Started 4 mo 12 days ago
Took 57 min on green-dragon-06

Success Build #9700 (Feb 10, 2021 2:03:20 AM)

Changes

Git (git http://labmaster3.local/git/llvm-project.git)

  1. [libcxx][test] Silence MSVC narrowing warning (detail)
  2. [AIX] Actually push back "-mabi=vec-extabi" when option is on. (detail)
  3. [VE] Add include for formatted_raw_ostream after 046cfb856517c6140d5e1c0989232e26d00b05b2 (detail)
  4. [GlobalISel] Remove hint instructions in generic InstructionSelect code. (detail)
  5. [CMake] Actually require python 3.6 or greater (detail)
  6. Revert "[CMake] Actually require python 3.6 or greater" (detail)
  7. [mlir] Preserve lexicographic order after loop collapsing. (detail)
  8. [clang-tidy] Fix linking tests to LLVMTestingSupport (detail)
  9. Emit metadata when instr. profiles hash mismatch occurs. (detail)
  10. [SVE] delete VectorType::getNumElements() (detail)
  11. [LowerConstantIntrinsics] Preserve Dominator Tree, if avaliable (detail)
  12. [ShadowStackGCLowering] Preserve Dominator Tree, if avaliable (detail)
  13. [ExpandMemCmpPass] Preserve Dominator Tree, if available (detail)
  14. [AMDGPU] Be more specific in needsFrameBaseReg (detail)
  15. [NFC][RISCV] Remove redundant pseudo instructions for vector load/store. (detail)
  16. [RISCV] Update the version number to v0.10 for vector. (detail)
  17. [flang] Support disabled alternative PARAMETER statement (detail)
  18. [RISCV]  Update the version number to v0.10 for vector. (detail)
  19. Update ElementsAttr::isValidIndex to handle ElementsAttr with a scalar. Scalar will have rank 0. (detail)
  20. [VFS] Combine VFSFromYamlDirIterImpl and OverlayFSDirIterImpl into a single implementation (NFC) (detail)
  21. [RISCV] Merge rv32 and rv64 vector fadd/fsub/fmul/fdiv sdnode tests into single tests files with 2 run lines. (detail)
  22. Use type alias for large tuples (detail)
  23. Detect Source Drift with Propeller. (detail)
  24. [X86] Fix tile config register spill issue. (detail)
  25. [llvm-objdump-macho] print per-second-level-page encodings for option --unwind-info (detail)
  26. [lld-macho][NFC] Add new option group for versions (detail)
  27. [NFC][VFS] Fix a build warning due to an extra semicolon (detail)
  28. [llvm] Use append_range (NFC) (detail)
  29. [llvm] Use llvm::lower_bound and llvm::upper_bound (NFC) (detail)
  30. [llvm] Use isa instead of dyn_cast (NFC) (detail)
  31. [ELF][test] Add --emit-relocs --gc-sections test for relocation section for a non-SHF_ALLOC section (detail)
  32. [LTO] Use lto::backend for code generation. (detail)
  33. [LTO] Add option enable NewPM with LTOCodeGenerator. (detail)
  34. [compiler-rt][tests] Define TARGET_FLAGS (detail)
  35. [ASTMatchers] Fix traversal below range-for elements (detail)
  36. [LoopUnswitch] Properly update MSSA if header has non-clobbering stores. (detail)
  37. Fix traversal with hasDescendant into lambdas (detail)
  38. [llvm-symbolizer] - Fix the crash in GNU output style with --no-inlines and missing input file. (detail)
  39. [RISCV] Update extractelt tests to sign-extend results (NFC) (detail)
  40. [ASTMatchers] Fix definition of decompositionDecl (detail)
  41. [TableGen] Use ListSeparator (NFC) (detail)
  42. [AMDGPU] Forward-declare AMDGPUTargetMachine (NFC) (detail)
  43. [llvm] Add missing header guards (NFC) (detail)
  44. [ConstraintElimination] Verify CS and DFSInStack are in sync.(NFC) (detail)
  45. [OpenMP][NVPTX] Refined CMake logic to choose compute capabilites (detail)
  46. [TableGen] Avoid a couple vector copies in ExpandHwModeBasedTypes. (detail)
  47. [TableGen] Use emplace_back to add to PatternsToMatch in GenerateVariants. Use std::move when adding to PatternsToMatch in AddPatternToMatch. (detail)
  48. [RISCV] Use MVT instead of EVT in RISCVISelDAGToDAG.cpp (detail)
  49. [flang] Improve shape & length characterization (detail)
  50. [mlir][AVX512] Fix result type of vp2intersect (detail)
  51. [RISCV] Add rv64 command line to bswap-ctlz-cttz-ctpop.ll. (detail)
  52. [docs] Clarify compile_flags.txt subtleties (detail)
  53. [clangd] Quote/escape argv included in log messages. (detail)
  54. [clangd] Remove references to old future-based API. NFC (detail)
  55. [libc++] Allow building with C++17. (detail)
  56. [dsymutil][DWARFLinker][NFC] make AddressManager not depending on the order of checks for relocations. (detail)
  57. Revert "[clangd] Quote/escape argv included in log messages." (detail)
  58. Reland [clangd] Quote/escape argv included in log messages. (detail)
  59. AMDGPU: Add missing consts (detail)
  60. RegAlloc: Fix assert if all registers in class reserved (detail)
  61. [VE] Fix compiler warnings (NFC) (detail)
  62. [TableGen] Use ListSeparator (NFC) (detail)
  63. [llvm] Drop unnecessary const from return types (NFC) (detail)
  64. [lldb] [Process/FreeBSDRemote] Introduce arm64 support (detail)
  65. [lldb] [Process/FreeBSDRemote] Introduce arm (32-bit) support (detail)
  66. [ConstraintElimination] Add tests for ICMP_EQ predicates. (detail)
  67. [LTOCodeGenerator] Use lto::Config for options (NFC). (detail)
  68. [LSan][RISCV] Enable LSan for RISCV64 (detail)
  69. Fix omitted kw in type alias printer (detail)
  70. [RISCV][LegalizeTypes] Try to expand BSWAP before promoting if the promoted BSWAP would expand anyway. (detail)
  71. CFG: Create scope for non-compound range-for body. (detail)
  72. [CMake][compiler-rt][RISCV] Support RISC-V cross-compilation (detail)
  73. [Sanitizer][RISCV] Fix FastUnwindTest (detail)
  74. [CodeGenPrepare] Also skip lifetime.end intrinsic when check return block in dupRetToEnableTailCallOpts. (detail)
  75. Fix test in "CFG: Create scope for non-compound range-for body." (detail)
  76. [OpenMP] Fix comment and assertion strings (NFC). (detail)
  77. [JITLink] Add missing symbols for ELF ehframe testcase, re-enable on Windows. (detail)
  78. [RISCV] Custom lower fshl/fshr with Zbt extension. (detail)
  79. Revert "[JITLink] Add missing symbols for ELF ehframe testcase, re-enable ...." (detail)
  80. [X86] Cleanup isel patterns to use 'vnot' instead of (xor X, immAllOnesV) to improve readability. NFC (detail)
  81. [PowerPC] Remove vnot_ppc and replace with the standard vnot. (detail)
  82. [Mips] Cleanup isel patterns to use 'vnot' instead of (xor X, immAllOnesV). NFCI (detail)
  83. [FPEnv] Intrinsic for setting rounding mode (detail)
  84. [TableGen] Don't commute isel patterns if it would put an immAllOnesV or immAllZerosV on the left hand side. (detail)
  85. [OpenMP] Fix python3 compatibility in openmp's lit.cfg (detail)
  86. [mlir][Linalg] Replace SimplePad with PadTensor in hoist-padding (detail)
  87. [mlir] Fix missing null termination in cuLinkAddData argument. (detail)
  88. Revert "[mlir][Linalg] Replace SimplePad with PadTensor in hoist-padding" (detail)
  89. [lldb][docs] Attempt to disable the generated GitHub button on the LLDB website (detail)
  90. [LoopPeel] Use llvm.experimental.noalias.scope.decl for duplicating noalias metadata as needed. (detail)
  91. [LoopUnswitch] Add shortcut if unswitched path is a no-op. (detail)
  92. [LoopUnswitch] Pacify compiler warnings. (detail)
  93. [VE] Change inetger constants 32-bit friendly (detail)
  94. [ARM] Simplify extract of VMOVDRR (detail)
  95. [X86][AVX] combineExtractWithShuffle - combine extracts from 256/512-bit vector shuffles. (detail)
  96. [Verifier] enable llvm.experimental.noalias.scope.decl dominance check. (detail)
  97. [ARM] Turn sext_inreg(VGetLaneu) into VGetLaneu (detail)
  98. [Branch-Rename] Fix some links (detail)
  99. [lldb/API] Expose Module::IsLoadedInTarget() to SB API (NFC) (detail)
  100. [AMDGPU][MC] Corrected error position for invalid operands (detail)
  101. [lldb] Remove a stray semicolon, fixing GCC warnings. NFC. (detail)
  102. Revert "[clang][cli] Port OpenMP-related LangOpts to marshalling system" (detail)
  103. [lldb] Use current execution context in SBDebugger (detail)
  104. NFC: Migrate SpeculativeExecution to work on InstructionCost (detail)
  105. [LV] Fix crash when computing max VF too early (detail)
  106. GlobalISel: check type size before getZExtValue()ing it. (detail)
  107. [lldb/test] Skip `SBTarget::IsLoaded` test on windows (NFC) (detail)
  108. [ConstraintElimination] Add tests for signed predicates. (detail)
  109. Revert "[Verifier] enable llvm.experimental.noalias.scope.decl dominance check." (detail)
  110. [SCEV] Bail out if URem operand cannot be zero-extended. (detail)
  111. [SVE][CodeGen] Remove performMaskedGatherScatterCombine (detail)
  112. [clangd] Rename: merge index/AST refs path-insensitively where needed (detail)
  113. [gn build] Port b63cd4db915c (detail)
  114. [clangd] Don't rely on builtin headers for document-link.test. (detail)
  115. [libc++] Fix for the Bug 41784 (detail)
  116. [APFloat] multiplySignificand - pass IEEEFloat as const reference. NFCI. (detail)
  117. [TableGen] Fix anonymous record self-reference in foreach and multiclass (detail)
  118. [clangd] Respect ReferencesParams.context.includeDeclarations (detail)
  119. [ARM] Simplify VMOVRRD from extracts of buildvectors (detail)
  120. [SimplifyCFG] NFC: Rename static methods to clang-tidy standards. (detail)
  121. NFC: Migrate SimplifyCFG to work on InstructionCost (detail)
