{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":215875945,"defaultBranch":"next","name":"llvm-project","ownerLogin":"apple","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2019-10-17T20:03:10.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/10639145?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1715761022.0","currentOid":""},"activityList":{"items":[{"before":"f090801a9651cf4f0d05cc361a2a1b14805b62bf","after":"2f6c0e6e180c81087c26f4afac2155ea70472ec6","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T09:29:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[flang][Alias Analysis] not all block arguments are dummy arguments (#92156)\n\nArguments to openmp regions should not be tagged as dummy arguments.\r\nThis is particularly unsafe because these openmp blocks will eventually\r\nbe inlined into the calling function, where they will trivially alias\r\nwith other values inside of the calling function.\r\n\r\nThis is probably a theoretical issue because the calls to openmp runtime\r\nfunction calls would act as barriers, preventing optimizations that are\r\ntoo aggressive. But a lot more thought would need to go into a bet like\r\nthat.\r\n\r\nThis came out of discussion on\r\nhttps://github.com/llvm/llvm-project/pull/92036","shortMessageHtmlLink":"[flang][Alias Analysis] not all block arguments are dummy arguments (l…"}},{"before":"ca4a405232cf170f20a2f111bf72beab82095935","after":"f090801a9651cf4f0d05cc361a2a1b14805b62bf","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T09:24:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[lldb] Disable unaligned-pc-sigbus.test on arm(32)\n\nI though the test could work there as well, but (of course) it does not,\nbecause the lowest bit just means \"run the code as thumb\".","shortMessageHtmlLink":"[lldb] Disable unaligned-pc-sigbus.test on arm(32)"}},{"before":"6479e3cb66895754089dc017a33478e9eb4b8d65","after":"ca4a405232cf170f20a2f111bf72beab82095935","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T08:54:51.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[analyzer] Refactor recognition of the errno getter functions (#91531)\n\nThere are many environments where `errno` is a macro that expands to\r\nsomething like `(*__errno())` (different standard library\r\nimplementations use different names instead of \"__errno\").\r\n\r\nIn these environments the ErrnoModeling checker creates a symbolic\r\nregion which will be used to represent the return value of this \"get the\r\nlocation of errno\" function.\r\n\r\nPreviously this symbol was only created when the checker was able to\r\nfind the declaration of the \"get the location of errno\" function; but\r\nthis commit eliminates the complex logic that was responsible for this\r\nand always creates the symbolic region when `errno` is not available as\r\na \"regular\" global variable.\r\n\r\nThis significantly simplifies a code and only introduces a minimal\r\nperformance reduction (one extra symbol) in the case when `errno` is not\r\ndeclared (neither as a variable nor as a function).\r\n\r\n\r\nIn addition to this simplification, this commit specifies that the\r\n`CallDescription`s for the \"get the location of errno\" functions are\r\nmatched in `CDM::CLibrary` mode. (This was my original goal, but I was\r\nsidetracked by resolving a FIXME above the `CallDescriptionSet` in\r\n`ErrnoModeling.cpp`.)\r\n\r\nThis change is very close to being NFC, but it fixes weird corner cases\r\nlike the handling of a C++ method that happens to be named \"__errno()\"\r\n(previously it could've been recognized as an errno location getter\r\nfunction).","shortMessageHtmlLink":"[analyzer] Refactor recognition of the errno getter functions (llvm#9…"}},{"before":"d12c48cad52798f4846dd8ef882af0f854118d16","after":"6479e3cb66895754089dc017a33478e9eb4b8d65","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T08:19:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[clang][Interp] Use proper type for non-primitive reference dummies","shortMessageHtmlLink":"[clang][Interp] Use proper type for non-primitive reference dummies"}},{"before":"7f3ac51b946bf6d6fa8c8443457ebee219879302","after":"d12c48cad52798f4846dd8ef882af0f854118d16","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T08:04:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[lldb/aarch64] Allow unaligned PC addresses below a trap handler (#92093)\n\nThe stack validation heuristic is counter-productive in this case, as\r\nthe unaligned address is most likely