Programming Languages Research Group: Git

2015-09-21	Simon Pilgrim	[DAGCombiner] Improve FMA support for interpolation...	commit \| commitdiff \| tree
2015-09-21	Simon Pilgrim	[DAGCombiner] Tidy up FMA combine helpers. NFCI.	commit \| commitdiff \| tree
2015-09-20	Simon Pilgrim	[X86][SSE] Intrinsics builtins test refresh. NFCI	commit \| commitdiff \| tree
2015-09-19	Simon Pilgrim	[X86][AVX2] Use general sext IR for vpmovsx stack folding...	commit \| commitdiff \| tree
2015-09-19	Simon Pilgrim	[X86][SSE] Vectorize CTTZ + CTTZ_ZERO_UNDEF	commit \| commitdiff \| tree
2015-09-19	Simon Pilgrim	[InstCombine] Use SimplifyDemandedVectorEltsLow helper...	commit \| commitdiff \| tree
2015-09-17	Simon Pilgrim	[InstCombine] Added vector demanded bits support for...	commit \| commitdiff \| tree
2015-09-13	Simon Pilgrim	[X86][MMX] Added shuffle decodes for MMX/3DNow! shuffles.	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[X86] Renamed lowerVectorShuffleAsUnpack NFCI.	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[X86] Moved lowerVectorShuffleWithUNPCK earlier to...	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[X86] Added i1 vector sextload tests	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[X86][FMA] Refreshed fma tests	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	Fixed unused variable warning.	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[InstCombine] CVTPH2PS Vector Demanded Elements + Constant...	commit \| commitdiff \| tree
2015-09-12	Simon Pilgrim	[X86][SSE] Use general sext IR for (v)pmovsx stack...	commit \| commitdiff \| tree
2015-09-06	Simon Pilgrim	[X86][MMX] Added missing stack folding tests for MMX...	commit \| commitdiff \| tree
2015-09-06	Simon Pilgrim	[X86][AVX512] Added 512-bit vector shift tests.	commit \| commitdiff \| tree
2015-09-05	Simon Pilgrim	[X86] Updated vector lzcnt tests. Added missing vec512...	commit \| commitdiff \| tree
2015-09-05	Simon Pilgrim	[X86] Updated vector tzcnt tests. Added vec512 tests.	commit \| commitdiff \| tree
2015-09-05	Simon Pilgrim	[X86] Updated vector popcnt tests. Added vec512 tests.	commit \| commitdiff \| tree
2015-09-04	Simon Pilgrim	[X86][AVX] Test tidyup + regeneration. NFCI.	commit \| commitdiff \| tree
2015-08-24	Simon Pilgrim	[X86][SSE] Added tests for zero-extension vector shuffles...	commit \| commitdiff \| tree
2015-08-23	Simon Pilgrim	[DAGCombiner] Fold CONCAT_VECTORS of bitcasted EXTRACT_SUBVECTOR	commit \| commitdiff \| tree
2015-08-21	Simon Pilgrim	Line endings fix.	commit \| commitdiff \| tree
2015-08-19	Simon Pilgrim	[DAGCombiner] Added SMAX/SMIN/UMAX/UMIN constant folding	commit \| commitdiff \| tree
2015-08-19	Simon Pilgrim	[DAGCombiner] Fold CONCAT_VECTORS of EXTRACT_SUBVECTOR...	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	[X86] Refreshed sign extension tests.	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	[X86][AVX] Added shuffle concatenation tests	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	Updated constants to give more useful min/max constant...	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	Fixed max/min typo in test names	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	[X86][SSE} Added constant SMAX/SMIN/UMAX/UMIN tests	commit \| commitdiff \| tree
2015-08-18	Simon Pilgrim	[X86][SSE] Added extra vector truncation tests.	commit \| commitdiff \| tree
2015-08-15	Simon Pilgrim	Updated broadcast stack folding test to avoid use of...	commit \| commitdiff \| tree
