Repository navigation
Conversation
…n flags `CMakeFactory.__call__` appended `-DCMAKE_CUDA_ARCHITECTURES=...` to `self.cmake_extra_flags` on every call for a GPU source, so the flag list grew with each program and the factory's fingerprint changed after the first call. Through `CompiledbFactory`, which passes its own list when it is not empty, the same could happen to the factory that names the program cache folders. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The DaCe backend has the device architecture in its build context id via `compiler.cuda.cuda_arch`; GTFN did not, so a folder built for another architecture (or without `CUDAARCHS`) was reused. `build_context_id` now adds the architecture for CUDA/HIP sources, resolved when the program is compiled, and the compiledb prototype name carries the architecture flag. CPU folder names are unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2 of 3 tasks
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
GTFN GPU program folders, and the compiledb prototype, are now keyed by the device architecture, as DaCe's already are via
compiler.cuda.cuda_arch. Before this, a folder built for another architecture was reused. CPU folder names are unchanged. Also fixesCMakeFactoryappending the architecture flag to its owncmake_extra_flagson every call, which changed its fingerprint after the first GPU program.Requirements
If this PR contains code authored by new contributors please make sure:
AUTHORS.mdfile adding the names of all the new contributors.🤖 Generated with Claude Code