On Mon, 2012-03-12 at 13:39 -0400, Chuck Lever wrote:
> Hi-
>
> On Mar 9, 2012, at 5:35 PM, Chuck Lever wrote:
>
> > diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
> > index c0d7220..747bf7a 100644
> > --- a/fs/nfs/nfs4proc.c
> > +++ b/fs/nfs/nfs4proc.c
> > @@ -3773,13 +3773,12 @@ wait_on_recovery:
> > return -EAGAIN;
> > }
> >
> > -static void nfs4_construct_boot_verifier(struct nfs_client *clp,
> > - nfs4_verifier *bootverf)
> > +static void nfs4_gen_boot_verifier(nfs4_verifier *bootverf)
> > {
> > __be32 verf[2];
> >
> > - verf[0] = htonl((u32)clp->cl_boot_time.tv_sec);
> > - verf[1] = htonl((u32)clp->cl_boot_time.tv_nsec);
> > + verf[0] = cpu_to_be32((u32)nfs_boot_time.tv_sec);
> > + verf[1] = cpu_to_be32((u32)nfs_boot_time.tv_nsec);
> > memcpy(bootverf->data, verf, sizeof(bootverf->data));
> > }
> >
>
>
> One other question I had about this. Do we really need to perform byte-swapping when constructing a boot verifier? It's supposed to be opaque.
No we don't need to do that.
--
Trond Myklebust
Linux NFS client maintainer
NetApp
Trond.Myklebust@xxxxxxxxxx
www.netapp.com
��.n��������+%������w��{.n�����{��w��� jg��� �����ݢj����G���� ���j:+v���w�m������w�������h�����٥
