[llvm-branch-commits] [llvm] [LoongArch] Custom legalize vector_shuffle to `[x]vpermi.w` (PR #164945)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 186429 tests passed * 4868 tests skipped https://github.com/llvm/llvm-project/pull/164945 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/c

[llvm-branch-commits] [llvm] [LoongArch] Custom legalize vector_shuffle to `[x]vpermi.w` (PR #164945)

https://github.com/zhaoqi5 updated https://github.com/llvm/llvm-project/pull/164945 >From f38bb557b1408dcd502b73f586cefe802b11738a Mon Sep 17 00:00:00 2001 From: Qi Zhao Date: Fri, 24 Oct 2025 17:00:29 +0800 Subject: [PATCH 1/2] [LoongArch] Custom legalize vector_shuffle to `[x]vpermi.w` ---

[llvm-branch-commits] [llvm] [dwarf] make dwarf fission compatible with RISCV relaxations 2/2 (PR #164813)

https://github.com/lenary approved this pull request. LGTM https://github.com/llvm/llvm-project/pull/164813 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

https://github.com/vikramRH edited https://github.com/llvm/llvm-project/pull/168833 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

@@ -527,14 +534,16 @@ void SIShrinkInstructions::shrinkMadFma(MachineInstr &MI) const { TII->removeModOperands(MI); MI.setDesc(TII->get(NewOpcode)); } + return true; } /// Attempt to shrink AND/OR/XOR operations requiring non-inlineable literals. /// For AND or

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

@@ -241,27 +241,30 @@ void SIShrinkInstructions::copyExtraImplicitOps(MachineInstr &NewMI, } } -void SIShrinkInstructions::shrinkScalarCompare(MachineInstr &MI) const { +bool SIShrinkInstructions::shrinkScalarCompare(MachineInstr &MI) const { if (!ST->hasSCmpK()) -re

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

https://github.com/vikramRH updated https://github.com/llvm/llvm-project/pull/168833 >From e1a745b13ea3c3784b67f25bd54a8c32334194f2 Mon Sep 17 00:00:00 2001 From: vikhegde Date: Thu, 20 Nov 2025 11:18:25 +0530 Subject: [PATCH 1/2] [AMDGPU] Make SIShrinkInstructions pass return valid changed st

[llvm-branch-commits] [llvm] [NPM] Schedule PhysicalRegisterUsageAnalysis before RegUsageInfoCollectorPass (PR #168832)

https://github.com/vikramRH updated https://github.com/llvm/llvm-project/pull/168832 >From a773367ede893b3806df8cdd8a6cadf9b7b08889 Mon Sep 17 00:00:00 2001 From: vikhegde Date: Tue, 18 Nov 2025 11:13:37 +0530 Subject: [PATCH] [NPM] Schedule PhysicalRegisterUsageAnalysis before RegUsageInfoCol

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

https://github.com/vikramRH updated https://github.com/llvm/llvm-project/pull/168833 >From e1a745b13ea3c3784b67f25bd54a8c32334194f2 Mon Sep 17 00:00:00 2001 From: vikhegde Date: Thu, 20 Nov 2025 11:18:25 +0530 Subject: [PATCH 1/2] [AMDGPU] Make SIShrinkInstructions pass return valid changed st

[llvm-branch-commits] [llvm] [NPM] Schedule PhysicalRegisterUsageAnalysis before RegUsageInfoCollectorPass (PR #168832)

https://github.com/vikramRH updated https://github.com/llvm/llvm-project/pull/168832 >From a773367ede893b3806df8cdd8a6cadf9b7b08889 Mon Sep 17 00:00:00 2001 From: vikhegde Date: Tue, 18 Nov 2025 11:13:37 +0530 Subject: [PATCH] [NPM] Schedule PhysicalRegisterUsageAnalysis before RegUsageInfoCol

[llvm-branch-commits] [llvm] [LoongArch] Perform DAG combine for MUL to generate `[x]vmulw{ev/od}` (PR #161368)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 186429 tests passed * 4868 tests skipped https://github.com/llvm/llvm-project/pull/161368 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/c

