Started 2 mo 16 days ago
Took 2 hr 15 min

Build #1983 (Sep 22, 2021 10:01:58 PM)

Changes
  1. Avoid building the entire tree and testing LLVM itself on MLIR builders (details / githubweb)
Changes
  1. [Polly] Add -polly-reschedule and -polly-postopts options. (details)
  2. tsan: reset destination range in Java heap move (details)
  3. tsan: uninline Enable/DisableIgnores (details)
  4. tsan: prepare for trace mapping removal (details)
  5. [lldb] Add --stack option to `target symbols add` command (details)
  6. [flang] Change complex type define in runtime for clang-cl (details)
  7. [InstCombine] Move InstCombineWorklist to Utils to allow reuse (NFC). (details)
  8. [clang][ASTImporter] Generic attribute import handling (first step). (details)
  9. [Utils] Replace llc with cat for tests (details)
  10. tsan: account for mid app range in mem profile (details)
  11. tsan: include MBlock/SyncObj stats into mem profile (details)
  12. tsan: make mem profile data more consistent (details)
  13. tsan: include internal allocator info in mem profile (details)
  14. tsan: move mem profile initialization into separate function (details)
  15. tsan: remove stale comment (details)
  16. tsan: write uptime in mem profile (details)
  17. [ARM] Add additional tests for VMOVL in tail predicated loops. (details)
  18. [AMDGPU] Divergence-driven instruction selection for mul i32 (details)
  19. [AMDGPU] Convert mac/fmac to mad/fma when folding output modifiers (details)
  20. [AArch64][SVE] Add missing load/store patterns for unpacked bfloat vectors. (details)
  21. [VectorCombine] Switch to using a worklist. (details)
  22. [LoopVectorize][X86] Add operands to make it more obvious what line the CHECK concerns (details)
  23. [SelectionDAG] Make WidenVecRes_Convert work for scalable vectors. (details)
  24. [hwasan] also omit safe mem[cpy|mov|set]. (details)
  25. Don't fold (select C, (gep Ptr, Idx), Ptr) if C is vector but Idx is scalar (details)
  26. Unbreak module builds by making InstructionWorklist.h non-modular (details)
  27. [ARM] Allow smaller VMOVL in tail predicated loops (details)
  28. [lldb] [Windows] Fix continuing from breakpoints and singlestepping on ARM/AArch64 (details)
  29. [Matrix] Emit assumption that matrix indices are valid. (details)
  30. Revert "[CodeGen] regenerate test checks; NFC" (details)
  31. Revert "[InstCombine] fold cast of right-shift if high bits are not demanded" (details)
  32. [Passes] Run vector-combine early with -fenable-matrix. (details)
  33. [gn build] (manually) port f8b1cc365786 (details)
  34. [gn build] Port 7a320b279d07 (details)
  35. [SelectionDAG] Add PromoteIntOp_INSERT_SUBVECTOR. (details)
  36. [lldb] JITLoaderGDB tests can use lli in ORC greedy mode (details)
  37. [SLP][NFC]Rename function in the test for better matching of the (details)
  38. [ELF][test] Restore important part of ICF alignment test (details)
  39. AArch64: use indivisible cmpxchg for 128-bit atomic loads at O0 (details)
  40. [SelectionDAG] Remove PromoteIntOp_EXTRACT_SUBVECTOR. (details)
  41. [AArch64][SVE] Add extract_subvector patterns for unpacked fp16 and bfloat types. (details)
  42. [mailmap] Add entry for myself (details)
  43. [OpenMP] Make sure the Thread ID function is not removed (details)
  44. [AArch64][SVE] NFC: Move extract_subvector tests around. (details)
  45. [InstCombine] Update InstCombine to use poison instead of undef for shufflevector's placeholder (1/3) (details)
  46. [Target][CodeGen] Remove default CostKind arguments on inner/impl TTI overrides (details)
  47. [Sanitizer] Add Windows header for _mkdir (details)
  48. [libc++][NFC] Add link to Discord channel from documentation (details)
  49. [InstCombine] Update InstCombine to use poison instead of undef for shufflevector's placeholder (2/3) (details)
  50. [OpenMP][Offloading] Use bitset to indicate execution mode instead of value (details)
  51. [libc++] Disallow volatile types in std::allocator (details)
  52. [InstCombine] Update InstCombine to use poison instead of undef for shufflevector's placeholder (3/3) (details)
  53. [SLP] getReductionCost - use explicit TTI::TCK_RecipThroughput CostKind. NFCI. (details)
  54. [ORC] Re-enable ELF DebugObjectManagerPlugin tests (details)
  55. [CodeGen] Remove redundant declaration MIRCanonicalizerID (NFC) (details)
  56. [SLP][NFC]Add a test to show an issue with incorrectly extracted (details)
  57. [CSSPGO] Set PseudoProbeInserter as a default pass. (details)
  58. [NFC] clang-format -i llvm/lib/Frontend/OpenMP/OMPIRBuilder.cpp (details)
  59. [RISCV] Optimize vp.store with an all ones mask to avoid a vmset. (details)
  60. [mlir][sparse] replace ad-hoc MemRef struct with CRunnerUtils definition (details)
  61. [OpenMP] Add function tracing debugging to device RTL (details)
  62. [ORC] DebugObjectManagerPlugin tests can use lli in ORC greedy mode (details)
  63. [HWASan] Use a single .weak binding in asm. (details)
  64. [libc] Add an implementation of bsearch. (details)
  65. [SimplifyCFG] Ignore free instructions when computing cost for folding branch to common dest (details)
