Jonathan Morton <[email protected]> writes: >> On 1 Jun, 2018, at 8:46 pm, Toke Høiland-Jørgensen <[email protected]> wrote: >> >> As you may have seen, I hit a lockup bug when testing Cake at high >> speeds. I am not quite sure what is happening, but we seem to be looping >> forever either in cake_dequeue() or cake_enqueue(), causing deadlocks. >> >> Unfortunately, I have no more time to debug this for now. As far as I >> can tell, it only seems to be happening as the shaped rate approaches or >> exceeds the link speed, or when running in unlimited mode. So for all I >> know it might be possible to trigger it at lower speeds as well... >> >> Anyone with any bright ideas? > > Is there a complete copy of the code somewhere I can look at it? I'm > still running the cobalt branch.
I've been trying to keep everything in sync. The upstream-4.18 branch is identical to what I've been submitting; the only thing missing from the cobalt branch should be the classification stuff. The tests I just did where I saw the lockup was the current cobalt branch. -Toke _______________________________________________ Cake mailing list [email protected] https://lists.bufferbloat.net/listinfo/cake
