Re: [PATCH v7 13/19] NFSD return nfs4_stid in nfs4_preprocess_stateid_op

[Olga Kornievskaia <olga.kornievskaia@xxxxxxxxx>]

On Wed, Oct 2, 2019 at 12:34 PM J. Bruce Fields <bfields@xxxxxxxxxx> wrote:
>
> On Wed, Oct 02, 2019 at 12:12:17PM -0400, Olga Kornievskaia wrote:
> > On Wed, Oct 2, 2019 at 11:52 AM J. Bruce Fields <bfields@xxxxxxxxxxxx> wrote:
> > >
> > > On Mon, Sep 16, 2019 at 05:13:47PM -0400, Olga Kornievskaia wrote:
> > > > @@ -1026,7 +1026,8 @@ static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
> > > >  static __be32
> > > >  nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> > > >                 stateid_t *src_stateid, struct nfsd_file **src,
> > > > -               stateid_t *dst_stateid, struct nfsd_file **dst)
> > > > +               stateid_t *dst_stateid, struct nfsd_file **dst,
> > > > +               struct nfs4_stid **stid)
> > > >  {
> > > >       __be32 status;
> > > >
> > > ...
> > > > @@ -1072,7 +1073,7 @@ static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
> > > >       __be32 status;
> > > >
> > > >       status = nfsd4_verify_copy(rqstp, cstate, &clone->cl_src_stateid, &src,
> > > > -                                &clone->cl_dst_stateid, &dst);
> > > > +                                &clone->cl_dst_stateid, &dst, NULL);
> > > >       if (status)
> > > >               goto out;
> > > >
> > > > @@ -1260,7 +1261,7 @@ static int nfsd4_do_async_copy(void *data)
> > > >
> > > >       status = nfsd4_verify_copy(rqstp, cstate, &copy->cp_src_stateid,
> > > >                                  &copy->nf_src, &copy->cp_dst_stateid,
> > > > -                                &copy->nf_dst);
> > > > +                                &copy->nf_dst, NULL);
> > > >       if (status)
> > > >               goto out;
> > > >
> > >
> > > So both callers pass NULL for the new stid parameter.  Looks like that's
> > > still true after the full series of patches, too.
> > >
> >
> > If you look at an earlier chunk it uses it (there is only a single
> > user of it: copy notify state)
>
> You're talking about nfs4_preprocess_stateid_op, the above is
> nfsd4_verify_copy.

I see. You are right, looks like after reworks the nfsd4_verify_copy
doesn't need a stid. Previously both copy stateid and copy_notify
stateids were tied to its parents but I've removed the link for the
copy stateid a while back. Should I now combine this patch with "NFSD
add COPY_NOTIFY operation" because that's the only caller of
nfsd4_preprocess_stateid_op that needs the stid.

>
> --b.
>
> > @@ -1034,14 +1035,14 @@ static __be32 nfsd4_do_lookupp(struct svc_rqst
> > *rqstp, struct svc_fh *fh)
> >                 return nfserr_nofilehandle;
> >
> >         status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
> > -                                           src_stateid, RD_STATE, src);
> > +                                           src_stateid, RD_STATE, src, NULL);
> >         if (status) {
> >                 dprintk("NFSD: %s: couldn't process src stateid!\n", __func__);
> >                 goto out;
> >         }
> >
> >         status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
> > -                                           dst_stateid, WR_STATE, dst);
> > +                                           dst_stateid, WR_STATE, dst, stid);
> >         if (status) {
> >                 dprintk("NFSD: %s: couldn't process dst stateid!\n", __func__);
> >                 goto out_put_src;
> >
> > > --b.