> In that case if I do umount after device has gone away, I can see above > WARN(). And it does seem to be coming from. > > WARN_ON_ONCE(write_inode_now(inode, true)) > > If we failed to write back inode, then warning about it sounds right? WARN() is for detecting kernel internal consistency problems, like potential bugs. It's not for handling IO errors or really everything that can happen on a non buggy kernel. > What's wrong with that? Should it be just a kernel log of level KERN_WARN > instead? Yes. Something like that. -Andi -- ak@xxxxxxxxxxxxxxx -- Speaking for myself only. -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
