{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":474219132,"defaultBranch":"main","name":"llvm-project","ownerLogin":"dayanruben","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2022-03-26T02:19:48.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/6559692?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1701487171.0","currentOid":""},"activityList":{"items":[{"before":"c675a58edec6d1a876a0d0e7d261f74764855b38","after":"bd6c358d01f6ebc3851996e2c29c47b08e992525","ref":"refs/heads/main","pushedAt":"2024-05-16T16:56:33.000Z","pushType":"push","commitsCount":19,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[lldb-dap] Update repository in package.json\n\nUse https://github.com/llvm/vscode-lldb instead of the monorepo, for\nconsistency with the other two extensions (mlir, clangd).","shortMessageHtmlLink":"[lldb-dap] Update repository in package.json"}},{"before":"398162ddbcf741c49e86bef2ef4aaa3fd0213916","after":"c675a58edec6d1a876a0d0e7d261f74764855b38","ref":"refs/heads/main","pushedAt":"2024-05-16T13:48:53.000Z","pushType":"push","commitsCount":178,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[TableGen][SubtargetEmitter] Early exit from loop in FindWriteResources and FindReadAdvance (#92202)\n\nThis gives us a 30% speed improvement in our downstream.","shortMessageHtmlLink":"[TableGen][SubtargetEmitter] Early exit from loop in FindWriteResourc…"}},{"before":"041baf2f60ac3e107399641aea04c77019e7eab8","after":"398162ddbcf741c49e86bef2ef4aaa3fd0213916","ref":"refs/heads/main","pushedAt":"2024-05-15T12:17:32.000Z","pushType":"push","commitsCount":103,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[libc] Fix typo in test message","shortMessageHtmlLink":"[libc] Fix typo in test message"}},{"before":"a6d7828f4c50c1ec7b0b5f61fe59d7a768175dcc","after":"041baf2f60ac3e107399641aea04c77019e7eab8","ref":"refs/heads/main","pushedAt":"2024-05-14T14:06:53.000Z","pushType":"push","commitsCount":105,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[mlir][ArmSME] Use liveness information in the tile allocator (#90448)\n\nThis patch rewrites the ArmSME tile allocator to use liveness\r\ninformation to make better tile allocation decisions and improve the\r\ncorrectness of the ArmSME dialect. This algorithm used here is a linear\r\nscan over live ranges, where live ranges are assigned to tiles as they\r\nappear in the program (chronologically). Live ranges release their\r\nassigned tile ID when the current program point is passed their end.\r\nThis is a greedy algorithm (which is mainly to keep the implementation\r\nrelatively straightforward), and because it seems to be sufficient for\r\nmost kernels (e.g. matmuls) that use ArmSME. The general steps of this\r\nare roughly from\r\nhttps://link.springer.com/content/pdf/10.1007/3-540-45937-5_17.pdf,\r\nthough there have been a few simplifications and assumptions made for\r\nour use case.\r\n\r\nHopefully, the only changes needed for a user of the ArmSME dialect is\r\nthat:\r\n\r\n- `-allocate-arm-sme-tiles` will no longer be a standalone pass \r\n  - `-test-arm-sme-tile-allocation` is only for unit tests \r\n- `-convert-arm-sme-to-llvm` must happen after `-convert-scf-to-cf` \r\n   - SME tile allocation is now part of the LLVM conversion\r\n\r\nBy integrating this into the `ArmSME -> LLVM` conversion we can allow\r\nhigh-level (value-based) ArmSME operations to be side-effect-free, as we\r\ncan guarantee nothing will rearrange ArmSME operations before we emit\r\nintrinsics (which could invalidate the tile allocation).\r\n\r\nThe hope is for ArmSME operations to have no hidden state/side effects\r\nand allow easily lowering dialects such as `vector` and `arith` to SME,\r\nwithout making assumptions about how the input IR looks, as the\r\nsemantics of the operations will be the same. That is no (new) side\r\neffects and the IR follows the rules of SSA (a value will never change).