Started 2 days 11 hr ago
Took 5 hr 31 min

Success Build #197 (Oct 27, 2020 1:37:59 AM)

Changes
  1. [LegalizeTypes] Legalize vector rotate operations (details)
  2. [DAG] Add BuildVectorSDNode::getRepeatedSequence helper to recognise multi-element splat patterns (details)
  3. [InstCombine] narrowFunnelShift - fold trunc/zext or(shl(a,x),lshr(b,sub(bw,x))) -> fshl(a,b,x) (PR35155) (details)
  4. Fix some signed/unsigned comparison gcc warnings from D87930 (details)
  5. [JITLink][ELF] PCRel32GOTLoad relocations are resolved like regular PCRel32 ones (details)
  6. [jitlink][ELF] Add zero-fill blocks for symbols in section SHN_COMMON (details)
  7. [LoopVersioning] Form dedicated exits for versioned loop to preserve simplify form (details)
  8. Revert "[jitlink][ELF] Add zero-fill blocks for symbols in section SHN_COMMON" (details)
  9. [NPM] Port -loop-versioning-licm to NPM (details)
  10. [Flang][Driver] Add infrastructure for basic frontend actions and file I/O (details)
  11. [JITLink][ELF] PCRel32GOTLoad edge offset can be smaller three (details)
  12. Reapply "[jitlink][ELF] Add zero-fill blocks for symbols in section SHN_COMMON" (details)
  13. [flang][tests] Fix Python bug in the lit config (details)
  14. [ARM] Remove some dead code. NFC (details)
  15. [AMDGPU][CostModel] Refine cost model for half- and quarter-rate instructions. (details)
  16. Unbreak the clang-interpreter example after 0aec49c8531bc5282b095730d34681455826bc2c (details)
  17. [X86] Add a stub for znver3 based on the little public information there is in AMD's manuals (details)
  18. [X86] Add a stub for Intel's alderlake. (details)
  19. [CostModel] remove cost-kind predicate for vector reduction costs (details)
  20. Revert "hwasan: Disable operator {new,delete} interceptors when interceptors are disabled." (details)
  21. [AArch64][GlobalISel] Fix -Wunused-variable. NFC (details)
  22. [asan][fuchsia] set current thread before reading thread state (details)
  23. [asan] Fix stack-use-after-free checks on non-main thread on Fuchsia (details)
  24. [X86] Produce R_X86_64_GOTPCRELX for test/binop instructions (MOV32rm/TEST32rm/...) when -Wa,-mrelax-relocations=yes is enabled (details)
  25. [ELF] Don't crash on R_X86_64_GOTPCRELX for test/binop instructions (details)
  26. Fix typo SSC -> SCC (details)
  27. [clangd] Add a TestWorkspace utility (details)
  28. For P0732R2, P1907R1: ensure that template parameter objects don't refer (details)
  29. [MemProf] Temporarily disable part of test (details)
  30. Revert "[CostModel] remove cost-kind predicate for vector reduction costs" (details)
  31. [X86] Use TargetConstant for FPDiff with X86::TC_RETURN. (details)
  32. [c++20] For P0732R2: Support string literal operator templates. (details)
  33. [InstCombine] Add test for or(fshl(),fshl()) bswap pattern. (details)
  34. [InstCombine] matchBSwapOrBitReversem - recognise or(fshl(),fshl()) bswap patterns. (details)
  35. [clang] Enable support for #pragma STDC FENV_ACCESS (details)
  36. Remove superfluous whitespace around if(). NFC. (details)
  37. [SLP] Add AArch64 tests with vectorizable compare/select patterns. (details)
  38. Correct LIT test failure detected on buildbot after mibintc committed rG2e204e23911b: [clang] Enable support for #pragma STDC FENV_ACCESS D87528 (details)
  39. [InstSimplify] add tests for logic-of-cmps with not op; NFC (details)
  40. [InstSimplify] peek through 'not' operand in logic-of-icmps fold (details)
  41. [IndVars] Regenerate test checks (NFC) (details)