  122. Revert rGce587529ad8b5 - "[APFloat] multiplySignificand - pass IEEEFloat as const reference. NFCI." (detail)
  123. [clangd] references: decls of overrides of x are refs to x, not decls (detail)
  124. [OpenMP] Fix seg fault in libomptarget when using Info with multiple threads (detail)
  125. [clangd] Remove some obsolete options that are now always on (detail)
  126. [compiler-rt] Fix AArch64 out-of-line atomics visibility on mach-o (detail)
  127. [AMDGPU] Fix crash with sgpr spills to vgpr disabled (detail)
  128. [AMDGPU] Fix release build after 0397dca0. (detail)
  129. [flang] Fix parsing of WRITE(I+J) with more accurate look-ahead (detail)
  130. [ConstraintElimination] Negate IR condition directly. (detail)
  131. [RISCV] Optimize (srl (and X, 0xffff), C) -> (srli (slli X, 16), 16 + C). (detail)
  132. [GlobalISel] Make sure G_ASSERT_ZEXT's src ends up with the same rc as dst (detail)
  133. [RISCV] Update comment text from D95774. NFC (detail)
  134. [RISCV] Add scalable vector support for floating point FMA instructions (detail)
  135. [LTO] Move part of gold devirt test to v1.16 directory (detail)
  136. [AArch64][GlobalISel] Emit G_ASSERT_ZEXT in assignValueToReg (detail)
  137. [libc++] Implements concept default_initializable. (detail)
  138. [OpenMP] Disable tests if FileCheck is not available in in-tree building (detail)
  139. [X86][SSE] LowerScalarImmediateShift - use APInt::getLowBitsSet for vXi8 ISD::SRL mask generation. NFCI. (detail)
  140. [X86][AVX] Add 'OK' tests cases for PR48877 (detail)
  141. [flang][directive] Enforce basic semantic check for all clauses (detail)
  142. [NFC] Disallow unused prefixes under clang/test/Driver (detail)
  143. [libc] Small adjustments to fenv tests (detail)
  144. [InstCombine] add tests for min/max with extend and constant operand; NFC (detail)
  145. [InstCombine] try to narrow min/max intrinsics with constant operand (detail)
  146. [clang][PATCH] Fix bug 48848 assertion related to recoverFromMSUnqualifiedLookup (detail)
  147. [elfabi] Fix tests which failed on different timezones (detail)
  148. [X86] Accept 64-bit GPRs for vextractps when using a register that requires EVEX. (detail)
  149. [NewPM][Unswitch] Add option to disable -O3 non-trivial unswitching (detail)
  150. [ARM] Flatten identity shuffles through vqdmulh nodes (detail)
  151. [libc][NFC] Add a death test API adaptation macro (detail)
  152. [libc++][format] Add basic_format_parse_context. (detail)
  153. [ConstantHoisting] Fix bug where constant materialization could insert into EH pad (detail)
  154. [WebAssembly] fixed wasm64 data segment init exp not 64-bit (detail)
  155. [mlir][Linalg] Replace SimplePad with PadTensor in hoist-padding (detail)
  156. [Loads] Plumb through TLI argument [NFC] (detail)
  157. [OpenMP] libomp: implement nteams-var and teams-thread-limit-var ICVs (detail)
  158. [mlir][spirv] Add support for OpImageType (detail)
  159. Revert "[mlir][spirv] Add support for OpImageType" (detail)
  160. [ConstraintElimination] Add support for EQ predicates. (detail)
  161. Avoid string comparisons on the fast path of MLIR Identifier lookup (NFC) (detail)
  162. Revert "[libc++][format] Add basic_format_parse_context." (detail)
  163. [LoopVectorize] improve IR fast-math-flags propagation in reductions (detail)
  164. NFC: Re-generate out-of-date matchers docs (detail)
  165. [tests] highlight cornercase w/deref hoisting from D95815 (detail)
  166. [mlir][spirv] Add support for OpImageType (detail)
  167. [clang-format] Add option to control the spaces in a line comment (detail)
  168. [CSSPGO] Tweaking inlining with pseudo probes. (detail)
  169. [AMDGPU] Set s-memtime-inst feature from clang (detail)
  170. [llvm-objcopy] -O binary: consider SHT_NOBITS sections to be empty (detail)
  171. [InstrProfiling] Use !associated metadata for counters, data and values (detail)
  172. [NFC][X86] Avoid redundant work inspecting callee (detail)
  173. [NFC][X86] Use CallBase interface to simplify code (detail)
  174. [flang][NFC] Rename complex type. (detail)
  175. [obj2yaml, yaml2obj] Use Hex64 for BBAddressMap fields. (detail)
  176. [lldb] Use the host architecture in TestAppleSimulatorOSType.py (detail)
  177. [x86] introduce no_callee_saved_registers attribute (detail)
  178. [OpenMP][Libomptarget] Remove possible harmful copy constructor call for RTLsTy (detail)
  179. [Utils] Add a switch controlling prefix warnings in UpdateTestChecks (detail)
  180. [mlir] Add custom directive hooks for printing mixed integer or value operands. (detail)
  181. AMDGPU: Fix dbg_value handling when forming soft clause bundles (detail)
  182. [LoopVectorize] Relax a FCmpInst assert to dyn_cast after D95690 (detail)
  183. [test] Fix unuses FileCheck prefixes in clang/test/Modules (detail)
  184. [test] Fix unuses FileCheck prefixes in lld (detail)
  185. [llvm] Use pop_back_val (NFC) (detail)
  186. [TableGen] Use ListSeparator (NFC) (detail)
  187. [TableGen] Use range-based for loops (NFC) (detail)
  188. [VFS] Add support to RedirectingFileSystem for mapping a virtual directory to one in the external FS. (detail)
  189. clang-extra: fix incorrect use of std::lock_guard by adding variable name (identified by MSVC [[nodiscard]] error) (detail)
  190. [test] Fix unused FileCheck prefixes in compiler-rt/test (detail)
  191. [AArch64] Homogeneous Prolog and Epilog Size Optimization (detail)
  192. [MLIR][TOSA] Comparison based elementwise operations for tosa-to-linalg (detail)
  193. [test] Fix unused FileCheck prefixes in lldb (detail)
  194. [test] Default clang/test to FileCheck --allow-unused-prefixes=false (detail)
  195. [SCEV] Apply loop guards to divisibility tests (detail)
  196. [test] Fix unused FileCheck prefixes in compiler-rt (detail)
  197. [ORC] Clear unused materializing info entries. (detail)
  198. [test] Fix unused FileCheck prefixes in clang-tidy and one llvm/test/Reduce test (detail)
  199. [test] Fix unused FileCheck prefixes in test/Reduce (detail)
  200. Revert "[AArch64] Homogeneous Prolog and Epilog Size Optimization" (detail)
  201. [NFC][Docs] Fix RAVFrontendAction doc's CMakelists.txt for Shared build (detail)
  202. [SelectionDAG] Prevent scalable vector warning from ComputeNumSignBits on extract_vector_elt on a scalable vector. (detail)
  203. [mlir][Standard] Extend n-D vector lowering to LLVM to [s|z]exti ops. (detail)
  204. [CSSPGO] Call site prioritized inlining for sample PGO (detail)
  205. [RISCV] Replace NoX0 SDNodeXForm with a ComplexPattern to do the selection of the VL operand. (detail)
  206. Added clamp i64 to i16 global isel pattern. (detail)
  207. clang-format (detail)
  208. Added comments. (detail)
  209. Added early exit. (detail)
  210. Renames (detail)
  211. Refactored the pattern matching. (detail)
  212. Fixed the lit tests and a bug in the implementation. (detail)
  213. Reverted unintended git-format change. (detail)
  214. Renamed identifiers in lit (detail)
  215. Move Combiner to PreLegalize step (detail)
  216. Move step to PreLegalizer (detail)
  217. Fixed tests. (detail)
  218. Code changes yielded from review. (detail)
  219. Resolve formatting changes. (detail)
  220. Updating formatting changes. (detail)
  221. Formatting changes. (detail)
  222. Formatting changes (detail)
  223. Added and used new target pseudo for v_cvt_pk_i16_i32, changes due to code review. (detail)
  224. Implemented a MED3_S32 GIR opcode. (detail)
  225. Removed the generic virtual register creations. Reworked the tests. (detail)
  226. Renamed med3 opcode, removed superfluous copy. (detail)
  227. Added missing includes. (detail)
  228. Reverted whitespace changes. (detail)
  229. Fixed includes. (detail)
  230. Fixed includes. (detail)
  231. Removed Diff file. (detail)
  232. [CSSPGO] Factor out common part for CSSPGO inline and AFDO inline (detail)
  233. Fix build break from D95024 (detail)
  234. [AMDGPU] Clarify calling conv about inactive lanes (detail)
  235. [AMDGPU] Mark epilog restores as frame-destroy (detail)
  236. [dllimport] Honor always_inline when deciding whether a dllimport function should be available for inlining (PR48925) (detail)
  237. [mlir] Keep track of region signature conversions as argument replacements (detail)
  238. Fold one-use variable into assert. NFCI. (detail)
  239. [SVE][LoopVectorize] Add masked load/store and gather/scatter support for SVE (detail)
  240. [flang][driver] Disallow non-existent input files in the frontend driver (detail)
  241. [ARM] Regenerate LowOverheadLoops mir tests. NFC (detail)
  242. Revert "Fix build break from D95024" (detail)
  243. Revert "[CSSPGO] Factor out common part for CSSPGO inline and AFDO inline" (detail)
  244. [ARM] Remove DLS lr, lr (detail)
  245. [OpenCL] Change extension handling for -fdeclare-opencl-builtins (detail)
  246. [X86][AVX] Add missing VEX_WIG tags from VPACKUSDW/VPHSUBD/VPCMPISTRI/VPCMPISTRM/VPCMPESTRI/VPCMPESTRM (detail)
  247. [mlir][Linalg] Refactor Linalg vectorization for better reuse and extensibility. (detail)
  248. [X86] Fix disassembly of x86-64 GDTLS code sequence (detail)
  249. [X86][AVX512] Support variable-index vector insertion on AVX512 targets (PR47924) (detail)