\r\n\r\nThe aim is correctness, so we have a base for working on optimizations.","shortMessageHtmlLink":"[mlir][ArmSME] Use liveness information in the tile allocator (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2268656980\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/90448\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/90448/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/90448\">llvm#9…</a>"}},{"before":"5df01ed79c47b0b10ea6be87251ec6e0484515a5","after":"a6d7828f4c50c1ec7b0b5f61fe59d7a768175dcc","ref":"refs/heads/main","pushedAt":"2024-05-13T19:54:42.000Z","pushType":"push","commitsCount":19,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[test] Use conventional -emit-llvm-only\n\nThis avoids 'Permission denied' when PWD is read-only.\nWhile here, change the triple from a Linux one to a generic ELF one.","shortMessageHtmlLink":"[test] Use conventional -emit-llvm-only"}},{"before":"a427aa9346295fe7dd3be5955214d28c8be2ad4a","after":"5df01ed79c47b0b10ea6be87251ec6e0484515a5","ref":"refs/heads/main","pushedAt":"2024-05-13T17:50:55.000Z","pushType":"push","commitsCount":164,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Revert \"[mlir][vector] Add Vector-dialect interleave-to-shuffle pattern, enable in VectorToSPIRV\" (#92006)\n\nReverts llvm/llvm-project#91800\r\n\r\nReason: https://lab.llvm.org/buildbot/#/builders/268/builds/13935","shortMessageHtmlLink":"Revert \"[mlir][vector] Add Vector-dialect interleave-to-shuffle patte…"}},{"before":"3be8e2c95d3dca5b2fdea889649a69dce8605e65","after":"a427aa9346295fe7dd3be5955214d28c8be2ad4a","ref":"refs/heads/main","pushedAt":"2024-05-10T20:06:17.000Z","pushType":"push","commitsCount":45,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[flang][Lower] Treat directives with nested evaluations as constructs (#91614)\n\nWhen generating block terminators in `genFIR(Evaluation)`, treat\r\n`Directives` with nested evaluations the same way as `Constructs` to\r\ndetermine the successor block.\r\n\r\nThis fixes https://github.com/llvm/llvm-project/issues/91526","shortMessageHtmlLink":"[flang][Lower] Treat directives with nested evaluations as constructs (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2288012130\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/91614\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/91614/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/91614\">…</a>"}},{"before":"200f3bd39562f4d605f13567398025d30fa27d61","after":"3be8e2c95d3dca5b2fdea889649a69dce8605e65","ref":"refs/heads/main","pushedAt":"2024-05-10T13:00:20.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[InstCombine] Prefer to keep power-of-2 constants when combining ashr exact and slt/ult of a constant (#86111)\n\nWe have flexibility in what constant to use when combining an `ashr\r\nexact` with a slt or ult of a constant, and it's not possible to revisit\r\nthis decision later in the compilation pipeline after the `ashr exact`\r\nis removed. Keeping a constant close to power-of-2 (pow2val + 1) should\r\nbe no worse than neutral, and in some cases may allow better codegen\r\nlater on for targets that can more cheaply generate power of 2 (which\r\nmay be selectable if converting back to setle/setge) or near power of 2\r\nconstants.\r\n\r\nAlive2 proofs:\r\n<https://alive2.llvm.org/ce/z/2BmPnq> and\r\n<https://alive2.llvm.org/ce/z/DtuhnR>","shortMessageHtmlLink":"[InstCombine] Prefer to keep power-of-2 constants when combining ashr…"}},{"before":"90501be35b2c4ad314a45634062e0dfe878d8621","after":"200f3bd39562f4d605f13567398025d30fa27d61","ref":"refs/heads/main","pushedAt":"2024-05-10T12:36:26.000Z","pushType":"push","commitsCount":68,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[Clang][Sema] access checking of friend declaration should not be delayed (#91430)\n\nattempt to fix https://github.com/llvm/llvm-project/issues/12361\r\nConsider this example:\r\n```cpp\r\nclass D {\r\n    class E{\r\n        class F{};\r\n        friend  void foo(D::E::F& q);\r\n        };\r\n    friend  void foo(D::E::F& q);\r\n    };\r\n\r\nvoid foo(D::E::F& q) {}\r\n```\r\nThe first friend declaration of foo is correct. After that, the second\r\nfriend declaration delayed access checking and set its previous\r\ndeclaration to be the first one. When doing access checking of `F`(which\r\nis private filed of `E`), we put its canonical declaration(the first\r\nfriend declaration) into `EffectiveContext.Functions`. Actually, we are\r\nstill checking the first one. This is incorrect due to the delayed\r\nchecking.