  42. [SCEV] Strength nowrap flags after constant folding (details)
  43. [SCEV] Separate out constant folding in mul expr creation (details)
  44. [SCEV] Always constant fold mul expression operands (details)
  45. Thread safety analysis: Consider global variables in scope (details)
  46. Thread safety analysis: Nullability improvements in TIL, NFCI (details)
  47. [SCEV] Strenthen nowrap flags after constant folding for mul exprs (details)
  48. [CostModel] fix operand/type accounting for fadd/fmul reductions (details)
  49. [CostModel] remove cost-kind predicate for some vector reduction costs (details)
  50. [X86] Use TargetConstant instead of Constant for operands to X86vaarg64. (details)
  51. [X86] Use TargetConstant for immediates for VASTART_SAVE_XMM_REGS. (details)
  52. [X86] VEX/EVEX prefix doesn't work for inline assembly. (details)
  53. [X86] Don't disassemble wbinvd with 0xf2 or 0x66 prefix. (details)
  54. [AMDGPU] Avoid offset register in MUBUF for direct stack object accesses (details)
  55. [clangd] Add dependency on remote index service proto (details)
  56. [IndVars] Use contextual knowledge when proving trivial conds (details)
  57. [NFC] Factor out common code into lambda for further improvement (details)
  58. [NFC] Remove unused funciton param (details)
  59. Fix broken build after previous commit (details)
  60. [VE] Support atomic fence (details)
  61. [Schedule] Add a MultiHazardRecognizer (details)
  62. [gn build] Port 61bc18de0b2 (details)
  63. [ARM][SchedModels] Convert IsLdrAm3NegRegOffPred to MCSchedPredicate (details)
  64. [ARM][SchedModels] Convert IsLdstsoMinusRegPred to MCSchedPredicate (details)
  65. [ARM][SchedModels] Get rid of IsLdrAm2ScaledPred (details)
  66. [VE] Support atomic load (details)
  67. [PowerPC] Add test case for pr47830. NFC. (details)
  68. [ARM][SchedModels] Rename and generalize predicate. NFC (details)
  69. [AMDGPU] Emit new pal metadata by default (details)
  70. [AsmPrinter] Add per BB instruction mix remark. (details)
  71. [VE] Add integer arithmetic vector instructions (details)
  72. [VE] Add vector comparison and min/max (details)
  73. [Annotation] Allows annotation to carry some additional constant arguments. (details)
  74. [flang] Tighten rules to resolve procedure as intrinsic procedure (details)
  75. Try to fix buildbots after d3205bbca3e0002d76282878986993e7e7994779 (details)
  76. [lldb] [Process/FreeBSDRemote] Fix #include for i386 compat (details)
  77. [lldb] [test/Register] Use initial state for write tests (details)
  78. [lldb] [Process/NetBSD] Set xs_xstate_bv correctly when setting regs (details)
  79. [lldb] [Process/Linux] Reuse NativeRegisterContextWatchpoint_x86 (details)
  80. [lldb] Fix bitfield "frame var" for pointers (pr47743) (details)
  81. [lldb] Modernize PseudoTerminal::Fork (details)
  82. [NFC][InstrRefLDV] Fix a typo (details)
  83. [llvm-mca] Add few memory instructions to cortex-a57 test (details)
  84. tsan: add mips64 support in lib/tsan/go/buildgo.sh (details)
  85. [InstCombine] Add rotate tests where the shift amount is zero extended after masking (details)
  86. [InstCombine] Add support for zext(and(neg(amt),width-1)) rotate shift amount patterns (details)
  87. [clang] Suppress "follow-up" diagnostics on recovery call expressions. (details)
  88. [MLIR][mlir-spirv-cpu-runner] A pass to emulate a call to kernel in LLVM (details)
  89. [clang-fuzzer] CreateAndRunJITFunc - fix use after move static analyzer warning. (details)