  250. [AMDGPU][MC] Corrected parsing of optional modifiers (detail)
  251. [mlir][Linalg] Fix unused variable warning in Release builds. NFC. (detail)
  252. [clangd] Report only decl of overridding method in xref. (detail)
  253. Revert "[test] Default clang/test to FileCheck --allow-unused-prefixes=false" (detail)
  254. [mlir][Linalg] Fix and properly test CodegenStrategy API (detail)
  255. [ASTMatchers] Fix matching after generic top-level matcher (detail)
  256. NFC: Migrate SimpleLoopUnswitch to work on InstructionCost (detail)
  257. NFC: Migrate SpeculateAroundPHIs to work on InstructionCost (detail)
  258. Revert "[lldb] Use current execution context in SBDebugger" (detail)
  259. [X86][SSE] LowerINSERT_VECTOR_ELT - pull out repeated EltSizeInBits calls. NFCI. (detail)
  260. [ARM] Select VINS from vector inserts (detail)
  261. [ASTMatchers] Add matchers for decomposition decls (detail)
  262. [clang-format] Add case aware include sorting. (detail)
  263. [ARM] Extra shuffle tests. NFC (detail)
  264. Revert "[InstrProfiling] Use !associated metadata for counters, data and values" (detail)
  265. [ASTMatchers] Ignore parts of BindingDecls which are not spelled in source (detail)
  266. [clangd] Fix race in Global CDB shutdown (detail)
  267. [AVR][clang] Fix a bug in AVR toolchain search paths (detail)
  268. [NFC][InstCombine] Add tests for  (~x) a>> y  -->  ~(x a>> y)  fold (PR48995) (detail)
  269. [InstCombine] Host inversion out of ashr's value operand (PR48995) (detail)
  270. [mlir] Put template specialization in the same namespace (detail)
  271. [clang][PATCH][NFC] Correct test case related to review D95482 (detail)
  272. [OpenCL] Add diagnostics for references to functions (detail)
  273. Ensure that the matcher is instantiated (detail)
  274. [ARM] Add MVE insert-of-extract pattern (detail)
  275. [PowerPC] Materialize 34 bit constants with pli on Power 10. (detail)
  276. [OpenMP][NFC] Adding FAQ Entry for errors with static libraries (detail)
  277. [CSSPGO] Factor out common part for CSSPGO inline and AFDO inline (detail)
  278. [AIX] Improve option processing for mabi=vec-extabi and mabi=vec=defaul (detail)
  279. [mlir] Fix scf.for single iteration canonicalization check (detail)
  280. [mlir][spirv] Define sp.VectorShuffle (detail)
  281. Revert "[mlir] Fix scf.for single iteration canonicalization check" (detail)
  282. [AMDGPU] Save fp/bp after csr saves (detail)
  283. [AMDGPU] Remove unused tmp register (detail)
  284. Fixed failing OpenCL test (detail)
  285. [ARM] Mark MVE_VMOV_to_lane_32 as isInsertSubregLike (detail)
  286. [ConstraintElimination] Add nicer way to dump constraints (NFC). (detail)
  287. [flang][NFC] Update #include and comment. (detail)
  288. [OpenMP] Fix sign comparison warnings from GCC (detail)
  289. [ARM] Correct some tablegen operand types. NFC (detail)
  290. [InlineFunction] Only update noalias scopes once for an instruction. (detail)
  291. [ExpandReductions] add test for fmin with FMF; NFC (detail)
  292. [yaml2obj/obj2yaml/llvm-readobj] Support SHF_GNU_RETAIN (detail)
  293. [MC] Support SHF_GNU_RETAIN as section flag 'R' (detail)
  294. [test] Add basic _Unwind_ForcedUnwind + exception tests (detail)
  295. [ConstraintElimination] Add test with pointer bitcast. (detail)
  296. [MC] Upgrade DWARF version to 5 upon .file 0 (detail)
  297. [debugserver] Fix -Winconsistent-missing-override warnings on arm64 (detail)
  298. [CSSPGO] Passing the clang driver switch -fpseudo-probe-for-profiling to the linker. (detail)
  299. [ELF] Delete unused --warn-ifunc-textrel (detail)
  300. [libcxx] Implement the space function for windows (detail)
  301. [lld-macho][NFC] refactor relocation handling (detail)
  302. [OpenMP] Fix iterations calculation for dependent counters. (detail)
  303. [debugserver] Fix -Wunused-function warnings on arm64 (detail)
  304. [RISCV] Use a ComplexPattern to merge isel patterns for vector load/store with GPR and FrameIndex addresses. (detail)
  305. [ConstraintElimination] Fix -Wunused-function in -DLLVM_ENABLE_ASSERTIONS=off build (detail)
  306. [debugserver] Fix -Wsign-compare warnings on arm64 (detail)
  307. [mlir][NFC] Use explicit `mlir` namespace in generated code (detail)
  308. [mlir][NFC] Add missing include guards to MlirOptMain.h (detail)
  309. [mlir] Allow to use constant lambda as callbacks for `TypeConverter` (detail)
  310. [mlir] Print more verbose message in case of type inference error (detail)
  311. [debugserver] Fix -Wmissing-field-initializers warnings on arm64 (detail)
  312. [mlir] Return new Operation from `Rewriter::replaceOpWithNewOp` (detail)
  313. [RISCV] Correct types in tablegen multiclasses found by D95874. (detail)
  314. [RISCV] Fix incorrect RVV sdiv/udiv lowering (detail)
  315. Fix CMake LLVM_TARGETS_TO_BUILD "Native" option to work with JIT (detail)
  316. [mlir] Delay adding the __resume function (detail)
  317. [MLIR] [CMake] Support building MLIR standalone (detail)
  318. Fix namespace for MLIR Async Runtime (detail)
  319. [test] Default clang/test to FileCheck --allow-unused-prefixes=false (detail)
  320. [flang] Detect UBOUND() error on assumed-size array (detail)
  321. [flang] Implement IEEE_SUPPORT_xxx inquiry functions (detail)
  322. Ensure that InstructionCost actually implements a total ordering (detail)
  323. [flang][NFC] Update #include. (detail)
  324. [CSSPGO] Introducing distribution factor for pseudo probe. (detail)
  325. [flang][NFC] Update the CG rewrite pass. (detail)
  326. [mlir][Pattern] Create a new IRRewriter class to enable sharing code with pattern rewrites (detail)
  327. Revert "Ensure that InstructionCost actually implements a total ordering" (detail)
  328. [lldb] Convert assertTrue(a == b) to assertEqual(a, b) (detail)
  329. Disable CFI in __get_elem to allow casting a pointer to uninitialized memory (detail)
  330. Revert "Fix namespace for MLIR Async Runtime" (detail)
  331. [libc++] Add new queues for specific macOS system versions (detail)
  332. [libcxx] Implement the current_path function for windows (detail)
  333. [libcxx] Implement the canonical function for windows (detail)
  334. [ConstraintElimination] Skip pointer casts. (detail)
  335. [dfsan] Clean TLS after sigaction callbacks (detail)
  336. Diagnose if a SLEB128 is too large to fit in an int64_t. (detail)
  337. [GWP-ASan] Add aligned allocations. (detail)
  338. [GWP-ASan] Add locking around unwinder for atfork protection. (detail)
  339. Revert "[GWP-ASan] Add aligned allocations." (detail)
  340. [AArch64] Homogeneous Prolog and Epilog Size Optimization (detail)
  341. [gn build] Port 4f58b1bd2974 (detail)
  342. [test][NewPM] Pin various tests to the legacy PM (detail)
  343. [flang] Add TypeAndShape::MeasureElementSizeInBytes() (detail)
  344. [FileCheck] Make --allow-unused-prefixes cl::ZeroOrMore (detail)
  345. [mlir][Vector] Add lowering to LLVM for vector.bitcast (detail)
  346. [mlir] Fix scf.for single iteration canonicalization check (detail)
  347. [libc] Add hardware implementations of ceil and ceilf for aarch64. (detail)
  348. [GlobalISel] Check if branches use the same MBB in matchOptBrCondByInvertingCond (detail)
  349. AMDGPU: Don't consider global pressure when bundling soft clauses (detail)
  350. AMDGPU: Fix adding extra operands for i128 asm constraints (detail)
  351. [lldb/test] Stop recording fundamental return types (NFC) (detail)
  352. [PowerPC] [NFC] fix wording typos (detail)
  353. [VFS] Fix Wreturn-type gcc warning (NFC) (detail)
  354. [yaml2obj][wasm] Support numeric relocation type (detail)
  355. [clang-format] Fix MSVC "unsafe use of type 'bool' in operation" warning (NFC) (detail)
  356. [CSSPGO] Fix MSVC initializing truncation warning (NFC) (detail)
  357. [Statepoint] Handle 'undef' operand tied to def (detail)
  358. [clang-tidy] Fix crash in readability-identifier-naming check (detail)
  359. [dfsan] Wrap memmove (detail)
  360. [NFC] Fix typo in comment (detail)
  361. [Docs] Update HowToSubmitABug (detail)
  362. [RISCV] Add new vector instructions in v0.10. (detail)
  363. [OpenMP] Add OpenMP offloading toolchain for AMDGPU (detail)
  364. [RISCV] Load/store vector mask types. (detail)
  365. [gn build] Port fcf03e728007 (detail)
  366. [CodeGen] Drop unnecessary const from return types (NFC) (detail)
  367. [Transforms/Utils] Use range-based for loops (NFC) (detail)
  368. [AsmPrinter] Use ListSeparator (NFC) (detail)
  369. [InstrProfiling] Use !associated metadata for counters, data and values (detail)
  370. [libcxx] Implement the permissions function for windows (detail)
  371. [libcxx] Implement the read_symlink function for windows (detail)
  372. [clang-include-fixer] Pre-reserve vector size. NFC (detail)
  373. [X86] Correct types in tablegen multiclasses found by D95874. (detail)
  374. [VPlan][NFC] Introduce constructors for VPIteration (detail)
  375. [LIBC] Add optimized memcpy routine for AArch64 (detail)
  376. [AArch64][SVE]Add cost model for broadcast shuffle (detail)
  377. Revert "[AMDGPU] Add a new Clamp Pattern to the GlobalISel Path." (detail)
  378. [clang][AVR] Improve avr-ld command line options (detail)
  379. [OpenCL] Add cl_khr_subgroup_ballot to TableGen BIFs (detail)
  380. [OpenCL] Add cl_khr_subgroup_non_uniform_vote to TableGen BIFs (detail)