2015-08-15	Simon Pilgrim	[DAGCombiner] Attempt to mask vectors before zero extension...	commit \| commitdiff \| tree
2015-08-14	Simon Pilgrim	Renamed min tests (typo)	commit \| commitdiff \| tree
2015-08-13	Simon Pilgrim	[AMDGPU] Use the general SMAX/SMIN/UMAX/UMIN pattern...	commit \| commitdiff \| tree
2015-08-13	Simon Pilgrim	[X86][SSE] Use the general SMAX/SMIN/UMAX/UMIN pattern...	commit \| commitdiff \| tree
2015-08-13	Simon Pilgrim	[X86][SSE] Tests for SMAX/SMIN/UMAX/UMIN vector instructions	commit \| commitdiff \| tree
2015-08-13	Simon Pilgrim	[InstCombine] SSE/AVX vector shifts demanded shift...	commit \| commitdiff \| tree
2015-08-12	Simon Pilgrim	Cleaned up test. NFCI.	commit \| commitdiff \| tree
2015-08-12	Simon Pilgrim	unused variable warning fix.	commit \| commitdiff \| tree
2015-08-12	Simon Pilgrim	[InstCombine] Move SSE/AVX vector blend folding to...	commit \| commitdiff \| tree
2015-08-10	Simon Pilgrim	[InstCombine] Move SSE2/AVX2 arithmetic vector shift...	commit \| commitdiff \| tree
2015-08-07	Simon Pilgrim	[InstCombine] Fix SSE2/AVX2 vector logical shift by...	commit \| commitdiff \| tree
2015-08-05	Simon Pilgrim	[InstCombine] Added more specific SSE2/AVX2 vector...	commit \| commitdiff \| tree
2015-08-05	Simon Pilgrim	Fixed line endings.	commit \| commitdiff \| tree
2015-08-04	Simon Pilgrim	[InstCombine] Split off SSE2/AVX2 vector shift tests.	commit \| commitdiff \| tree
2015-08-04	Simon Pilgrim	[InstCombine] Moved SSE vector shift constant folding...	commit \| commitdiff \| tree
2015-08-02	Simon Pilgrim	[X86][SSE] Refreshed sse2 vector shift tests	commit \| commitdiff \| tree
2015-08-01	Simon Pilgrim	Remove trailing whitespace. NFCI.	commit \| commitdiff \| tree
2015-08-01	Simon Pilgrim	Use SDValue bool check. NFCI.	commit \| commitdiff \| tree
2015-08-01	Simon Pilgrim	[DAGCombiner] Convert constant AND masks to shuffle...	commit \| commitdiff \| tree
2015-07-29	Simon Pilgrim	[X86][SSE] Keep 32-bit target i64 vector shifts on...	commit \| commitdiff \| tree
2015-07-29	Simon Pilgrim	[X86][SSE] Vectorize i64 ASHR operations	commit \| commitdiff \| tree
2015-07-28	Simon Pilgrim	[X86][SSE] Use bitmasks instead of shuffles where possible.	commit \| commitdiff \| tree
2015-07-27	Simon Pilgrim	[X86][SSE] Added shuffle tests to demonstrate missed...	commit \| commitdiff \| tree
2015-07-27	Simon Pilgrim	Fixed signed/unsigned comparison warning.	commit \| commitdiff \| tree
2015-07-27	Simon Pilgrim	[InstCombine][X86][SSE] Replace sign/zero extension...	commit \| commitdiff \| tree
2015-07-27	Simon Pilgrim	[X86] Reordered lowerVectorShuffleAsBitMask before...	commit \| commitdiff \| tree
2015-07-26	Simon Pilgrim	[X86][SSE] Refreshed vector bit count tests.	commit \| commitdiff \| tree
2015-07-26	Simon Pilgrim	[X86][AVX2] Refreshed avx2 conversion tests	commit \| commitdiff \| tree
2015-07-25	Simon Pilgrim	[InstCombine][SSE4A] Standardized references to Length...	commit \| commitdiff \| tree
2015-07-25	Simon Pilgrim	[InstCombine] Split off SSE4a tests.	commit \| commitdiff \| tree
2015-07-25	Simon Pilgrim	[X86][SSE] Added additional vector sign/zero load extension...	commit \| commitdiff \| tree
2015-07-25	Simon Pilgrim	[X86][SSE] Added additional vector sign/zero extension...	commit \| commitdiff \| tree
2015-07-19	Simon Pilgrim	[X86][SSE] Tidied up vector CTLZ/CTTZ. NFCI.	commit \| commitdiff \| tree
2015-07-19	Simon Pilgrim	[X86][SSE] Reordered cast vectorization costs. NFCI.	commit \| commitdiff \| tree
