Started 15 days ago
Took 5 hr 53 min

Success Build #257 (May 29, 2021 1:37:04 AM)

Changes
  1. Removed abandoned lldb-sphinx-docs builder. (details / githubweb)
  2. [sanitizer] Switch to patched QEMU (details / githubweb)
  3. [sanitizer] Fix QEMU checkout (details / githubweb)
Changes
  1. [lldb][NFC] Remove misleading ModulePass base class for IRForTarget (details)
  2. [NFC][LoopIdiom] Add tests for 'arithmetic right-shift until zero' idiom (details)
  3. [LoopIdiom] Support 'arithmetic right-shift until zero' idiom (details)
  4. [lldb] Fix that LLDB doesn't print NaN's sign on Darwin (details)
  5. [Dexter] Add DexDeclareFile command to Dexter (details)
  6. [AArch64][SVE] Add fixed length codegen for FP_TO_{S,U}INT/{S,U}INT_TO_FP (details)
  7. [AMDGPU][Libomptarget] Mark lambda_by_value test as XFAIL (details)
  8. [NFC][LoopIdiom] Add tests for 'left-shift until zero' idiom (details)
  9. [LoopIdiom] Support 'left-shift until zero' idiom (details)
  10. [dexter] Change --source-root-dir and add --debugger-use-relative-paths (details)
  11. [NFC] [Coroutines] Remove unused variable: UnreachableCache (details)
  12. [Dexter] Remove erroneously added diff file (details)
  13. [mlir] Disallow certain transfer ops in VectorToSCF (details)
  14. [InstCombine] add tests for cast-of-shuffle; NFC (details)
  15. [InstCombine] canonicalize cast before unary shuffle (details)
  16. [mlir][doc] Fix links and references in documentation of Rationale (details)
  17. [mlir] Support buffer hoisting on allocas (details)
  18. [mlir][doc] Fix links and references in documentation of Dialects (details)
  19. [VectorCombine] Use constant range info for index scalarization legality. (details)
  20. [Headers][WASM] adjust test that runs the optimizer; NFC (details)
  21. [VectorCombine] Add test that combines load & store scalarization. (details)
  22. [InstSimplify] Transform X * Y % Y --> 0 (details)
  23. [AMDGPU] More accurate names for dpp operand types (details)
  24. [clang][ARM] When handling multiple -mimplicit-it mark all as used (details)
  25. Reland "Do not create LLVM IR `constant`s for objects with dynamic initialisation" (details)
  26. [X86][SSE] Regenerate vector shift codegen tests. NFCI. (details)
  27. [CostModel][X86] Improve accuracy of vXi64 vector non-uniform shift costs on AVX2+ targets (details)
  28. [AMDGPU] Allow no-modifier operands in cvtDPP (details)
  29. [libc++] [test] Format some C++20 iterator_traits tests. NFCI. (details)
  30. [libc++] [test] Make iter_difference_t.pass.cpp into a .compile.pass.cpp. NFCI. (details)
  31. [SystemZ][z/OS] Implement getHostCPUName for z/OS (details)
  32. BPF: Add more relocation kinds (details)
  33. [SystemZ] Return true from preferZeroCompareBranch(). (details)
  34. [scudo] Rework dieOnMapUnmapError (details)
  35. [mlir][CAPI][test] Change casts and fprintf format strings from long to intptr_t (details)
  36. [VectorCombine] Remove unneeded InsertPointGuard (NFCI). (details)
  37. [CostModel][X86] Improve accuracy of 256-bit non-uniform vector shifts on AVX1 (details)
  38. [X86][Atom] Fix vector PSHUFB resource/throughputs (details)
  39. [AIX][AsmPrinter] Print Symbol in comments for TOC load (details)
  40. [sparse][mlir] simplify sparse runtime support library (details)
  41. [libc++] Try to fix the oss-fuzz failure (details)
  42. [SCEV] Remove unused parameter from computeBECount [NFC] (details)
  43. [llvm-reduce] Exit when input module is malformed (details)
  44. [SCEV] Cleanup doesIVOverflowOnX checks [NFC] (details)
  45. Revert "[OpaquePtr] Make atomicrmw work with opaque pointers" (details)
  46. [InstCombine] avoid infinite loop from vector select transforms (details)
  47. [AMDGPU][GlobalISel] Stop foldInsertEltToCmpSelect from changing reg banks (details)
  48. [flang] Fix recent regression (proc. dummy arg on ENTRY) (details)
  49. [CSSPGO][llvm-profgen] Change default cold threshold for context merging (details)
  50. [mlir][linalg] Update Linalg.md (NFC). (details)
  51. [Hexagon] Improve argument packing in vector shuffle selection (details)
  52. [scudo] Consistent setting of SCUDO_DEBUG (details)
  53. [scudo] Fix CHECK implementation (details)