[llvm-branch-commits] [llvm] AMDGPU: Fix treating unknown mem operands as uniform (PR #168980)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 186432 tests passed * 4864 tests skipped https://github.com/llvm/llvm-project/pull/168980 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/c

[llvm-branch-commits] [llvm] [AMDGPU] Allow hazard checks for WMMA co-exec (PR #168805)

https://github.com/arsenm approved this pull request. https://github.com/llvm/llvm-project/pull/168805 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] VectorCombine: Improve the insert/extract fold in the narrowing case (PR #168820)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168820 From b43293a52737145abd98c771156b633ab3e94167 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 18:00:32 -0800 Subject: [PATCH] VectorCombine: Improve the insert/extract fo

[llvm-branch-commits] [llvm] VectorCombine: Improve the insert/extract fold in the narrowing case (PR #168820)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168820 From b43293a52737145abd98c771156b633ab3e94167 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 18:00:32 -0800 Subject: [PATCH] VectorCombine: Improve the insert/extract fo

[llvm-branch-commits] [llvm] VectorCombine: Improve the insert/extract fold in the narrowing case (PR #168820)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168820 From b43293a52737145abd98c771156b633ab3e94167 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 18:00:32 -0800 Subject: [PATCH] VectorCombine: Improve the insert/extract fo

[llvm-branch-commits] [llvm] VectorCombine: Fold chains of shuffles fed by length-changing shuffles (PR #168819)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168819 From aaee5f6315322448fbeaba415aaef288a1faeb66 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 17:59:11 -0800 Subject: [PATCH] VectorCombine: Fold chains of shuffles fed b

[llvm-branch-commits] [llvm] VectorCombine: Fold chains of shuffles fed by length-changing shuffles (PR #168819)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168819 From aaee5f6315322448fbeaba415aaef288a1faeb66 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 17:59:11 -0800 Subject: [PATCH] VectorCombine: Fold chains of shuffles fed b

[llvm-branch-commits] [llvm] VectorCombine: Fold chains of shuffles fed by length-changing shuffles (PR #168819)

https://github.com/nhaehnle updated https://github.com/llvm/llvm-project/pull/168819 From aaee5f6315322448fbeaba415aaef288a1faeb66 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Wed, 19 Nov 2025 17:59:11 -0800 Subject: [PATCH] VectorCombine: Fold chains of shuffles fed b

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168914 >From 4be9e5ba071d75258949617539d352aa2d2b5171 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:14:33 -0500 Subject: [PATCH] AMDGPU: Handle invariant when lowering global loads Global with

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168914 >From 4be9e5ba071d75258949617539d352aa2d2b5171 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:14:33 -0500 Subject: [PATCH] AMDGPU: Handle invariant when lowering global loads Global with

[llvm-branch-commits] [clang] [AMDGPU] Removal of language sensitive option for _Float16 and half( 'e') handling (PR #168037)

https://github.com/ranapratap55 closed https://github.com/llvm/llvm-project/pull/168037 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 3053 tests passed * 7 tests skipped https://github.com/llvm/llvm-project/pull/168972 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bi

[llvm-branch-commits] [compiler-rt] cc5837d - Revert "[UBSan] [compiler-rt] add preservecc variants of handlers (#168643)"

Author: Florian Mayer Date: 2025-11-20T15:59:54-08:00 New Revision: cc5837d4fad0578ed67bebc32555766f2364b7e4 URL: https://github.com/llvm/llvm-project/commit/cc5837d4fad0578ed67bebc32555766f2364b7e4 DIFF: https://github.com/llvm/llvm-project/commit/cc5837d4fad0578ed67bebc32555766f2364b7e4.diff

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

@@ -1590,21 +1590,31 @@ parseBoundsCheckingOptions(StringRef Params) { Options.Rt = { /*MinRuntime=*/false, /*MayReturn=*/true, + /*HandlerPreserveAllRegs=*/false, fmayer wrote: Done https://github.com/llvm/llvm-project/pull

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

evelez7 wrote: See HTML output here: https://erickvelez.com/clang-doc-mustache-output/pr168972/GlobalNamespace/ https://github.com/llvm/llvm-project/pull/168972 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

llvmbot wrote: @llvm/pr-subscribers-clang-tools-extra Author: Erick Velez (evelez7) Changes The HTML spec states that only `

