Started 1 day 1 hr ago
Took 5 hr 24 min

Unstable Build #189 (Sep 29, 2020 1:37:40 AM)

Changes
  1. [X86] Cleanup check-prefixes for vector-mul.ll tests (details)
  2. [mlir] Fix passes.md's naming & add missing (details)
  3. [mlir] Updates to generate dialect rather than op docs (details)
  4. [mlir] Fix capitalization typo (details)
  5. [gn build] update TODO (details)
  6. Revert "[DSE] Switch to MemorySSA-backed DSE by default." (details)
  7. [InstCombine] Add basic vector test coverage for icmp_eq/ne zero combines (details)
  8. [flang] SAVE statement should not apply to nested scoping units (details)
  9. Thread safety analysis: Improve documentation for ASSERT_CAPABILITY (details)
  10. [X86] Add 64-bit target tests (details)
  11. [ConstraintElimination] Internalize function/class and delete an implied condition. NFC (details)
  12. Fix sphinx warnings in AttributeReference, NFC (details)
  13. [NFC][PowerPC] Change PPCSubTarget (introduced from D87671) to Subtarget (details)
  14. Internalize functions from various tools. NFC (details)
  15. [LLVM-C] Turn a ShuffleVector Constant Into a Getter. (details)
  16. [Machinesink] add one more profitable loop related pattern (details)
  17. Update Kaleidoscope: Change headers (details)
  18. [X86] Add more test cases to inline-asm-flag-output.ll. NFC (details)
  19. [DivRemPairs] Use DenseMapBase::find instead of operator[]. NFC (details)
  20. [AArch64][GlobalISel] Use the look-through constant helper for the shift s32->s64 custom legalization. (details)
  21. [Legalize][X86] Improve nnan fmin/fmax vector reduction (details)
  22. [AArch64][GlobalISel] Promote scalar G_SHL constant shift amounts to s64. (details)
  23. Fix MIPS and MIPS64 ABI to use ConstString in their register info arrays. (details)
  24. [llvm-objcopy][NFC] refactor error handling. part 2. (details)
  25. Typo fix; NFC (details)
  26. [InstSimplify] add tests for fcmp with casted op; NFC (details)
  27. [ValueTracking] enhance isKnownNeverInfinity to understand sitofp (details)
  28. [CostModel] add cl option to check size and latency costs; NFC (details)
  29. [X86] Regenerate i128 sdiv tests and add i686 coverage. (details)
  30. [X86] Add some basic i128 udiv test coverage (details)
  31. [CVP] Pass context instruction when narrowing div/rem (details)
  32. [LVI] Require context instruction in external API (NFCI) (details)
  33. [LVI] Clarify getValueAt/getValueInBlock doc comments (NFC) (details)
  34. [CVP] Make srem test more robust (NFC) (details)
  35. [InstCombine] Simplify code. NFCI. (details)
  36. [NewPM] Port ConstraintElimination to the new pass manager (details)
  37. [LVI][CVP] Use block value when simplifying icmps (details)
  38. [CVP] Remove unnecessary block splits in tests (NFC) (details)
  39. [OpenMP] Add Missing _static Director for OpenMP Documentation (details)
  40. [BuildLibCalls] Add noalias for strcat and stpcpy (details)
  41. [mlir][openacc] Remove detach and delete operands from acc.data (details)
  42. [mlir][openacc] Switch to assembly format for acc.data (details)
  43. [mlir][openacc] Add if, deviceptr operands and default attribute (details)
  44. Canonicalize declaration pointers when forming APValues. (details)
  45. Add profiling support for APValues. (details)
  46. [Test] Add tests where we can replace condition with invariants (details)
  47. Revert "Reland [CodeGen] emit CG profile for COFF object file" (details)
  48. [PowerPC] Add tests for `select` patterns. NFC. (details)
  49. [SVE] Replace / operator in TypeSize/ElementCount with divideCoefficientBy (details)
  50. [Coroutines] Reuse storage for local variables with non-overlapping lifetimes (details)
  51. [lldb] Remove nothreadallow from SWIG's __str__ wrappers to work around a Python>=3.7 crash (details)
  52. [ARM] Expand cannotInsertWDLSTPBetween to the last instruction (details)
  53. [ARM][MVE] tail-predication: overflow checks for elementcount, cont'd (details)
  54. [Coroutines] Remove unused includes. NFC. (details)
  55. [obj2yaml][yaml2obj] - Stop recognizing SHT_MIPS_ABIFLAGS on non-MIPS targets. (details)
  56. [lldb] Reference STL types in import-std-module tests (details)
  57. [yaml2obj][obj2yaml] - Add a support for SHT_ARM_EXIDX section. (details)
  58. [InstCombine] Add tests for vector rotate by constants with undefs. (details)
  59. [AMDGPU] Split R600 and GCN bfi patterns (details)
  60. [AMDGPU] Make bfi patterns divergence-aware (details)
  61. [AMDGPU] Add bfi immediate pattern (details)
  62. [PowerPC] Clean-up mayRaiseFPException bits (details)
  63. [lldb][NFC] Minor cleanup in CxxModuleHandler::tryInstantiateStdTemplate (details)