  54. [LoopIdiom] 'arithmetic right-shift until zero': don't turn potentially infinite loops into finite ones (details)
  55. [mlir] Add an optional distributionTypes attribute to TiledLoopOp. (details)
  56. [Matrix] Factor and distribute transposes across multiplies (details)
  57. [mlir][doc] Fix links and references in documentation of Tutorials (details)
  58. [InstSimplify] allow undef element match in vector select condition value (details)
  59. [AMDGPU] Lower kernel LDS into a sorted structure (details)
  60. [mlir] Use unique_function in AbstractOperation fields (details)
  61. Add pre-commit tests for [D102215](https://reviews.llvm.org/D102215). (details)
  62. [SelectionDAG] Propagate scoped AA metadata when lowering mem intrinsics. (details)
  63. [CVP] Guard against poison in common phi value transform (PR50399) (details)
  64. Revert "[NFC][scudo] Let disableMemoryTagChecksTestOnly to fail" (details)
  65. [lld-macho][nfc] Misc code cleanup (details)
  66. [lld-macho][nfc] clang-format everything (details)
  67. [lld-macho][nfc] Rename MergedOutputSection to ConcatOutputSection (details)
  68. [lld-macho][nfc] Sort OutputSections based on explicit order of command-line inputs (details)
  69. [lld-macho][nfc] Remove unnecessary parameterization of section sort (details)
  70. [gn build] Port 33706191d88d (details)
  71. [SCEV] Cache operands used in BEInfo (NFC) (details)
  72. Revert "[AIX] Avoid structor alias; die before bad alias codegen" (details)
  73. [Matrix] Use LLVM_DEBUG for a debug flag (details)
  74. [mlir] Add a copy constructor to FailureOr (details)
  75. [JITLink][MachO][arm64] Use a more descriptive test name. (details)
  76. [JITLink][MachO][arm64] Build GOT entries for defined symbols too. (details)
  77. [NFC] Fix 'unused' warning (details)
  78. [AMDGPU] Fix unused variable warning. NFC. (details)
  79. [TSAN][CMake] Add support to run lit on individual tests (details)
  80. [docs] Explain address spaces a bit more in opaque pointers doc (details)
  81. [sanitizer][test] s/A<10>/A<7>/ to fix "WARNING: Symbolizer buffer too small" which is somehow a hard error on s390x (details)
  82. [Hexagon] Remove unused function from HexagonISelDAGToDAGHVX.cpp (details)
  83. [docs] [CMake] Change recommendations for how to use LLVM_DEFINITIONS (details)
  84. [libcxx] Make the visibility attributes consistent for __narrow_to_utf8/__widen_from_utf8 (details)
  85. [libcxx] [test] Explain an XFAIL LIBCXX-WINDOWS-FIXME and convert into UNSUPPORTED (details)
  86. [libcxxabi] Use ASan interface header for declaration. NFC (details)
  87. [Canonicalize] Fully parameterize the pass based on config options. NFC. (details)
  88. [flang] Implement checks for defined input/output procedures (details)
  89. [Canonicalize] Switch the default setting to "top down". (details)
  90. [libcxx] Fix the function name in exceptions from create_directories (details)
  91. [libc++] [P0619] Hide not1 and not2 under _LIBCPP_ENABLE_CXX20_REMOVED_NEGATORS. (details)
  92. Revert "[LoopDeletion] Break backedge if we can prove that the loop is exited on 1st iteration" (details)
  93. [Internalize] Rename instead of removal if a to-be-internalized comdat has more than one member (details)
  94. [ARM] Extra predicated tests for VMULH. NFC (details)
  95. [libc++] Install GCC 11 on CI builders (details)
  96. [libomptarget][nfc] Move hostcall required test to rtl (details)
  97. [Toy] Update tests to pass with top-down canonicalize pass. NFC (details)
  98. [MLIR Core] Cache the empty StringAttr like we do for empty dictionaries. NFC. (details)
  99. [NFC][SCUDO] Fix unittest for -gtest_repeat=10 (details)
  100. [NFC][MLIR][TOSA] Replaced tosa linalg.indexed_generic lowerings with linalg.index (details)
  101. [gcov] Silence warning: comparison of integers of different signs (details)
  102. [benchmark] Silence 'suggest override' and 'missing override' warnings (details)
  103. LLVM Detailed IR tests for introduction of flag -fsanitize-address-detect-stack-use-after-return-mode. (details)
  104. [sanitizer] Let glibc aarch64 use O(1) GetTls (details)
  105. [OpaquePtr] Create new bitcode encoding for atomicrmw (details)