` can be children of `

`. Nested `
` tags in an unordered list must be children of `
• `. --- Full diff: https://github.com/llvm/llvm-project/pull/168972

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

https://github.com/evelez7 ready_for_review https://github.com/llvm/llvm-project/pull/168972 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

evelez7 wrote: > [!WARNING] > This pull request is not mergeable via GitHub because a downstack PR is > open. Once all requirements are satisfied, merge this PR as a stack href="https://app.graphite.com/github/pr/llvm/llvm-project/168972?utm_source=stack-comment-downstack-mergeability-warning";

[llvm-branch-commits] [clang-tools-extra] [clang-doc] `` must be nested in `` (PR #168972)

https://github.com/evelez7 created https://github.com/llvm/llvm-project/pull/168972 The HTML spec states that only `` can be children of ``. Nested `` tags in an unordered list must be children of ``. >From 627d5a5b0b25e3b4b8b0ddcc92cd6f3de03ff335 Mon Sep 17 00:00:00 2001 From: Erick Velez Dat

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168645 >From 54f9084ddba1711dd20ff231605e7418cf4f19cf Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Tue, 18 Nov 2025 17:39:33 -0800 Subject: [PATCH 1/6] fmt Created using spr 1.3.7 --- clang/lib/CodeGen/BackendUt

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168645 >From 54f9084ddba1711dd20ff231605e7418cf4f19cf Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Tue, 18 Nov 2025 17:39:33 -0800 Subject: [PATCH 1/6] fmt Created using spr 1.3.7 --- clang/lib/CodeGen/BackendUt

[llvm-branch-commits] [clang] [UBsan] add -fsanitize-handler-preserve-all-regs flag (PR #168644)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168644 >From d2231031d7ecfffc97709d6bf2236ead4147aab8 Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Wed, 19 Nov 2025 18:01:59 -0800 Subject: [PATCH 1/3] rename Created using spr 1.3.7 --- clang/include/clang/Basi

[llvm-branch-commits] [clang] [UBsan] add -fsanitize-handler-preserve-all-regs flag (PR #168644)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168644 >From d2231031d7ecfffc97709d6bf2236ead4147aab8 Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Wed, 19 Nov 2025 18:01:59 -0800 Subject: [PATCH 1/3] rename Created using spr 1.3.7 --- clang/include/clang/Basi

[llvm-branch-commits] [llvm] [AMDGPU] Allow hazard checks for WMMA co-exec (PR #168805)

rampitec wrote: Testcase added. https://github.com/llvm/llvm-project/pull/168805 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [AMDGPU] Allow hazard checks for WMMA co-exec (PR #168805)

llvmbot wrote: @llvm/pr-subscribers-backend-amdgpu Author: Stanislav Mekhanoshin (rampitec) Changes Now we are just inserting V_NOP instrtuctions, try to schedule something into the shadow. It is still somewhat imprecise, for example AdvanceCycle() will use TII.getNumWaitStates() anyway,

[llvm-branch-commits] [llvm] [AMDGPU] Allow hazard checks for WMMA co-exec (PR #168805)

https://github.com/rampitec ready_for_review https://github.com/llvm/llvm-project/pull/168805 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] [AMDGPU] Allow hazard checks for WMMA co-exec (PR #168805)

https://github.com/rampitec updated https://github.com/llvm/llvm-project/pull/168805 >From e602eabcd42a393bd2dc0ea66ef950b6158bb61f Mon Sep 17 00:00:00 2001 From: Stanislav Mekhanoshin Date: Mon, 17 Nov 2025 14:47:40 -0800 Subject: [PATCH] [AMDGPU] Allow hazard checks for WMMA co-exec Now we a

[llvm-branch-commits] [clang] [llvm] [PowerPC] Add AMO load signed builtins (PR #168747)

https://github.com/maryammo updated https://github.com/llvm/llvm-project/pull/168747 >From 962119bdda50304f82ffb4569a262594945951cd Mon Sep 17 00:00:00 2001 From: Maryam Moghadas Date: Wed, 19 Nov 2025 16:46:34 + Subject: [PATCH 1/3] [PowerPC] Add AMO load signed builtins This commit adds

