I won't apply this version, please send v2 with the changes below.

On Mon, 16 Nov 2026 16:41:15 +0100, Alexander Ivanov 
<[email protected]> wrote:
> persists and a lost netlink message is not fatal. Only ENOSPC writes
> and flushes are held for now; the op and errno are reported so that

- Item 3 of the cover letter reply: ENOSPC reads are held too at this
  patch.

Why: qcow2_try_hold_error() of patch 3 holds any BLK_STS_NOSPC, read
or write; the read side changes only with the policy of patch 7. The
same goes for "reads are not held" below, so the udev probe can be
held here.

> when the held list turns non-empty. A periodic retry that fails
> again repopulates the list, so the event is repeated while the error

- Item 4 of the cover letter reply: say that this applies to every
  dm-qcow2 device.

Why: with the default policy a full CT disk now gets a change uevent
and a blkid probe every 20 seconds, which the message does not
mention.

-- 
Konstantin Khorenko,
Virtuozzo Linux Kernel Team

_______________________________________________
Devel mailing list
[email protected]
https://lists.openvz.org/mailman/listinfo/devel

Reply via email to