Added VPERM optimization for AVX2 shuffles
[oota-llvm.git] / test /
2012-04-15 Elena DemikhovskyAdded VPERM optimization for AVX2 shuffles
2012-04-14 Duncan SandsRename "fpaccuracy" metadata to the more generic "fpmat...
2012-04-14 Hal FinkelFix an error in BBVectorize important for vectorizing...
2012-04-14 Hal FinkelEnhance BBVectorize to more-properly handle pointer...
2012-04-13 Richard SmithFix X86 codegen for 'atomicrmw nand' to generate *x...
2012-04-13 Hal FinkelAdd support to BBVectorize for vectorizing selects.
2012-04-13 Evan ChengOn Darwin targets, only use vfma etc. if the source...
2012-04-13 Dan GohmanConsider ObjC runtime calls objc_storeWeak and others...
2012-04-13 Sylvestre LedruCatch the Python exception when subprocess.Popen is...
2012-04-13 Dan GohmanUse the new Use-aware dominates method to apply the...
2012-04-13 Dan GohmanDon't move objc_autorelease calls past autorelease...
2012-04-12 Sirish PandeDisable Hexagon test temporarily.
2012-04-12 Preston GurdThis patch improves the MCJIT runtime dynamic loader...
2012-04-12 Craig TopperFix 128-bit ptest intrinsics to take v2i64 instead...
2012-04-11 Akira HatanakaRevert changes that were accidentally committed.
2012-04-11 Akira HatanakaFix string that is being checked.
2012-04-11 Akira HatanakaEmit neg.s or neg.d only if -enable-no-nans-fp-math...
2012-04-11 Akira HatanakaEmit abs.s or abs.d only if -enable-no-nans-fp-math...
2012-04-11 Kevin EnderbyFixed a case of ARM disassembly getting an assert on...
2012-04-11 Akira HatanakaFix bugs in lowering of FCOPYSIGN nodes.
2012-04-11 Jim GrosbachARM 'vuzp.32 Dd, Dm' is a pseudo-instruction.
2012-04-11 Jim GrosbachARM 'vzip.32 Dd, Dm' is a pseudo-instruction.
2012-04-11 Evan ChengAdd more fused mul+add/sub patterns. rdar://10139676
2012-04-11 Nadav RotemReapply 154396 after fixing a test.
2012-04-11 Evan ChengClean up ARM fused multiply + add/sub support some...
2012-04-11 Evan ChengMatch (fneg (fma) to vfnma. rdar://10139676
2012-04-11 Charles DavisAdd retw and lretw instructions. Also, fix Intel syntax...
2012-04-11 Evan ChengMerge fma.ll into fusedMAC.ll
2012-04-11 Kevin EnderbyFix ARM disassembly of VLD instructions with writebacks...
2012-04-11 Jim GrosbachARM add missing Thumb1 two-operand aliases for shift...
2012-04-11 Evan ChengFix a number of problems with ARM fused multiply add...
2012-04-11 Jakob Stoklund OlesenFix test to be register assignment invariant.
2012-04-10 Owen AndersonMove the constant-folding support for FP_ROUND in Selec...
2012-04-10 Kostya Serebryany[tsan] two more compile-time optimizations:
2012-04-10 Evan ChengHandle llvm.fma.* intrinsics. rdar://10914096
2012-04-10 Duncan SandsAdd a comment noting that the fdiv -> fmul conversion...
2012-04-10 Eric ChristopherTemporarily revert this patch to see if it brings the...
2012-04-10 Kostya Serebryany[tsan] compile-time instrumentation: do not instrument...
2012-04-10 Eric ChristopherTo ensure that we have more accurate line information...
2012-04-10 Jim GrosbachARM fix cc_out operand handling for t2SUBrr instructions.
2012-04-10 Nadav RotemModify the code that lowers shuffles to blends from...
2012-04-10 Anton KorobeynikovTransform div to mul with reciprocal only when fp imm...
2012-04-10 Duncan SandsExpress the number of ULPs in fpaccuracy metadata as...
2012-04-10 Andrew TrickFix 12513: Loop unrolling breaks with indirect branches.
2012-04-10 Evan ChengAdd proper checks.
2012-04-10 Evan ChengFix a long standing tail call optimization bug. When...
2012-04-10 Rafael EspindolaDon't try to zExt just to check if an integer constant...
2012-04-09 Lang HamesTest case for PR12495.
2012-04-09 Akira HatanakaHave TargetLowering::getPICJumpTableRelocBase return...
