Mirror intel/llvm commits #2811
Merged
Conversation
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
…class (#19387) For intel/llvm#18644 Most UR adapters had their own reference counting of some sort. This adds a new RefCount class and refactors adapter code so all adapters can share the same code for reference counting. This PR handles HIP. I have kept the pre-existing reference counting members inside stream_queue.hpp, temporarily renaming it as RefCountOld as the CUDA adapter also uses this and want to only change one adapter per PR. I will revert this once the last of the HIP and CUDA PRs are merged.
Unless something has gone terribly wrong, the compiler used to compile cpp files will be a cpp compiler.
… class (#19287) For intel/llvm#18644 Most UR adapters had their own reference counting of some sort. This adds a new RefCount class and refactors adapter code so all adapters can share the same code for reference counting. This PR handles CUDA and I will open more PRs for each adapter in turn. I have kept the pre-existing reference counting members inside `stream_queue.hpp`, temporarily renaming it as `RefCountOld` as the HIP adapter also uses this and want to only change one adapter per PR. I will revert this once the last of the HIP and CUDA PRs are merged.
Support pool query via urUSMPoolGetInfoExp for information about reserved or in-use memory and their peak sizes.
* Previously we only clean shadow once, this may cause false positive reports sometimes. So, we need to clean shadow of allocated memory before each kernel launch. * Allow internal managed queue to be created with out of order property for better performance. * Updated one test for better code coverage.
The current implementation unnecessarily cleans up the freelist when enqueueing new USM allocations. This creates a potential scenario where active allocations present in the freelist could be incorrectly freed.
The change making L0 v2 adapter default for BMG broke handling of UR_ADAPTERS_FORCE_LOAD. Depending on the platform, urAdapterGet could return 0 available adapter even if force loadded.
Unified Runtime -> intel/llvm Repo Move NoticeInformationThe source code of Unified Runtime has been moved to intel/llvm under the unified-runtime top-level directory, The code will be mirrored to oneapi-src/unified-runtime and the specification will continue to be hosted at oneapi-src.github.io/unified-runtime. The contribution guide will be updated with new instructions for contributing to Unified Runtime. PR MigrationAll open PRs including this one will be marked with the Should you wish to continue with your PR you will need to migrate it to intel/llvm. If your PR should remain open and not be closed automatically, you can remove the This is an automated comment. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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.
Automated changes by create-pull-request GitHub action