On Mon, Mar 23, 2020 at 09:50:34AM -0400, Chuck Lever wrote:
>
>
> > On Mar 23, 2020, at 3:55 AM, Vasily Averin <vvs@xxxxxxxxxxxxx> wrote:
> >
> > New struct nfsd4_blocked_lock allocated in find_or_allocate_block()
> > does not initialised nbl_list and nbl_lru.
> > If conflock allocation fails rollback can call list_del_init()
> > access uninitialized fields and corrupt memory.
> >
> > Fixes: 76d348fadff5 ("nfsd: have nfsd4_lock use blocking locks for v4.1+ lock")
> > Signed-off-by: Vasily Averin <vvs@xxxxxxxxxxxxx>
> > ---
> > fs/nfsd/nfs4state.c | 32 +++++++++++++++-----------------
> > 1 file changed, 15 insertions(+), 17 deletions(-)
> >
> > diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
> > index 369e574c5092..176ef8d24fae 100644
> > --- a/fs/nfsd/nfs4state.c
> > +++ b/fs/nfsd/nfs4state.c
> > @@ -6524,6 +6524,13 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
> > goto out;
> > }
> >
> > + conflock = locks_alloc_lock();
> > + if (!conflock) {
> > + dprintk("NFSD: %s: unable to allocate lock!\n", __func__);
> > + status = nfserr_jukebox;
> > + goto out;
> > + }
>
> Nit: What do people think about removing this dprintk() as part of the fix?
I don't think we want a dprintk every place we kmalloc. All for
removing them.--b.