  64. [clangd] Add a tweak for filling in enumerators of a switch statement. (details)
  65. [gn build] Port 018066d9475 (details)
  66. [X86] Simplify broadcast mask detection with isUndefOrEqual helper. (details)
  67. [X86] Flip isShuffleEquivalent argument order to match isTargetShuffleEquivalent (details)
  68. [AArch64] Generate .note.gnu.property based on module flags. (details)
  69. [SCEV] Also use info from assumes in applyLoopGuards. (details)
  70. [InstCombine] matchRotate - allow undef in uniform constant rotation amounts (PR46895) (details)
  71. [ARM][MVE] Enable tail-predication by default (details)
  72. [clang] Don't emit "no member" diagnostic if the lookup fails on an invalid record decl. (details)
  73. [clang-tidy] IncludeInserter: allow <> in header name (details)
  74. [llvm-readobj/elf] - Fix the PREL31 relocation computation used for dumping arm32 unwind info (-u). (details)
  75. [AMDGPU] Reformat SITargetLowering::isSDNodeSourceOfDivergence. NFC. (details)
  76. [NFC][ARM] Factor out some logic for LoLoops. (details)
  77. [InstCombine] matchRotate - force splat of uniform constant rotation amounts (PR46895) (details)
  78. [TableGen] Improved messages in PseudoLoweringEmitter. (details)
  79. [ValueTracking] Fix analyses to update CxtI to be phi's incoming edges' terminators (details)
  80. [SVE] Lower fixed length VECREDUCE_[UMAX|UMIN] to Scalable (details)
  81. [ARM] Added more patterns to generate SSAT/USAT with shift (details)
  82. [llvm] Fix unused variable in non-debug configurations (details)
  83. [ARM][LowOverheadLoops] Cleanup and re-arrange (details)
  84. [AMDGPU] Reformat AMDGPUTargetLowering::isSDNodeAlwaysUniform. NFC. (details)
  85. [clang][codegen] Annotate `correctly-rounded-divide-sqrt-fp-math` fn-attr for OpenCL only. (details)
  86. Revert "Add the ability to write target stop-hooks using the ScriptInterpreter." (details)
  87. [clangd] Add a trained DecisionForest for code completion. (details)
  88. [ubsan] nullability-arg: Fix crash on C++ member pointers (details)
  89. [unittests] Preserve LD_LIBRARY_PATH in crash recovery test (details)
  90. [AArch64] Reuse map iterator instead of double lookup. NFC (details)
  91. Add FunctionType to MLIR C and Python bindings. (details)
  92. [clangd] Use Decision Forest to score code completions. (details)
  93. [InstCombine] Add basic trunc(shr(trunc(x),c)) tests (details)
  94. [GlobalISel] Combine (xor (and x, y), y) -> (and (not x), y) (details)
  95. [AArch64][GlobalISel] Support shifted register form in emitTST (details)
  96. [CostModel] split handling of intrinsics from other calls (details)
  97. [CostModel] move early exit for free intrinsics (details)
  98. [AArch64][GlobalISel] Infer whether G_PHI is going to be a FPR in regbankselect (details)
  99. [WebAssembly] Use wasm::Signature for in ObjectWriter (NFC) (details)
  100. [InstCombine] Add trunc(shr(trunc(x),c)) non-uniform vector tests (details)
  101. [AddressSanitizer] Copy type metadata to prevent miscompilation (details)
  102. [clangd] Rename evaluate() to evaluateHeuristics() (details)
  103. Revert "[AArch64][GlobalISel] Add selection support for <8 x s16>  G_INSERT_VECTOR_ELT with GPR scalar." (details)
  104. [AArch64] reuse another map iterator. NFC (details)
  105. [mlir] [VectorOps] changes to printing support for integers (details)
  106. scudo: Re-order Allocator fields for improved performance. NFCI. (details)
  107. [python][tests] Fix string comparison with "is" (details)
  108. [CostModel] fill in arguments as part of intrinsic attribute constructor (details)
  109. [PowerPC] Legalize v256i1 and v512i1 and implement load and store of these types (details)
  110. [lldb] Enable markdown support for documentation (details)
  111. Once we've found a firmware binary and loaded it, don't search more (details)
  112. [CostModel] remove hack for intrinsic cost based on cost type (details)
  113. [wasm] Move WasmTraits.h to BinaryFormat (details)
  114. [libc++] Fix heap UaF issue in coroutine test (details)
  115. [libc++] Add UNSUPPORTED markup to atomic test in single-threaded mode (details)
  116. [libc++] Replace uses of __libcpp_allocate by std::allocator<> (details)
  117. [COFF] Aliases resolve directly to defined external targets (details)
  118. [InstCombine] Regenerate cast tests. NFC. (details)
  119. [X86] Use inlineasm flag output for the _bittest* intrinsics. (details)
  120. [mlir] [VectorOps] Relaxed restrictions on vector.reduction types even more (details)
  121. Revert "Revert "[AArch64][GlobalISel] Add selection support for <8 x s16>  G_INSERT_VECTOR_ELT with GPR scalar."" (details)
  122. [lldb] Add print_function import (details)