  90. [llvm-ar][Object] Fix detection of need for 64-bit archive symbol tables (details)
  91. [DAGCombine] Add test case showing incorrect DAGCombine optimization (details)
  92. [MLIR][mlir-spirv-cpu-runner] A SPIR-V cpu runner prototype (details)
  93. [clangd] Fix remote index build failures due to lack of proto dependency (details)
  94. [LV] Add cost-model test for AArch64 select costs. (details)
  95. [AMDGPU] Make more use of printNamedBit in AMDGPUInstPrinter. NFC. (details)
  96. [InstCombine] Add bswap test pattern using bitreverse intrinsic (details)
  97. [lldb][NFC] Make GetShellSafeArgument simpler and faster (details)
  98. [clangd] NFC: Update FIXME comment regarding lack of c/dtor support (details)
  99. [InstCombine] collectBitParts - add bitreverse intrinsic support. (details)
  100. Fix SBError::SetErrorToGenericError (details)
  101. [TableGen] [tests] Change integer ranges to use new '...' punctuation (details)
  102. [analyzer] [NFC] Simplify SVal::getAsLocSymbol function using existing functions (details)
  103. [VE] Support atomic store (details)
  104. [VE] Add vector logical instructions (details)
  105. [VE] Add vector shift instructions (details)
  106. [X86] Use mtriple instead of march in MIR tests (details)
  107. [AArch64] Add 2 cases where insertelement lowering could be improved. (details)
  108. [OpenMP] changing OMP rtl to use shared memory instead of env variable (details)
  109. [InstCombine] Add bswap test pattern using truncates (details)
  110. [libc++] Add a CI jobs to test the Standalone builds (details)
  111. [SVE][CodeGen][DAGCombiner] Fix TypeSize warning in redundant store elimination (details)
  112. Revert "[SVE][CodeGen][DAGCombiner] Fix TypeSize warning in redundant store elimination" (details)
  113. [SVE][CodeGen][DAGCombiner] Fix TypeSize warning in redundant store elimination (details)
  114. Unconfuse gcc5.3 after 2e204e23911b1f / D87528 (details)
  115. [libc++] Fix indentation of buildkite-pipeline.yml (details)
  116. [lld][ELF] Don't write output to the test directory. NFC. (details)
  117. [clang][unittest] Don't hardcode the string "Assertion" (details)
  118. [AMDGPU] Avoid unused variable warning in Release builds. NFC. (details)
  119. Fix issue in cortex-a57 sched model (details)
  120. [mlir][Linalg] Add basic support for TileAndFuse on Linalg on tensors. (details)
  121. [ARM][SchedModels] Convert IsLdstsoScaledNotOptimalPred to MCSchedPredicate (details)
  122. [lldb][PDB] Add ObjectFile PDB plugin (details)
  123. [DebugInfo][dexter] Add dexter tests for escaped locals (details)
  124. [SVE][AArch64] Fix TypeSize warning in loop vectorization legality (details)
  125. [SVE][AArch64] Fix TypeSize warning in GEP cost analysis (details)
  126. [SVE][InstCombine] Fix TypeSize warning in canReplaceGEPIdxWithZero (details)
  127. [SVE] Fix TypeSize warning in llvm::getGEPInductionOperand (details)
  128. [mlir][vector] Update doc strings for insert_map/extract_map and fix insert_map semantic (details)
  129. [AArch64] Extend tests for insertelement improvements. (details)
  130. SourceManager: Return non-const references in getOrCreateContentCache and related, NFC (details)
  131. [libc++abi] Make sure we can run the tests in Standalone mode (details)
  132. [Clang][CodeGen] fix failed assertion (details)
  133. [lldb][NativePDB] fix test load-pdb.cpp (details)
  134. [libc++] NFC: Consistent indentation for buildkite-pipeline.yml (details)
  135. [AMDGPU] Fix VC warning about singed/unsigned comparison. NFC. (details)