the thing that caused the signal in\r\nthe first place.","shortMessageHtmlLink":"[lldb/aarch64] Allow unaligned PC addresses below a trap handler (llv…"}},{"before":"73324cbc9c5892541aa82d466799748b435ece29","after":"7f3ac51b946bf6d6fa8c8443457ebee219879302","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T07:54:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[clang][Interp] Only accept constant variables in c++98","shortMessageHtmlLink":"[clang][Interp] Only accept constant variables in c++98"}},{"before":"4688df68f9d022dd8bc102675a9e86ad274355d6","after":"73324cbc9c5892541aa82d466799748b435ece29","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T07:04:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[ASan] Remove COMPILER_RT_ASAN_SHADOW_SCALE_DEFINITION. (#91832)\n\nSince the set of COMPILER_RT_ASAN_SHADOW_SCALE_DEFINITION is removed in\r\ncommit 8421fa5d536aadf42c0e54c566bc439a40ebdb8e,\r\ncleanup the use of COMPILER_RT_ASAN_SHADOW_SCALE_DEFINITION.","shortMessageHtmlLink":"[ASan] Remove COMPILER_RT_ASAN_SHADOW_SCALE_DEFINITION. (llvm#91832)"}},{"before":"45726c1a3a3d89ff9f6ebe657c3cb7bcd59b88db","after":"4688df68f9d022dd8bc102675a9e86ad274355d6","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T06:59:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"Avoid partial munmap (#92109)\n\n...which caused issues like\r\n\r\n> ==42==ERROR: AddressSanitizer failed to deallocate 0x32 (50) bytes at\r\naddress 0x117e0000 (error code: 28)\r\n> ==42==Cannot dump memory map on emscriptenAddressSanitizer: CHECK\r\nfailed: sanitizer_common.cpp:81 \"((0 && \"unable to unmmap\")) != (0)\"\r\n(0x0, 0x0) (tid=288045824)\r\n> #0 0x14f73b0c in __asan::CheckUnwind()+0x14f73b0c\r\n(this.program+0x14f73b0c)\r\n> #1 0x14f8a3c2 in __sanitizer::CheckFailed(char const*, int, char\r\nconst*, unsigned long long, unsigned long long)+0x14f8a3c2\r\n(this.program+0x14f8a3c2)\r\n> #2 0x14f7d6e1 in __sanitizer::ReportMunmapFailureAndDie(void*,\r\nunsigned long, int, bool)+0x14f7d6e1 (this.program+0x14f7d6e1)\r\n> #3 0x14f81fbd in __sanitizer::UnmapOrDie(void*, unsigned\r\nlong)+0x14f81fbd (this.program+0x14f81fbd)\r\n> #4 0x14f875df in __sanitizer::SuppressionContext::ParseFromFile(char\r\nconst*)+0x14f875df (this.program+0x14f875df)\r\n> #5 0x14f74eab in __asan::InitializeSuppressions()+0x14f74eab\r\n(this.program+0x14f74eab)\r\n> #6 0x14f73a1a in __asan::AsanInitInternal()+0x14f73a1a\r\n(this.program+0x14f73a1a)\r\n\r\nwhen trying to use an ASan suppressions file under Emscripten: Even\r\nthough it would be considered OK by SUSv4, the Emscripten runtime states\r\n\"We don't support partial munmapping\" (see\r\n\r\n\r\n\"Implement MAP_ANONYMOUS on top of malloc in STANDALONE_WASM mode\r\n(#16289)\").\r\n\r\nCo-authored-by: Stephan Bergmann ","shortMessageHtmlLink":"Avoid partial munmap (llvm#92109)"}},{"before":"d6ee7e8481fbaee30f37d82778ef12e135db5e67","after":"45726c1a3a3d89ff9f6ebe657c3cb7bcd59b88db","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T06:44:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[LLVM] Make sanitizers respect the disable_santizer_instrumentation attribute. (#91732)\n\n`disable_sanitizer_instrumetation` is attached to functions that shall\r\nnot be instrumented e.g. ifunc resolver because those run before\r\neverything is initialised.\r\nSome sanitizer already handles this attribute, this patch adds it to\r\nDataFLow and Coverage too.","shortMessageHtmlLink":"[LLVM] Make sanitizers respect the disable_santizer_instrumentation a…"}},{"before":"e6216906f528b948018b883068cef0fd4157bfd1","after":"d6ee7e8481fbaee30f37d82778ef12e135db5e67","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T06:39:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[SystemZ] Handle address clobbering in splitMove(). (#92105)\n\nWhen expanding an L128 (which is used to reload i128) it is\r\npossible that the quadword destination register clobbers an\r\naddress register. This patch adds an assertion against the case\r\nwhere both of the expanded parts clobber the address, and in the\r\ncase where one of the expanded parts do so puts it last.