From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Steve Wise" Subject: RE: [PATCH v1 05/13] xprtrdma: Don't drain CQs on transport disconnect Date: Wed, 2 Jul 2014 14:50:32 -0500 Message-ID: <006001cf962e$e20df020$a629d060$@opengridcomputing.com> References: <20140623223201.1634.83888.stgit@manet.1015granger.net> <20140623223942.1634.89063.stgit@manet.1015granger.net> <53B45D7B.4020705@opengridcomputing.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Content-Language: en-us Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: 'Devesh Sharma' , 'Chuck Lever' , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org > -----Original Message----- > From: Devesh Sharma [mailto:Devesh.Sharma-iH1Dq9VlAzfQT0dZR+AlfA@public.gmane.org] > Sent: Wednesday, July 02, 2014 2:43 PM > To: Steve Wise; Chuck Lever; linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; linux-nfs@vg= er.kernel.org > Subject: RE: [PATCH v1 05/13] xprtrdma: Don't drain CQs on transport = disconnect >=20 > > -----Original Message----- > > From: Steve Wise [mailto:swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org] > > Sent: Thursday, July 03, 2014 12:59 AM > > To: Devesh Sharma; Chuck Lever; linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; linux- > > nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org > > Subject: Re: [PATCH v1 05/13] xprtrdma: Don't drain CQs on transpor= t > > disconnect > > > > On 7/2/2014 2:06 PM, Devesh Sharma wrote: > > > This change is very much prone to generate poll_cq errors because= of > > > un-cleaned completions which still point to the non-existent QPs.= On > > > the new connection when these completions are polled, the poll_cq= will fail > > because old QP pointer is already NULL. > > > Did anyone hit this situation during their testing? > > > > Hey Devesh, > > > > iw_cxgb4 will silently toss CQEs if the QP is not active. >=20 > Ya, just now checked that in mlx and cxgb4 driver code. On the other = hand ocrdma is asserting > a BUG-ON for such CQEs causing system panic. > Out of curiosity I am asking, how this change is useful here, is it r= educing the re-connection > time...Anyhow rpcrdma_clean_cq was discarding the completions (flush/= successful both) > Well, I don't think there is anything restricting an application from d= estroying the QP with pending CQEs on its CQs. So it definitely shoul= dn't cause a BUG_ON() I think. I'll have to read up in the Verbs spec= s if destroying a QP kills all the pending CQEs... =20 > > > > > > >> -----Original Message----- > > >> From: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org [mailto:linux-rdma- > > >> owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org] On Behalf Of Chuck Lever > > >> Sent: Tuesday, June 24, 2014 4:10 AM > > >> To: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org > > >> Subject: [PATCH v1 05/13] xprtrdma: Don't drain CQs on transport > > >> disconnect > > >> > > >> CQs are not destroyed until unmount. By draining CQs on transpor= t > > >> disconnect, successful completions that can change the r.frmr.st= ate > > >> field can be missed. >=20 > Still those are missed isn=E2=80=99t it....Since those successful com= pletions will still be dropped after re- > connection. Am I missing something to > understanding the motivation... >=20 > > >> > > >> Signed-off-by: Chuck Lever > > >> --- > > >> net/sunrpc/xprtrdma/verbs.c | 5 ----- > > >> 1 file changed, 5 deletions(-) > > >> > > >> diff --git a/net/sunrpc/xprtrdma/verbs.c > > >> b/net/sunrpc/xprtrdma/verbs.c index 3c7f904..451e100 100644 > > >> --- a/net/sunrpc/xprtrdma/verbs.c > > >> +++ b/net/sunrpc/xprtrdma/verbs.c > > >> @@ -873,9 +873,6 @@ retry: > > >> dprintk("RPC: %s: rpcrdma_ep_disconnect" > > >> " status %i\n", __func__, rc); > > >> > > >> - rpcrdma_clean_cq(ep->rep_attr.recv_cq); > > >> - rpcrdma_clean_cq(ep->rep_attr.send_cq); > > >> - > > >> xprt =3D container_of(ia, struct rpcrdma_xprt, rx_ia); > > >> id =3D rpcrdma_create_id(xprt, ia, > > >> (struct sockaddr *)&xprt->rx_data.addr); > > @@ -985,8 +982,6 @@ > > >> rpcrdma_ep_disconnect(struct rpcrdma_ep *ep, struct rpcrdma_ia *= ia) > > >> { > > >> int rc; > > >> > > >> - rpcrdma_clean_cq(ep->rep_attr.recv_cq); > > >> - rpcrdma_clean_cq(ep->rep_attr.send_cq); > > >> rc =3D rdma_disconnect(ia->ri_id); > > >> if (!rc) { > > >> /* returns without wait if not connected */ > > >> > > >> -- > > >> To unsubscribe from this list: send the line "unsubscribe linux-= rdma" > > >> in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More > > majordomo > > >> info at http://vger.kernel.org/majordomo-info.html > > > N r y b X =C7=A7v ^ )=DE=BA{.n + { " ^n r z =1A = h & =1E G h =03 > > > ( =E9=9A=8E =DD=A2j" =1A =1Bm z =DE=96 f h ~ mml=3D=3D -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" i= n the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html