  136. Revert "Ensure that checkInitIsICE is called exactly once for every variable" (details)
  137. Revert "Revert "Ensure that checkInitIsICE is called exactly once for every variable"" (details)
  138. [mlir] Convert MemRefReinterpretCastOp to LLVM. (details)
  139. [clang-tidy] Silence unused variable warning in Release builds. NFCI. (details)
  140. Make sure Objective-C category support in IncludeSorter handles top-level imports (details)
  141. [ARM][SchedModels] Move IsLDMBaseRegInListPred to ARMSchedule.td. NFC (details)
  142. Add release tarballs for libclc (details)
  143. SourceManager: Fix an SLocEntry memory regression introduced with FileEntryRef (details)
  144. [clangd] Add back dependency on proto generated targets (details)
  145. Fix SROA with a PHI mergig values from a same block (details)
  146. [BitCode] decode nossp fn attr (details)
  147. llvm-reduce: Test reduction for D88684 ( ee6e25e4391a6d3ac0a3c89615474e512f44cda6 ) (details)
  148. [GWP-ASan] Refactor memory mapping functions (details)
  149. [libc++] NFC: Minor refactoring in filesystem_test_helper.h to ease readability (details)
  150. [InstCombine] reduce code duplication in icmp intrinsic folds; NFC (details)
  151. [InstCombine] add tests for ctpop at bitwidth limit; NFC (details)
  152. [InstCombine] add folds for icmp+ctpop (details)
  153. [AIX] Also error on -G for link-only step (details)
  154. [MemProf] Decouple memprof build from COMPILER_RT_BUILD_SANITIZERS (details)
  155. [libc++] Get rid of <iostream> in a filesystem test (details)
  156. Avoid unnecessary uses of `MDNode::getTemporary`, NFC (details)
  157. Test to check backtraces with machine function splitting. (details)
  158. [NFC] Remove max_align.c LIT testcase (details)
  159. [libunwind] Add -Wno-dll-attribute-on-redeclaration when building for windows (details)
  160. Prepend "__uniq" to symbol names hash with -funique-internal-linkage-names. (details)
  161. Run test only if X86 target is available (details)
  162. [AMDGPU] Use flat scratch instructions where available (details)
  163. [mlir] NFC: properly align IR in comments (details)
  164. [libc++] Remove the reliance of several <random> tests on <iostream> (details)
  165. [NFC] Fixing comment heading for MachineStableHash.h. (details)
  166. [nfc] [lldb] Refactor DWARFUnit::GetDIE (details)
  167. [mlir] Do not print back 0 alignment in LLVM dialect 'alloca' op (details)
  168. [TargetRegisterInfo] Fix a couple of typos in the comments (details)
  169. [CodeView] Emit static data members as S_CONSTANTs. (details)
  170. [MLIR] Fix AttributeInterface declaration. (details)
  171. IR: Clarify ownership of ConstantDataSequentials, NFC (details)
  172. [cmake] Add LLVM_UBSAN_FLAGS, to allow overriding UBSan flags (details)
  173. [mlir] Document 'ParentOneOf' with the HasParent trait (details)
  174. Fixed release build after D89170 (details)
  175. Explicitly check for entry basic block, rather than relying on MachineBasicBlock::pred_empty. (details)
  176. [mlir][Linalg] Miscalleneous enhancements to cover more fusion cases. (details)
  177. [DebugInfo] Expose Fortran array debug info attributes through DIBuilder. (details)
  178. [profile] Suppress spurious 'expected profile to require unlock' warning (details)
  179. [PowerPC] Implement Set Boolean Condition Instructions (details)
  180. Fix checking for C++98 ICEs in C++11-and-later mode to not consider use (details)
  181. [clang][NFC] Rearrange Comment Token and Lexer fields to reduce padding (details)