\r\n\r\nFixes #91437","shortMessageHtmlLink":"[SystemZ] Handle address clobbering in splitMove(). (llvm#92105)"}},{"before":"d7bb0723fe79d2b75d41789d2ffadda3567dd94e","after":"e6216906f528b948018b883068cef0fd4157bfd1","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T06:34:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[MLIR][LLVM] Improve atomic verifier to properly support larger types (#92120)\n\nThis commit extends the verifier for atomics to properly verify larger\r\ntypes. Beforehand, the verifier strictly rejected larger integer types,\r\nwhile it now consults the data layout to determine if their bitsize is a\r\npower of two. This behavior reflects what LLVM's verifier is checking\r\nfor.","shortMessageHtmlLink":"[MLIR][LLVM] Improve atomic verifier to properly support larger types ("}},{"before":"0980f715cf7c3d78be6ba64e902bd2dfad3ebc75","after":"d7bb0723fe79d2b75d41789d2ffadda3567dd94e","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T05:44:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"InstCombine: Emit ldexp intrinsic in exp2->ldexp combine (#92039)\n\nPrefer to emit the intrinsic over a libcall in the\r\nintrinsic or no-math-errno case.","shortMessageHtmlLink":"InstCombine: Emit ldexp intrinsic in exp2->ldexp combine (llvm#92039)"}},{"before":"24180ea0c295856a696992f072c36259a266226b","after":"0980f715cf7c3d78be6ba64e902bd2dfad3ebc75","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T05:29:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[libc][errno] Remove previously added errno numbers (#92163)\n\nIntroduced in https://github.com/llvm/llvm-project/pull/91150. Not\r\nneeded anymore as https://github.com/llvm/llvm-project/pull/92041 fixed\r\nthe root cause. `ENAMETOOLONG` and `EOVERFLOW` are well defined in\r\n``.\r\n\r\nPost mortem: Due to the previously missing inclusion of\r\n`` (fixed with\r\nhttps://github.com/llvm/llvm-project/pull/92041), I misinterpreted an\r\nundefined macro issue during the development of\r\nhttps://github.com/llvm/llvm-project/pull/91150 as being caused by a\r\nmissing definition rather than by the missing inclusion of the linux\r\nheader. I realized too late that `ENAMETOOLONG` and `EOVERFLOW` were\r\ncorrectly defined in `` and that it was my missing\r\ninclusion that caused the problem.","shortMessageHtmlLink":"[libc][errno] Remove previously added errno numbers (llvm#92163)"}},{"before":"15397583e3d85eb1f1a051de26eb409aaedd3b54","after":"24180ea0c295856a696992f072c36259a266226b","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T05:19:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[analyzer] Treat break, continue, goto, and label statements as trivial in WebKit checkers. (#91873)\n\nAlso allow CXXBindTemporaryExpr, which creates a temporary object with a\r\nnon-trivial destructor, and add a few more std and WTF functions to the\r\nexplicitly allowed list.","shortMessageHtmlLink":"[analyzer] Treat break, continue, goto, and label statements as trivi…"}},{"before":"74d91d9acebdbc81306c8a0ee547002e7e8084f9","after":"15397583e3d85eb1f1a051de26eb409aaedd3b54","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T05:04:58.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"Revert \"[polly] Port polly tests to use NPM\" (#92215)\n\nReverts llvm/llvm-project#90632.\r\n\r\nCausing failures on buildbots that dynamically load polly. Reverting\r\nwhile we sort it out.","shortMessageHtmlLink":"Revert \"[polly] Port polly tests to use NPM\" (llvm#92215)"}},{"before":"77047e3cd2edd2b870982fc92f505cbb7fd764cd","after":"74d91d9acebdbc81306c8a0ee547002e7e8084f9","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T04:49:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[polly] Port polly tests to use NPM (#90632)\n\nEven as the NPM has been in use by Polly for a while now, the\r\nmajority of the tests continue using the LPM passes. This patch\r\nports the tests to use the NPM passes (for example, by replacing\r\na flag such as -polly-detect with -passes=polly-detect following\r\nthe NPM syntax for specifying passes) with some exceptions for\r\nsome missing features in the new passes. Additionally, the lit\r\nsubstitution %loadPolly is replaced by the substitution of what\r\nwas %loadNPMPolly and %loadNPMPolly is removed.","shortMessageHtmlLink":"[polly] Port polly tests to use NPM (llvm#90632)"}},{"before":"d3455f4ddd16811401fa153298fadd2f59f6914e","after":"77047e3cd2edd2b870982fc92f505cbb7fd764cd","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T04:34:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[RISCV] Make vsetvli in test not loop invariant. NFC (#92094)\n\nThe middle end will remove the inner vsetvli otherwise, and it's more\r\ntypical to set the AVL to the remaining VL.