[llvm-branch-commits] [llvm] AMDGPU: Account for read/write register intrinsics for AGPR usage (PR #161988)

https://github.com/arsenm created https://github.com/llvm/llvm-project/pull/161988 Fix the special case intrinsics that can directly reference a physical register. There's no reason to use this. >From e1402218d01f9269d181d69a289e80aa68d84c01 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date:

[llvm-branch-commits] [llvm] AMDGPU: Stop inferring amdgpu-agpr-alloc on irrelevant targets (PR #161957)

llvmbot wrote: @llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) Changes This only matters for subtargets with configurable AGPR allocation. --- Patch is 159.43 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/161957.diff 25

[llvm-branch-commits] [flang] [llvm] [mlir] [Flang][MLIR][OpenMP] Add explicit shared memory (de-)allocation ops (PR #161862)

llvmbot wrote: @llvm/pr-subscribers-mlir-llvm @llvm/pr-subscribers-flang-codegen Author: Sergio Afonso (skatrak) Changes This patch introduces the `omp.alloc_shared_mem` and `omp.free_shared_mem` operations to represent explicit allocations and deallocations of shared memory across threa

[llvm-branch-commits] [llvm] [Github] Add stefanp-synopsys to release uploaders (PR #148643)

tstellar wrote: release/20.x branch is frozen now so closing this. https://github.com/llvm/llvm-project/pull/148643 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-comm

[llvm-branch-commits] [llvm] [Github] Add stefanp-synopsys to release uploaders (PR #148643)

https://github.com/tstellar closed https://github.com/llvm/llvm-project/pull/148643 ___ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

[llvm-branch-commits] [libc] [llvm] [libc][math] Refactor exp2f16 implementation to header-only in src/__support/math folder. (PR #161993)

https://github.com/bassiounix created https://github.com/llvm/llvm-project/pull/161993 None >From 8bf93446dccaff1c82aa9a3a10ca54049176 Mon Sep 17 00:00:00 2001 From: bassiounix Date: Sun, 5 Oct 2025 06:48:10 +0300 Subject: [PATCH] [libc][math] Refactor exp2f16 implementation to header-only

[llvm-branch-commits] [llvm] AMDGPU: Stop using the wavemask register class for SCC cross class copies (PR #161801)

https://github.com/rampitec approved this pull request. That's a scalar condition, so really does not depend on wave size. LGTM. https://github.com/llvm/llvm-project/pull/161801 ___ llvm-branch-commits mailing list [email protected] ht

[llvm-branch-commits] [llvm] AMDGPU: Remove LDS_DIRECT_CLASS register class (PR #161762)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/161762 >From 53a6a5b9e3adcabc51e7eff0a21642f33859b946 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Fri, 3 Oct 2025 10:21:10 +0900 Subject: [PATCH] AMDGPU: Remove LDS_DIRECT_CLASS register class This is a singlet

[llvm-branch-commits] [llvm] TableGen: Support target specialized pseudoinstructions (PR #159880)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/159880 >From 74bae94efb361527a6ad5db59e127014dc0c65c3 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Mon, 15 Sep 2025 22:41:07 +0900 Subject: [PATCH] TableGen: Support target specialized pseudoinstructions Allow a

[llvm-branch-commits] [llvm] CodeGen: Make target overrides of PointerLikeRegClass mandatory (PR #159882)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/159882 >From a8f90376a5de997c3361afe8a87b28a10c6a40a5 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Tue, 16 Sep 2025 14:54:15 +0900 Subject: [PATCH] CodeGen: Make target overrides of PointerLikeRegClass mandatory

[llvm-branch-commits] [llvm] CodeGen: Make all targets override pseudos with pointers (PR #159881)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/159881 >From 64f0e18b0875e636705e09541ca99c6665ea19b7 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Tue, 16 Sep 2025 14:55:52 +0900 Subject: [PATCH] CodeGen: Make all targets override pseudos with pointers This e

[llvm-branch-commits] [llvm] AMDGPU: Stop using aligned VGPR classes for addRegisterClass (PR #158278)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/158278 >From bad7300404b215bc299b68524cd666f6752925c7 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Fri, 12 Sep 2025 20:45:56 +0900 Subject: [PATCH] AMDGPU: Stop using aligned VGPR classes for addRegisterClass Th

[llvm-branch-commits] [llvm] AMDGPU: Remove wrapper around TRI::getRegClass (PR #159885)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/159885 >From bcaa5f77561898d1c0a83e7be20e5673b0932781 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Wed, 17 Sep 2025 21:14:02 +0900 Subject: [PATCH] AMDGPU: Remove wrapper around TRI::getRegClass This shadows the

[llvm-branch-commits] [llvm] AMDGPU: Stop using aligned VGPR classes for addRegisterClass (PR #158278)

https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/158278 >From bad7300404b215bc299b68524cd666f6752925c7 Mon Sep 17 00:00:00 2001 From: Matt Arsenault Date: Fri, 12 Sep 2025 20:45:56 +0900 Subject: [PATCH] AMDGPU: Stop using aligned VGPR classes for addRegisterClass Th