\r\nCreating a new scope to indicate we are parsing a friend declaration and\r\ndoing access checking in time.","shortMessageHtmlLink":"[Clang][Sema] access checking of friend declaration should not be del…"}},{"before":"6f2997cefc1e32c11a891ede2e3a2d73310e6ce1","after":"90501be35b2c4ad314a45634062e0dfe878d8621","ref":"refs/heads/main","pushedAt":"2024-05-09T18:06:40.000Z","pushType":"push","commitsCount":292,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[flang] Accept interoperable types without BIND(C) (#91363)\n\nA derived type that meets (most of) the requirements of an interoperable\r\ntype but doesn't actually have the BIND(C) attribute can be accepted as\r\nan interoperable type, with optional warnings.","shortMessageHtmlLink":"[flang] Accept interoperable types without BIND(C) (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2283888357\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/91363\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/91363/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/91363\">llvm#91363</a>)"}},{"before":"a13c5140a2a26923f3e7bf3684409425ff54de6f","after":"6f2997cefc1e32c11a891ede2e3a2d73310e6ce1","ref":"refs/heads/main","pushedAt":"2024-05-07T12:38:22.000Z","pushType":"push","commitsCount":269,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[libc++][NFC] Remove unused include from <__type_traits/is_equality_comparable.h> (#90950)","shortMessageHtmlLink":"[libc++][NFC] Remove unused include from &lt;__type_traits/is_equality_c…"}},{"before":"427c5bfd39ebb9d008b621370579444fbf2a60d7","after":"a13c5140a2a26923f3e7bf3684409425ff54de6f","ref":"refs/heads/main","pushedAt":"2024-05-03T13:35:38.000Z","pushType":"push","commitsCount":143,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[OpenACC] Implement firstprivate clause for compute constructs\n\nThis clause is pretty nearly copy/paste from private, except that it\ndoesn't support 'loop', and thus 'kernelsloop' for appertainment.","shortMessageHtmlLink":"[OpenACC] Implement firstprivate clause for compute constructs"}},{"before":"0647b2a3caed3215588477dfc7aaa6bd3b4f8076","after":"427c5bfd39ebb9d008b621370579444fbf2a60d7","ref":"refs/heads/main","pushedAt":"2024-05-02T12:18:36.000Z","pushType":"push","commitsCount":110,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Revert \"[clang][Interp] Fix locals created in ExprWithCleanups\"\n\nThis reverts commit 1c7673b91d4d3bab4e296f5c67751d3879fb21a2.\n\nUnfortunately, this one is broken because de04e6cd90b891215f1dfc83ec886d037a7c2ed0\nhas been reverted.","shortMessageHtmlLink":"Revert \"[clang][Interp] Fix locals created in ExprWithCleanups\""}},{"before":"a1423ba4278775472523fed074de6dbdfd01898a","after":"0647b2a3caed3215588477dfc7aaa6bd3b4f8076","ref":"refs/heads/main","pushedAt":"2024-05-01T13:25:58.000Z","pushType":"push","commitsCount":71,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[gn build] Port df241b19c952","shortMessageHtmlLink":"[gn build] Port <a class=\"commit-link\" data-hovercard-type=\"commit\" data-hovercard-url=\"https://github.com/dayanruben/llvm-project/commit/df241b19c952b904eec755d9f090737aed437986/hovercard\" href=\"https://github.com/dayanruben/llvm-project/commit/df241b19c952b904eec755d9f090737aed437986\"><tt>df241b1</tt></a>"}},{"before":"431be8626696da6059eee26b9a14e1dc9fc0c37f","after":"a1423ba4278775472523fed074de6dbdfd01898a","ref":"refs/heads/main","pushedAt":"2024-04-30T19:43:51.000Z","pushType":"push","commitsCount":377,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[mlir][tensor] Fix integration tests that uses reshape ops. (#90649)\n\nDue to generalization introduced in\r\nhttps://github.com/llvm/llvm-project/pull/90040","shortMessageHtmlLink":"[mlir][tensor] Fix integration tests that uses reshape ops. (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2272264476\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/90649\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/90649/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/90649\">llvm#90649</a>)"}},{"before":"f5953f46aa0a664461584b78c14cb141a3be2b9d","after":"431be8626696da6059eee26b9a14e1dc9fc0c37f","ref":"refs/heads/main","pushedAt":"2024-04-26T11:34:29.000Z","pushType":"push","commitsCount":110,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Revert \"[RISCV] Support RISCV Atomics ABI attributes (#84597)\"\n\nThis reverts commit 9221f3af8f832d990be986c05d964ad37e5a2356.