  106. [LTT] Handle merged llvm.assume when dropping type tests (details)
  107. [lldb] Avoid format string in LLDB_SCOPED_TIMER (details)
  108. Revert "[lldb] Avoid format string in LLDB_SCOPED_TIMER" (details)
  109. [OpaquePtr] Make atomicrmw work with opaque pointers (details)
  110. [libcxx][iterator] adds `std::ranges::advance` (details)
  111. [gn build] Port 36d0fdf9ac3b (details)
  112. [clang-format][NFC] correctly sort StatementAttributeLike-macros' IO.map (details)
  113. [ARM] Add patterns for vmulh (details)
  114. [NFC][object] Change the input parameter of the method isDebugSection. (details)
  115. [mlir] Fold complex.re(complex.create) and complex.im(complex.create) (details)
  116. [InstCombine] Fold extractelement + vector GEP with one use (details)
  117. [mlir] Simplify folding code (NFC) (details)
  118. [mlir] LocalAliasAnalysis: Assume allocation scope to function scope if cannot determine better (details)
  119. [HIP] Adjust check in hip-include-path.hip test case (details)
  120. Fix warning introduced by 9c766f4090d19e3e2f56e87164177f8c3eba4b96 (details)
  121. [NFC][X86] clang-format X86TTIImpl::getInterleavedMemoryOpCostAVX2() (details)
  122. [Test] Add test on unrolling to make sure it won't fail (details)
  123. [X86][Atom] Fix vector variable shift resource/throughputs (details)
  124. [CostModel][X86] Remove old testshift* tests (details)
  125. AArch64: support post-indexed stores to bfloat types. (details)
  126. [Test] Add simplified versions of tests for loop deletion that don't need context (details)
  127. [RISCV] Pre-commit fixed-length mask vselect tests (details)
  128. Revert "[AMDGPU][GlobalISel] Stop foldInsertEltToCmpSelect from changing reg banks" (details)
  129. [AMDGPU][GlobalISel] Stop foldInsertEltToCmpSelect from changing reg banks (details)
  130. [NFC] Add CHECK lines for unordered FP reductions (details)
  131. [AMDGPU][Libomptarget] Move Kernel/Symbol info tables to RTLDeviceInfoTy (details)
  132. [SCEV] Add tests with signed predicates for applyLoopGuards. (details)
  133. [X86][SLM] Fix vector PSHUFB + variable shift resource/throughputs (details)
  134. [lldb][NFC] Use C++ versions of the deprecated C standard library headers (details)
  135. [AMDGPU][Libomptarget] Delete g_atmi_initialized (details)
  136. [AMDGPU][Libomptarget] Inline atmi_init/atmi_finalize (details)
  137. [MC][NFCI] Factor out ELF section unique ID calculation (details)
  138. [MC][ELF] Emit unique sections for different flags (details)
  139. [Docs] Updated the content of getting started documentation under llvm/lib/MC (details)
  140. [MC] Move elf-unique-sections-by-flags.ll to X86/ (details)
  141. [OpenCL] Include header for atomic-ops test (details)
  142. [Test] Add Loop Deletion test with irreducible CFG (details)
  143. [AArch64] Generate LD1 for anyext i8 or i16 vector load (details)
  144. [mlir] Fold complex.create(complex.re(op), complex.im(op)) (details)
  145. Return "[LoopDeletion] Break backedge if we can prove that the loop is exited on 1st iteration" (details)
  146. Revert "Return "[LoopDeletion] Break backedge if we can prove that the loop is exited on 1st iteration"" (details)
  147. [InstCombine] avoid 'tmp' usage in test file; NFC (details)
  148. [InstCombine] avoid 'tmp' usage in test files; NFC (details)
  149. [InstCombine] add fmul tests with shared operand; NFC (details)
  150. Return "[LoopDeletion] Break backedge if we can prove that the loop is exited on 1st iteration" (try 2) (details)
  151. [LoopVectorize] Enable strict reductions when allowReordering() returns false (details)
  152. [SLP][NFC]Add a test for multiple uses of insertelement instruction, (details)
  153. [MCA][InOrderIssueStage] Fix LastWriteBackCycle computation. (details)
  154. [X86][SSE] Regenerate some tests to expose the rip relative vector/broadcast loads (details)
  155. [ARM] Extra test for reverted WLS memset. NFC (details)
  156. [CostModel][AArch64] Add tests for bitreverse. NFC. (details)
  157. [X86][AMX] Fix a bug on tile config. (details)
  158. [clang-cl] Add driver support for /std:c++20 and bump /std:c++latest (PR50465) (details)
