http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c1 Takashi Iwai <tiwai@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |tiwai@suse.com --- Comment #1 from Takashi Iwai <tiwai@suse.com> --- AFAIK, the error message from blkdev_get_no_open() itself should be present already in 5.18, at least. There no significant change around that code in 5.19. Maybe the problem is rather what triggers this. Something about the loop device setup? -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c3 Alessandro Sturniolo <alessandro.sturniolo@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |alessandro.sturniolo@gmail. | |com --- Comment #3 from Alessandro Sturniolo <alessandro.sturniolo@gmail.com> --- I've encountered the same issue after I've updated my workstation when I come back from my vacations, so I don't know, effectively, when this issue has begin to appear. My previuos updates are dated July 29, and previously the issue is not present. Now, when I shutdown my workstation, some times it shutdown in seconds, some times in minutes, and some times I have to force power off with magic sysrq keys. It not seems related to the uptime of the workstation. I've encountered this issue also after only a few minutes of uptime. Every times that the workstation doesn't shut down immediately, I can see a flood of messages on the Virtual Terminal 10 (CTRL+Alt+F10). The message is "block device autoloading is deprecated and will be removed" repeated continuosly. My setup have some software RAID-1 (on SSD disks) and RAID-6 disks (on mechanical disks), some NFS shares mounted via autofs, some paths as /tmp, /var/log, /var/cache, and others, binded from the SSD to the HDD to preserve SSD lifespan, Docker, VirtualBox, and a lot of other services, that I don't know if they can be relevants or not. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c5 --- Comment #5 from Takashi Iwai <tiwai@suse.com> --- (In reply to Daniel Wagner from comment #4)

I found this:

https://lore.kernel.org/linux-block/20220225181440.1351591-1-hch@lst.de/

This is the upstream commit 451f0b6f4c44, which is already in 5.18 kernel.

Was the BLOCK_LEGACY_AUTOLOAD always enabled?

The kconfig was introduced in 5.18 (commit fbdee71bb5d8). And the behavior before that version was same as CONFIG_BLOCK_LEGACY_AUTOLOAD=y but it was without spewing the message. So, the message is new, but the behavior is same. (And note that those bdev changes were already in 5.18, not new in 5.19.) -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c7 --- Comment #7 from David Boyd <David.Boyd49@twc.com> --- I have booted with 5.18.15 kernel and the reported behavior is not present. I have also booted with 5.19.0, 5.19.1 and, most recently, 5.19.2 and each kernel version show the reported problem. Hope that helps. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c10 --- Comment #10 from David Boyd <David.Boyd49@twc.com> --- Created attachment 861146 --> http://bugzilla.opensuse.org/attachment.cgi?id=861146&action=edit VirtualBox snapshot of trace output Jan, I applied the patched kernel that you posted. The output flooded the screen. I don't have a good method available to capture the output to submit back to you. But, I noticed that each trace section was checking md devices. That got me to thinking. I created a Tumbleweed VirtualBox VM with mirrored disks to determine if that was a factor. The VM with your kernel does, in fact, exhibit the same behaviour as the "real machine". I have included this snapshot of part of the trace output. Please let me know if this helps and also if there is anything else that I can do to assist. David. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c14 --- Comment #14 from Frank Kr�ger <fkrueger@mailbox.org> --- (In reply to David Boyd from comment #13)

It appears that this bug has made its way to the LKML. I guess that means there is not much more that I/we can do but wait.

Presumably one of the kernel experts will provide a kernel with the reverted commit in their home repo for testing. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1202534 http://bugzilla.opensuse.org/show_bug.cgi?id=1202534#c22 --- Comment #22 from Walter Haidinger <walter.haidinger@gmx.at> --- *** Bug 1202589 has been marked as a duplicate of this bug. *** -- You are receiving this mail because: You are on the CC list for the bug.