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kparzysz wrote: I guess another point would be that there is actually no ambiguity here. The unstructured loops could be parsed as a whole, but perhaps it's simpler or more convenient to parse them as a sequence of statements. They cannot span construct boundaries (even in F77), so in that se

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

https://github.com/vitalybuka approved this pull request. https://github.com/llvm/llvm-project/pull/168645 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

@@ -1590,21 +1590,31 @@ parseBoundsCheckingOptions(StringRef Params) { Options.Rt = { /*MinRuntime=*/false, /*MayReturn=*/true, + /*HandlerPreserveAllRegs=*/false, vitalybuka wrote: These ifs do not scale well :) https://git

[llvm-branch-commits] [clang] [UBsan] add -fsanitize-handler-preserve-all-regs flag (PR #168644)

https://github.com/vitalybuka approved this pull request. https://github.com/llvm/llvm-project/pull/168644 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [clang] [UBsan] add -fsanitize-handler-preserve-all-regs flag (PR #168644)

@@ -2656,6 +2656,11 @@ defm sanitize_minimal_runtime : BoolOption<"f", "sanitize-minimal-runtime", PosFlag, NegFlag>, Group; +defm sanitize_handler_preserve_all_regs +: BoolOption<"f", "sanitize-handler-preserve-all-regs", + CodeGenOpts<"SanitizeHandl

[llvm-branch-commits] [clang] [AMDGPU] Removal of language sensitive option for _Float16 and half( 'e') handling (PR #168037)

https://github.com/ranapratap55 edited https://github.com/llvm/llvm-project/pull/168037 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168914 >From 954dc93ab5509d84007758c688a0cfb499308bc5 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:14:33 -0500 Subject: [PATCH] AMDGPU: Handle invariant when lowering global loads Global with

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168914 >From 954dc93ab5509d84007758c688a0cfb499308bc5 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:14:33 -0500 Subject: [PATCH] AMDGPU: Handle invariant when lowering global loads Global with

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168917 Unicorn! · GitHub body { background-color: #f1f1f1; margin: 0; font-family: "Helvetica Neue", Helvetica, Arial, sans-serif; } .container { margin: 50px aut

[llvm-branch-commits] [clang] [AMDGPU] Removal of 'e' handling (PR #168037)

https://github.com/shiltian approved this pull request. LGTM, with the fix @arsenm suggested. https://github.com/llvm/llvm-project/pull/168037 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailma

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168645 >From 54f9084ddba1711dd20ff231605e7418cf4f19cf Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Tue, 18 Nov 2025 17:39:33 -0800 Subject: [PATCH 1/3] fmt Created using spr 1.3.7 --- clang/lib/CodeGen/BackendUt

[llvm-branch-commits] [clang] [llvm] [UBSan] Use -fsanitize-handler-preserve-all-regs in codegen (PR #168645)

https://github.com/fmayer updated https://github.com/llvm/llvm-project/pull/168645 >From 54f9084ddba1711dd20ff231605e7418cf4f19cf Mon Sep 17 00:00:00 2001 From: Florian Mayer Date: Tue, 18 Nov 2025 17:39:33 -0800 Subject: [PATCH 1/2] fmt Created using spr 1.3.7 --- clang/lib/CodeGen/BackendUt

[llvm-branch-commits] [clang] [AMDGPU] Removal of 'e' handling (PR #168037)

arsenm wrote: Can you fix the title to be clear about what e means? It means nothing without a lot more context https://github.com/llvm/llvm-project/pull/168037 ___ llvm-branch-commits mailing list [email protected] https://lists.llv

[llvm-branch-commits] [clang] [AMDGPU] Removal of 'e' handling (PR #168037)

llvmbot wrote: @llvm/pr-subscribers-clang Author: Rana Pratap Reddy (ranapratap55) Changes Removing the 'e' handling for the amdgcn builtins as we decided to use _Float16 for both HIP/C++ and OpenCL --- Patch is 53.61 KiB, truncated to 20.00 KiB below, full version: https://github.com