  381. [Flang][OpenMP] Add semantic checks for invalid branches into or out of OpenMP constructs (detail)
  382. [clangd] Report xref for base methods. (detail)
  383. [clang][AVR][NFC] Fix a typo (detail)
  384. Fix MLIR Async Runtime DLL on Windows (detail)
  385. [OpenCL] Fix address space in binding of initializer lists to referencs (detail)
  386. [Sema] Fix -Warray-bounds false negative when casting an out-of-bounds array item (detail)
  387. [OpenCL][Docs] Fix command line flag in the example. (detail)
  388. [X86][SSE] Support variable-index float/double vector insertion on SSE41+ targets (PR47924) (detail)
  389. [LoopVectorize] add test for fake min/max; NFC (detail)
  390. AMDGPU: Move handling of allocation of fixed ABI inputs (detail)
  391. [test] Use host platform specific error message substitution in lit tests - continued (detail)
  392. Revert "[ConstantFold] Fold more operations to poison" (detail)
  393. [AMDGPU] Fix multiclass template parameter types. NFC. (detail)
  394. [ConstraintElimination] Add some tests with conds in loop header. (detail)
  395. [FPEnv][X86] Platform builtins edition: clang should get from the AST the metadata for constrained FP builtins (detail)
  396. [flang] Add -fsyntax-only to f18; retain -fparse-only synonym (detail)
  397. [Hexagon] Add ELF flags for Hexagon V68 (detail)
  398. [dfsan] Clean TLS after signal callbacks (detail)
  399. [flang][NFCI] Relpace `LoopOp` Op with `DoLoopOp` Op in FIR Dialect (detail)
  400. [DWARF] Location-less inlined variables should not have DW_TAG_variable (detail)
  401. [VPlan] Manage induction value creation using VPValues. (detail)
  402. Revert "[DWARF] Location-less inlined variables should not have DW_TAG_variable" (detail)
  403. [OpenMP] Fixed an issue that taskwait doesn't work on detachable task (detail)
  404. [flang] Enable FIR types and Dialect round trip tests (detail)
  405. [DAGCombiner] Remove (sra (shl X, C), C) if X has more than C sign bits. (detail)
  406. [AMDGPU] Added -mcpu to couple more tests. NFC. (detail)
  407. Revert "AMDGPU: Don't consider global pressure when bundling soft clauses" (detail)
  408. [scudo/standalone] Use zx_system_get_page_size() on Fuchsia (detail)
  409. [gwp_asan] Use zx_system_get_page_size() on Fuchsia (detail)
  410. [lldb] Fix debugserver-entitlements.plist path (detail)
  411. RegisterCoalescer: Prune undef subranges from copy pairs in loops (detail)
  412. [lld-macho] Associate each Symbol with an InputFile (detail)
  413. [lld-macho] Force-loading should share code path with regular archive loads (detail)
  414. [lld-macho] Remove stray ehFrame change (detail)
  415. [lld-macho] Rename VERSION CONTROL to VERSION TARGETING in helptext (detail)
  416. [sanitizer_common] Use zx_system_get_page_size() on Fuchsia (detail)
  417. [libc][NFC] Move generic math implementations to the generic directory. (detail)
  418. [dfsan] Test IGN and DFL for sigaction (detail)
  419. [dfsan] Wrap strcat (detail)
  420. [flang] Fix calls to LBOUND() intrinsic for arrays with lower bounds not 1 (detail)
  421. RegisterCoalescer: Fix not setting undef on coalesced subregister uses (detail)
  422. Set GPU context before {cu,hip}MemHostRegister. (detail)
  423. [GlobalISel] Combine narrowScalar of G_ADD and G_SUB. NFC (detail)
  424. Delete CUDA context after linking device code. (detail)
  425. [NFC][CUDA] Refactor registering device variable (detail)
  426. [SampleFDO][NFC] Detach SampleProfileLoader from SampleCoverageTracker (detail)
  427. [clang-tblgen] AnnotateAttr::printPretty has spurious comma when no variadic argument is specified (detail)
  428. [Hexagon] Add LLVM instruction definitions for Hexagon V68 (detail)
  429. [mlir] Add gpu async integration test. (detail)
  430. Fix overflowing signed left shift, found by ubsan buildbot. (detail)
  431. [asan] Fix pthread_create interceptor (detail)
  432. [NewPM][HelloWorld] Move HelloWorld to Utils (detail)
  433. [lld-macho] Try to fix Windows build (detail)
  434. Add API for adding arguments to blocks (detail)
  435. [libc++] Rationalize our treatment of contiguous iterators and __unwrap_iter(). (detail)
  436. [Coverage] Propogate counter to condition of conditional operator (detail)
  437. [lldb] Honor the CPU type & subtype when launching on macOS (detail)
  438. [lldb] Check for both Lua 5.3 and 5.4 error messages in the tests. (detail)
  439. [GlobalISel] Add sext(constant) -> constant artifact combine. (detail)
  440. PR49020: Diagnose brace elision in designated initializers in C++. (detail)
  441. Turn on the new pass manager by default (detail)
  442. Revert "[LTOCodeGenerator] Use lto::Config for options (NFC)." (detail)
  443. Revert "[LTO] Add option enable NewPM with LTOCodeGenerator." (detail)
  444. Revert "[LTO] Use lto::backend for code generation." (detail)
  445. PR44325 (and duplicates): don't issue -Wzero-as-null-pointer-constant (detail)
  446. [clang-tidy] Use new mapping matchers (detail)
  447. [Docs] Add some documentation for constructor homing, a debug info optimization (-fuse-ctor-homing) (detail)
  448. [libc++] [P0879] constexpr std::sort (detail)
  449. [AArch64][GlobalISel] Emit G_ASSERT_ZEXT in assignValueToAddress for ZExt params (detail)
  450. Revert "[InstrProfiling] Use !associated metadata for counters, data and values" (detail)
  451. [AArch64][GlobalISel] Change store value type from p0 -> s64 to import patterns (detail)
  452. [CSSPGO][llvm-profgen] Pseudo probe based CS profile generation (detail)
  453. [lldb] Rollback to using i386 for the watch simulator (detail)
  454. [OpenMPIRBuilder] Implement collapseLoops. (detail)
  455. [OpenMP][NVPTX] Take functions in `deviceRTLs` as `convergent` (detail)
  456. Make the folder more robust against op fold() methods that generate a type mismatch (detail)
  457. [MLIR] Fix building unittests in in-tree build (detail)
  458. [CSSPGO][llvm-profgen] Compress recursive cycles in calling context (detail)
  459. [CSSPGO][llvm-profgen] Aggregate samples on call frame trie to speed up profile generation (detail)
  460. [test] Pin spir-codegen.ll to legacy PM (detail)
  461. Fix the guaranteed alignment of memory returned by malloc/new on Darwin (detail)
  462. [Support] Drop unnecessary const from return types (NFC) (detail)
  463. [TableGen] Use ListSeparator (NFC) (detail)
  464. [Transforms/IPO] Use range-based for loops (NFC) (detail)
  465. [NFC][Coroutine] Remove redundant comment (detail)
  466. [lldb] Convert more assertTrue to assertEqual (NFC) (detail)
  467. [NFC] Fix the noprofile attribute comment (detail)
  468. [ASTReader] Always rebuild a cached module that has errors (detail)
  469. Revert "[CSSPGO][llvm-profgen] Aggregate samples on call frame trie to speed up profile generation" (detail)
  470. Revert "[CSSPGO][llvm-profgen] Compress recursive cycles in calling context" (detail)
  471. [CSSPGO][llvm-profgen] Compress recursive cycles in calling context (detail)
  472. [hip][cuda] Enable extended lambda support on Windows. (detail)
  473. [mlir][Linalg] NFC - Extract a standalone LinalgInterfaces (detail)
  474. Fix miscompile when performing template instantiation of non-dependent (detail)
  475. Give this test a target triple. (detail)
  476. [mlir][Linalg] Generalize the definition of a Linalg contraction. (detail)
  477. [Syntax] Support condition for IfStmt. (detail)
  478. [clangd] Detect rename conflicits within enclosing scope (detail)
  479. [AMDGPU] Save all lanes for reserved VGPRs (detail)
  480. [Support] Indent multi-line descr of enum cli options. (detail)
  481. [clang][cli] Command line round-trip for HeaderSearch options (detail)
  482. [clang-cl] Remove the /fallback option (detail)
  483. [mlir] Apply source materialization in case of transitive conversion (detail)
  484. [lldb] Make TestLocalVariables.py compatible with the new pass manager (detail)
  485. [mlir] make vector to llvm conversion truly partial (detail)
  486. [ARM] Handle f16 in GeneratePerfectShuffle (detail)
  487. Re-land D94976 after revert in e29552c5aff6 (detail)
  488. [mlir][Linalg] Drop SliceOp (detail)
  489. [ElementCount] NFC: Set 'const' qualifier for getWithIncrement/Decrement. (detail)
  490. [gn build] (manually) port 0609f257dc2e2c3 (detail)
  491. Revert "[hip][cuda] Enable extended lambda support on Windows." (detail)
  492. [X86] Remove stale TODO comment. NFC. (detail)
  493. [flang][driver] Add forced form flags and -ffixed-line-length (detail)
  494. [clang] Add AddClang.cmake to the list of the CMake modules that are installed (detail)
  495. [AVR] Remove an assertion that causes generic CodeGen tests to fail (detail)
  496. AMDGPU: Add support for amdgpu-unsafe-fp-atomics attribute (detail)
  497. [mlir] Return scf.parallel ops resulted from tiling. (detail)
  498. [OpenCL] Fix default address space in template argument deduction. (detail)
  499. [ConstraintElimination] Support conditions from loop preheaders (detail)
  500. [OpenCL][Docs] Link page explaining tooling for offline compilation. (detail)
  501. NFC: Migrate LoopUnrollPass to work on InstructionCost (detail)
  502. [flang][driver] Add PrescanAction frontend action (nfc) (detail)
  503. [Dexter] Avoid infinite loop in dbgeng driver (detail)
  504. [AVR] Fix 14 Generic CodeGen tests by making address space explicit or optional (detail)
  505. [AVR] Add 'XFAIL' to the remaining failing Generic CodeGen tests for AVR (detail)
  506. [X86] Use VT::changeVectorElementType helper where possible. NFCI. (detail)
  507. [libc++] Rename include/support to include/__support (detail)
  508. [AVR] Fix up a few accidentally-regressed Generic CodeGen tests recently broken (detail)
  509. [InstCombine] add tests for demanded/known bits of shifted constant; NFC (detail)
  510. [Hexagon] Add clang builtin definitions for Hexagon V68 (detail)
  511. [libc++] Adds a make_string test helper function. (detail)
  512. [Hexagon] Add -mv68 option to driver (detail)
  513. [flang][driver] Add support for `-J/-module-dir` (detail)
  514. [clang][Arm] Fix handling of -Wa,-march= (detail)
  515. [libc++] Fix libcxx build on 32bit architectures with 64bit time_t defaults e.g. riscv32 (detail)
  516. [libc++] Make feature-test macros consistent with availability macros (detail)
  517. [CSSPGO][llvm-profgen] Aggregate samples on call frame trie to speed up profile generation (detail)
  518. [libc++] Remove the ability to use braced-init for filesystem paths (detail)
  519. [flang][driver] Fix reading from stdin when using `-test-io` (detail)
  520. [lldb] Simplify the logic to detect compiler flag support (detail)
  521. [mlir][Linalg] Introduce a ContractionOpInterface (detail)
  522. [AMDGPU][GlobalISel] Use scalar min/max instructions (detail)
  523. [AMDGPU][GlobalISel] Fix v2s16 right shifts (detail)
  524. [mlir][ODS] Small fixes for ODS classes (detail)
  525. [llvm-objdump] --source: drop the warning when there is no debug info (detail)
  526. [mlir][ODS] Add explicit namespace to `ViewLikeInterface` definition (detail)
  527. [examples] Fix Target does not support MC emission in ParallelJIT (detail)
  528. [ELF] Allow R_386_GOTOFF from .debug_info (detail)
  529. [ELF] Make SHF_GNU_RETAIN sections GC roots (detail)
  530. [mlir][ODS] Use StringLiteral instead of StringRef when applicable (detail)
  531. Remove dead code from Linalg vectorization to fix GCC warning (NFC) (detail)
  532. [OpenMP][deviceRTLs] Drop `assert` in common parts of `deviceRTLs` (detail)
  533. [PS4] Allow triple to reflect the new company name. (detail)
  534. [AMDGPU] Add f16 to i1 CodeGen patterns. (detail)
  535. [NFC][llvm-mca] Fix compiler warning (detail)
  536. Reland "Ensure that InstructionCost actually implements a total ordering" (detail)
  537. [ExpandReductions] fix FMF requirement for fmin/fmax (detail)
  538. [mlir] Silence GCC warnings (detail)
  539. Remove overzealous verifier check on DW_OP_LLVM_entry_value and improve the documentation (detail)
  540. Revert "[mlir][ODS] Use StringLiteral instead of StringRef when applicable" (detail)
  541. [lldb-vscode] correctly use Windows macros (detail)
  542. [CSSPGO][llvm-profgen] Merge and trim profile for cold context to reduce profile size (detail)
  543. Print the "no plugin" warning only when there is no plugin (detail)
  544. [MemorySSA] Don't treat lifetime.end as NoAlias (detail)
  545. [libc++] Add format check to CI (detail)
  546. [TargetLowering] Use LegalOnly operand to isOperationLegalOrCustom to simplify some code. NFC (detail)
  547. [clangd] Improve name conflict detection (detail)
  548. [flang][NFC] Rename IntType to IntegerType. (detail)
  549. Correct some confused diagnostic terminology (detail)
  550. [ARM] Do not emit ldrexd/strexd on Cortex-M chips (detail)
  551. Fix miscomputation of dependence for elaborated types that are (detail)
  552. [CodeComplete] Guess type for designated initializers (detail)
  553. [HIP] Add __managed__ macro to header (detail)
  554. [mlir] Small touchups to LangRef attribute section (detail)
  555. [sanitizer] Add suffix to report file name (detail)
  556. Revert "[BuildLibcalls, Attrs] Support more variants of C++'s new, add attributes for C++'s delete" (detail)
  557. Don't infer attributes on '::operator new'. (detail)
  558. Revert "[libc++] [P0879] constexpr std::nth_element, and rewrite its tests." (detail)
  559. [flang][msvc] Reapply "Explicitly reference "this" inside closure" (detail)
  560. [flang] Remove dims type and gendims op. (detail)
  561. [CUDA][HIP] Fix checking dependent initalizer (detail)
  562. [mlir] Mark LogicalResult as LLVM_NODISCARD (detail)
  563. [asan] Add %d variable to external_symbolizer_path option, so that user can specify paths relative to the location of the binary. (detail)
  564. Fix a test case after committing D94563. (detail)
  565. [CSSPGO][llvm-profgen] Fix bug with parsing hybrid sample trace line (detail)
  566. Fix test case from D94563. (detail)
  567. [RISCV] Add i8/i16 test cases to div.ll and i8/i16/i64 to rem.ll. NFC (detail)
  568. fix -Werror build issue. (detail)
  569. [AST] Update LVal before evaluating lambda decl fields. (detail)
  570. [OpenMP][libomptarget] Fixed an issue that device sync is skipped if the kernel doesn't have any argument (detail)
  571. [LV] Unconditionally branch from middle to scalar preheader if the scalar loop must execute (detail)
  572. DebugInfo: Temporarily work around -gsplit-dwarf + LTO .debug_gnu_pubnames regression after D94976 (detail)
  573. [sanitizer] Fix the test on android (detail)
  574. Add missing test update from 3e5ce49 (detail)
  575. [WebAssembly] Support single-floating-point immediate value (detail)
  576. [WebAssembly] Use single-threaded mode when -matomics isn't enabled. (detail)
  577. [TargetLowering] Use Align in allowsMisalignedMemoryAccesses. (detail)
  578. [X86] Generate unaligned access for fixed slots in unaligned stack (detail)
  579. [RISCV] Split vrgather intrinsics into separate vrgather.vv and vrgather.vx intrinsics. (detail)
  580. [RISCV] Use LLVMScalarOrSameVectorWidth to make avoid needing to mention the index type for vrgatherei16 intrinsics. (detail)
  581. [ARM][WebAssembly] Fix incorrect MCOperand::createDFPImm after D96091 (detail)
  582. [MC] Add createFPImm/isFPImm/setFPImm to smooth migration from FPImm to DFPImm after D96091 (detail)
  583. [VE] Fix allowsMisalignedMemoryAccesses after D96097 (detail)
  584. [MC] Add isFPImm after D96091 (detail)
  585. [Darwin] Switch to new logging api for sanitizers (detail)
  586. LLVMgold.so: Fix tests after D95380 (detail)
  587. [IR] Drop unnecessary const from return types (NFC) (detail)
  588. [GlobalISel] Use ListSeparator (NFC) (detail)
  589. [Transforms/Scalar] Use range-based for loops (NFC) (detail)
  590. [flang] Adapt mlir based error status handling in `tco` tool (detail)
  591. [Test] Add more tests demonstrating oddities in behavior of LSR (detail)
  592. [lldb/test] Reduce API test tools configuration boilerplate (detail)
  593. [lldb/cmake] Reduce duplication in generation lldb-dotest (detail)
  594. [OpenCL] Introduce new language options for OpenCL keywords. (detail)
  595. [mlir][linalg] Linalg.fill on tensor should not have side-effects (detail)
  596. [llvm-objdump] Fix missing first line of license in header file (detail)
  597. [lldb][NFC] Minor comment and inlining fixes for Args (detail)
  598. [CostModel] Remove VF from IntrinsicCostAttributes (detail)
  599. [RISCV] Add patterns for scalable-vector fsqrt (detail)
  600. [RISCV] Optimize sign-extended EXTRACT_VECTOR_ELT nodes (detail)
  601. [RISCV] Support scalable-vector integer reduction intrinsics (detail)
  602. [NFC] inline variable (detail)
  603. [NFC] Simplify expression (detail)
  604. [clangd] Take into account what is in the index (symbols, references, etc.) at indexes merge (detail)
  605. [AArch64][SVE] Coalesce ptrue instrinsic calls where possible (detail)
  606. DWARFDebugFrame.h - fix Wdocumentation warning. NFCI. (detail)
  607. TypeMismatchCheck.cpp - fix Wdocumentation warning. NFCI. (detail)
  608. CodeGenPassBuilder.h - fix Wdocumentation warning. NFCI. (detail)
  609. IROutliner.cpp - fix Wdocumentation warnings. NFCI. (detail)
  610. SampleProfile.cpp - fix Wdocumentation warning. NFCI. (detail)
  611. AArch64LowerHomogeneousPrologEpilog.cpp - fix Wdocumentation warning. NFCI. (detail)
  612. IROutliner.cpp - fix Wdocumentation warning. NFCI. (detail)
  613. [mlir][Linalg] Add option to anchor on func name in TestLinalgCodegenStrategy (detail)
  614. [Test] Add tests demonstrating missing case of usub with overflow usage (detail)
  615. Revert "[LV] Unconditionally branch from middle to scalar preheader if the scalar loop must execute" (detail)
  616. [mlir][Linalg] NFC - Refactor vectorization to be more composable (detail)
  617. [mlir][Linalg] Fix spurious test change (detail)
  618. Treat opencl_unroll_hint subject errors as semantic rather than parse errors (detail)
  619. [mlir] Add `const` qualifiers to `AffineMap` methods (detail)
  620. [LoopVectorize][ARM] Regenerate mve-gather-scatter-tailpred.ll test (detail)
  621. Revert "[PowerPC] [Clang] Enable float128 feature on P9 by default" (detail)
  622. [clangd] Deduplicate scopes in IncludeFixer queries (detail)
  623. [mlir] Turn Linalg to LLVM into a partial conversion (detail)
  624. [mlir][Linalg] NFC - Improve usage of mlir::linalg::isaContractionOpInterface (detail)