  182. [lldb][NativePDB] fix test load-pdb.cpp (details)
  183. [NFC] Use [MC]Register in RegAllocPBQP & RegisterCoalescer (details)
  184. SourceManager: Use the same fake SLocEntry whenever it fails to load (details)
  185. [mlir] Add a conversion pass between PDL and the  PDL Interpreter Dialect (details)
  186. [mlir][Pattern] Refactor the Pattern class into a "metadata only" class (details)
  187. [mlir][NFC] Move around the code related to PatternRewriting to improve layering (details)
  188. [mlir][Pattern] Add a new FrozenRewritePatternList class (details)
  189. IR: Add a comment at missing std::make_unique calls from b2b7cf39d596b1528cd64015575b3f5d1461c011, NFC (details)
  190. [NPM] Port -slsr to NPM (details)
  191. [AMDGPU] Move WQM Pass after MI Scheduler (details)
  192. Teach `-fsanitize=fuzzer` to respect `-static` and `-static-libstdc++` when adding C++ standard libraries. (details)
  193. Update email addresses in CODE_OWNERS. (details)
  194. IR: Simplify two loops walking ConstantDataSequential, NFC (details)
  195. SourceManager: clang-format the SrcMgr namespace, NFC (details)
  196. [LSR] fix typo in comments and rename for a new added hook. (details)
  197. [Clang][OpenMP] Fixed an issue of segment fault when using target nowait (details)
  198. [mlir][StorageUniquer] Refactor parametric storage to use sharded dense sets (details)
  199. [mlir][SymbolTable] Use Identifier instead of StringRef when looking up symbol name attributes (details)
  200. Port -objc-arc-apelim to NPM (details)
  201. Port -objc-arc-expand to NPM (details)
  202. [PlaceSafepoints] Pin tests to legacy PM (details)
  203. [AlwaysInliner] Pass callee AAResults to InlineFunction() (details)
  204. [CostModel][X86] teach TTI calculate cost of chain of vector inserts/extracts more precisely and correctly:In each 128-lane, if there is at least one index is demanded and not all indices are demanded... (details)
  205. Revert "[AlwaysInliner] Pass callee AAResults to InlineFunction()" (details)
  206. Use uint64_t for branch weights instead of uint32_t (details)
  207. Reland [AlwaysInliner] Pass callee AAResults to InlineFunction() (details)
  208. Fix calls to (p)read on macOS when size > INT32_MAX (details)
  209. [Clang][OpenMP] Avoid unnecessary privatization of mapper array when there is no user defined mapper (details)
  210. [IndVars] Remove monotonic checks with unknown exit count (details)
  211. [GVN LoadPRE] Add an option to disable splitting backedge (details)
  212. [libTooling] Recognize sccache as a compiler wrapper in compilation database commands (details)
  213. [clangd] Increase the TooMany limit for index-based textual navigation to 5 (details)
  214. [NFC] Factor away lambda's redundant parameter (details)
  215. [mlir] Fix TypeID lookup in GDB pretty printers. (details)
  216. [X86] Encode global address in small code model (details)
  217. [X86] Alternate implementation of D88194. (details)
  218. [Syntax] Disallow invalid Node operations (details)
  219. [Test] One more range check test (details)
  220. [clangd] Fix proto deps, for real this time. (details)
  221. Updating llvm.mlir test to match recent IR change (details)

Started by timer

This run spent:

  • 4 ms waiting;
  • 5 hr 31 min build duration;
  • 5 hr 31 min total from scheduled to completion.
Revision: 0fc1aa22ee6ac337a5d51fa5666c9cd61da61b07
  • refs/remotes/origin/master
Revision: e8ba87e92b857c14b7eb5466c4266a9e09a1f5fb
  • refs/remotes/origin/master
Test Result (no failures)