> On Nov 23, 2022, at 2:00 PM, Anders Blomdell <anders.blomdell@xxxxxxxxxxxxxx> wrote: > > Make nfsd_splice_actor work with reads with a non-zero offset that doesn't end on a page boundary. > > This was found when virtual machines with nfs-mounted qcow2 disks failed to boot properly (originally found > on v6.0.5, fix also needed and tested on v6.0.9 and v6.1-rc6). > > Signed-off-by: Anders Blomdell <anders.blomdell@xxxxxxxxxxxxxx> > Link: https://bugzilla.redhat.com/show_bug.cgi?id=2142132 > Fixes: bfbfb6182ad1 "nfsd_splice_actor(): handle compound pages" > Cc: stable@xxxxxxxxxxxxxxx # v6.0+ "b4 am" and other tooling failed to recognize the diff content in this email, so I've applied Al's version to nfsd's for-rc, after running some regression tests. See: https://git.kernel.org/pub/scm/linux/kernel/git/cel/linux.git/log/?h=for-rc I will see that a pull request is sent to Linus before 6.1-rc7. Thanks for reporting and chasing this down. > -- a/fs/nfsd/vfs.c > +++ b/fs/nfsd/vfs.c > @@ -869,12 +869,13 @@ nfsd_splice_actor(struct pipe_inode_info *pipe, struct pipe_buffer *buf, > struct splice_desc *sd) > { > struct svc_rqst *rqstp = sd->u.data; > - struct page *page = buf->page; // may be a compound one > + // buf->page may be a compound one > unsigned offset = buf->offset; > + struct page *first = buf->page + offset / PAGE_SIZE; > + struct page *last = buf->page + (offset + sd->len - 1) / PAGE_SIZE; > > - page += offset / PAGE_SIZE; > - for (int i = sd->len; i > 0; i -= PAGE_SIZE) > - svc_rqst_replace_page(rqstp, page++); > + for (struct page *page = first; page <= last; page++) > + svc_rqst_replace_page(rqstp, page); > if (rqstp->rq_res.page_len == 0) // first call > rqstp->rq_res.page_base = offset % PAGE_SIZE; > rqstp->rq_res.page_len += sd->len; > > -- > Anders Blomdell Email: anders.blomdell@xxxxxxxxxxxxxx > Department of Automatic Control > Lund University Phone: +46 46 222 4625 > P.O. Box 118 > SE-221 00 Lund, Sweden -- Chuck Lever