  625. [ObjC][ARC] Use operand bundle 'clang.arc.rv' instead of explicitly (detail)
  626. Revert "[ObjC][ARC] Use operand bundle 'clang.arc.rv' instead of explicitly" (detail)
  627. [mlir][vector] Add patterns to cast away leading 1-dim (detail)
  628. Add an API to simplify setting TraversalKind in clang-tidy matchers (detail)
  629. [clang-tidy] Simplify implementation of container-size-empty (detail)
  630. We can only move static allocas into the resume entry points (detail)
  631. [ObjC][ARC] Use operand bundle 'clang.arc.rv' instead of explicitly (detail)
  632. [mlir][spirv] Add more vector conversion patterns (detail)
  633. [mlir][vector] Add constant folding for fp16 to fp32 bitcast (detail)
  634. [OpenCL] Add cl_khr_subgroup_shuffle* to TableGen BIFs (detail)
  635. [RGT][ProfileData] Correct a test assertion (detail)
  636. [Test] Fix test case demonstrating oddities in behavior of LSR. (detail)
  637. [ASTMatchers] Fix segfault when Action is nullptr (detail)
  638. [OpenCL] Add macro definitions of OpenCL C 3.0 features (detail)
  639. LoopUnrollAndJam] Only allow loops with single exit(ing) blocks (detail)
  640. Recommit of a2fdf9d4d734732a6fa9288f1ffdf12bf8618123. (detail)
  641. Drop LLVM_i1, LLVM_i8, LLVM_i32, LLVM_IntBase types (detail)
  642. [lldb] Emit type annotation in SWIG generated Python code. (detail)
  643. [clangd] Trace queue state for each TUScheduler action. (detail)
  644. [MSP430] Fix multiclass template parameter types. NFC. (detail)
  645. Revert "Revert "[libc++] [P0879] constexpr std::nth_element, and rewrite its tests."" (detail)
  646. [gn build] Turn on new pass manager by default (detail)
  647. [AMDGPU] Add Fiji target in fptosi/fptoui instruction-select MIR tests. (detail)
  648. [ELF] Improve --icf=safe diagnostic (detail)
  649. [dfsan] Add thread registration (detail)
  650. [CodeView][DebugInfo] Update the code for removing template arguments from the display name of a codeview function id. (detail)
  651. [DAGCombiner][SVE] Fix invalid use of getVectorNumElements() in visitSRA. (detail)
  652. [DebugInfo] Fix error from D95893, where I accidentally used an (detail)
  653. Stop traping on sNaN in __builtin_isnan (detail)
  654. [libc] Add aarch64 flavors of floor, round, sqrt and trunc. (detail)
  655. [flang][NFC] Update comments. (detail)
  656. [flang][NFC] Make KindTy consistent and consistently used. (detail)
  657. [WebAssembly] locals can now be indirect in DWARF (detail)
  658. [Codegen][ReplaceWithVecLib] add pass to replace vector intrinsics with calls to vector library (detail)
  659. [Reassociate] add test for shl+or; NFC (detail)
  660. [PhaseOrdering] add test to show combined result of reassociate+instcombine+vectorizers; NFC (detail)
  661. [lic++][docs] Explain noexcept policy for narrow contracts. (detail)
  662. [NVPTX][NewPM] Temporarily disable NVPTX passes in new PM pipeline (detail)
  663. [RISCV] Correct types in tablegen multiclasses found by D95874. (detail)
  664. Revert "[Codegen][ReplaceWithVecLib] add pass to replace vector intrinsics with calls to vector library" (detail)
  665. [CUDA][HIP] Fix device variable linkage (detail)
  666. [libc++] Further improve the contiguous-iterator story, and fix some bugs. (detail)
  667. [flang][fir] Add FIR's vector type. (detail)
  668. [libcxx] Use the posix code for directory_entry::__do_refresh (detail)
  669. [libcxx] Implement temp_directory_path using GetTempPath on windows (detail)
  670. Various minor fixes for python 3 (detail)
  671. Don't emit coverage mapping for excluded functions (detail)
  672. Allow SmallPtrSet to be used with a std::insert_iterator (detail)
  673. [gn build] enable new pass manager more, follow-up to 39ceb5c9cfe (detail)
  674. [WebAssembly] Prevent data inside text sections in assembly (detail)
  675. [clang][emscripten] Add builtin define for __EMSCRIPTEN_PTHREADS__ (detail)
  676. [libc++abi] Disable _Unwind_ForcedUnwind + exception tests for ARM EHABI (detail)
  677. [mlir][vector] Add pattern to shuffle bitcast ops (detail)
  678. [libc][aarch64] Enable a bunch of math functions. (detail)
  679. [debugserver] Fix more warnings in DNBArchImplARM64 (detail)
  680. [lldb] Re-enable TestExprsChar on arm64 (detail)
  681. [WebAssembly] ensure .functype applies to right label in assembler (detail)
  682. [lld-macho] Fix alignment & layout to match ld64 and satisfy kernel & codesign (detail)
  683. SPARCv9: recognize SIR trap instruction (detail)
  684. Have stripDebugInfo() also strip !llvm.loop annotations from all (detail)
  685. Add a link to the LLVM Dev recording from the MLIR tutorial landing page (detail)
  686. Rework ExecutionEngine::invoke() to make it more friendly to use from C++ (detail)
  687. [CSSPGO] Use merged base profile for hot threshold calculation (detail)
  688. [AsmPrinter] __patchable_function_entries: Set SHF_LINK_ORDER for binutils 2.36 and above (detail)
  689. [AMDGPU] Drop unnecessary const from a return type (NFC) (detail)
  690. [llvm] Fix header guards (NFC) (detail)
  691. [Transforms/Instrumentation] Use range-based for loops (NFC) (detail)
  692. [ELF] --gc-sections: collect unused SHF_LINK_ORDER .gcc_except_table (detail)
  693. .gcc_except_table: Set SHF_LINK_ORDER if binutils>=2.36, and drop unneeded unique ID for -fno-unique-section-names (detail)
  694. [MLIR] [affine-loop-fusion] Fix a bug about non-result ops in affine-loop-fusion (detail)
  695. NFC: Migrate LoopFlatten to work on InstructionCost. (detail)
  696. [AArch64] Correct some tablegen operand types. NFC (detail)
  697. NFC: Migrate LoopIdiomRecognize to work on InstructionCost (detail)
  698. [WebAssembly] Fix call unwind mismatches (detail)
  699. [WebAssembly] Fix catch unwind mismatches (detail)
  700. [Hexagon] Fix multiclass template parameter types. NFC. (detail)
  701. [Sparc] Fix multiclass template parameter types. NFC. (detail)
  702. [PowerPC] Fix multiclass template parameter types. NFC. (detail)
  703. [X86][AVX] canonicalizeLaneShuffleWithRepeatedOps - merge VPERMILPD ops with different low/high masks. (detail)
  704. [WebAssembly] Handle EH terminate pads for cleanup (detail)
  705. [gn build] Port be0efa1f2368 (detail)
  706. [WebAssembly] Update InstPrinter and AsmParser for new EH instructions (detail)
  707. [LTT] Don't attempt to lower type tests used only by assumes (detail)
  708. [LSR] Add test for multi-edge latch (NFC) (detail)
  709. [AssumptionCache] Do not track llvm.assume calls (PR49043) (detail)
  710. [AssumptionCache] Avoid dangling llvm.assume calls in the cache (detail)
  711. [X86] Don't pass a 1 to the second argument of ISD::FP_ROUND in LowerFCOPYSIGN. (detail)
  712. [IR] Drop unnecessary const from return types (NFC) (detail)
  713. [TableGen] Use ListSeparator (NFC) (detail)
  714. [Analysis] Use range-based for loops (NFC) (detail)
  715. [ELF][test] Improve aarch64 tests (detail)
  716. [lldb] Use assertIn/NotIn over assertTrue/False (NFC) (detail)
  717. [Polly] Fix build after AssumptionCache change (D96168) (detail)
  718. [ConstraintElimination] Extend test coverage. (detail)
  719. [TableGen] Make all the fields in PatternToMatch private. NFCI (detail)
  720. [ConstraintElimination] Require GEPs to be inbounds for decomposition. (detail)
  721. [X86][AVX] Fold extract_subvector(splat, c) -> extract_subvector(splat, 0) (detail)
  722. [DAG] SelectionDAG::isSplatValue - handle OR/XOR cases (detail)
  723. [ASTMatchers] Add static constructor for ArgKinds of Matchers (detail)
  724. Make it possible to store NodeKinds in ArgKind (detail)
  725. [Reassociate] allow or->add with shl operands (detail)
  726. [ASTMatchers ]Make MatcherDescriptors indicate the node type they match (detail)
  727. [ASTMatchers] Make it possible to build mapAnyOf through the registry (detail)
  728. [ASTMatchers] Change internal method API (detail)
  729. [ASTMatchers] Extract parsing of bind token from the bind id (detail)
  730. Add clang-query support for mapAnyOf (detail)
  731. Add code complete support for mapAnyOf (detail)
  732. [clang-tidy] fix modernize-loop-convert to retain needed array-like operator[] (detail)
  733. [SelectionDAG] Drop unnecessary const from a return type (NFC) (detail)
  734. [TableGen] Use ListSeparator (NFC) (detail)
  735. [Transforms/Utils] Use range-based for loops (NFC) (detail)
  736. [ConstraintElimination] Add additional tests. (detail)
  737. [LV] Replace some uses of VectorLoopValueMap with VPTransformState (NFC) (detail)
  738. [ConstraintElimination] Decompose zext for unsigned compares. (detail)
  739. ELFObjectWriter: Don't sort local symbols (detail)
  740. [MC][test] Improve STT_FILE tests (detail)
  741. ELFObjectWriter: Make STT_FILE precede associated local symbols (detail)
  742. ELFObjectWriter: Don't de-duplicate STT_FILE symbols (detail)
  743. [scan-build-py] Update scan-build-py to allow outputing as SARIF (detail)
  744. [Test] Add failing test for PR49087 (detail)
  745. [mlir][ODS] Allow to specify custom namespace for `NativeOpTrait` (detail)
  746. [elf-core] Improve reading memory from core file (detail)
  747. [clang][cli] Generate and round-trip analyzer options (detail)
  748. Reorder MLIRContext location in BuiltinAttributes.h (detail)
  749. Revert "Reorder MLIRContext location in BuiltinAttributes.h" (detail)