  159. [SystemZ][z/OS] Validate symbol names for z/OS for printing without quotes (details)
  160. [AMDGPU] Fix function pointer argument bug in AMDGPU Propagate Attributes pass. (details)
  161. [SystemZ][z/OS] Enable the AllowAtInName attribute for the HLASM dialect (details)
  162. [MCA] Add a test for PR50483. NFC (details)
  163. [clangd] New ParsingCallback for semantics changes (details)
  164. [SystemZ] Support i128 inline asm operands. (details)
  165. [libc++] Fix concepts tests with GCC (details)
  166. [unroll] Use value domain for symbolic execution based cost model (details)
  167. [mlir][openacc] Translate UpdateOp to LLVM IR (details)
  168. [libc++][NFC] Move format_error to its own header. (details)
  169. [gn build] Port 16342e39947b (details)
  170. [libc++][NFC] Move basic_format_parse_context to its own header. (details)
  171. [mlir] Add `distributionTypes` to LinalgTilingOptions. (details)
  172. [libc++][format] Adds availability macros for std::format. (details)
  173. [gn build] Port de9df3f5b952 (details)
  174. [libomptarget][nfc][amdgpu] Remove atmi_status_t type (details)
  175. [AMDGPU][NFC] Remove non-existing function header (details)
  176. [RISCV] Enable cross basic block aware vsetvli insertion (details)
  177. [lldb] Remove cache in get_demangled_name_without_arguments (details)
  178. [libunwind] Inform ASan that resumption is noreturn (details)
  179. [lldb] add LLDB_SKIP_DSYM option (details)
  180. [DebugInfo] Limit the number of values that may be referenced by a dbg.value (details)
  181. [SLP]Fix vectorization of insertelements with multiple uses. (details)
  182. [scudo] Get rid of initLinkerInitialized (details)
  183. [RISCV] Don't propagate VL/VTYPE across inline assembly in the Insert VSETVLI pass. (details)
  184. [RISCV] Remove --riscv-no-aliases from RVV tests (details)
  185. [Scudo] Make -fsanitize=scudo use standalone. Migrate tests. (details)
  186. [SCEV] Extract out a helper for computing trip multiples (details)
  187. [RISCV] Optimize SEW=64 shifts by splat on RV32. (details)
  188. [SCEV] Add a utility for converting from "exit count" to "trip count" (details)
  189. [llvm-mc] Add -M to replace -riscv-no-aliases and -riscv-arch-reg-names (details)
  190. Revert "[Scudo] Make -fsanitize=scudo use standalone. Migrate tests." (details)
  191. [mlir] Make StripDebugInfo strip out block arguments locs (details)
  192. [SCEV] Generalize getSmallConstantTripCount(L) for multiple exit loops (details)
  193. [libomptarget][nfc][amdgpu] Refactor uses of KernelInfoTable (details)
  194. [AMDGPU][MC][GFX90A] Corrected DS_GWS opcodes (details)
  195. [AMDGPU] Fix kernel LDS lowering for constants (details)
  196. [SimplifyCFG] Use make_early_inc_range() while deleting instructions (details)
  197. [WebAssembly] Add TargetInstrInfo::getCalleeOperand (details)
  198. [SCEV] Compute trip multiple for multiple exit loops (details)
  199. [NFC][X86][Costmodel] Add some more interleaved load/store test with i16 element type (details)
  200. [NFC][Codegen][X86] Add a few more interleaved load/store patterns w/ i16 element type (details)
  201. [libomptarget][nfc][amdgpu] Factor out setting upper bounds (details)
  202. [pstl] Fix -Wundef errors in the test suite (details)
  203. [DebugInstrRef][1/3] Track PHI values through register allocation (details)
  204. [CostModel][AArch64] Add floating point arithmetic tests. NFC. (details)
  205. [pstl] Workaround more errors in the test suite (details)
  206. [libc++] Add a job testing on GCC 11 (details)
  207. Update documentation for InlineModel features. (details)
  208. [RISCV][NFC] Fix some whitespace nits in MC test RUN lines (details)
  209. [RISCV] Use X0 as destination of inserted vsetvli when possible. (details)
  210. [llvm-readobj] Optimize printing stack sizes to linear time. (details)
  211. Refactor mutation strategies into a standalone library (details)
  212. [libcxx][docs] Take mutex for common_iterator, common_view, and empty_view. (details)
  213. [AArch64] Support llvm-mc/llvm-objdump -M no-aliases (details)
  214. [flang][docs] Initial documentation for the Fortran LLVM Test Suite. (details)
  215. [libcxx][nfc] Fix the ASAN bots: update expected.pass.cpp. (details)
