> On Sep 5, 2020, at 10:03 AM, bfields@xxxxxxxxxxxx wrote: > > From: "J. Bruce Fields" <bfields@xxxxxxxxxx> > > Since p points at raw xdr data, there's no guarantee that it's NULL > terminated, so we should give a length. And probably escape any special > characters too. > > Reported-by: Zhi Li <yieli@xxxxxxxxxx> > Signed-off-by: J. Bruce Fields <bfields@xxxxxxxxxx> I sent a patch a couple months ago to remove this dprintk: https://lore.kernel.org/linux-nfs/20200708201029.22129.31971.stgit@xxxxxxxxxxxxxxxxxxxxx/T/#u However you might want to apply Bruce's patch first, so it can be backported to stable. > --- > net/sunrpc/rpcb_clnt.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c > index c27123e6ba80..4a67685c83eb 100644 > --- a/net/sunrpc/rpcb_clnt.c > +++ b/net/sunrpc/rpcb_clnt.c > @@ -982,8 +982,8 @@ static int rpcb_dec_getaddr(struct rpc_rqst *req, struct xdr_stream *xdr, > p = xdr_inline_decode(xdr, len); > if (unlikely(p == NULL)) > goto out_fail; > - dprintk("RPC: %5u RPCB_%s reply: %s\n", req->rq_task->tk_pid, > - req->rq_task->tk_msg.rpc_proc->p_name, (char *)p); > + dprintk("RPC: %5u RPCB_%s reply: %*pE\n", req->rq_task->tk_pid, > + req->rq_task->tk_msg.rpc_proc->p_name, len, (char *)p); > > if (rpc_uaddr2sockaddr(req->rq_xprt->xprt_net, (char *)p, len, > sap, sizeof(address)) == 0) > -- > 2.26.2 > -- Chuck Lever
