On Wednesday 30 July 2008 19:43, Miklos Szeredi wrote: > Jens, > > Please apply or ack this for 2.6.27. > > [v3: respun against 2.6.27-rc1] > > Thanks, > Miklos > > ---- > From: Miklos Szeredi <mszeredi@xxxxxxx> > > Brian Wang reported that a FUSE filesystem exported through NFS could > return I/O errors on read. This was traced to splice_direct_to_actor() > returning a short or zero count when racing with page invalidation. > > However this is not FUSE or NFSD specific, other filesystems (notably NFS) > also call invalidate_inode_pages2() to purge stale data from the cache. > > If this happens while such pages are sitting in a pipe buffer, then > splice(2) from the pipe can return zero, and read(2) from the pipe can > return ENODATA. > > The zero return is especially bad, since it implies end-of-file or > disconnected pipe/socket, and is documented as such for splice. But > returning an error for read() is also nasty, when in fact there was no > error (data becoming stale is not an error). Hmm, the PageError case is a similar one which cannot be avoided, so it kind of indicates to me that the splice async API is slightly lacking (and provides me with some confirmation about my dislike of removing ClearPageUptodate from invalidate...) Returning -EIO at the pipe read I don't think quite make sense because it is conceptually an IO error for the splicer, not the reader (who is reading from a pipe, not from the file causing the error). It seems like the right way to fix this would be to allow the splicing process to be notified of a short read, in which case it could try to refill the pipe with the unread bytes... -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