2012-04-09 Chad RosierWhen performing a truncating store, it's possible to...
2012-04-09 Rafael EspindolaPattern match a setcc of boolean value with 0 as a...
2012-04-09 Nadav RotemLower some x86 shuffle sequences to the vblend family...
2012-04-09 Nadav RotemFix a bug in the lowering of broadcasts: ConstantPools...
2012-04-09 Chandler CarruthCleanup and relax a restriction on the matching of...
2012-04-09 Chandler CarruthFold 15 tiny test cases into a single file that impleme...
2012-04-08 Duncan SandsOnly have codegen turn fdiv by a constant into fmul...
2012-04-08 Chandler CarruthTeach LLVM about a PIE option which, when enabled on...
2012-04-08 Chandler CarruthTeach InstCombine to nuke a common alloca pattern ...
2012-04-08 Nadav RotemAVX2: Build splat vectors by broadcasting a scalar...
2012-04-08 Bill WendlingRemove old 'grep' lines.
2012-04-08 Bill WendlingFileCheckize these testcases.
2012-04-07 Nadav Rotem1. Remove the part of r153848 which optimizes shuffle...
2012-04-07 Duncan SandsConvert floating point division by a constant into...
2012-04-07 Chandler CarruthFix ValueTracking to conclude that debug intrinsics...
2012-04-07 Benjamin KramerSCEV: When expanding a GEP the final addition to the...
2012-04-07 Sean HuntMake the test for r154235 more platform-independent...
2012-04-07 Sean HuntOutput UTF-8-encoded characters as identifier character...
2012-04-06 Akira HatanakaAdd lines in global-address.ll to test N32 and N64...
2012-04-06 Jakob Stoklund OlesenAllow negative immediates in ARM and Thumb2 compares.
2012-04-06 Chandler CarruthSink the collection of return instructions until after...
2012-04-06 Chandler CarruthTweak this test to ensure the inliner did indeed fire...
2012-04-06 Craig TopperTest case for PR12413
2012-04-06 Craig TopperAllow 256-bit shuffles to be split if a 128-bit lane...
2012-04-06 Craig TopperAdd the tests that were supposed to go with r153935...
2012-04-06 Chandler CarruthActually finish this sentence in the comment the way...
2012-04-06 Chandler CarruthSink the return instruction collection until after...
2012-04-05 Jim GrosbachARM assembly aliases for add negative immediates using...
2012-04-05 Akira HatanakaReapply test case in 154038, this time with triple...
2012-04-05 Eric ChristopherPatch to set is_stmt a little better for prologue lines...
2012-04-05 Jakob Stoklund OlesenDon't break the IV update in TLI::SimplifySetCC().
2012-04-05 Dan GohmanFix accidentally inverted logic from r152803, and make the
2012-04-05 Silviu BarangaAdded support for unpredictable ADC/SBC instructions...
2012-04-05 Silviu BarangaAdded support for handling unpredictable arithmetic...
2012-04-05 James MolloyAn oversight when applying the patches for r150956...
2012-04-05 Jim GrosbachARM assembly aliases for two-operand V[R]SHR instructions.
2012-04-05 Jim GrosbachARM assembly parsing for 'msr' plain 'cpsr' operand.
2012-04-05 Jakob Stoklund OlesenPass the right sign to TLI->isLegalICmpImmediate.
2012-04-04 Akira HatanakaReapply 154038 without the failing test.
2012-04-04 Owen AndersonRevert r154038. It was causing make check failures.
2012-04-04 Akira HatanakaFix LowerGlobalAddress to produce instructions with...
2012-04-04 Akira HatanakaFix LowerConstantPool to produce instructions with...
2012-04-04 Jakob Stoklund OlesenImplement ARMBaseInstrInfo::commuteInstruction() for...
2012-04-04 Akira HatanakaFix LowerBlockAddress to produce instructions with...
2012-04-04 Hongbin ZhengAdd testcase for r154007, when a function has the optsi...
2012-04-04 Rafael EspindolaAlways compute all the bits in ComputeMaskedBits.
2012-04-03 Michael J. SpencerAdd YAML parser to Support.
2012-04-03 Pete CooperAdd VSELECT to LegalizeVectorTypes::ScalariseVectorResu...
2012-04-03 Eric ChristopherFix thinko check for number of operands to be the one...
2012-04-03 Nadav RotemAdd an additional testcase which checks ops with multip...
2012-04-03 Craig TopperAdd support for AVX enhanced comparison predicates...
next