[llvm-branch-commits] [clang] [AMDGPU] Removal of 'e' handling (PR #168037)

https://github.com/ranapratap55 ready_for_review https://github.com/llvm/llvm-project/pull/168037 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kparzysz wrote: > Is the point that unlike in Fortran 77 loops, there is no ambiguity in > parsing OpenMP loop constructs and hence we can parse it all together? I guess the point would be that we preserve the "ambiguity", just wrap it in an OpenMP construct. The unstructured loops are simply

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 186426 tests passed * 4867 tests skipped https://github.com/llvm/llvm-project/pull/168917 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/c

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 186426 tests passed * 4867 tests skipped https://github.com/llvm/llvm-project/pull/168914 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/c

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kiranchandramohan wrote: Is the point that unlike in Fortran 77 loops, there is no ambiguity in parsing OpenMP loop constructs and hence we can parse it all together? https://github.com/llvm/llvm-project/pull/168884 ___ llvm-branch-commits mailing lis

[llvm-branch-commits] [llvm] AMDGPU: Add baseline test for split/widen invariant loads (PR #168913)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 166204 tests passed * 2852 tests skipped * 1 test failed ## Failed Tests (click on a test name to see its output) ### LLVM LLVM.CodeGen/AMDGPU/load-global-invariant.ll ``` Exit Code: 1 Command Output (stdout): -- # RUN: at lin

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

llvmbot wrote: @llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) Changes This isn't quite a constant pool, but probably close enough for this purpose. We just need some known invariant value address. The aliasing queries against the real kernarg base pointer will falsely

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

https://github.com/arsenm ready_for_review https://github.com/llvm/llvm-project/pull/168917 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

arsenm wrote: > [!WARNING] > This pull request is not mergeable via GitHub because a downstack PR is > open. Once all requirements are satisfied, merge this PR as a stack href="https://app.graphite.com/github/pr/llvm/llvm-project/168917?utm_source=stack-comment-downstack-mergeability-warning";

[llvm-branch-commits] [llvm] AMDGPU: Use ConstantPool as source value for DAG lowered kernarg loads (PR #168917)

https://github.com/arsenm created https://github.com/llvm/llvm-project/pull/168917 This isn't quite a constant pool, but probably close enough for this purpose. We just need some known invariant value address. The aliasing queries against the real kernarg base pointer will falsely report no alia

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kparzysz wrote: > Would this work fine for loops like the following? > > ``` > !$omp do > DO 10, I = 1, N >... loop body statements ... > 10A(I) = B(I) + C(I) > ``` Yes. ``` subroutine f(a, b, c) integer :: i, n integer :: a(:), b(:), c(:) !$omp do DO 10, I = 1, N continue

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

llvmbot wrote: @llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) Changes Global with invariant should be treated identically to constant. --- Full diff: https://github.com/llvm/llvm-project/pull/168914.diff 2 Files Affected: - (modified) llvm/lib/Target/AMDGPU/SIISelL

[llvm-branch-commits] [llvm] AMDGPU: Add baseline test for split/widen invariant loads (PR #168913)

llvmbot wrote: @llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) Changes --- Full diff: https://github.com/llvm/llvm-project/pull/168913.diff 1 Files Affected: - (added) llvm/test/CodeGen/AMDGPU/load-global-invariant.ll (+70) ``diff diff --git a/llvm/test/

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm ready_for_review https://github.com/llvm/llvm-project/pull/168914 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] AMDGPU: Add baseline test for split/widen invariant loads (PR #168913)

https://github.com/arsenm ready_for_review https://github.com/llvm/llvm-project/pull/168913 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

arsenm wrote: > [!WARNING] > This pull request is not mergeable via GitHub because a downstack PR is > open. Once all requirements are satisfied, merge this PR as a stack href="https://app.graphite.com/github/pr/llvm/llvm-project/168914?utm_source=stack-comment-downstack-mergeability-warning";

[llvm-branch-commits] [llvm] AMDGPU: Add baseline test for split/widen invariant loads (PR #168913)

arsenm wrote: > [!WARNING] > This pull request is not mergeable via GitHub because a downstack PR is > open. Once all requirements are satisfied, merge this PR as a stack href="https://app.graphite.com/github/pr/llvm/llvm-project/168913?utm_source=stack-comment-downstack-mergeability-warning";