  216. [libFuzzer] Add missing FuzzerBuiltinsMsvc.h include. (details)
  217. [lldb] Fix gnu_libstdcpp's update methods (details)
  218. [mlir][python] Provide "all passes" registration module in Python (details)
  219. Revert "Refactor mutation strategies into a standalone library" (details)
  220. [mlir] Add n-D vector lowering to LLVM for cast ops (details)
  221. [scudo] Build scudo_standalone on Android and Fuchsia. (details)
  222. [tests] Add some basic coverage of multiple exit unrolling (details)
  223. [Hexagon] Restore handling of expanding shuffles (details)
  224. [MLIR] Add support for empty IVs to affine.parallel (details)
  225. Revert "[scudo] Build scudo_standalone on Android and Fuchsia." (details)
  226. [AArch64][GlobalISel] Legalize non-power-of-2 vector elements for G_STORE. (details)
  227. Emit correct location lists with basic block sections. (details)
  228. [GlobalISel] Don't emit lost debug location remarks when legalizing tail calls (details)
  229. [ORC-RT] Add endianness support to the ORC runtime. (details)
  230. Fix unit test after 324af79dbc6066 (details)
  231. The compiler is crashing when compiling a coroutine intrinsic without (details)
  232. [RS4GC] Introduce intrinsics to get base ptr and offset (details)
  233. [Fuchsia][CMake] Add missing include path. (details)
  234. Fix non-global-value-max-name-size not considered by LLParser (details)
  235. [libcxx][iterator] adds `std::ranges::next` (details)
  236. [libcxx][iterator] adds `std::ranges::prev` (details)
  237. [gn build] Port 0dc7fd1bc167 (details)
  238. [gn build] Port 857fa7b7b187 (details)
  239. [llvm-objdump] Print the DEBUG type under `--section-headers`. (details)
  240. [NFCI][LoopDeletion] Only query SCEV about loop successor if another successor is also in loop (details)
  241. [NFC] Formatting fix (details)
  242. [NFCI] Lazily evaluate SCEVs of PHIs (details)
  243. [mlir] Add a pass to distribute linalg::TiledLoopOp. (details)
  244. [docs] llvm-objdump: Mention -M no-aliases is supported on AArch64 (details)
  245. [mlir] Add TestLinalgDistribution.cpp to cmake build. (details)
  246. [GlobalISel] Implement splitting of G_SHUFFLE_VECTOR. (details)
  247. [NFC] Reuse existing variables instead of re-requesting successors (details)
  248. [NFCI][LoopDeletion] Do not call complex analysis for known non-zero BTC (details)
  249. [mlir] Support dialect-wide canonicalization pattern registration (details)
  250. [AMDGPU][GlobalISel] Allow amdgpu_gfx calling conv (details)
  251. [SelectionDAG][RISCV] Don't unroll 0/1-type bool VSELECTs (details)
  252. [Matrix] Include matrix pipeline for new PM in new-pm-defaults.ll. (details)
  253. [lit][test] Improve testing of use_llvm_tool (details)
  254. Add triples to a bunch of x86-specific tests that currently fail on PPC (details)
  255. [clang-format] [NFC] realign documentation in Format.h... (details)
  256. [ARM] Extra test for reverted WLS memset. NFC (details)
  257. [OpenMP]Add support for workshare loop modifier in lowering (details)
  258. [AMDGPU][Libomptarget][NFC] Remove atmi_mem_place_t (details)
  259. Revert "[OpenMP]Add support for workshare loop modifier in lowering" (details)
  260. Add --quiet option to llvm-gsymutil to suppress output of warnings. (details)
  261. [mlir][Linalg] Add comprehensive bufferization support for subtensor (5/n) (details)
  262. Add support for #elifdef and #elifndef (details)
  263. [VPlan] Do not sink uniform recipes in sinkScalarOperands. (details)
  264. [RISCV] Allow passing fixed-length vectors via the stack (details)
  265. [DAGCombine][RISCV] Don't try to trunc-store combined vector stores (details)
  266. Fix -Wswitch warning; NFC (details)
  267. AMDGPU/GlobalISel: Remove redundant parameter from function (details)
  268. AMDGPU/GlobalISel: Lower constant-32-bit zextload/sextload consistently (details)
  269. Speculatively fix a -Woverloaded-virtual diagnostic; NFC (details)
  270. Speculatively fix this harder and with improved spelling capabilities. (details)
  271. Reimplement __builtin_unique_stable_name- (details)
  272. Reuse temporary files for print-changed=diff (details)
