Hi Timo!

>Rebasing on 6.15 now shows that these still have not landed upstream,
do you know what the plan there is?

Originally, when we've got a request from a partner to apply this
patches I've contacted with the original author (Dr. Hannes Reinecke) to
see if there is any plan to land this to upstream. He told me that these
patches are preliminary patches for the full solution currently under
discussion at NVMexpress.org (TP4129). And they actively working on
this. It will take a long time so it makes sense to go with current
temporary way.

Also, we can see that SUSE kernel still has this patches applied:
https://github.com/SUSE/kernel-source/blob/rpm-6.12.0-160000.9/patches.suse/nvme-tcp-do-not-terminate-commands-when-in-resetting.patch
https://github.com/SUSE/kernel-source/blob/rpm-6.12.0-160000.9/patches.suse/nvme-tcp-add-recovery_delay-to-sysfs.patch
https://github.com/SUSE/kernel-source/blob/rpm-6.12.0-160000.9/patches.suse/nvme-tcp-delay-error-recovery-until-the-next-kato.patch
https://github.com/SUSE/kernel-source/blob/rpm-6.12.0-160000.9/patches.suse/nvme-tcp-make-err_work-a-delayed-work.patch

I would keep them as they are very small and simple to rebase and at the
same time can potentially prevent a real issue. If you need any help
from my side like reviewing a rebased version or anything else - please
let me know.

Kind regards,
Alex

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/2070048

Title:
  nvme-tcp: fix retry logic

Status in linux package in Ubuntu:
  Confirmed

Bug description:
  SRU Justification

  [Impact]

  * Request to port some out-of-upstream-tree patches to fix NVMe over
  TCP request retry logic issues

  [Fix]

  * Port patches from LKML:
  https://lore.kernel.org/all/20230908100049.80809-1-h...@suse.de/

  [Test Plan]

  * It was tested internally in LXD team using a custom kernel build 
(linux-hwe-6.5 - 6.5.0-41.41~22.04.2ubuntu1nvmefixes) for Jammy:
  https://launchpad.net/~mihalicyn/+archive/ubuntu/kernel-nvme-fixes/+packages

  * SUSE kernel has these patches applied for a long time:
  
https://github.com/SUSE/kernel-source/blob/rpm-6.4.0-150600.21/patches.suse/nvme-tcp-do-not-terminate-commands-when-in-resetting.patch
  
https://github.com/SUSE/kernel-source/blob/rpm-6.4.0-150600.21/patches.suse/nvme-tcp-add-recovery_delay-to-sysfs.patch
  
https://github.com/SUSE/kernel-source/blob/rpm-6.4.0-150600.21/patches.suse/nvme-tcp-delay-error-recovery-until-the-next-kato.patch
  
https://github.com/SUSE/kernel-source/blob/rpm-6.4.0-150600.21/patches.suse/nvme-tcp-make-err_work-a-delayed-work.patch

  [Regression potential]

  * Limited to nvme-tcp block driver only

  [Other Info]

  Internal BugLink:
  https://warthogs.atlassian.net/browse/LXD-1120

  Related LXD feature:
  https://github.com/canonical/lxd/pull/12304

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2070048/+subscriptions


-- 
Mailing list: https://launchpad.net/~kernel-packages
Post to     : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help   : https://help.launchpad.net/ListHelp

Reply via email to