Re: [PATCH v4 1/2] NFSD: Return nfserr_serverfault if splice_ok but buf->pages have data

[Anna Schumaker <anna@kernel.org>]

On Tue, Sep 13, 2022 at 4:12 PM Chuck Lever III <chuck.lever@oracle.com> wrote:
>
>
>
> > On Sep 13, 2022, at 11:01 AM, Anna Schumaker <anna@kernel.org> wrote:
> >
> > From: Anna Schumaker <Anna.Schumaker@Netapp.com>
> >
> > This was discussed with Chuck as part of this patch set. Returning
> > nfserr_resource was decided to not be the best error message here, and
> > he suggested changing to nfserr_serverfault instead.
> >
> > Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
> > ---
> > fs/nfsd/nfs4xdr.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
>
> I've applied this one for nfsd for-next. Thanks!
>
> As I mentioned, 2/2 looks OK, and I'll apply it to my private
> tree for testing while we work out why it's a little slower.
>
>
> > diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
> > index 1e9690a061ec..01dd73ed5720 100644
> > --- a/fs/nfsd/nfs4xdr.c
> > +++ b/fs/nfsd/nfs4xdr.c
> > @@ -3994,7 +3994,7 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
> >       }
> >       if (resp->xdr->buf->page_len && splice_ok) {
> >               WARN_ON_ONCE(1);
> > -             return nfserr_resource;
> > +             return nfserr_serverfault;
>
> Odd, I couldn't find a definition for nfserr_serverfault when
> I asked for this patch last week, but this one-liner seems to
> compile correctly. Oh well!

I found it in fs/nfsd/nfsd.h with `git grep`: #define
nfserr_serverfault      cpu_to_be32(NFSERR_SERVERFAULT)

Anna

>
>
> >       }
> >       xdr_commit_encode(xdr);
> >
> > --
> > 2.37.3
> >
>
> --
> Chuck Lever
>
>
>