  273. Correct the 'KEYALL' mask. (details)
  274. Hopefully fix the Clang sphinx doc build. (details)
  275. [OpenMP]Add support for workshare loop modifier in lowering (details)
  276. [VP][SelectionDAG] Add a target-configurable EVL operand type (details)
  277. Disable misc-no-recursion checking in Clang (details)
  278. VirtRegMap: Preserve LiveDebugVariables (details)
  279. [Flang][Openmp] Fortran specific semantic checks for Allocate directive (details)
  280. [OpenCL][NFC] Fix typos in test (details)
  281. [X86][SSE] Regenerate some tests to expose the rip relative vector/broadcast loads (details)
  282. [CostModel][X86] AVX512 truncation ops are slower than cost models indicate. (details)
  283. AMDGPU/GlobalISel: Fix broken test run line (details)
  284. AMDGPU/GlobalISel: Use IncomingValueAssigner for implicit return (details)
  285. [libc++] Deprecate std::iterator and remove it as a base class (details)
  286. [libc++] NFC: Parenthesize expression to satisfy GCC 11 (details)
  287. Revert "Emit correct location lists with basic block sections." (details)
  288. Thread safety analysis: Factor out function for merging locks (NFC) (details)
  289. Thread safety analysis: Allow exlusive/shared joins for managed and asserted capabilities (details)
  290. [AIX] Add -lc++abi and -lunwind for linking (details)
  291. GlobalISel: Do not change register types in lowerLoad (details)
  292. [RISCV] Add a test case showing incorrect call-conv lowering (details)
  293. [HIP] Check compatibility of -fgpu-sanitize with offload arch (details)
  294. [mlir][gpu] Relax restriction on MMA store op to allow chain of mma ops. (details)
  295. [SPE] Disable strict-fp for SPE by default (details)
  296. [LoopUnrollAndJam] Change LoopUnrollAndJamPass to LoopNest pass (details)
  297. [mlir] Async reference counting for block successors with divergent reference counted liveness (details)
  298. [Clang] Enable __has_feature(coverage_sanitizer) (details)
  299. [mlir] Add error state and error propagation to async runtime values (details)
  300. [X86] Pre-commit tests for D103192. NFC (details)
  301. [X86] Fold (shift undef, X)->0 for vector shifts by immediate. (details)
  302. [mlir] Async: Add error propagation support to async groups (details)
  303. [mlir][NFC] Don't outline kernel in MMA integration tests (details)
  304. [RISCV] Teach vsetvli insertion to use vsetvl x0, x0 form when we can tell that VLMAX and AVL haven't changed. (details)
  305. [CostModel][X86] Improve accuracy of sext/zext to 256-bit vector costs on AVX1 targets (details)
  306. [NFC][X86][Codegen] Re-autogenerate check lines in a few tests to remove noise from future changes (details)
  307. Revert "[libc++] NFC: Parenthesize expression to satisfy GCC 11" (details)
  308. [NFC][scudo] Rename internal function (details)
  309. MC: mark `dump` with `LLVM_DUMP_METHOD` (details)
  310. [mlir] AsyncRefCounting: check that LivenessBlockInfo is not nullptr (details)
  311. [mlir] Update cmake variable post D102976 (details)
  312. [NFC][scudo] Check zeros on smaller allocations (details)
  313. [libc++] NFC: Refactor raw_storage_iterator test to use UNSUPPORTED markup (details)
  314. [RISCV] Add a test showing missed opportunity to avoid a vsetvli in a loop. (details)
  315. [lldb][intel-pt] Remove old plugin (details)
  316. [mlir:Async] Convert assertions to async errors only inside async functions (details)
  317. [analyzer] RetainCountChecker: Disable reference counting for OSMetaClass. (details)
  318. Support stripping indirectly referenced DILocations from !llvm.loop metadata (details)
  319. [NFC][X86][Codegen] vector-interleaved-store-i16-stride-5.ll: precisely match the actual IR (details)
  320. [x86] add tests for extend of vector compare; NFC (details)
  321. Replace 'magic static' with a member variable for SCYL kernel names (details)
  322. [libc++] NFC: Make it easier for vendors to extend the run-buildbot script (details)
  323. [clang] [MinGW] Don't mark emutls variables as DSO local (details)
  324. [libcxx] [test] Convert an XFAIL LIBCXX-WINDOWS-FIXME into UNSUPPORTED with explanation (details)
  325. [mlir] Add support for querying the ModRef behavior from the AliasAnalysis class (details)
