google / google/gapid

The logic to getting a queue for image data recovering and priming is wrong in some cases

Open
#1,700 0 comments 0 reactions 0 assignees View on GitHub
triaged
Dominant language
Go
Stars
2.2k
Forks
329
PR merge metrics
No merged PRs in 30d

Description

For MEC, at the starting point of capturing, If the image is never used, its `LastBoundQueue` pointer will be nil. But we randomly find a queue for it to recover its data in the trace side. This may change the ownership of the image in later execution of the app, and may cause an issue.

On the replay side we have a similar issue.

The right way to fix the replay side queue selection is: Walk through the following commands to see where the image is first used, prime it on THAT queue.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.