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant when lowering global loads (PR #168914)

https://github.com/arsenm created https://github.com/llvm/llvm-project/pull/168914 Global with invariant should be treated identically to constant. >From a406bd24da1125b6683e7524ace5b1e33d905822 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:14:33 -0500 Subject: [PATCH

[llvm-branch-commits] [llvm] AMDGPU: Add baseline test for split/widen invariant loads (PR #168913)

https://github.com/arsenm created https://github.com/llvm/llvm-project/pull/168913 None >From dd3f33991462d8a61a7ae7d71426bc7c9398f1ab Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Thu, 20 Nov 2025 12:09:35 -0500 Subject: [PATCH] AMDGPU: Add baseline test for split/widen invariant loads

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kiranchandramohan wrote: Would this work fine for loops like the following? ``` !$omp do DO 10, I = 1, N ... loop body statements ... 10A(I) = B(I) + C(I) ``` https://github.com/llvm/llvm-project/pull/168884 ___ llvm-branch-commits mailing list

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kiranchandramohan wrote: I have not followed the full patch but the two step approach was probably necessary since the Fortran loop parser itself operated in that way. Would changing this only for OpenMP loops cause a problem? https://github.com/llvm/llvm-project/pull/168884 __

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

https://github.com/tblah approved this pull request. LGTM, thanks https://github.com/llvm/llvm-project/pull/168884 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commi

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

https://github.com/kparzysz updated https://github.com/llvm/llvm-project/pull/168885 >From 72c046f755c26e7696ccff0ba4e25701293be726 Mon Sep 17 00:00:00 2001 From: Krzysztof Parzyszek Date: Wed, 19 Nov 2025 14:41:07 -0600 Subject: [PATCH 1/2] [flang][OpenMP] Better diagnostics for invalid or mi

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

@@ -389,6 +394,7 @@ class OmpStructureChecker : public OmpStructureCheckerBase { std::vector loopStack_; // Scopes for scoping units. std::vector scopeStack_; + bool invalidState_{false}; // Set during visiting OpenMPMisplacedEndDirective kparzysz wrote:

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

@@ -74,8 +81,6 @@ do i = 1, N enddo !$omp end parallel do - !ERROR: The END PARALLEL DO directive must follow the DO loop associated with the loop construct - !$omp end parallel do kparzysz wrote: This patch deletes the pre-existing handling of an end

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

@@ -74,8 +81,6 @@ do i = 1, N enddo !$omp end parallel do - !ERROR: The END PARALLEL DO directive must follow the DO loop associated with the loop construct - !$omp end parallel do tblah wrote: Why was this removed? Are all of these re-added in the fo

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

https://github.com/tblah edited https://github.com/llvm/llvm-project/pull/168885 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

https://github.com/tblah approved this pull request. LGTM with one nit https://github.com/llvm/llvm-project/pull/168885 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

@@ -389,6 +394,7 @@ class OmpStructureChecker : public OmpStructureCheckerBase { std::vector loopStack_; // Scopes for scoping units. std::vector scopeStack_; + bool invalidState_{false}; // Set during visiting OpenMPMisplacedEndDirective tblah wrote: W

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant loads when considering if a load can be scalar (PR #168787)

https://github.com/jayfoad approved this pull request. This is saying that invariant loads can be selected to SMEM instructions regardless of any "no clobber" check, right? I think that's safe... https://github.com/llvm/llvm-project/pull/168787 ___ ll

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant loads when considering if a load can be scalar (PR #168787)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168787 >From cd383966fb4d7d56ac5a61f37c4970d6c232c7cc Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Wed, 19 Nov 2025 17:34:24 -0500 Subject: [PATCH] AMDGPU: Handle invariant loads when considering if a load can b

[llvm-branch-commits] [llvm] AMDGPU: Handle invariant loads when considering if a load can be scalar (PR #168787)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/168787 >From cd383966fb4d7d56ac5a61f37c4970d6c232c7cc Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Wed, 19 Nov 2025 17:34:24 -0500 Subject: [PATCH] AMDGPU: Handle invariant loads when considering if a load can b