  750. [clang][cli] Generate and round-trip preprocessor options (detail)
  751. [ARM] i16 insert-of-extract to VINS pattern (detail)
  752. [clang][index] Mark file as C++ in parse-all-comments test (detail)
  753. [modules] Put Frontend/OpenMP headers into a Clang module to fix the module build (detail)
  754. [clang][cli] Report result of ParseLangArgs (detail)
  755. [LLDB] Disable TestExprsChar.py, still fails on Arm/Linux (detail)
  756. Use remote regnums in expedited list, value regs and invalidate regs (detail)
  757. [clang][cli] NFC: Mirror CreateFromArgs order in generateCC1CommandLine (detail)
  758. Revert "[clang][cli] Report result of ParseLangArgs" (detail)
  759. [WebAssembly] Fix multiclass template parameter types. NFC. (detail)
  760. Revert "Revert "Reorder MLIRContext location in BuiltinAttributes.h"" (detail)
  761. [Analysis] Remove unused functions from TargetLibraryInfo (detail)
  762. [RISCV] Support the scalable-vector fadd reduction intrinsic (detail)
  763. [DebugInfo] Re-engineer a test to be stricter, add XFails (detail)
  764. [LLDB] Fix buildbot regression in symbol-binding.test (detail)
  765. [OpenCL] Do not enforce ASTContext for OCL2Qual (detail)
  766. [ValueTypes] Fix size of nxv1f16 (32 -> 16). (detail)
  767. [AMDGPU][MC][GFX10] Improved errors reporting for invalid MIMG NSA operands (detail)
  768. [ARM] Add some float Biquad cases showing difficult shuffling. NFC (detail)
  769. [ARM] Testing for stored extracted values. NFC (detail)
  770. Follow up to e05c10380ce7: add aarch64 to test XFails (detail)
  771. [AMDGPU][MC] Corrected error position for invalid dim modifiers (detail)
  772. [ARM] Make v2f64 scalar_to_vector legal (detail)
  773. [MLIR][LangRef doc] Fix a small typo. (detail)
  774. [AMDGPU]: Fixes an invalid clamp selection pattern. (detail)
  775. [lldb] Debugger: reuse ExecutionContextRef to create ExecutionContext from Target (detail)
  776. Reland "[lldb] Make CommandInterpreter's execution context the same as debugger's one" (detail)
  777. [lldb/tests] Un-skip TestGuiBasicDebug.py on Arm and AArch64 (detail)
  778. [lldb/tests] Removed add_test_categories decorator for python API tests, NFC (detail)
  779. Reapply "[clang][cli] Report result of ParseLangArgs" (detail)
  780. [clang][CodeComplete] Fix crash on ParenListExprs (detail)
  781. Revert "[Sema] Fix -Warray-bounds false negative when casting an out-of-bounds array item" (detail)
  782. Revert "DebugInfo: Temporarily work around -gsplit-dwarf + LTO .debug_gnu_pubnames regression after D94976" (detail)
  783. Revert "Re-land D94976 after revert in e29552c5aff6" (detail)
  784. [AMDGPU] Generate test checks and add GFX10 test coverage (detail)
  785. [ConstraintElimination] Improve index handing during constraint building. (detail)
  786. [MLIR][SPIRV] NFC: Split serialization code among multiple files. (detail)
  787. [CodeGen][AArch64] Add TargetInstrInfo hook to modify the TailDuplicateSize default threshold (detail)
  788. [RISCV] Use std::make_tuple to make some toolchains happy again (detail)
  789. [DAG] visitVECTOR_SHUFFLE - move shuffle legality check into MergeInnerShuffle lamda. NFCI. (detail)
  790. Introduce -print-changed=[diff | diff-quiet] which show changes in patch-like format (detail)
  791. AArch64: use a constpool for blockaddress(...) on MachO (detail)
  792. [libc++] Add a wait step in the BuildKite pipeline to shield macOS builders (detail)
  793. [AArch64AsmParser] Fix type-limits warning for VectorIndex. (detail)
  794. [OpenCL] Fix pipe type printing in arg info metadata (detail)
  795. [FE] Manipulate the first byte of guard variable type in both load and store operation (detail)
  796. [flang][fir] Update FIR's character type. (detail)
  797. [llvm-objdump] Support PLT decoding for aarch64_be (detail)
  798. [ELF] Support aarch64_be (detail)
  799. [LLDB] Fix `Wunused-result` warning (detail)
  800. [lldb] [Process/FreeBSDRemote] Introduce mips64 support (detail)
  801. [AMDGPU] Use named unified buffer format constant. NFC. (detail)
  802. [RISCV] Use SplatPat/SplatPat_simm5 to handle PseudoVMV_V_X_/PseudoVMV_V_I_ selection as well. (detail)
  803. [Sanitizer] Fix failing sanitizer tests (detail)
  804. [RISCV] Make scalable vector FMA commutable for register allocation. (detail)
  805. [ConstraintElimination] Decompose a few more GEP indices. (detail)
  806. [mlir] Drop deprecated syntax for LLVM dialect types (detail)
  807. [ELF] Inspect -EL & -EB for OUTPUT_FORMAT(default, big, little) (detail)
  808. [RISCV] Add initial support for converting fixed vectors to scalable vectors during lowering to use RVV instructions. (detail)
  809. [lld-macho] Emit personalities in compact unwind (detail)
  810. [lld-macho] Emit LSDA info in compact unwind (detail)
  811. [mlir][Linalg] Fix padding related bugs. (detail)
  812. [RISCV] Add support for fixed vector FMA. (detail)
  813. [RISCV] Add support for splat fixed length build_vectors using RVV. (detail)
  814. Revert "[Utils] Add a switch controlling prefix warnings in UpdateTestChecks" (detail)
  815. [GWP-ASan] Add aligned allocations. (detail)
  816. [RISCV] Use _COMMUTABLE fma pseudos for fixed vectors. (detail)
  817. [lld-macho] Try to make ubsan happy (detail)
  818. [dfsan] Refactor visitCallBase (detail)
  819. [libomptarget][amdgcn] Fix language linkage post D95300, drop use of assert (detail)
  820. [flang][NFC] Update comments. (detail)
  821. [scudo/standalone] Use .arch_extension memtag, not mte (detail)
  822. DebugInfo/Symbolize: Allow STT_NOTYPE/STT_GNU_IFUNC symbols for .symtab symbolization (detail)
  823. [AArch64] Use '//' as comment string for MSVC assembly (detail)
  824. Revert "[Test] Add failing test for PR49087" (detail)
  825. [AArch64][GlobalISel] Support the 'returned' parameter attribute. (detail)
  826. [CMake] [MinGW] Enable use of LLVM_USE_SANITIZER in a MinGW environment (detail)
  827. [gn build] reformat all gn files (detail)
  828. [SimpleLoopUnswitch] Don't non-trivially unswitch loops that are unsafe to clone (detail)
  829. [ARM] One-off identity shuffle (detail)
  830. [Verifier] Allow DW_TAG_class_type/DW_TAG_union_type to have no filename (detail)
  831. [FileCheck] Default --allow-unused-prefixes to false (detail)
  832. [test] Fix unused check prefixes (detail)
  833. [flang][NFC] Add comment. (detail)
  834. [NVPTX][NewPM] Re-enable NVVMReflectPass (detail)
  835. AMDGPU/GlobalISel: Remove dead check prefixes (detail)
  836. AMDGPU: Stop adding stack passed wide arguments to call conv handler (detail)
  837. GlobalISel: Use correct calling convention in handleAssignments (detail)
  838. [RISCV] Initial support of LoopVectorizer for RISC-V Vector. (detail)
  839. Renovate CMake files in the `llvm-exegesis` tool. (detail)
  840. Renovate CMake file for the `llvm-cfi-verify` tool (detail)
  841. [flang][fir] Add OpaqueAttr. (detail)
  842. [CUDA][HIP] Pass -fgpu-rdc to host clang -cc1 (detail)
  843. Make sure a module file with errors produced via '-fallow-pcm-with-compiler-errors' can be loaded when using implicit modules (detail)
  844. Revert "Renovate CMake files in the `llvm-exegesis` tool." (detail)
  845. [SPARC] Recognize and handle the %lm(sym) operator (detail)
  846. Fix failure in cuda-external-tools.cu (detail)
  847. [MLIR][NFC] Fix std.copysign op documentation (detail)
  848. [TableGen] Use return value from EmitVBRValue instead of calling GetVBRSize on the same value. Consistently use unsigned for child sizes. NFCI (detail)
  849. [MC][WebAssembly] Fix provisional values for data alias relocations (detail)
  850. [lld-macho] Add ARM64 target arch (detail)
  851. [gn build] Port 87104faac433 (detail)
  852. [llvm-objcopy][test] Stablize build-id-link-dir.test (detail)
  853. PR48587: is_constant_evaluated() should not evaluate to true during a (detail)
  854. PR48606: The lifetime of a constexpr heap allocation always started (detail)
  855. Revert "[CostModel] Remove VF from IntrinsicCostAttributes" (detail)
  856. [lldb] Fix crash in FormatEntity for mangled-name (detail)
  857. [CUDA][HIP] Add -fuse-cuid (detail)
  858. [lld][WebAssembly] Fix typo in function name (detail)
  859. [MLIR] Add C API for navigating up the IR tree (detail)
  860. Return "[Test] Add failing test for PR49087" (detail)
  861. [Transforms/Utils] Drop unnecessary const from a return type (NFC) (detail)
  862. [TableGen] Use ListSeparator (NFC) (detail)
  863. [Transforms] Use range-based for loops (NFC) (detail)
  864. Mark 4 tests added in 6d766c8bf9df as requiring an x86 backend as they fail when it is not present. (detail)
  865. [test] Add REQUIRES: x86-registered-target to DebugInfo/Symbolize/ELF llvm-mc tests (detail)
  866. [test] Drop redundant REQUIRES: x86-registered-target (detail)
  867. [MLIR][AVX512] Add integration test for vp2intersect (detail)
  868. [Docs] Fix Typo (detail)
  869. [RISCV] Use whole register load/store for generic load/store. (detail)
  870. [NFC] [Coroutine] Remove Unused Variables (detail)
  871. [ELF] Rewriting the path of sample profile file for --reproduce response.txt (detail)
  872. [-Wcompletion-handler] Support checks with builtins (detail)
  873. [ELF] Drop Android specific workaround -m aarch64_elf64_le_vec (detail)
