[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
[Expired for linux-signed-hwe-5.8 (Ubuntu) because there has been no activity for 60 days.] ** Changed in: linux-signed-hwe-5.8 (Ubuntu) Status: Incomplete => Expired -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
As expected 5.11 did crash. Unfortunately the vmcore is not useful as it filled up disk space so it is incomplete and I had to remove it to free up some space. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
BTW
For 5.11 (or oem-5.10 on Focal) it has only 3914d88f7608e6c ("xsk: Respect
device's headroom and tailroom on generic xmit path") applied, so I expect you
will see this issue with it.
For anything older than 5.10, none of these four patches were applied.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1933245
Title:
mlx5_core: Error cqe on cqn leads to hang
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
Thanks for looking into this. I will try edge HWE. I did try it week or two ago but it was still at 5.8. I will need to let the 5.11 kernel run for couple of days to be sure that it is properly fixed. Btw the same problem also exists on 5.4 kernel line. I do not have vmcore from that kernel version as it was not setup back then and afterwards I have tried hwe kernel to see if it was fixed. What I am trying to say is that I get that it is not straightforward backport but we still should think about fixing as it affects LTS release. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
Considering the complexity of these and the fact that 5.8 is missing other vital parts to get these applied, e.g.: * f09ced4053 xsk: Fix race in SKB mode transmit with shared cq * 1c1efc2af1 xsk: Create and free buffer pool independently from umem I think it's quite unlikely to get this fixed there. Please give 21.04 Hirsute (5.11 kernel) a try, or linux-generic-hwe-20.04-edge on Focal (linux-hwe-5.11) and let us know if you can reproduce this issue. Thanks ** Changed in: linux-signed-hwe-5.8 (Ubuntu) Assignee: Po-Hsu Lin (cypressyew) => (unassigned) ** Changed in: linux-signed-hwe-5.8 (Ubuntu) Status: New => Incomplete -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
They're in the upstream tree: https://github.com/torvalds/linux/commit/ab5bd583b9289666e918f9e5f672d33ccdfd49b2 https://github.com/torvalds/linux/commit/c2ff53d8049f30098153cd2d1299a44d7b124c57 https://github.com/torvalds/linux/commit/3914d88f7608e6c2e80e344474fa289370c32451 https://github.com/torvalds/linux/commit/9c8f21e6f8856a96634e542a58ef3abf27486801 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
I think this patchset is the fix: https://lore.kernel.org/bpf/20210218204908.5455-1-aloba...@pm.me/ -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
** Changed in: linux-signed-hwe-5.8 (Ubuntu) Assignee: (unassigned) => Po-Hsu Lin (cypressyew) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1933245] Re: mlx5_core: Error cqe on cqn leads to hang
adding dmesg from vmcore ** Attachment added: "dmesg.202106161938" https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+attachment/5506276/+files/dmesg.202106161938 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1933245 Title: mlx5_core: Error cqe on cqn leads to hang To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1933245/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
