This repository has been archived by the owner on Mar 21, 2024. It is now read-only.
Clear CUDA global error state after CUDA API calls #1020
Merged
Conversation
This file contains 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
…e by calling cudaGetLastError. Otherwise, if the CUDA API call is followed directly by a kernel launch, checking for a synchronous error during the kernel launch by calling cudaGetLastError may potentially return the error code from the CUDA API call. This type of error leakage is very subtle and difficult to trace. Bug 2720132
|
So... isn't there a danger of silently ignoring an asynchronous error if we do this this way? Shouldn't we actually be checking the return value of the |
|
No. Asynchronous errors are sticky; they cannot be cleared. |
|
This passed internal CI. Accepting. |
brycelelbach
deleted the
bug/nvbug-2720132-clear-global-cuda-error-state
branch
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
After making a CUDA API call, always clear the global CUDA error state by calling
cudaGetLastError. Otherwise, if the CUDA API call is followed directly by a kernel launch, checking for a synchronous error during the kernel launch by callingcudaGetLastErrormay potentially return the error code from the CUDA API call. This type of error leakage is very subtle and difficult to trace.This is a fix for Bug 2720132.