\r\n\r\nThis also prevents the test from showing up as a regression in #91319","shortMessageHtmlLink":"[RISCV] Make vsetvli in test not loop invariant. NFC (llvm#92094)"}},{"before":"a71e2b9d0f287e4927e51d6764f90e492ba136e1","after":"d3455f4ddd16811401fa153298fadd2f59f6914e","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T04:14:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[libc][docs] Fix outdated code review section, as per #91934 (#92051)\n\nAs in the title, fixes #91934","shortMessageHtmlLink":"[libc][docs] Fix outdated code review section, as per llvm#91934 (llv…"}},{"before":"1a58e88690c1a48d1082b4ee6b759f5dc49a7144","after":"a71e2b9d0f287e4927e51d6764f90e492ba136e1","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T04:04:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[libc][errno] Remove non asm generic error number (#92172)\n\nThe following small thing caught my eye:\r\n\r\n1) `EILSEQ` is not part of the generic asm error number macros. See the\r\n[full list of generic asm errno\r\ncodes](https://github.com/torvalds/linux/blob/4b95dc87362aa57bdd0dcbad109ca5e5ef3cbb6c/include/uapi/asm-generic/errno-base.h).\r\nAFAIK the generic asm errno numbers are common between different\r\noperating systems and architectures. `EILSEQ` is not part of this common\r\nset of errno's.\r\n\r\n2) `EILSEQ`'s value is wrong. During the addition of `EILSEQ` in\r\nhttps://reviews.llvm.org/D151129, the value `35` was probably chosen as\r\nits the consecutive number. This is not correct. The actual values can\r\nbe looked up for example here:\r\n* [For Linux\r\nkernel](https://github.com/search?q=repo%3Atorvalds%2Flinux+EILSEQ&type=code&p=1):\r\n`EILSEQ = 84` (uapi; i.e. x86_64), `EILSEQ = 88` (mips), `EILSEQ = 47`\r\n(parisc)\r\n* [For Darwin\r\nkernel](https://github.com/apple-oss-distributions/xnu/blob/main/bsd/sys/errno.h#L237):\r\n`EILSEQ = 92`","shortMessageHtmlLink":"[libc][errno] Remove non asm generic error number (llvm#92172)"}},{"before":"72b2c37de6a4bbc2b2d2cda49293684b7cc71508","after":"1a58e88690c1a48d1082b4ee6b759f5dc49a7144","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T03:44:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[RISCV] Move RISCVInsertVSETVLI to after phi elimination (#91440)\n\nSplit off from #70549, this patch moves RISCVInsertVSETVLI to after phi\r\nelimination where we exit SSA and need to move to LiveVariables.\r\n\r\nThe motivation for splitting this off is to avoid the large scheduling\r\ndiffs from moving completely to after regalloc, and instead focus on\r\nconverting the pass to work on LiveIntervals.\r\n\r\nThe two main changes required are updating VSETVLIInfo to store VNInfos\r\ninstead of MachineInstrs, which allows us to still check for PHI defs in\r\nneedVSETVLIPHI, and fixing up the live intervals of any AVL operands\r\nafter inserting new instructions.\r\n\r\nOn O3 the pass is inserted after the register coalescer, otherwise we\r\nend up with a bunch of COPYs around eliminated PHIs that trip up\r\nneedVSETVLIPHI.\r\n\r\nCo-authored-by: Piyou Chen ","shortMessageHtmlLink":"[RISCV] Move RISCVInsertVSETVLI to after phi elimination (llvm#91440)"}},{"before":"2ece5cc2bb1b4cc787e33e24a6582043d441a572","after":"72b2c37de6a4bbc2b2d2cda49293684b7cc71508","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T03:34:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[clang-format][NFC] Clean up TokenAnnotator::mustBreakBefore()","shortMessageHtmlLink":"[clang-format][NFC] Clean up TokenAnnotator::mustBreakBefore()"}},{"before":"eb103104ef08ebc2d0de63db0592e76b294cf8bb","after":"2ece5cc2bb1b4cc787e33e24a6582043d441a572","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T03:09:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[ExceptionDemo] Correct and update example ExceptionDemo (#69485)\n\nThe ExceptionDemo example was no longer compiling (since llvm 14 at\r\nleast). The PR makes the example work with the current API and also\r\ntransition from MCJIT to ORC.\r\nFixes #63702","shortMessageHtmlLink":"[ExceptionDemo] Correct and update example ExceptionDemo (llvm#69485)"}},{"before":"13b265c7b5c6a989427639e33893c158f737480b","after":"eb103104ef08ebc2d0de63db0592e76b294cf8bb","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T02:54:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"Revert \"[Serialization] Read the initializer for interesting static variables before consuming it\"\n\nThis reverts commit 11b059145d177ee287c7ada9864addf8d083c160.