auto-detect GPU arch and adapt XSched preemption level, remove hardco…#1
Open
kylin1019 wants to merge 1 commit intoXpuOS:xschedfrom
Open
auto-detect GPU arch and adapt XSched preemption level, remove hardco…#1kylin1019 wants to merge 1 commit intoXpuOS:xschedfrom
kylin1019 wants to merge 1 commit intoXpuOS:xschedfrom
Conversation
…eleration, auto-detects GPU architecture, adapts preemption level, and runs stably without CUDA errors
Collaborator
|
Hi @kylin1019, Thank you for your contribution — your manual fix is indeed correct. As you also mentioned in XpuOS/xsched#23, automatic XQueue creation may be an even cleaner and more efficient way to integrate XSched into llama.cpp. We are therefore considering adding more APIs to xsched so that an XQueueHandle can be obtained directly from a cudaStream. This would allow us to remove the stream construction logic and the mapping between cudaStream and XQueueHandle. Also, for consistency and easier collaboration, please use English for code comments in future updates. related to XpuOS/xsched#23 |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Enable automatic GPU architecture detection and dynamic XSched preemption level adaptation for llama.cpp. Removed hardcoded preemption levels to support full compatibility across NVIDIA GPU platforms and eliminate runtime crashes.