2015-12-11[X86][SSE] Update the cost table for integer-integer conversions on SSE2/SSE4.1.Cong Hou1-2/+79
2015-12-02AVX-512: Updated cost of FP/SINT/UINT conversion operationsElena Demikhovsky1-11/+61
2015-11-19Pointers in Masked Load, Store, Gather, Scatter intrinsicsElena Demikhovsky1-8/+4
2015-10-28[X86] A small fix in X86/X86TargetTransformInfo.cpp: check a value type is si...Cong Hou1-1/+2
2015-10-28Remove templates from CostTableLookup functions. All instantiations had the s...Craig Topper1-30/+25
2015-10-27Convert cost table lookup functions to return a pointer to the entry or nullp...Craig Topper1-101/+74
2015-10-25Scalarizer for masked.gather and masked.scatter intrinsics.Elena Demikhovsky1-0/+27
2015-10-25Remove two unnecessary conversions from MVT to EVT. NFCCraig Topper1-2/+2
2015-10-22Partially reverted changes from r250686Elena Demikhovsky1-2/+4
2015-10-19Removed parameter "Consecutive" from isLegalMaskedLoad() / isLegalMaskedStore().Elena Demikhovsky1-12/+9
2015-10-17[CostModel] Fixed AVX integer shift costsSimon Pilgrim1-12/+36
2015-10-06Fix Clang-tidy modernize-use-nullptr warnings in source directories and gener...Hans Wennborg1-4/+5
2015-10-06[X86] Teach constant hoisting that ANDs with 64-bit immediates in the range 0...Craig Topper1-1/+7
2015-09-30[X86][XOP] Added support for the lowering of 128-bit vector shifts to XOP shi...Simon Pilgrim1-15/+61
2015-08-05[TTI] Make the cost APIs in TargetTransformInfo consistently use 'int'Chandler Carruth1-64/+56
2015-07-29Rename hasCompatibleFunctionAttributes->areInlineCompatible basedEric Christopher1-2/+2
2015-07-29[X86][SSE] Vectorize i64 ASHR operationsSimon Pilgrim1-2/+3
2015-07-19[X86][SSE] Reordered cast vectorization costs. NFCI.Simon Pilgrim1-47/+48
2015-07-18[X86][SSE] Updated SHL/LSHR i64 vectorization costs.Simon Pilgrim1-3/+3
2015-07-14Prune trailing whitespaces and CRs.NAKAMURA Takumi1-23/+23
2015-07-12[X86][SSE] Vectorized v4i32 non-uniform shifts.Simon Pilgrim1-23/+23
2015-07-09Make TargetLowering::getPointerTy() taking DataLayout as an argumentMehdi Amini1-13/+13
2015-07-06[X86][SSE] Vectorized i64 uniform constant SRA shiftsSimon Pilgrim1-0/+3
2015-07-02Implement TargetTransformInfo::hasCompatibleFunctionAttributes for X86.Eric Christopher1-0/+15
generated by cgit v1.1 at 2025-10-18 01:07:19 +0000