  123. Fix a think-o with the numerical suffixes in the docs for init_priority. (details)
  124. [gn build] Re-run CompletionModelCodegen when input json files change (details)
  125. [CMake][AIX] Limit tools in external project build (details)
  126. [GlobalISel] Add support for lowering of vector G_SELECT and use for AArch64. (details)
  127. [X86] Add tests for D87883. NFC (details)
  128. [X86] Add support for calling SimplifyDemandedBits on the input of PDEP with a constant mask. (details)
  129. [libc++] Fix constexpr dynamic allocation on GCC 10 (details)
  130. [libcxx] Don't pass -s to libtool (details)
  131. [mlir][shape] Make conversion passes more consistent. (details)
  132. [EHStreamer] Simplify sharedTypeIDs with std::mismatch (details)
  133. [CVP] Allow two transforms in one invocation (details)
  134. Guard `find_library(tensorflow_c_api ...)` by checking for TENSORFLOW_C_LIB_PATH to be set by the user (details)
  135. [clang] Selectively ena/disa-ble format-insufficient-args warning (details)
  136. [scudo][standalone] Remove unused atomic_compare_exchange_weak (details)
  137. [mlir][Affine][VectorOps] Fix super vectorizer utility (D85869) (details)
  138. Attempt to clear some msan errors in the libcxx atomic tests. (details)
  139. [clang][driver][AIX] Set compiler-rt as default rtlib (details)
  140. BPF: move AbstractMemberAccess and PreserveDIType passes to EP_EarlyAsPossible (details)
  141. BuildVectorType with a dependent (array) type is crashing the compiler  - Fix for PR-47542 (details)
  142. [RegisterCoalescer] Pass Undefs to extendToIndices() (details)
  143. [clang] Update warning-wall.c test (details)
  144. Ensure that we don't compute linkage for an anonymous class too early if (details)
  145. [gn build] Port 54d9f743c8b (details)
  146. Remove dependency from LLVM Dialect on the OpenMP dialect (details)
  147. [clangd] When finding refs for a renaming alias, do not return refs to underlying decls (details)
  148. [mlir][openacc] Add acc.data operation verifier (details)
  149. Skip -fPIE for AMDGPU and HIP toolchain (details)
  150. [AArch64][GlobalISel] Scalarize <2 x s64> G_MUL since we don't have native support for it. (details)
  151. Recommit "[HIP] Change default --gpu-max-threads-per-block value to 1024" (details)
  152. [HIP] Return non-zero value for invalid target ID (details)
  153. BPF: explicitly specify bpfel triple for certain tests (details)
  154. [MLIR][OpenMP] Removed the ambiguity in flush op assembly syntax (details)
  155. [OpenMP][FIX] Verify compatible types for declare variant calls (details)
  156. [IndVars] Remove exiting conditions that are trivially true/false (details)
  157. [NFC] Use assert instead of checking the guaranteed condition (details)
  158. [Docs][NewPM] Add note about required passes (details)
  159. Revert "[OpenMP][FIX] Verify compatible types for declare variant calls" (details)
  160. [X86] Add known bits test for PEXT. NFC (details)
  161. [X86] Add computeKnownBits support for PEXT. (details)
  162. [Driver] Filter out <libdir>/gcc and <libdir>/gcc-cross if they do not exists (details)
  163. This reduces code duplication between CGObjCMac.cpp and Mangle.cpp (details)
  164. [NFC][ARM] Comments and lambdas (details)
  165. Recommit "[SCCP] Do not replace deref'able ptr with un-deref'able one." (details)
  166. Revert "Recommit "[SCCP] Do not replace deref'able ptr with un-deref'able one."" (details)
  167. [IsKnownNonZero] Handle the case with non-constant phi nodes (details)
  168. Revert "OpaquePtr: Add type to sret attribute" (details)
Changes
  1. Add flang out of tree buildbot (details)
  2. Added mail notifier for fedora-llvm-x86_64 builder (details)
  3. Increase number of jobs from 32 to 64 for fedora-llvm-x86_64 worker (details)
  4. mail notifier for x86_64-fedora-clang builder (details)
  5. Removed unused TestBuilders. (details)

Started by timer

This run spent:

  • 6 ms waiting;
  • 5 hr 24 min build duration;
  • 5 hr 24 min total from scheduled to completion.
Revision: eb9f7c28e5fe6d75fed3587023e17f2997c8024b
  • refs/remotes/origin/master
Revision: 05481260c40e502d68e8d523b66eb8e23641c8b9
  • refs/remotes/origin/master
Test Result (2 failures / ±0)Show all failed tests >>>

Identified problems

Link error

A link command has failed:
Indication 1

Compile Error

This build failed because of a compile error. Below is a list of all errors in the build log:
Indication 2

Regression test failed

This build failed because a regression test in the test suite FAILed. See the test report for details.
Indication 3

Ninja target failed

Below is a link to the first failed ninja target.
Indication 4