  66. Add document numbers for the C99 status page. (details)
  67. [NFCI][CodeGen, AArch64] Fix inconsistent TargetCostKind types. (details)
  68. [llvm-profgen] An option to dump disasm of specified symbols (details)
  69. [mlir][linalg] Fix interchange initialization in fusion on tensors. (details)
  70. Don't diagnose unused but set when the Cleanup attribute is used. (details)
  71. [Analysis] improve function matching for strlen libcall (details)
  72. Change error for storage-class to mean linkage, fix lang-linkage diag (details)
  73. Mark CFG as preserved in TypePromotion and InterleaveAccess passes (details)
  74. [lldb] Remove IRExecutionUnit::CollectFallbackNames (details)
  75. DebugInfo: Add (initially no-op) -gsimple-template-names={simple,mangled} (details)
  76. [OpenMP] Fix KeepAlive usage (details)
  77. [ELF][AArch64] Refine and fix the condition when BTI/PAC PLT needs bti c (details)
  78. [Analysis] reduce code for isOnlyUsedInZeroEqualityComparison; NFC (details)
  79. [ValueTracking] fix isOnlyUsedInZeroEqualityComparison with no users (details)
  80. [mlir][sparse] generalize reduction support in sparse compiler (details)
  81. [libc++][test] Remove disable_missing_braces_warning.h from tests (details)
  82. [libc++][NFCI] Remove uses of _LIBCPP_INLINE_VAR (details)
  83. Reset operation when canceling root update transaction (details)
  84. [mlir:DataFlowAnalysis] Reprocess the arguments of already executable edges (details)
  85. [JumpThreading] Add test with free instructions (NFC) (details)
  86. [InstCombine] add test for shift-shift with extra use; NFC (details)
  87. [InstCombine] add one-use check to shift-shift transform (details)
  88. [libc++][NFC] Add missing whitespace in <compare> (details)
  89. [mlir][Linalg] Teach constant -> generic op fusion to handle scalar constants. (details)
  90. [Clang] Ignore BTFTag attr if used as a type attribute (details)
  91. [OpenMP][Offloading] Change `bool IsSPMD` to `int8_t Mode` in `__kmpc_target_init` and `__kmpc_target_deinit` (details)
  92. Fix bug for Ops with default valued attributes and successors/variadic regions. (details)
  93. [RISCV] Add test cases showing failure to use ADDIW before fcvt.s.w/fcvt.d.w/fcvt.h.w. NFC (details)
  94. [RISCV] Add fcvt.s.w(u)/fcvt.d.w(u)/fcvt.h.w(u) to hasAllNBitUsers (details)
  95. [RISCV] Remove stale FIXMEs from float-convert.ll and double-convert.ll. NFC (details)
  96. Diagnose -Wunused-value based on CFG reachability (details)
  97. [flang] Catch error: base of DATA statement object can't be a pointer (details)
  98. [WebAssembly] Add relaxed-simd feature (details)
  99. [mlir][sparse] add sparse kernels test to sparse compiler test suite (details)
  100. [llvm-profgen] Use hot threshold for context merging and trimming (details)
  101. [NFC] Remove trailing spaces from some files (details)
  102. [compiler-rt][test] Add REQUIRES for checking static libc++abi (details)
  103. [Polly] Implement user-directed loop distribution/fission. (details)
  104. [Polly] Dissolve Isl test directory. NFC. (details)
  105. [libc++] Fix __wrap_iter to be a proper contiguous iterator. (details)
  106. [NFC][InstCombine] Fix inconsistent comments (details)
  107. [llvm-profgen] Fix a bug of assertion (details)
  108. [llvm-profgen] Fix an out-of-range error during unwinding (details)
  109. [llvm-profgen] Fix a dangling vector reference in CS line number based generator (details)
  110. [RISCV] Update mir tests. (details)
  111. [NFC] Remove redundant setOperationAction. (details)
  112. [MLIR] [Python] Make Attribute and Type hashable (details)
  113. [X86][FP16] Change the order of the operands in complex FMA intrinsics to allow swap between the mul operands. (details)
  114. [demangle] Support for ISO/IEC TS 18661 binary floating point type (details)
  115. [AArch64][SVE][InstCombine] Eliminate redundant chains of tuple get/set (details)
  116. [OpenMP][NFC] Clang format OMPIRBuilder (details)
  117. [OpenMP][NFC] Improve debug output (details)
  118. [OpenMP][NFC] Precommit test that exposes a bug in our optnone handling (details)
  119. [OpenMP][NFC] Rerun the test check update script on all OpenMP-Opt tests (details)
  120. [OpenMP][NFC] Repair test that contained nested kernels (details)
  121. [OpenMP][FIX] Be more deliberate about invalidating the AAKernelInfo state (details)
  122. [libc++][spaceship] Implement std::pair::operator<=> (details)
Changes
  1. Avoid building the entire tree and testing LLVM itself on MLIR builders (details)

Started by timer

This run spent:

  • 41 min waiting;
  • 2 hr 15 min build duration;
  • 2 hr 15 min total from scheduled to completion.
Revision: fbdda46df1702d87909e66856796ffaefb5c0b41
Repository: https://github.com/llvm/llvm-zorg.git
  • refs/remotes/origin/main
Revision: f4abdb0c074bd7b7b89d9828d449078d16d0e1c3
Repository: http://labmaster3.local/git/llvm-project.git
  • refs/remotes/origin/main
Revision: fbdda46df1702d87909e66856796ffaefb5c0b41
Repository: http://labmaster3.local/git/llvm-zorg.git
  • refs/remotes/origin/main
Test Result (no failures)