On Tue, Nov 01, 2022 at 08:32:30PM -0700, Eric Biggers wrote:
> On Tue, Nov 01, 2022 at 03:00:31PM -0400, Stefan Hajnoczi wrote:
> > /* Let's try to use the logical blocksize for the alignment. */
> > - if (probe_logical_blocksize(fd, &bs->bl.request_alignment) < 0) {
> > - bs->bl.request_alignment = 0;
> > + if (!bs->bl.request_alignment) {
> > + if (probe_logical_blocksize(fd, &bs->bl.request_alignment) < 0) {
> > + bs->bl.request_alignment = 0;
> > + }
> > }
> >
> > #ifdef __linux__
> > - /*
> > - * The XFS ioctl definitions are shipped in extra packages that might
> > - * not always be available. Since we just need the XFS_IOC_DIOINFO
> > ioctl
> > - * here, we simply use our own definition instead:
> > - */
> > - struct xfs_dioattr {
> > - uint32_t d_mem;
> > - uint32_t d_miniosz;
> > - uint32_t d_maxiosz;
> > - } da;
> > - if (ioctl(fd, _IOR('X', 30, struct xfs_dioattr), &da) >= 0) {
> > - bs->bl.request_alignment = da.d_miniosz;
> > - /* The kernel returns wrong information for d_mem */
> > - /* s->buf_align = da.d_mem; */
> > + if (!bs->bl.request_alignment) {
>
> This patch changes the fallback code to make the request_alignment value from
> probe_logical_blocksize() override the value from XFS_IOC_DIOINFO. Is that
> intentional?
Thanks for pointing out the bug. That was not intentional. Will fix.
> > + if (ioctl(fd, _IOR('X', 30, struct xfs_dioattr), &da) >= 0) {
> > + bs->bl.request_alignment = da.d_miniosz;
> > + /* The kernel returns wrong information for d_mem */
> > + /* s->buf_align = da.d_mem; */
>
> Has this bug been reported to the XFS developers (linux-...@vger.kernel.org)?
Paolo: Do you remember if you reported this when you wrote commit
c25f53b06eba ("raw: Probe required direct I/O alignment")?
Stefan
signature.asc
Description: PGP signature
