driver::op::submit_with() panic
#145
Comments
ollie-etl
pushed a commit
to ollie-etl/tokio-uring
that referenced
this issue
Oct 24, 2022
Merged
Noah-Kennedy
pushed a commit
that referenced
this issue
Oct 24, 2022
Fixes #145. A data race was observed in Pr #144, where with high concurrency in user-space, and single threaded sqpoll in the ring, we could trigger a panic when submitting an entry to the queue The speculation is that the submission queue is full, but no ops have yet been executed and placed in completion queue by the ring. The submit call therefore submits the queue, but doesn't free any sqe's . The fix, which is not very elegant, does busy-polling on full. Co-authored-by: ollie-etl <Oliver Bunting@etlsystems.com>
Oh, you are already running with sqpoll. Very interesting. I haven't tried that and have no idea what problems you may be running into there. Well I guess with the commit from yesterday it's working for you. I wonder if you have something trying to call tick periodically to try and keep up with the kernel's results or if tick is only being called when the thread is about to park itself? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Pr #144 can trigger This panic.
It can be reliably triggered if:
.setup_sqpoll()is specified on the ring builderI speculate this can occur under other conditions, but the above is an easy way to fill the submission queue without draining it (i.e, if the ring poll thread is not scheduled whilst simultaneously having loads of things adding entries)
The text was updated successfully, but these errors were encountered: