summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Target/X86/X86IntrinsicsInfo.h
Commit message (Expand)AuthorAgeFilesLines
* [AVX-512] Remove vinsert intrinsics and autoupgrade to native shufflevectors....Craig Topper2017-01-031-25/+1
* [AVX-512] Remove masked pmuldq and pmuludq intrinsics and autoupgrade them to...Craig Topper2016-12-271-12/+0
* [AVX-512] Add 512-bit unmasked intrinsics for pmuldq and pmuludq so we can ad...Craig Topper2016-12-271-0/+2
* Added a template for building target specific memory node in DAG.Elena Demikhovsky2016-12-211-0/+73
* [X86] Remove masking from 512-bit VPERMIL intrinsics in preparation for being...Craig Topper2016-12-111-4/+2
* [X86] Remove masking from 512-bit PSHUFB intrinsics in preparation for being ...Craig Topper2016-12-101-2/+1
* [AVX-512] Remove 128/256 masked vpermil instrinsics and autoupgrade to a sele...Craig Topper2016-12-101-8/+0
* [X86] Use X86ISD::CVTTP2SI and X86ISD::CVTTP2UI for lowering 128-bit cvttps2q...Craig Topper2016-12-101-2/+2
* [AVX-512] Correctly preserve the passthru semantics of the FMA scalar intrinsicsCraig Topper2016-12-091-10/+10
* [X86] Generalize CVTTPD2DQ/CVTTPD2UDQ and CVTDQ2PD/CVTUDQ2PD opcodes. NFCI Simon Pilgrim2016-11-241-5/+5
* [AVX-512] Remove intrinsics for valignd/q and autoupgrade them to native shuf...Craig Topper2016-11-231-12/+0
* [AVX-512] Replace masked 16-bit element variable shift intrinsics with new un...Craig Topper2016-11-181-9/+9
* [X86][AVX512] Autoupgrade lossless i32/u32 to f64 conversion intrinsics with ...Simon Pilgrim2016-11-161-12/+0
* [X86][AVX512] Removing llvm x86 intrinsics for _mm_mask_move_{ss|sd} intrinsics.Ayman Musa2016-11-161-4/+0
* [AVX-512] Remove and autoupgrade masked dword/qword variable shift intrinsics...Craig Topper2016-11-141-8/+0
* revert commit r286761, some builds failed on Win platformsIgor Breger2016-11-131-0/+4
* [X86][AVX512] Removing llvm x86 intrinsics for _mm_mask_move_{ss|sd} intrinsics.Ayman Musa2016-11-131-4/+0
* [AVX-512] Add unmasked intrinsics for variable shifts of dwords and qwords.Craig Topper2016-11-131-0/+8
* [AVX-512] Remove the remaining masked shift by immediate or by single value. ...Craig Topper2016-11-121-22/+0
* [AVX-512] Add unmasked version of shift by immediate and shift by single elem...Craig Topper2016-11-121-0/+22
* [AVX-512] Allow legacy cvtpd2dq intrinsics to select EVEX encoded instruction...Craig Topper2016-11-101-0/+2
* [AVX-512][X86] Convert avx_cvtt_ps2dq_256 and sse2_cvttps2dq intrinsics to IS...Craig Topper2016-11-101-0/+2
* [X86] Convert int_x86_avx_cvtt_pd2dq_256 to fp_to_sint using the intrinsics t...Craig Topper2016-11-101-0/+1
* [AVX-512] Add lowering to cvttpd2udq/cvttps2udq for fptoui v2f64/2f32 to 2i32Craig Topper2016-11-091-1/+1
* [X86] Lower AVX512 and SSE intrinsics for CVTTPD2DQ to X86ISD::CVTTPD2DQ.Craig Topper2016-11-091-1/+2
* [AVX-512] Remove masked pmovzx/pmovsx builtins and autoupgrade them to select...Craig Topper2016-11-071-72/+0
* [AVX-512] Remove 128/256 masked pshufb intrinsics. Autoupgrade them to legacy...Craig Topper2016-11-071-4/+0
* [AVX-512] Remove intrinsics for 128/256-bit masked variable shift. Instead up...Craig Topper2016-11-061-10/+0
* [AVX-512] Remove intrinsics for 128/256-bit masked shift by immediate. Instea...Craig Topper2016-11-061-16/+0
* [AVX-512] Remove intrinsics for 128/256-bit masked shift by single element in...Craig Topper2016-11-061-16/+0
* [AVX-512] Lower AVX cvtpd2ps intrinsic to ISD::FP_ROUND so it can use EVEX in...Craig Topper2016-11-061-3/+5
* [AVX-512] Lower SSE/AVX cvtdq2ps intrinsics directly to ISD::SINT_TO_FP so th...Craig Topper2016-11-061-0/+2
* [X86] Use intrinsics table for PMADDUBSW and PMADDWD so that we can use the l...Craig Topper2016-10-301-0/+4
* [X86] Use intrinsics table for VPMULHRSW intrincis so that the legacy intrins...Craig Topper2016-10-291-0/+2
* [AVX-512] Add scalar vfmsub/vfnmsub mask3 intrinsicsCraig Topper2016-10-261-0/+4
* [AVX-512] Remove masked pmin/pmax intrinsics and autoupgrade to native IR.Craig Topper2016-10-241-48/+0
* [AVX-512] Fix the operand order for vpermi2var_qi intrinsics to match the oth...Craig Topper2016-10-161-3/+3
* [AVX-512] Don't use two opcodes for INTR_TYPE_SCALAR_MASK_RM. The handling wa...Craig Topper2016-09-251-20/+20
* [AVX-512] Split X86ISD::VFPROUND and X86ISD::VFPEXT into separate opcodes for...Craig Topper2016-09-231-4/+4
* [AVX-512] Add separate ISD opcodes for each form of CVT instructions. Don't r...Craig Topper2016-09-231-68/+60
* [AVX-512] Use different ISD opcodes for some of the scalar intrinsic lowering...Craig Topper2016-09-231-14/+14
* [AVX-512] Split the 3 different usages of the X86ISD::FSETCC opcode into 3 di...Craig Topper2016-09-211-4/+4
* [AVX-512] Don't add an additional rounding mode operand to the avx512 vcvtps2...Craig Topper2016-09-211-3/+3
* [AVX-512] Don't lower avx512 vcvtps2ph/vcvtph2ps nodes to ISD::FP16_TO_FP/ISD...Craig Topper2016-09-211-6/+6
* [AVX-512] Don't lower CVTPD2PS intrinsics to ISD::FP_ROUND with an X86 roundi...Craig Topper2016-09-181-3/+3
* [AVX-512] Stop lowering avx512_mask_sqrt intrinsics to ISD:FSQRT with a secon...Craig Topper2016-09-181-2/+2
* [X86] Remove the VCVTSI2SD32 with rounding intrinsic. It's not used by clang ...Craig Topper2016-09-141-1/+0
* [X86] Remove masked shufpd/shufps intrinsics and autoupgrade to native vector...Craig Topper2016-09-131-12/+0
* [X86] Remove unused encoding from IntrinsicType enum.Craig Topper2016-09-061-1/+1
* [AVX-512] Remove 128-bit and 256-bit masked floating point add/sub/mul/div in...Craig Topper2016-09-041-16/+0
OpenPOWER on IntegriCloud