On Mon, 18 Jul 2022 at 21:17, Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote: > > On Mon, Jul 18, 2022 at 12:04 PM Christian Kohlschütter > <christian@xxxxxxxxxxxxxxxx> wrote: > > > > The regression in question caused overlayfs to erroneously return ENOSYS when one lower filesystem (e.g., davfs2) returned this upon checking extended attributes (there were two relevant submissions triggering this somewhere around 5.15, 5.16) > > Well, if that's the case, isn't the proper fix to just fix davfs2? > > If ENOSYS isn't a valid error, and has broken apps that want to just > ignore "no fattr support", then it's a davfs2 bug, and fixing it there > will presumably magically just fix the ovl case too? Libfuse returns ENOSYS for requests which the filesystem doesn't support. Hence this is not a davfs bug, davfs knows nothing about itoctls, the userspace filesystem is just returning the error value that is used to mean "no support for this type of request". So this is a bug in the kernel part of fuse, that doesn't catch and convert ENOSYS in case of the ioctl request. Fixing it in fuse should fix it in overlayfs as well, as you say. Thanks, Miklos