2015-07-19	Simon Pilgrim	[DAGCombiner] Fixed minor typo that was missed in D9097.	commit \| commitdiff \| tree
2015-07-19	Simon Pilgrim	Remove TargetInstrInfo::canFoldMemoryOperand	commit \| commitdiff \| tree
2015-07-19	Simon Pilgrim	Use SDValue bool check. NFCI.	commit \| commitdiff \| tree
2015-07-18	Simon Pilgrim	[X86][SSE] Updated SHL/LSHR i64 vectorization costs.	commit \| commitdiff \| tree
2015-07-18	Simon Pilgrim	[X86][SSE] Added additional fp/int tests.	commit \| commitdiff \| tree
2015-07-18	Simon Pilgrim	Refreshed tests.	commit \| commitdiff \| tree
2015-07-18	Simon Pilgrim	Refreshed tests and reordered in descending integer...	commit \| commitdiff \| tree
2015-07-18	Simon Pilgrim	Tidyup shufflevector calls - don't repeat inputs if...	commit \| commitdiff \| tree
2015-07-16	Simon Pilgrim	Fix spelling. NFCI.	commit \| commitdiff \| tree
2015-07-16	Simon Pilgrim	[X86][SSE] Added nounwind attribute to vector shift...	commit \| commitdiff \| tree
2015-07-16	Simon Pilgrim	[X86][SSE] Updated vector conversion test names.	commit \| commitdiff \| tree
2015-07-15	Simon Pilgrim	[X86][SSE] Added i686/SSE2 vector shift tests.	commit \| commitdiff \| tree
2015-07-12	Simon Pilgrim	[X86][SSE] Tidied up vector extend/truncation tests...	commit \| commitdiff \| tree
2015-07-12	Simon Pilgrim	[X86][SSE] (V)PMINSB is commutable.	commit \| commitdiff \| tree
2015-07-12	Simon Pilgrim	Trim trailing whitespaces. NFC.	commit \| commitdiff \| tree
2015-07-12	Simon Pilgrim	[X86][SSE] Vectorized v4i32 non-uniform shifts.	commit \| commitdiff \| tree
2015-07-08	Simon Pilgrim	[X86][SSE] Vector shift test cleanup. NFC.	commit \| commitdiff \| tree
2015-07-08	Simon Pilgrim	[X86][SSE] Added (V)ROUNDSD + (V)ROUNDSS stack folding...	commit \| commitdiff \| tree
2015-07-06	Simon Pilgrim	[X86][AVX] Add support for shuffle decoding of vperm2f128...	commit \| commitdiff \| tree
2015-07-06	Simon Pilgrim	[X86][SSE] Vectorized i64 uniform constant SRA shifts	commit \| commitdiff \| tree
2015-07-06	Simon Pilgrim	[X86][SSE4A] Shuffle lowering using SSE4A EXTRQ/INSERTQ...	commit \| commitdiff \| tree
2015-07-06	Simon Pilgrim	[X86][SSE] Use the general SMAX/SMIN/UMAX/UMIN opcodes...	commit \| commitdiff \| tree
2015-07-06	Simon Pilgrim	[X86][SSE] Added missing stack folding test for SQRTSD...	commit \| commitdiff \| tree
2015-07-05	Simon Pilgrim	[X86][SSE3] Just use an explicit SSE3 target attribute...	commit \| commitdiff \| tree
2015-07-05	Simon Pilgrim	[X86][SSE2] Just use an explicit SSE2 target attribute...	commit \| commitdiff \| tree
2015-07-04	Simon Pilgrim	[X86][SSE] Improved i8/i16 to f64 uint2fp vector conversions	commit \| commitdiff \| tree
2015-07-03	Simon Pilgrim	[X86] Added 32-bit builds to fp<->int tests.	commit \| commitdiff \| tree
2015-07-03	Simon Pilgrim	whitespace tidyup. NFC.	commit \| commitdiff \| tree
2015-07-03	Simon Pilgrim	[X86][SSE] Sign extension for target vector sizes less...	commit \| commitdiff \| tree
2015-07-03	Simon Pilgrim	[X86][SSE] Sign extension for target vector sizes less...	commit \| commitdiff \| tree
2015-06-24	Simon Pilgrim	[X86][AVX] Added full set of 256-bit vector shift tests.	commit \| commitdiff \| tree
2015-06-23	Simon Pilgrim	[X86][SSE] Added full set of 128-bit vector shift tests.	commit \| commitdiff \| tree
2015-06-22	Simon Pilgrim	[X86][FMA4] FMA4 ops can perform unaligned folded loads.	commit \| commitdiff \| tree
next

C/C++ LLVM-based compilers that forbids OOTA behaviors

RSS Atom