Commits · e94134052fda15bcac2d237304abbfeb866d6a27 · Roger Ferrer / llvm-epi · GitLab

Dec 04, 2021
- [InstSimplify] Add logic 'or' fold to -1 · e9413405
  Mehrnoosh Heidarpour authored Dec 03, 2021
  e9413405
- [llvm-c] Avoid deprecated APIs in tests · 573a9bc4
  Nikita Popov authored Dec 04, 2021
  573a9bc4
- [CodeGen] Use range-based for loops (NFC) · ca2f5389
  Kazu Hirata authored Dec 04, 2021
  
  ca2f5389
- [NewPM] Test more options in pipeline test (NFC) · 8bd62119
  Nikita Popov authored Dec 04, 2021
  8bd62119
- [NewPM] Fix MergeFunctions scheduling · ae7f4680
  Nikita Popov authored Dec 04, 2021
  ae7f4680
- Attributor: Fix typo in function name · a25111c9
  Matt Arsenault authored Dec 04, 2021
  
  a25111c9
- Revert "[DwarfDebug] Support emitting function-local declaration for a lexical block" · a9616048
  Kristina Bessonova authored Dec 04, 2021
  a9616048
- AMDGPU: Enable fixed function ABI by default · 729bf9b2
  Matt Arsenault authored Aug 14, 2021
  729bf9b2
- [BasicAA] Add atomic mem intrinsic tests. · 89f0f277
  Florian Hahn authored Dec 04, 2021
  
  89f0f277
- AMDGPU: Assume all amdhsa kernarg passed implicit arguments by default · 2959e082
  Matt Arsenault authored Oct 25, 2021
  2959e082
- AMDGPU: Optimize out implicit kernarg argument allocation if unused · ae0ba7de
  Matt Arsenault authored Oct 25, 2021
  ae0ba7de
- [DwarfDebug] Support emitting function-local declaration for a lexical block · ee691970
  Kristina Bessonova authored Dec 04, 2021
  ee691970
- [DwarfDebug] Move emission of global vars, types and imports to endModule() · 79d31329
  Kristina Bessonova authored Dec 04, 2021
  79d31329
- [Passes] Move AggressiveInstCombine after InstCombine · c34d157f
  Anton Afanasyev authored Nov 01, 2021
  c34d157f
- [AMDGPU] Change llvm.amdgcn.image.bvh.intersect.ray to take vec3 args · 2774bad1
  Jay Foad authored Dec 02, 2021
  2774bad1
- [IR,TableGen] Add support for vec3 intrinsic arguments · c8e84c7a
  Jay Foad authored Dec 02, 2021
  c8e84c7a
- [AMDGPU] Generate checks for llvm.amdgcn.image.bvh.intersect.ray · bc7dacf5
  Jay Foad authored Dec 02, 2021
  bc7dacf5
- [PhaseOrdering] Add test for incorrect merge function scheduling · 5b94037a
  Nikita Popov authored Dec 03, 2021
  5b94037a
- [Test][PhaseOrdering] Precommit test for PR52289 · 728736e7
  Anton Afanasyev authored Dec 04, 2021
  
  728736e7
- [CodeGen] Use range-based for loops (NFC) · 3aed2822
  Kazu Hirata authored Dec 03, 2021
  
  3aed2822
- Split the locking of the queue and the threads vector in the ThreadPool implementation · e8469718
  Mehdi Amini authored Dec 03, 2021
  e8469718
- Fix build for ThreadPool when using -DLLVM_ENABLE_THREADS=OFF · b28f317c
  Mehdi Amini authored Dec 04, 2021
  b28f317c
- [LICM] Remove profile driven restriction on hoisting · 1a25d0bf
  Philip Reames authored Dec 03, 2021
  1a25d0bf
Dec 03, 2021
- [gn build] (semiautomatically) port 98bb1986 · 6c11b872
  Nico Weber authored Dec 03, 2021
  
  6c11b872
- ThreadPool: grow the pool only as needed · 728b982b
  Benoit Jacob authored Dec 03, 2021
  728b982b
- [DebugInfo] Check DIEnumerator bit width when comparing for equality · 93a20ece
  Arthur Eubanks authored Dec 03, 2021
  93a20ece
- [test-release.sh] Do not run chrpath on AIX. · 97eb3bb8
  Amy Kwan authored Dec 03, 2021
  97eb3bb8
- [Passes] Adjust SLPVectorizer placement in test. · 31413c45
  Florian Hahn authored Dec 03, 2021
  31413c45
- [Passes] Improve opt-pipeline-vector-passes.ll test. · 5da920bf
  Florian Hahn authored Dec 03, 2021
  5da920bf
- [CFG] Handle calls with funclet bundle · 181c4ba4
  Choongwoo Han authored Dec 03, 2021
  181c4ba4
- [AMDGPU] Fixed incomplete definitions in twoaddr-fma.mir. NFC. · e1d63068
  Stanislav Mekhanoshin authored Dec 03, 2021
  
  e1d63068
- [AMDGPU] Kill def when folding immediate in two-addr pass · 3b17cb15
  Stanislav Mekhanoshin authored Dec 02, 2021
  3b17cb15
- [DAG] PromoteIntRes_FunnelShift - rename shift Amount variable to Amt to... · ebf52719
  Simon Pilgrim authored Dec 03, 2021
  ebf52719
- [funcattrs] Fix a bug in recently introduced writeonly argument inference · 7b54de5f
  Philip Reames authored Dec 03, 2021
  7b54de5f
- [IR][AutoUpgrade] Merge x86 mask load intrinsic upgrades. NFC. · 74cc0fa1
  Simon Pilgrim authored Dec 03, 2021
  74cc0fa1
- [ARM] Separate ARM autoupgrade code into a separate function · 08035000
  David Green authored Dec 03, 2021
  08035000
- [ARM] Replace if's with a switch, NFC · 11f67f5a
  David Green authored Dec 03, 2021
  11f67f5a
- [gn build] Build with Fission on non-mac non-win when using lld · 1217b4b4
  Nico Weber authored Dec 03, 2021
  1217b4b4
- [MemoryLocation] Move DSE intrinsic handling to MemoryLocation. (NFC) · ead3979a
  Florian Hahn authored Dec 03, 2021
  ead3979a
- [DebugInfo] Attempt to preserve more information during tail duplication · 98a021fc
  Stephen Tozer authored Dec 03, 2021
  98a021fc