  326. [RISCV] Fix typo, use addImm instead of addReg. (details)
  327. [PDB] Enable parallel ghash type merging by default (details)
  328. [PowerPC] Added multiple PowerPC builtins (details)
  329. [MCA] Refactor the InOrderIssueStage stage. NFCI (details)
  330. [mlir-lsp-server] Add support for processing split files (details)
  331. [sanitizer] Android ELF TLS is supported from Q (API 29) (details)
  332. [mlir][capi] fix build issue with "all passes" registration (details)
  333. [NFC][X86][Codegen] Re-autogenerate a few tests to reduce noise in future changes (details)
  334. [SanCov] Properly set ABI parameter attributes (details)
  335. [RISCV] Teach VSETVLI insertion to look through PHIs to prove we don't need to insert a vsetvli. (details)
  336. [ConstFold] Simplify a load's GEP operand through local aliases (details)
  337. [MCA] Minor changes to the InOrderIssueStage. NFC (details)
  338. [NFC][libObject] clang-format Archive{.h,.cpp} (details)
  339. Fix comment to reflect what the method is doing (NFC) (details)
  340. [dfsan] Add a flag about whether to propagate offset labels at gep (details)
  341. [AIX] Enable stackprotect feature (details)
  342. [PDB] Fix ubsan complaint about memcpy from null pointer (details)
  343. [clang-cl] Bump default -fms-compatibility-version to 19.14 (details)
  344. Pass -gcodeview-ghash when using clang-cl and lld-link (details)
  345. Revert "[NFCI] Lazily evaluate SCEVs of PHIs" (details)
  346. [AArch64][GlobalISel] Legalize oversize G_EXTRACT_VECTOR_ELT sources. (details)
  347. [LoopNest][LoopFlatten] Change LoopFlattenPass to LoopNest pass (details)
  348. [LoopNest][LoopFlatten] Change LoopFlattenPass to LoopNest pass (details)
  349. [MLIR][SPIRV] Use getAsmResultName(...) hook for ConstantOp. (details)
  350. Revert "[LoopNest][LoopFlatten] Change LoopFlattenPass to LoopNest pass" (details)
  351. Revert "[LoopNest][LoopFlatten] Change LoopFlattenPass to LoopNest pass" (details)
  352. [ConstantFolding] Fix -Wunused-variable warning (NFC) (details)
  353. [clang] [MinGW] Fix gcc version detection/picking (details)
  354. Revert "[scudo] Check if MADV_DONTNEED zeroes memory" (details)
  355. [NFC][scudo] Re-enable check in the test (details)
  356. [cmake][scudo] Add missing headers (details)
  357. [WebAssembly][CodeGen] IR support for WebAssembly local variables (details)
  358. [AMDGPU] Precommit test (details)
  359. [AMDGPU] Fix function calls with flat scratch (details)
  360. Revert "[LAA] Support pointer phis in loop by analyzing each incoming pointer." (details)
  361. [NFC][Transforms][Utils] remove useless variable in CloneBasicBlock (details)
  362. ARM: support mandatory tail calls for tailcc & swifttailcc (details)
  363. SwiftTailCC: teach verifier musttail rules applicable to this CC. (details)
  364. Revert "[WebAssembly][CodeGen] IR support for WebAssembly local variables" (details)
  365. [mlir] Don't elide the last op if there is no terminator (details)
  366. SwiftAsync: add Clang attribute to apply the LLVM `swiftasync` one. (details)
  367. [RISCV] Add tests for fixed vector conversions between fp to/from i1 (details)
  368. [RISCV][NFC] Merge identical RV32 and RV64 test checks (details)
  369. Fix "enumerator 'llvm::TargetStackID::WasmLocal' in switch of enum 'llvm::TargetStackID::Value' is not handled" MSVC warnings. NFCI. (details)
  370. [x86] add test for extend of legal vector compare; NFC (details)
  371. [Matrix] Move C++ matrix cast checks to TryStaticCast. (details)
  372. [gn build] (semi-manually) port bd04d78d649b (details)
  373. [clang] NFC: Replace std::pair by a struct in InitHeaderSearch (details)
  374. [AArch64] Add additional vector load scalarization tests for D103077. (details)
  375. [LLD][COFF] Reduce the maximum size of the GHASH table (details)
  376. [libc++] Add all indirect callable concepts and projected (details)
  377. [gn build] Port 58b29a4efc22 (details)
  378. [gn] Make ubsan errors fatal, as in cmake (details)
  379. [IR] Make TypeFinder aware of DIArgList values (details)
  380. [LoopNest][LoopFlatten] Change LoopFlattenPass to LoopNest pass (details)
