summaryrefslogtreecommitdiff
path: root/test
AgeCommit message (Expand)AuthorFilesLines
2015-12-15[WinEH] Use operand bundles to describe call sitesDavid Majnemer27-190/+143
2015-12-15Test cleanup -- remove duplicate run linesXinliang David Li1-4/+0
2015-12-15AMDGPU/SI: Select constant loads with non-uniform addresses to MUBUF instruct...Tom Stellard1-37/+40
2015-12-15[Sparc] Fix handling of double incoming arguments on sparc little-endian.James Y Knight1-0/+191
2015-12-15[X86] MOVPC32r should only emit CFI adjustments when neededMichael Kuperstein1-0/+30
2015-12-15[SimplifyCFG] allow speculation of exactly one expensive instruction (PR24818)Sanjay Patel2-43/+23
2015-12-15[llvm-profdata] Add support for weighted merge of profile data (2nd try)Nathan Slingerland7-1/+142
2015-12-15AMDGPU: mark ldexp LibCalls as unavailableNicolai Hahnle1-10/+9
2015-12-15[X86] Smaller code for materializing 32-bit 1 and -1 constantsHans Wennborg1-0/+100
2015-12-15[Hexagon] Preprocess mapped instructions before lowering to MCKrzysztof Parzyszek4-6/+6
2015-12-15AMDGPU/SI: Add llvm.amdgcn.mbcnt.* intrinsicsTom Stellard1-0/+24
2015-12-15AMDGPU/SI: Add llvm.amdgcn.v.interp.p[12] intrinsicsTom Stellard1-0/+30
2015-12-15Bitcasts between FP and INT values using direct movesNemanja Ivanovic2-2/+119
2015-12-15Do not try to use i8 and i16 versions of FP_TO_U/SINT soft float library callsMichael Kuperstein1-0/+36
2015-12-15Improve the successor list update in TailDuplication.cpp.Cong Hou1-1/+1
2015-12-15Type legalizer for masked gather and scatter intrinsics.Elena Demikhovsky2-95/+1638
2015-12-15[ShrinkWrapping] Do not choose restore point inside loops.Quentin Colombet1-0/+142
2015-12-15[WebAssembly] Remove .import printing.Dan Gohman1-32/+0
2015-12-15WebAssembly: test global array indexingJF Bastien1-0/+9
2015-12-15Instcombine: destructor loads of structs that do not contains paddingMehdi Amini3-84/+108
2015-12-15[llvm-readobj] s/FunctionName/LinkageName/ for codeview dumpingReid Kleckner1-18/+18
2015-12-15[PGO] make profile prefix even shorter and more readableXinliang David Li14-92/+92
2015-12-15[X86] Add relaxtion logic for SBB instructions.Quentin Colombet1-0/+16
2015-12-14[PGO] Shorten profile symbol prefixesXinliang David Li14-92/+92
2015-12-14Use diagnostic handler in the LLVMContextRafael Espindola1-1/+1
2015-12-14[X86] Add relaxtion logic for ADC instructions.Quentin Colombet1-0/+16
2015-12-14[WebAssembly] Add type prefixes to call instructionsDan Gohman6-21/+21
2015-12-14[WebAssembly] Implement a new algorithm for placing BLOCK markersDan Gohman1-11/+622
2015-12-14Revert "Don't create unnecessary PHIs"Reid Kleckner2-200/+4
2015-12-14[X86] Part 2 to fix x86-64 fp128 calling convention.Chih-Hung Hsieh8-4/+929
2015-12-14add fast-math-flags to 'call' instructions (PR21290)Sanjay Patel7-22/+44
2015-12-14Start implementing FDE dumping when printing the eh_frame.Pete Cooper1-1/+6
2015-12-14Print the eh_frame section in MachoDump.Pete Cooper2-0/+18
2015-12-14[ConstantFold] Fix bitcast to gep constant folding transform.David Majnemer1-0/+4
2015-12-14[MergeFunctions] Use II instead of CI for InvokeInst; NFCSanjoy Das1-0/+30
2015-12-14Teach MergeFunctions about operand bundlesSanjoy Das1-0/+34
2015-12-14Teach haveSameSpecialState about operand bundlesSanjoy Das1-0/+21
2015-12-14[PGO] Value profiling text format reader/writer supportXinliang David Li5-0/+180
2015-12-14[IR] Remove terminatepadDavid Majnemer8-188/+15
2015-12-14FastISel needs to remove dead code when it bails out.Paul Robinson1-0/+147
2015-12-14[Power PC] llvm soft float support for ppc32Petar Jovanovic1-0/+50
2015-12-14AMDGPU: Use generic bitreverse intrinsicMatt Arsenault2-28/+115
2015-12-14AMDGPU: Fix splitting vector loads with existing offsetsMatt Arsenault1-0/+104
2015-12-14[InstCombine] fold trunc ([lshr] (bitcast vector) ) --> extractelement (PR25543)Sanjay Patel1-11/+7
2015-12-14[sanitizer] [msan] VarArgHelper for AArch64Adhemerval Zanella1-0/+75
2015-12-14Don't create unnecessary PHIsJames Molloy2-4/+200
2015-12-14[llvm-dwp] Deduplicate type unitsDavid Blaikie3-0/+35
2015-12-14[llvm-dwp] Remove some unused test codeDavid Blaikie2-5/+0
2015-12-13I Added a triple flag for x86-evenDirective test.Michael Zuckerman1-1/+1
2015-12-13Revert r255460, which still causes test failures on some platforms.Cong Hou2-70/+1