\n\nAs reported\n<https://github.com/llvm/llvm-project/pull/84597#issuecomment-2079128332>\nand confirmed by me locally, adding these attributes causes current GNU\nld to segfault when processing the input. Reverted so we can discuss\nthe best next step.","shortMessageHtmlLink":"Revert \"[RISCV] Support RISCV Atomics ABI attributes (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2177026413\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/84597\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/84597/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/84597\">llvm#84597</a>)\""}},{"before":"ed962a66c5a2257100ca20624521ef51c0abcfb3","after":"f5953f46aa0a664461584b78c14cb141a3be2b9d","ref":"refs/heads/main","pushedAt":"2024-04-25T17:41:19.000Z","pushType":"push","commitsCount":24,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Fix lldb build failure caused by 39adc8f42329\n\nWe changed the name of one of the types, which is consumed by LLDB.  My\npatch local build + CI didn't catch it, but a build bot did!  This\ncommit fixes it by updating the name in LLDB.","shortMessageHtmlLink":"Fix lldb build failure caused by <a class=\"commit-link\" data-hovercard-type=\"commit\" data-hovercard-url=\"https://github.com/dayanruben/llvm-project/commit/39adc8f423297c5741bb731bb8b1e545d558502c/hovercard\" href=\"https://github.com/dayanruben/llvm-project/commit/39adc8f423297c5741bb731bb8b1e545d558502c\"><tt>39adc8f</tt></a>"}},{"before":"deafb36f87a3541715854d4a620a4cfd6b1ac672","after":"ed962a66c5a2257100ca20624521ef51c0abcfb3","ref":"refs/heads/main","pushedAt":"2024-04-25T14:00:50.000Z","pushType":"push","commitsCount":9,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[libc++] Implement LWG4023 (#87513)\n\nThis patch implements LWG4023 by adding explicit assertions for the\r\nadded preconditions and also fixes a few tests that were violating these\r\npreconditions.","shortMessageHtmlLink":"[libc++] Implement LWG4023 (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2223342524\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/87513\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/87513/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/87513\">llvm#87513</a>)"}},{"before":"2575cd8a900283368e451cdbc27a2e4db0722af9","after":"deafb36f87a3541715854d4a620a4cfd6b1ac672","ref":"refs/heads/main","pushedAt":"2024-04-25T12:10:26.000Z","pushType":"push","commitsCount":74,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[flang][OpenMP] Add OpenMP versions to some tests (#89295)\n\nSome constructs used in the tests are only allowed in certain OpenMP\r\nspec versions. Add a flag with the minimum required OpenMP version\r\n(other than the default version) to these tests that need it.","shortMessageHtmlLink":"[flang][OpenMP] Add OpenMP versions to some tests (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2251448527\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/89295\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/89295/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/89295\">llvm#89295</a>)"}},{"before":"96b2e35a58819eb2fbe1821650e35a1f0e085bd7","after":"2575cd8a900283368e451cdbc27a2e4db0722af9","ref":"refs/heads/main","pushedAt":"2024-04-24T19:05:23.000Z","pushType":"push","commitsCount":43,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Add ics link for Floating Point WG (#82545)\n\nThis adds a link to an ics file for the LLVM Floating Point WG line in\r\nthe Getting Involved page.","shortMessageHtmlLink":"Add ics link for Floating Point WG (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2147811001\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/82545\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/82545/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/82545\">llvm#82545</a>)"}},{"before":"79d4d165638b7587937fc60431e0865fd73c9334","after":"96b2e35a58819eb2fbe1821650e35a1f0e085bd7","ref":"refs/heads/main","pushedAt":"2024-04-24T13:40:15.000Z","pushType":"push","commitsCount":16,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[CostModel][AArch64] Improve fixed-width vector costs for get.active.lane.mask (#89068)\n\nWhen SVE is available we can lower calls to get.active.lane.mask using\r\nthe SVE whilelo instruction, however in practice since vXi1 types are\r\nnot legal for NEON we often end up expanding the predicate into a vector\r\nof integers, e.g. v4i1 -> v4i32. This usually happens when we have to\r\nkeep the predicate live out of the block, for example when the predicate\r\nis the incoming value to a PHI node in a tail-folded vector loop.