This repository has been archived by the owner on Dec 15, 2022. It is now read-only.
Synchronize deletion of mInterface in NSFW destructor #5
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
We're seeing intermittent crashes on Windows calling mInterface->isWatching() in StartWorker::Execute on a background thread. The only unsynchronized access to mInterface that we can find takes place in the destructor of NSFW itself, making us wonder if interleaved calls to start and stop might be causing the following race condition. * Start is called and sets the persistent handle * Stop is called * Before the StopWorker's HandleOkCallback runs and clears the persistent handle, we call start again * Start spawns a worker thread * Stop's ok callback clears the persistent handle * The garbage collector runs and runs the destructor on the NSFW object, which clears the mInterface field * The thread accesses the mInterface pointer, which has now been deleted We don't have proof that this is what is happening, but if it is, I'm hoping that locking the field access in the destructor would prevent it.
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Outdated suggestions cannot be applied.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.
We're seeing intermittent crashes on Windows calling
mInterface->isWatching()inStartWorker::Executeon a background thread. The only unsynchronized access to mInterface that we can find takes place in the destructor of NSFW itself, making us wonder if interleaved calls to start and stop might be causing the following race condition.NSFWfrom being garbage collectedHandleOkCallbackruns and clears thepersistent handle, we call start again
NSFWfree to be collected as garbageNSFWobject,which clears the
mInterfacefieldStartWorkeraccesses themInterfacepointer on a background thread, which now points to a deleted objectWe don't have proof that this is what is happening, but if it is, I'm hoping that locking the field access in the destructor would prevent it.
I want to do some more local testing on Windows to see if I can reproduce the crash again without this change, then spend 2-3x more time running with this change to see if it avoids the crash. 😓
cc @as-cii @rafeca