Commits · f0c4e70e95d94f1d585058c5ad18098e5924d06d · Roger Ferrer / llvm-epi · GitLab

Sep 28, 2019
- [SampleFDO] Create a separate flag profile-accurate-for-symsinlist to handle · f0c4e70e
  Wei Mi authored Sep 27, 2019
  f0c4e70e
Sep 27, 2019
- [NFC][PhaseOrdering] Add end-to-end tests for the 'two shifts by sext' problem · 9c604a0d
  Roman Lebedev authored Sep 27, 2019
  9c604a0d
- [InstSimplify] add tests for fma/fmuladd with undef operand; NFC · 1b40402a
  Sanjay Patel authored Sep 27, 2019
  1b40402a
- [InstCombine] Simplify shift-by-sext to shift-by-zext · 269f1bea
  Roman Lebedev authored Sep 27, 2019
  269f1bea
- [SLPVectorizer][X86] Regenerate arith-fp tests · 756f5cfc
  Simon Pilgrim authored Sep 27, 2019
  756f5cfc
- [NFC][InstCombine] Revisit shift-by-signext tests · 09564804
  Roman Lebedev authored Sep 27, 2019
  09564804
- Revert "[LoopInfo] Limit the iterations to check whether a loop has dedicated · 9c8efeda
  Wei Mi authored Sep 27, 2019
  9c8efeda
- Revert [SLP] Fix for PR31847: Assertion failed: (isLoopInvariant(Operands[i],... · f98d2c09
  Jordan Rupprecht authored Sep 26, 2019
  f98d2c09
Sep 26, 2019
- [LoopFusion] Add ability to fuse guarded loops · 50bc6104
  Kit Barton authored Sep 26, 2019
  50bc6104
- [Unroll] Do NOT unroll a loop with small runtime upperbound · 1128fa09
  Zhaoshi Zheng authored Sep 26, 2019
  1128fa09
- [InstCombine][NFC] Add tests for shift-by-signext · 86b40b0b
  Roman Lebedev authored Sep 26, 2019
  86b40b0b
- [InstCombine][NFC] Regenerate load-cmp.ll test · d1ef2e48
  Roman Lebedev authored Sep 26, 2019
  d1ef2e48
- [NFC] Precommit tests for D68089 · f1a5a931
  David Bolvansky authored Sep 26, 2019
  f1a5a931
- [InstCombine] Use m_Zero instead of isNullValue() when checking if a GEP index... · 46721bb7
  Craig Topper authored Sep 26, 2019
  46721bb7
- [LoopInfo] Limit the iterations to check whether a loop has dedicated exits · 67d93f0d
  Wei Mi authored Sep 26, 2019
  67d93f0d
- Handle successor's PHI node correctly when flattening CFG merges two if-regions · d98cb81c
  Jakub Kuderski authored Sep 26, 2019
  d98cb81c
- [InstCombine] Don't assume CmpInst has been visited in getFlippedStrictnessPredicateAndConstant · 163c54d2
  Bjorn Pettersson authored Sep 26, 2019
  163c54d2
- [SLPVectorizer][X86] Add SSE common check prefix to let us merge SSE2+SLM checks · fc82c7a1
  Simon Pilgrim authored Sep 26, 2019
  fc82c7a1
- [CostModel][X86] Fix SLM <2 x i64> icmp costs · d7f0207d
  Simon Pilgrim authored Sep 26, 2019
  d7f0207d
- [InstCombine] foldUnsignedUnderflowCheck(): one last pattern with 'sub' (PR43251) · a2fa03af
  Roman Lebedev authored Sep 25, 2019
  a2fa03af
- [NFC][InstCombine] Tests for 'base u<= offset && (base - offset) != 0' pattern (PR43251) · ca524621
  Roman Lebedev authored Sep 25, 2019
  ca524621
- [InstSimplify] Handle more 'A </>/>=/<= B &&/|| (A - B) !=/== 0' patterns (PR43251) · 914a3d1c
  Roman Lebedev authored Sep 25, 2019
  914a3d1c
- [NFC][InstSimplify] More exaustive test coverage for 'A </>/>=/<= B &&/|| (A -... · 26606bec
  Roman Lebedev authored Sep 25, 2019
  26606bec
Sep 25, 2019
- [InstSimplify] Match 1.0 and 0.0 for both operands in SimplifyFMAMul · d663efe2
  Florian Hahn authored Sep 25, 2019
  d663efe2
- [InstCombine] Fold (A - B) u>=/u< A --> B u>/u<= A iff B != 0 · 23646952
  Roman Lebedev authored Sep 25, 2019
  23646952
- [NFC][InstCombine] Add tests for (X - Y) < X --> Y <= X iff Y != 0 · dfda7d2d
  Roman Lebedev authored Sep 25, 2019
  dfda7d2d
- [InstCombine] Limit FMul constant folding for fma simplifications. · f3ab99dc
  Florian Hahn authored Sep 25, 2019
  f3ab99dc
Sep 24, 2019
- [GCRelocate] Add a peephole to canonicalize base pointer relocation · d9629b88
  Philip Reames authored Sep 24, 2019
  d9629b88
- [InstCombine] (a+b) < a && (a+b) != 0 -> (0-b) < a iff a/b != 0 (PR43259) · 45fd1e9d
  Roman Lebedev authored Sep 24, 2019
  45fd1e9d
- [InstCombine] (a+b) <= a && (a+b) != 0 -> (0-b) < a (PR43259) · 5b881f35
  Roman Lebedev authored Sep 24, 2019
  5b881f35
- [Debuginfo] dbg.value points to undef value after Induction Variable Simplification. · 49f3c2b6
  Alexey Lapshin authored Sep 24, 2019
  49f3c2b6
- [LV] Forced vectorization with runtime checks and OptForSize · 0fcb3afb
  Sjoerd Meijer authored Sep 24, 2019
  0fcb3afb
- [InstCombine] Fold a shifty implementation of clamp-to-allones. · a4dd98f2
  Huihui Zhang authored Sep 24, 2019
  a4dd98f2
- [InstCombine] Fold a shifty implementation of clamp-to-zero. · 89521997
  Huihui Zhang authored Sep 24, 2019
  89521997
- [NFC][InstCombine] Add tests for shifty implementation of clamping. · 5b5f1c8e
  Huihui Zhang authored Sep 23, 2019
  5b5f1c8e
- HotColdSplitting: invalidate the AssumptionCache on split · 082f895b
  Saleem Abdulrasool authored Sep 23, 2019
  082f895b
- [SampleFDO] Treat names in profile as not cold only when profile symbol list · 22fd8853
  Wei Mi authored Sep 23, 2019
  22fd8853
Sep 23, 2019
- [InstCombine] Annotate strndup calls with dereferenceable_or_null · 48db0272
  David Bolvansky authored Sep 23, 2019
  48db0272
- [SLC] Convert some strndup calls to strdup calls · 8d520161
  David Bolvansky authored Sep 23, 2019
  8d520161
- [InstCombine] dropRedundantMaskingOfLeftShiftInput(): pat. c/d/e with mask (PR42563) · 0a51e1f6
  Roman Lebedev authored Sep 23, 2019
  0a51e1f6