From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id A169DC433EF for ; Fri, 11 Mar 2022 17:09:27 +0000 (UTC) Received: from localhost ([::1]:42958 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1nSila-0004Rp-9z for qemu-devel@archiver.kernel.org; Fri, 11 Mar 2022 12:09:26 -0500 Received: from eggs.gnu.org ([209.51.188.92]:53046) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nSikx-0003ks-DX for qemu-devel@nongnu.org; Fri, 11 Mar 2022 12:08:47 -0500 Received: from us-smtp-delivery-44.mimecast.com ([205.139.111.44]:41778) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nSiku-0002MR-D9 for qemu-devel@nongnu.org; Fri, 11 Mar 2022 12:08:46 -0500 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-472-w_lohQNkP0K3Q6PuAkkHqQ-1; Fri, 11 Mar 2022 12:08:41 -0500 X-MC-Unique: w_lohQNkP0K3Q6PuAkkHqQ-1 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id CA7941091DA0; Fri, 11 Mar 2022 17:08:40 +0000 (UTC) Received: from bahia (unknown [10.39.192.168]) by smtp.corp.redhat.com (Postfix) with ESMTP id E943B8752F; Fri, 11 Mar 2022 17:08:39 +0000 (UTC) Date: Fri, 11 Mar 2022 18:08:38 +0100 From: Greg Kurz To: Christian Schoenebeck Subject: Re: [PATCH 5/6] 9pfs: fix 'Twalk' to only send error if no component walked Message-ID: <20220311180838.3a8c9f74@bahia> In-Reply-To: <3685932.ieBdHTVF6f@silver> References: <1785989.mTmxrQOLcC@silver> <20220311173541.06d6e96e@bahia> <3685932.ieBdHTVF6f@silver> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: kaod.org Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Received-SPF: softfail client-ip=205.139.111.44; envelope-from=groug@kaod.org; helo=us-smtp-delivery-44.mimecast.com X-Spam_score_int: -18 X-Spam_score: -1.9 X-Spam_bar: - X-Spam_report: (-1.9 / 5.0 requ) BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_SOFTFAIL=0.665, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=no autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: qemu-devel@nongnu.org Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" On Fri, 11 Mar 2022 17:44:54 +0100 Christian Schoenebeck wrote: > On Freitag, 11. M=C3=A4rz 2022 17:35:41 CET Greg Kurz wrote: > > On Thu, 10 Mar 2022 10:13:33 +0100 > >=20 > > Christian Schoenebeck wrote: > > > On Mittwoch, 9. M=C3=A4rz 2022 18:57:39 CET Christian Schoenebeck wro= te: > > > > Current implementation of 'Twalk' request handling always sends an > > > > 'Rerror' > > > >=20 > > > > response if any error occured. The 9p2000 protocol spec sais though= : > > > > " > > > > If the first element cannot be walked for any reason, Rerror is > > > > returned. > > > > Otherwise, the walk will return an Rwalk message containing nwqid= qids > > > > corresponding, in order, to the files that are visited by the nwq= id > > > > successful elementwise walks; nwqid is therefore either nwname or= the > > > >=20 > > > > index of the first elementwise walk that failed. > > > >=20 > > > > " > > > > =20 > > > > http://ericvh.github.io/9p-rfc/rfc9p2000.html#anchor33 > > > >=20 > > > > For that reason we are no longer leaving from an error path in func= tion > > > > v9fs_walk(), unless really no path component could be walked > > > > successfully or if the request has been interrupted. > > > >=20 > > > > Local variable 'nvalid' counts and reflects the number of path > > > > components > > > > successfully processed by background I/O thread, whereas local vari= able > > > > 'name_idx' subsequently counts and reflects the number of path > > > > components > > > > eventually accepted successfully by 9p server controller portion. > > > >=20 > > > > New local variable 'any_err' is an aggregate variable reflecting wh= ether > > > > any error occurred at all, while already existing variable 'err' on= ly > > > > reflects the last error. > > > >=20 > > > > Despite QIDs being delivered to client in a more relaxed way now, i= t is > > > > important to note though that fid still must remain uneffacted if a= ny > > > > error > > >=20 > > > Typo: should be "unaffected". > > >=20 > > > > occurred. > > > >=20 > > > > Signed-off-by: Christian Schoenebeck > > > > --- > > > >=20 > > > > hw/9pfs/9p.c | 29 +++++++++++++++++++++-------- > > > > 1 file changed, 21 insertions(+), 8 deletions(-) > > > >=20 > > > > diff --git a/hw/9pfs/9p.c b/hw/9pfs/9p.c > > > > index 6cdc566866..8ccd180608 100644 > > > > --- a/hw/9pfs/9p.c > > > > +++ b/hw/9pfs/9p.c > > > > @@ -1766,7 +1766,7 @@ static void coroutine_fn v9fs_walk(void *opaq= ue) > > > >=20 > > > > { > > > > =20 > > > > int name_idx, nvalid; > > > > g_autofree V9fsQID *qids =3D NULL; > > > >=20 > > > > - int i, err =3D 0; > > > > + int i, err =3D 0, any_err =3D 0; > > > >=20 > > > > V9fsPath dpath, path; > > > > P9ARRAY_REF(V9fsPath) pathes =3D NULL; > > > > uint16_t nwnames; > > > >=20 > > > > @@ -1832,6 +1832,7 @@ static void coroutine_fn v9fs_walk(void *opaq= ue) > > > >=20 > > > > * driver code altogether inside the following block. > > > > */ > > > > =20 > > > > v9fs_co_run_in_worker({ > > > >=20 > > > > + nvalid =3D 0; > > > >=20 > > > > if (v9fs_request_cancelled(pdu)) { > > > > =20 > > > > err =3D -EINTR; > > > > break; > > > >=20 > > > > @@ -1842,7 +1843,7 @@ static void coroutine_fn v9fs_walk(void *opaq= ue) > > > >=20 > > > > break; > > > > =20 > > > > } > > > > stbuf =3D fidst; > > > >=20 > > > > - for (nvalid =3D 0; nvalid < nwnames; nvalid++) { > > > > + for (; nvalid < nwnames; nvalid++) { > > > >=20 > > > > if (v9fs_request_cancelled(pdu)) { > > > > =20 > > > > err =3D -EINTR; > > > > break; > > > >=20 > > > > @@ -1874,12 +1875,13 @@ static void coroutine_fn v9fs_walk(void *op= aque) > > > >=20 > > > > /* > > > > =20 > > > > * Handle all the rest of this Twalk request on main thread ..= . > > > > */ > > > >=20 > > > > - if (err < 0) { > > > > + if ((err < 0 && !nvalid) || err =3D=3D -EINTR) { > > > >=20 > > > > goto out; > > > > =20 > > > > } > > > >=20 > > > > + any_err |=3D err; > > > >=20 > > > > err =3D stat_to_qid(pdu, &fidst, &qid); > > > >=20 > > > > - if (err < 0) { > > > > + if (err < 0 && !nvalid) { > > > >=20 > > > > goto out; > > > > =20 > > > > } > > > > stbuf =3D fidst; > > > >=20 > > > > @@ -1888,20 +1890,30 @@ static void coroutine_fn v9fs_walk(void *op= aque) > > > >=20 > > > > v9fs_path_copy(&dpath, &fidp->path); > > > > v9fs_path_copy(&path, &fidp->path); > > > >=20 > > > > - for (name_idx =3D 0; name_idx < nwnames; name_idx++) { > > > > + for (name_idx =3D 0; name_idx < nvalid; name_idx++) { > > > >=20 > > > > if (!same_stat_id(&pdu->s->root_st, &stbuf) || > > > > =20 > > > > strcmp("..", wnames[name_idx].data)) > > > > =20 > > > > { > > > > =20 > > > > stbuf =3D stbufs[name_idx]; > > > > err =3D stat_to_qid(pdu, &stbuf, &qid); > > > > if (err < 0) { > > > >=20 > > > > - goto out; > > > > + break; > > > >=20 > > > > } > > > > v9fs_path_copy(&path, &pathes[name_idx]); > > > > v9fs_path_copy(&dpath, &path); > > > > =20 > > > > } > > > > memcpy(&qids[name_idx], &qid, sizeof(qid)); > > > > =20 > > > > } > > > >=20 > > > > + any_err |=3D err; > > > > + if (any_err) { > > >=20 > > > Not sure if there is ever the case err > 0, but as we are already > > > comparing > > > for "if (err < 0)" everywhere, we should probably also do the same > > > comparison for the aggregate error variable here, right? > >=20 > > It seems that you could drop any_err and just check name_idx !=3D nwnam= es ? >=20 > Mmm, what about the special case 'Twalk nwnames=3D0' (i.e. fid cloning), = that=20 > implementation would then skip errors, no? >=20 Ouch you're right... honestly, v9fs_walk() is really a mess and it is getting harder to fix. What about having a totally separate path for the cloning case (as a preparatory patch) ? > > > if (any_err < 0) { > > > =20 > > > ... > > > >=20 > > > > + if (!name_idx) { > > > > + /* don't send any QIDs, send Rlerror instead */ > > > > + goto out; > > > > + } else { > > > > + /* send QIDs (not Rlerror), but fid MUST remain unaffe= cted > > > > */ > > > > + goto send_qids; > > > > + } > > > > + } > > > >=20 > > > > if (fid =3D=3D newfid) { > > > > =20 > > > > if (fidp->fid_type !=3D P9_FID_NONE) { > > > > =20 > > > > err =3D -EINVAL; > > > >=20 > > > > @@ -1919,8 +1931,9 @@ static void coroutine_fn v9fs_walk(void *opaq= ue) > > > >=20 > > > > newfidp->uid =3D fidp->uid; > > > > v9fs_path_copy(&newfidp->path, &path); > > > > =20 > > > > } > > > >=20 > > > > - err =3D v9fs_walk_marshal(pdu, nwnames, qids); > > > > - trace_v9fs_walk_return(pdu->tag, pdu->id, nwnames, qids); > > > > +send_qids: > > > > + err =3D v9fs_walk_marshal(pdu, name_idx, qids); > > > > + trace_v9fs_walk_return(pdu->tag, pdu->id, name_idx, qids); > > > >=20 > > > > out: > > > > put_fid(pdu, fidp); > > > > if (newfidp) { >=20 >=20