\r\nCurrently in such cases the intrinsic call has a cost of 1, which is far\r\ntoo low when considering the extra instructions required to expand the\r\npredicate. This patch fixes that by basing the cost on the number of\r\nlane moves required for expansion. This is required for a follow-on\r\npatch that adds the cost of the intrinsic call to the vectorisation cost\r\nmodel, so that we can teach the vectoriser to make better choices.","shortMessageHtmlLink":"[CostModel][AArch64] Improve fixed-width vector costs for get.active.…"}},{"before":"0c032fd5425d853dfc577e607b9c179d811cec19","after":"79d4d165638b7587937fc60431e0865fd73c9334","ref":"refs/heads/main","pushedAt":"2024-04-24T12:43:48.000Z","pushType":"push","commitsCount":83,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[mlir][python] extend LLVM bindings (#89797)\n\nAdd bindings for LLVM pointer type.","shortMessageHtmlLink":"[mlir][python] extend LLVM bindings (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2259370523\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/89797\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/89797/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/89797\">llvm#89797</a>)"}},{"before":"31af5e9001508dd2e58d2232e900adba01896736","after":"0c032fd5425d853dfc577e607b9c179d811cec19","ref":"refs/heads/main","pushedAt":"2024-04-23T19:51:42.000Z","pushType":"push","commitsCount":40,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[RISCV] Use SHL_ADD in remaining strength reduce cases for MUL (#89789)\n\nThe interesting bit is the zext folding. This is the first case where we\r\nend up with a profitable fold of shNadd (zext x), y to shNadd.uw x, y.\r\nSee zext_mul68 from rv64zba.ll.\r\n\r\nThe test differences are cases where we can legally fold (only because\r\nthere's no one use check). These are not profitable or harmful, but we\r\ncan't a oneuse check without breaking the zext_mul68 case.\r\n\r\nNote that XTHeadBa doesn't appear to have the equivalent patterns so\r\nthis only shows up in Zba.","shortMessageHtmlLink":"[RISCV] Use SHL_ADD in remaining strength reduce cases for MUL (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2259268156\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/89789\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/89789/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/89789\">llvm#…</a>"}},{"before":"fe28a0e4823f59d193a14a07d12e3f60555350e0","after":"31af5e9001508dd2e58d2232e900adba01896736","ref":"refs/heads/main","pushedAt":"2024-04-23T13:07:51.000Z","pushType":"push","commitsCount":156,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"AtomicExpand: Emit or with constant on RHS\n\nThis will save later code from commuting it.","shortMessageHtmlLink":"AtomicExpand: Emit or with constant on RHS"}},{"before":"5b4ed0dc9fac9144efe74da77517dcff46e731e9","after":"fe28a0e4823f59d193a14a07d12e3f60555350e0","ref":"refs/heads/main","pushedAt":"2024-04-22T13:10:26.000Z","pushType":"push","commitsCount":132,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[LAA] Document reasoning in multiple places in isDependent (NFC). (#89381)\n\nAs suggested in https://github.com/llvm/llvm-project/pull/88039, add\r\nextra documentation for reasoning in isDependent.