  381. Revert "Fix "enumerator 'llvm::TargetStackID::WasmLocal' in switch of enum 'llvm::TargetStackID::Value' is not handled" MSVC warnings. NFCI." (details)
  382. [clang][Parse] Add parsing support for C++ attributes on using-declarations (details)
  383. [mlir] Add missing namespace to createCanonicalizerPass. (details)
  384. [libc++] NFC: Move unwrap_iter to its own header (details)
  385. [LoopDeletion] Add test with potentially infinite sub-loop. (details)
  386. [PhaseOrdering] add test for late simplifycfg with LTO; NFC (details)
  387. [PassManager] unify late simplifycfg options between regular and LTO pipelines (details)
  388. Revert "[NFCI][LoopDeletion] Do not call complex analysis for known non-zero BTC" (details)
  389. Revert "[NFC] Reuse existing variables instead of re-requesting successors" (details)
  390. Revert "[NFC] Formatting fix" (details)
  391. Revert "[NFCI][LoopDeletion] Only query SCEV about loop successor if another successor is also in loop" (details)
  392. Revert "Return "[LoopDeletion] Break backedge if we can prove that the loop is exited on 1st iteration" (try 2)" (details)
  393. [VectorCombine] Add variants of multi-extract tests with assumes. (details)
  394. [VectorCombine] Check indices for all extracts we scalarize. (details)
  395. [SelectionDAG] Fix typo in assert. NFC (details)
  396. [VP] Make getMaskParamPos/getVectorLengthParamPos return unsigned. Lowercase function names. (details)
  397. Revert "[clang][Parse] Add parsing support for C++ attributes on using-declarations" (details)
  398. [RISCV] Add octuple to LMULInfo tablegen class, remove octuple_from_str. NFCI (details)
  399. [gn build] Port 9968896cd62a (details)
  400. [libc++] [test] Constexpr-ify a couple of insert-iterator tests. (details)
  401. [gn build] manually port 982e3c05108b6 (check-lld needs dsymutil) (details)
  402. [AArch64][RISCV] Make sure isel correctly honors failure orderings. (details)
  403. [clang-format] [docs] Regenerate style options documentation. (details)
  404. [dsymutil tests] Try to make eh_frames.test run on other platforms (details)
  405. [NFC] Remove confusing info about MainLoop VF/UF from debug message (details)
  406. [RISCV] Pre-commit test cases for D103211. NFC (details)
  407. [LoopUnroll] Clean up exit folding (NFC) (details)
  408. [LoopUnroll] Add store to unreachable latch test (NFC) (details)
  409. [RISCV] Add separate MxList tablegen classes for widening/narrowing and sext.zext.vf2/4/8. NFC (details)
  410. [LoopUnroll] Use changeToUnreachable() (NFC) (details)
  411. [lld-macho][test] Simplify --allow-empty with count 0 (details)
  412. [CGAtomic] Delete outdated code comparing success/failure ordering for cmpxchg. (details)
  413. [Verifier] Significantly speed up IsolatedFromAbove checking. NFC. (details)
  414. [clang-format] New BreakInheritanceList style AfterComma (details)
  415. [AArch64][GlobalISel] Fix a crash during selection of a G_ZEXT(s8 = G_LOAD) (details)
  416. [Internalize] Simplify comdat renaming with noduplicates after D103043 (details)
  417. [lld/mac] Don't crash on -order_file with assembly inputs on arm64 (details)
  418. [RISCV] Enable interleaved vectorization for RVV (details)
  419. [InstrProfiling][test] Fix stale tests (details)
  420. [InstrProfiling][test] Fix stale linkage.ll (details)
  421. [AtomicExpandPass][AArch64] Promote xchg with floating-point types to integer ones (details)
  422. [LoopUnroll] Make DomTree explicitly required (NFC) (details)
  423. Revert "[libc++] NFC: Move unwrap_iter to its own header" (details)
  424. [gn build] Port b13edf6e907b (details)
Changes
  1. Removed abandoned lldb-sphinx-docs builder. (details)
  2. [sanitizer] Switch to patched QEMU (details)
  3. [sanitizer] Fix QEMU checkout (details)

Started by timer

This run spent:

  • 6 ms waiting;
  • 5 hr 53 min build duration;
  • 5 hr 53 min total from scheduled to completion.
Revision: b423773c90d5046a6f9eef0e6ed48a1a377282e4
  • refs/remotes/origin/main
Revision: 42910da585d368208365247e6616f80880d5cc3e
  • refs/remotes/origin/main
Revision: b423773c90d5046a6f9eef0e6ed48a1a377282e4
  • refs/remotes/origin/main
Test Result (no failures)