[llvm-branch-commits] [clang] [llvm] [PowerPC] Add AMO load signed builtins (PR #168747)

https://github.com/maryammo edited https://github.com/llvm/llvm-project/pull/168747 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 4063 tests passed * 202 tests skipped https://github.com/llvm/llvm-project/pull/168884 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

github-actions[bot] wrote: # :penguin: Linux x64 Test Results * 4063 tests passed * 202 tests skipped https://github.com/llvm/llvm-project/pull/168885 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

kparzysz wrote: PR stack: 1. https://github.com/llvm/llvm-project/pull/168884 2. https://github.com/llvm/llvm-project/pull/168885 (this PR) https://github.com/llvm/llvm-project/pull/168885 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

kparzysz wrote: PR stack: 1. https://github.com/llvm/llvm-project/pull/168884 (this PR) 2. https://github.com/llvm/llvm-project/pull/168885 https://github.com/llvm/llvm-project/pull/168884 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

https://github.com/kparzysz updated https://github.com/llvm/llvm-project/pull/168884 >From aaea8e6b47808333581569d40e3721050be52bf1 Mon Sep 17 00:00:00 2001 From: Krzysztof Parzyszek Date: Fri, 14 Nov 2025 15:52:11 -0600 Subject: [PATCH 1/2] [flang][OpenMP] Implement loop nest parser Previousl

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

github-actions[bot] wrote: :warning: C/C++ code formatter, clang-format found issues in your code. :warning: You can test this locally with the following command: ``bash git-clang-format --diff origin/main HEAD --extensions h,cpp -- flang/include/flang/Parser/parse-tree.h flang/

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

llvmbot wrote: @llvm/pr-subscribers-flang-parser Author: Krzysztof Parzyszek (kparzysz) Changes Add two more AST nodes, one for a misplaced end-directive, and one for an invalid string following the OpenMP sentinel (e.g. "!$OMP XYZ"). Emit error messages when either node is encountered i

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

llvmbot wrote: @llvm/pr-subscribers-flang-fir-hlfir Author: Krzysztof Parzyszek (kparzysz) Changes Previously, loop constructs were parsed in a piece-wise manner: the begin directive, the body, and the end directive were parsed separately. Later on in canonicalization they were all coale

[llvm-branch-commits] [flang] [flang][OpenMP] Better diagnostics for invalid or misplaced directives (PR #168885)

https://github.com/kparzysz created https://github.com/llvm/llvm-project/pull/168885 Add two more AST nodes, one for a misplaced end-directive, and one for an invalid string following the OpenMP sentinel (e.g. "!$OMP XYZ"). Emit error messages when either node is encountered in semantic analys

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

llvmbot wrote: @llvm/pr-subscribers-flang-openmp Author: Krzysztof Parzyszek (kparzysz) Changes Previously, loop constructs were parsed in a piece-wise manner: the begin directive, the body, and the end directive were parsed separately. Later on in canonicalization they were all coalesce

[llvm-branch-commits] [flang] [flang][OpenMP] Implement loop nest parser (PR #168884)

https://github.com/kparzysz created https://github.com/llvm/llvm-project/pull/168884 Previously, loop constructs were parsed in a piece-wise manner: the begin directive, the body, and the end directive were parsed separately. Later on in canonicalization they were all coalesced into a loop con

[llvm-branch-commits] [llvm] [LV] Mask off possibly aliasing vector lanes (PR #100579)

@@ -8974,11 +8982,104 @@ void LoopVectorizationPlanner::attachRuntimeChecks( assert((!CM.OptForSize || CM.Hints->getForce() == LoopVectorizeHints::FK_Enabled) && "Cannot SCEV check stride or overflow when optimizing for size"); -VPlanTransforms::a

[llvm-branch-commits] [llvm] [LV] Mask off possibly aliasing vector lanes (PR #100579)

@@ -702,6 +703,29 @@ Value *VPInstruction::generate(VPTransformState &State) { {PredTy, ScalarTC->getType()}, {VIVElem0, ScalarTC}, nullptr, Name); } + // Count the number of bits set in each lane and redu

[llvm-branch-commits] [llvm] [AMDGPU] Make SIShrinkInstructions pass return valid changed state (PR #168833)

https://github.com/vikramRH edited https://github.com/llvm/llvm-project/pull/168833 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

  1   2   >