[Bugfix]: Fix concurrency edge case in Gaussian - #183
Open
daniellegillai wants to merge 2 commits into
Open
Conversation
1 task
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
The current implementation of Gaussian has a concurrency edge case: if the operator is called on GPU, and then the same operator object is called on CPU, it is possible for the CPU-side call to start modifying the CPU memory held in
m_hostKernelMembefore the GPU-side call has finished copying it over, asynchronously, to GPU memory.Also added a note to the header file clarifying the enforced serialization of this operator.
Technical Details
Call
hipEventSynchronize(m_completionEvent)on both backends (rather than just GPU), so that a CPU call waits for any previous GPU-side operator call to complete its asynchronous GPU work (including the mem copy) before touching the CPU memory.Test Plan
rocCV CTests tests.
Test Result
All CTests pass.