Enhancement - Support AMDGPU_TARGETS in ROCm CMake builds#793
Open
Enhancement - Support AMDGPU_TARGETS in ROCm CMake builds#793
Conversation
polarG
added
enhancement
Contributor
There was a problem hiding this comment.
Pull request overview
Adds support for driving ROCm/HIP GPU architecture selection from the AMDGPU_TARGETS environment variable during CMake configuration, to make ROCm micro-benchmark builds work on hosts where hipcc GPU auto-detection is unavailable (e.g., CI without a GPU).
Changes:
- Read
AMDGPU_TARGETSfrom the environment and convert it to a CMake list for use during configuration. - Cache
AMDGPU_TARGETSand (for CMake ≥ 3.21) setCMAKE_HIP_ARCHITECTURESwhen the env var is provided. - Preserve the prior “auto-detection” path when
AMDGPU_TARGETSis not set, and add status logging.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #793 +/- ##
=======================================
Coverage 85.69% 85.69%
=======================================
Files 103 103
Lines 7890 7890
=======================================
Hits 6761 6761
Misses 1129 1129
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
added 2 commits
April 27, 2026 21:49
- Clarify comment: micro-benchmarks are CXX-only via hipcc; CMAKE_HIP_ARCHITECTURES is set only for compatibility with HIP-language projects, not required here. - Normalize whitespace via STRIP + REGEX REPLACE [ \t]+ to avoid empty list elements from leading/trailing or repeated whitespace. - Add FORCE to the cache set so a changed AMDGPU_TARGETS env value overrides any stale cached value on reconfigure. - Replace DEFINED ENV check with non-empty HIP_ARCH_LIST check so an empty or whitespace-only AMDGPU_TARGETS falls back to hipcc auto-detection.
- Broaden the separator regex to accept whitespace, ',', and ';' so values like 'gfx908,gfx90a' or CMake-list 'gfx908;gfx90a' are parsed correctly instead of becoming a single bogus token. - Drop CACHE STRING ... FORCE: use a normal directory-scoped variable for AMDGPU_TARGETS so we do not pollute the global CMake cache or override the variable in nested projects. The env var is re-read on every reconfigure, so this still wins over stale state.
Contributor
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Allow specifying AMD GPU architectures via the AMDGPU_TARGETS environment variable in rocm_common.cmake. This fixes builds on machines without a GPU (e.g., CI) where hipcc auto-detection fails. When unset, the previous auto-detection behavior is preserved.