\n\nThe premerge bot is broken.","shortMessageHtmlLink":"Revert \"[Serialization] Read the initializer for interesting static v…"}},{"before":"11b059145d177ee287c7ada9864addf8d083c160","after":"13b265c7b5c6a989427639e33893c158f737480b","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T02:45:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[X86][MC] Support Intel FRED and LKGS instructions. (#91909)\n\nSpec reference: https://cdrdv2.intel.com/v1/dl/getContent/678938","shortMessageHtmlLink":"[X86][MC] Support Intel FRED and LKGS instructions. (llvm#91909)"}},{"before":"6642cc60a21f857bddde8ec2c81008a83a54b4d3","after":"11b059145d177ee287c7ada9864addf8d083c160","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T02:35:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[Serialization] Read the initializer for interesting static variables before consuming it\n\nClose https://github.com/llvm/llvm-project/issues/91418\n\nSince we load the variable's initializers lazily, it'd be problematic\nif the initializers dependent on each other. So here we try to load the\ninitializers of static variables to make sure they are passed to code\ngenerator by order. If we read any thing interesting, we would consume\nthat before emitting the current declaration.","shortMessageHtmlLink":"[Serialization] Read the initializer for interesting static variables…"}},{"before":"71fbbb69d63c461f391cbabf1e32cd9977c4ce68","after":"6642cc60a21f857bddde8ec2c81008a83a54b4d3","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T02:14:59.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[SCEV] Add tests for ule/sle exit counts (NFC)","shortMessageHtmlLink":"[SCEV] Add tests for ule/sle exit counts (NFC)"}},{"before":"4d1ecf192313b612090d60181937eff03c1a966b","after":"71fbbb69d63c461f391cbabf1e32cd9977c4ce68","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T01:49:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[IR] Move GlobalValue::getGUID() out of line (NFC)\n\nAvoid including MD5.h in a core IR header.","shortMessageHtmlLink":"[IR] Move GlobalValue::getGUID() out of line (NFC)"}},{"before":"03c7458a3603396d2d0e1dee43399d3d1664a264","after":"4d1ecf192313b612090d60181937eff03c1a966b","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T01:29:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[Transforms] Preserve inbounds attribute of transformed GEPs when flattening loops (#86961)\n\nWhen flattening the loop, if the GEP was inbound, it should stay\r\ninbound, because the only thing that changed is how the pointers are\r\ncalculated, not the elements being accessed.\r\n\r\nProof: https://alive2.llvm.org/ce/z/dApMpQ","shortMessageHtmlLink":"[Transforms] Preserve inbounds attribute of transformed GEPs when fla…"}},{"before":"dfdc3dcbe7f38bde64bc83a74b9c3451c50e1ad4","after":"03c7458a3603396d2d0e1dee43399d3d1664a264","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T01:10:01.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"Revert \"[ctx_profile] Profile reader and writer\" (#92199)\n\nReverts llvm/llvm-project#91859\r\n\r\nBuildbot failures.","shortMessageHtmlLink":"Revert \"[ctx_profile] Profile reader and writer\" (llvm#92199)"}},{"before":"c5cd049566a795ba5de88dfbb2eb563cad4a9d8a","after":"dfdc3dcbe7f38bde64bc83a74b9c3451c50e1ad4","ref":"refs/heads/llvm.org/main","pushedAt":"2024-05-15T01:04:59.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"swift-ci","name":null,"path":"/swift-ci","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15467072?s=80&v=4"},"commit":{"message":"[ctx_profile] Profile reader and writer (#91859)\n\nUtility converting a profile coming from `compiler_rt` to bitstream, and\r\na reader.\r\n\r\n`PGOCtxProfileWriter::write` would be used as the `Writer` parameter for\r\n`__llvm_ctx_profile_fetch` API. This is expected to happen in user code,\r\nfor example in the RPC hanler tasked with collecting a profile, and\r\nwould look like this:\r\n\r\n```\r\n// set up an output stream \"Out\", which could contain other stuff\r\n{\r\n // constructing the Writer will start the section, in Out, containing\r\n // the collected contextual profiles.\r\n PGOCtxProfWriter Writer(Out);\r\n __llvm_ctx_profile_fetch(&Writer, +[](void* W, const ContextNode &N) {\r\n reinterpret_cast(W)->write(N);\r\n });\r\n // Writer going out of scope will finish up the section.\r\n}\r\n```\r\n\r\nThe reader produces a data structure suitable for maintenance during IPO\r\ntransformations.","shortMessageHtmlLink":"[ctx_profile] Profile reader and writer (llvm#91859)"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAESloaQQA","startCursor":null,"endCursor":null}},"title":"Activity · apple/llvm-project"}