On Wed, Mar 04, 2020 at 02:37:38PM +0100, Jan Kara wrote:

Hi!

On Mon 02-03-20 11:55:44, He Zhe wrote:

quoted

Since the following commit
https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git/commit/?h=for-5.5/disk-revalidate&id=6917d0689993f46d97d40dd66c601d0fd5b1dbdd
until now(v5.6-rc4),

If we start udisksd service of systemd(v244), systemd-udevd will scan
/dev/hdc (the cdrom device created by default in qemu(v4.2.0)).
systemd-udevd will endlessly run and cause OOM.

Thanks for report! The commit you mention has this:

There is a small behavior change in that we now send the kevent change
notice also if we were not invalidating but no partitions were found, which
seems like the right thing to do.

And apparently this confuses systemd-udevd because it tries to open
/dev/hdc in response to KOBJ_CHANGE event on that device and the open calls
rescan_partitions() which generates another KOBJ_CHANGE event.  So I'm
afraid we'll have to revert to the old behavior of not sending KOBJ_CHANGE
event when there are no partitions found. Christoph?

Looks like it.  Let me figure out how to best do that.