  874. [clang][cli] Generate and round-trip language options (detail)
  875. [gn build] Port 40c261c41c4c (detail)
  876. Revert "Follow up to e05c10380ce7: add aarch64 to test XFails" (detail)
  877. Revert "[DebugInfo] Re-engineer a test to be stricter, add XFails" (detail)
  878. [clang][codegen] Remember string used to create llvm::Regex for optimization remarks (detail)
  879. [clang][cli] Generate and round-trip CodeGen options (detail)
  880. [-Wcompletion-handler][NFC] Remove unexpected warnings on Windows (detail)
  881. [clangd] Fix false positive in local rename collision detetction (detail)
  882. [NFC,Clang] Add LTO Driver Tsan tests (detail)
  883. [NFC,Clang] Add LTO Driver DFsan tests (detail)
  884. [NFC,Clang] Add LTO Driver MSan,KMsan tests (detail)
  885. [NFC,Clang] Add SanCov Driver tests (detail)
  886. [NFC,Clang] Add more Asan Driver tests (detail)
  887. Improve STRICT_FSETCC codegen in absence of no NaN (detail)
  888. Revert STRICT_FCMP nonan optimisation (detail)
  889. [AVR] Fix global references to function symbols (detail)
  890. [libomptarget][amdgcn] Tolerate deadstripped env variable (detail)
  891. [RISCV] Fix shared libs build (detail)
  892. [DAGCombine] Do not remove masking argument to FP16_TO_FP for some targets (detail)
  893. [X86] Always assign reassoc flag for intrinsics *reduce_add/mul_ps/pd. (detail)
  894. [lld/mac] Implement -u flag (detail)
  895. [mlir][linalg] Verify indexing map required attributes (detail)
  896. [mlir][linalg] Define a depthwise 2-D convolution op (detail)
  897. [clang-tidy] fix modernize-use-nullptr false positive with spaceship operator comparisons (detail)
  898. [mlir] Add stacksave, stackrestore to llvm dialect (detail)
  899. [X86][SSE] foldShuffleOfHorizOp - add SHUFPS v4f32 handling (detail)
  900. [clang][cli] Store InputKind in FrontendOptions (detail)
  901. Try to fix compiler-rt tests after 87dbdd2e3bb63b on mac and win (detail)
  902. [clang][cli] Look up program action argument in a table (detail)
  903. [clang][cli] Store unsigned instead of OptSpecifier in table (detail)
  904. [mlir][linalg] Fix one more missing NoSideEffect in linalg tensor op (detail)
  905. [mlir][linalg] Fix depthwise conv C++ symbol to be consistent (detail)
  906. [OpenCL] Fix types with signed prefix in arginfo metadata. (detail)
  907. [clangd] Implement semanticTokens modifiers (detail)
  908. [clang][cli] Generate and round-trip Frontend options (detail)
  909. [lldb] Inline invariant params to AppleThreadPlanStepThrough (NFC) (detail)
  910. Revert "[ObjC][ARC] Use operand bundle 'clang.arc.rv' instead of explicitly" (detail)
  911. [libomptarget][amdgcn] Tolerate deadstripped device_state variable (detail)
  912. [mlir] add support for verification in integration tests (detail)
  913. [flang][fir] Add shape, shape_shift, and slice types. (detail)
  914. [clangd] Add semanticTokens modifiers for function/class/file/global scope (detail)
  915. [llvm-readobj] Print empty line between note sections in GNU mode (detail)
  916. [yaml2obj] Handle NT_* string values in for ELF note types (detail)
  917. [llvm-readelf] Print raw ELF note contents if we can't parse it (detail)
  918. [llvm-readobj] Add support for decoding FreeBSD ELF notes (detail)
  919. [MLIR] Replace dialect registration hooks with dialect handle (detail)
  920. [dfsan] Refactor loadShadow (detail)
  921. [libc++] Remove C++11 work-arounds in src. (detail)
  922. [libc++] Require C++20 to build the benchmarks. (detail)
  923. [RISCV] Remove SRO* and SLO* instructions from bitmanip. (detail)
  924. [ValueTracking] improve analysis for "C << X" and "C >> X" (detail)
  925. [IROutliner] Adding instruction strings to IRSimilarityPrinting diagnostics. (detail)
  926. Revert "[mlir] add support for verification in integration tests" (detail)
  927. [mlir][Linalg] Add a build method for linalg.pad_tensor (detail)
  928. [InstCombine] convert assumes to operand bundles (detail)
  929. [RISCV] Make the min and max vector width command line options more consistent and check their relationship to each other. (detail)
  930. GlobalISel: Fix using wrong calling convention for callees (detail)
  931. AMDGPU: Fix verifier error with argument passed in CSR SGPR (detail)
  932. Revert "[libc++] Require C++20 to build the benchmarks." (detail)
  933. [CUDA, NVPTX] Allow targeting sm_86 GPUs. (detail)
  934. [Attributor][FIX] Do not create UB by introducing a `noundef undef` (detail)
  935. [WebAssembly] Use data sections by default (detail)
  936. [mlir][spirv] Add support for sampled image type (detail)
  937. [clangd] Expose more dependent-name detail via semanticTokens (detail)
  938. [libc++] Add `noexcept` to `string::find` and similar members. (detail)
  939. [mlir][IR] Remove the concept of `OperationProperties` (detail)
  940. [lldb] [Process/FreeBSDRemote] Introduce powerpc support (detail)
  941. [libc++] [LWG2993] reference_wrapper<T> conversion from U&& (detail)
  942. [lld][WebAssembly] Allow --export of optional start/stop symbols (detail)
  943. [MLIR] Add context accessor to identifier (detail)
  944. PR48545: Access check the inherited constructor, not the inheriting (detail)
  945. [AIX][llvm][support] Implement getHostCPUName (detail)
  946. [AArch64][GlobalISel] Allow vector load legalization into 128-bit-wide types (detail)
  947. [VPlan] Use VPUser to manage CondBit (detail)
  948. [llvm-objdump][test] Fix --prefix tests for system-windows (detail)
  949. [mlir] Add initial support for an alias analysis framework in MLIR (detail)
  950. [mlir][OpFormatGen] Refactor `type_ref` into a more general `ref` directive (detail)
  951. [lld][WebAssembly] Fix segfault in map file support (detail)
  952. [flang][fir] Updates to internal name uniquer. (detail)
  953. Add NoSideEffect trait to shape.split_at and shape.concat (detail)
  954. Add documentation for the extra_args parameter to breakpoint commands. (detail)
  955. Remove trailing spaces after \ in comments. (detail)
  956. [LLDB] Remove uneeded CopyType from BlockPointerSyntheticFrontEnd (detail)
  957. [Polly] Remove use of -O3 in regression test. (detail)
  958. Revert "[InstCombine] convert assumes to operand bundles" (detail)
  959. Fix side-effect detection in LLVMIRIntrinsicGen. (detail)
  960. Add LLVMIR Dialect counterparts of @llvm.maximum and @llvm.minimum. (detail)
  961. Use internal_dict everywhere we refer to the python session dict in docs. (detail)
  962. Fix xray fdr mode to allow multiple flushes (detail)
  963. Fix JSON formatting when converting to trace event format (detail)
  964. [clang][cli] Fix gcc "enumeral and non-enumeral type in conditional expression" warning (NFC) (detail)
  965. [libcxx] adds concept `std::convertible_to` (detail)
  966. [MLIR] Update affine.for unroll utility for iter_args support (detail)
  967. [MLIR] NFC Fix vector transforms build warnings (detail)
  968. [clang][driver] Only warn once about invalid library values (detail)
  969. [Test] Two more tests on usub (detail)
  970. Fix deprecated usage of `mallinfo` (detail)
  971. [Polly] Port ForwardOpTree to the NewPM. (detail)
  972. [Polly] Port DeLICM to the NewPM. (detail)
  973. [Polly] Port PruneUnprofitable to the NewPM. (detail)
  974. [Polly] Register pass-instrumentation for NewPM's Scop level. (detail)
  975. [Polly] Add TargetTransformInfo to ScopPass standard analysis results. (detail)
  976. [Polly] Port IslScheduleOptimizer to the NewPM. (detail)
  977. [Polly] Improve Simplify pass PM integration. (detail)
  978. [Polly] Make NewPM's IslAstAnalysis more similar to the legacy IslAstInfoWrapperPass. (detail)
  979. [Polly] Make the NewPM pass pipeline more similar to the legacy's. (detail)
  980. [TableGen] Use ListSeparator (NFC) (detail)
  981. [TableGen] Drop unnecessary const from return types (NFC) (detail)
  982. [SelectionDAG] Use range-based for loops (NFC) (detail)
  983. [mlir][shape] Generalize broadcast to a variadic number of shapes (detail)
  984. Revert "[analyzer] RetainCountChecker: Add a suppression for OSSymbols." (detail)
  985. [AArch64][GlobalISel] Fold selects fed by G_PTR_ADD (detail)
  986. [WebAssembly] Enable loop unrolling (detail)
  987. [ValueTypes] Add MVT for nxv1bf16. (detail)
  988. [LoopVectorize] NFC: Change computeFeasibleMaxVF to operate on ElementCount. (detail)
  989. [OpenCL] Add cl_khr_subgroup_non_uniform_arithmetic to TableGen BIFs (detail)
  990. [OpenCL] Add cl_khr_subgroup_clustered_reduce to TableGen BIFs (detail)
  991. [flang][driver] Add support for -fopenmp and -fopenacc (detail)

Git (git http://labmaster3.local/git/llvm-test-suite.git)

  1. [test-suite] SPEC2017 CPU CactuBSSN floating point tests. (detail)
  2. [test-suite] SPEC2017 CPU Fotonik3d floating point tests. (detail)

Started by user Azharuddin Mohammed

This run spent:

  • 42 min waiting;
  • 57 min build duration;
  • 1 hr 39 min total from scheduled to completion.
Revision: efb2e01b23f098f82bab3701c983d41711a2a3bb
  • refs/remotes/origin/main
Revision: f48d431f44610e339d00a33d57564c6029c4ff43
  • refs/remotes/origin/main
Revision: 6d48a1a53fabae8715091a9aca128cc4a5d9b0a6
  • refs/remotes/origin/main
Clang (LLVM based): No warnings
Test Result (no failures)