\r\n\r\nPR: https://github.com/llvm/llvm-project/pull/89381","shortMessageHtmlLink":"[LAA] Document reasoning in multiple places in isDependent (NFC). (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2252921358\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/89381\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/89381/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/89381\">ll…</a>"}},{"before":"0a8cd1ed1f4f35905df318015b0dbcb69d81d7c2","after":"5b4ed0dc9fac9144efe74da77517dcff46e731e9","ref":"refs/heads/main","pushedAt":"2024-04-19T18:47:33.000Z","pushType":"push","commitsCount":36,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[test] Fix test. NFC\n\n- Add the missing token in @llvm.coro.end","shortMessageHtmlLink":"[test] Fix test. NFC"}},{"before":"215eee60497489ae0cc7cc78c0d8b8270e057a70","after":"0a8cd1ed1f4f35905df318015b0dbcb69d81d7c2","ref":"refs/heads/main","pushedAt":"2024-04-19T13:15:08.000Z","pushType":"push","commitsCount":101,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[OpenMP] Use half of available logical processors for collapse tests (#88319)\n\nThe new collapse test cases define `MAX_THREADS` to be 256 and use all\r\navailable threads/logical processors on the system. This triples the\r\ntesting time on an AIX machine that has 128 logical processors. This\r\npatch changes to use half of available logical processors to avoid over\r\nsubscribing because there are other libomp tests running at the same\r\ntime, including 2 other such collapse tests.","shortMessageHtmlLink":"[OpenMP] Use half of available logical processors for collapse tests (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2236409243\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/88319\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/88319/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/88319\">l…</a>"}},{"before":"6254b6dd892728306ddfe657f10eb43a3799864d","after":"215eee60497489ae0cc7cc78c0d8b8270e057a70","ref":"refs/heads/main","pushedAt":"2024-04-18T16:13:10.000Z","pushType":"push","commitsCount":485,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"Revert \"[flang][build] Fixed paths discrovery for the out-of-tree build. (#87822)\"\n\nThis reverts commit 920298456037b9ed3ab14cb646ef6d3bf95d2c2b.","shortMessageHtmlLink":"Revert \"[flang][build] Fixed paths discrovery for the out-of-tree bui…"}},{"before":"ab037c4ff3452a680efb758ccacb6ee210ce333b","after":"6254b6dd892728306ddfe657f10eb43a3799864d","ref":"refs/heads/main","pushedAt":"2024-04-15T11:34:17.000Z","pushType":"push","commitsCount":177,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[VPlan] Version VPValue names in VPSlotTracker.  (#81411)\n\nThis patch restructures the way names for printing VPValues are handled.\r\nIt moves the logic to generate names for printing to VPSlotTracker.\r\n\r\nVPSlotTracker will now version names of the same underlying value if it\r\nis used by multiple VPValues, by adding a .V suffix to the name.\r\n\r\nThis fixes cases where at the moment the same name is printed for\r\ndifferent VPValues.\r\n\r\nPR: https://github.com/llvm/llvm-project/pull/81411","shortMessageHtmlLink":"[VPlan] Version VPValue names in VPSlotTracker. (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2128993545\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/81411\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/81411/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/81411\">llvm#81411</a>)"}},{"before":"38f8fcea3455ef1d77faf9746579f52b4e18dacc","after":"ab037c4ff3452a680efb758ccacb6ee210ce333b","ref":"refs/heads/main","pushedAt":"2024-04-12T12:21:00.000Z","pushType":"push","commitsCount":62,"pusher":{"login":"dayanruben","name":"Dayan Ruben","path":"/dayanruben","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6559692?s=80&v=4"},"commit":{"message":"[DAG] computeKnownBits - add ISD::ABDU/ISD::ABDS handling #84905 (#88253)\n\nResolve #84905","shortMessageHtmlLink":"[DAG] computeKnownBits - add ISD::ABDU/ISD::ABDS handling <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2181533450\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/84905\" data-hovercard-type=\"issue\" data-hovercard-url=\"/llvm/llvm-project/issues/84905/hovercard\" href=\"https://github.com/llvm/llvm-project/issues/84905\">llvm#84905</a> (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2235279120\" data-permission-text=\"Title is private\" data-url=\"https://github.com/llvm/llvm-project/issues/88253\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/llvm/llvm-project/pull/88253/hovercard\" href=\"https://github.com/llvm/llvm-project/pull/88253\">l…</a>"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAES-FydAA","startCursor":null,"endCursor":null}},"title":"Activity · dayanruben/llvm-project"}