On Fri, 2016-03-25 at 14:42 -0700, Dan Williams wrote:
> On Fri, Mar 25, 2016 at 1:59 PM, Verma, Vishal L
> <vishal.l.verma@xxxxxxxxx> wrote:
> >
> > On Fri, 2016-03-25 at 03:45 -0700, Christoph Hellwig wrote:
> > >
> > > On Thu, Mar 24, 2016 at 05:17:30PM -0600, Vishal Verma wrote:
> > > >
> > > >
> > > > dax_do_io (called for read() or write() for a dax file system)
> > > > may
> > > > fail
> > > > in the presence of bad blocks or media errors. Since we expect
> > > > that
> > > > a
> > > > write should clear media errors on nvdimms, make dax_do_io fall
> > > > back to
> > > > the direct_IO path, which will send down a bio to the driver,
> > > > which
> > > > can
> > > > then attempt to clear the error.
> > > Leave the fallback on -EIO to the callers please.ÂÂThey generally
> > > call
> > > __blockdev_direct_IO anyway, so it should actually become simpler
> > > that
> > > way.
> > I thought of this, but made the retrying happen in the wrapper so
> > that
> > it can be centralized. If the callers were to become responsible
> > for
> > the retry, then any new callers of dax_do_io might not realize they
> > are
> > responsible for retrying, and hit problems.
> That's their prerogative otherwise you are precluding an alternate
> handling of a dax_do_io() failure.ÂÂMaybe a fs or upper layer can
> recover in a different manner than re-submit the I/O to the
> __blockdev_direct_IO path.
I'm happy to make the change, but we don't preclude that -- __dax_do_io
is still exported and available..
|