On Thu, 2024-10-10 at 11:33 -0400, cel@xxxxxxxxxx wrote:
> From: Chuck Lever <chuck.lever@xxxxxxxxxx>
>
> NFSv4 LOCK operations should not avoid the set of authorization
> checks that apply to all other NFSv4 operations. Also, the
> "no_auth_nlm" export option should apply only to NLM LOCK requests.
> It's not necessary or sensible to apply it to NFSv4 LOCK operations.
>
> The replacement MAY bit mask,
> "NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE", comes from the access
> bits that are set in nfsd_permission() when the caller has set
> NFSD_MAY_LOCK.
>
> Reported-by: NeilBrown <neilb@xxxxxxx>
> Signed-off-by: Chuck Lever <chuck.lever@xxxxxxxxxx>
> ---
> fs/nfsd/nfs4state.c | 7 +++----
> 1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
> index 9c2b1d251ab3..3f2c11414390 100644
> --- a/fs/nfsd/nfs4state.c
> +++ b/fs/nfsd/nfs4state.c
> @@ -7967,11 +7967,10 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> if (check_lock_length(lock->lk_offset, lock->lk_length))
> return nfserr_inval;
>
> - if ((status = fh_verify(rqstp, &cstate->current_fh,
> - S_IFREG, NFSD_MAY_LOCK))) {
> - dprintk("NFSD: nfsd4_lock: permission denied!\n");
> + status = fh_verify(rqstp, &cstate->current_fh, S_IFREG,
> + NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE);
> + if (status != nfs_ok)
> return status;
> - }
> sb = cstate->current_fh.fh_dentry->d_sb;
>
> if (lock->lk_is_new) {
I would say this warrants a comment as to why you're not using
NFSD_MAY_LOCK here, but Neil's renaming patch takes care of that.
Reviewed-by: Jeff Layton <jlayton@xxxxxxxxxx>
