From: andros@netapp.com
To: trond.myklebust@netapp.com
Cc: bfields@redhat.com, linux-nfs@vger.kernel.org,
Andy Adamson <andros@netapp.com>
Subject: [PATCH_V4 2/9] NFS use svc_create_xprt for NFSv4.1 callback service
Date: Fri, 17 Dec 2010 13:20:03 -0500 [thread overview]
Message-ID: <1292610010-19084-3-git-send-email-andros@netapp.com> (raw)
In-Reply-To: <1292610010-19084-2-git-send-email-andros@netapp.com>
From: Andy Adamson <andros@netapp.com>
Fixes a bug where the svc_drop svc_xprt_put() call would try to free the
bc_xprt where the TCP transport xpo_free routine hits a NULL svc_sock pointer.
SVC_DROP should release the xprt, but not free it. svc_xprt_free should call
the back channel free routine.
Signed-off-by: Andy Adamson <andros@netapp.com>
---
fs/nfs/callback.c | 30 +++++++++++++++++++++---------
1 files changed, 21 insertions(+), 9 deletions(-)
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 93a8b3b..06f0ad9 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -177,30 +177,42 @@ nfs41_callback_svc(void *vrqstp)
struct svc_rqst *
nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt)
{
- struct svc_xprt *bc_xprt;
- struct svc_rqst *rqstp = ERR_PTR(-ENOMEM);
+ struct svc_rqst *rqstp;
+ int ret;
dprintk("--> %s\n", __func__);
- /* Create a svc_sock for the service */
- bc_xprt = svc_sock_create(serv, xprt->prot);
- if (!bc_xprt)
+
+ /* register the back channel transport class */
+ svc_init_bc_xprt_sock();
+
+ /*
+ * Create an svc_sock for the back channel service that shares the
+ * fore channel connection.
+ * Returns the input port (0) and sets the svc_serv bc_xprt on success
+ */
+ ret = svc_create_xprt(serv, "tcp-bc", &init_net, PF_INET, 0,
+ SVC_SOCK_ANONYMOUS);
+ if (ret < 0) {
+ rqstp = ERR_PTR(ret);
goto out;
+ }
/*
* Save the svc_serv in the transport so that it can
* be referenced when the session backchannel is initialized
*/
- serv->bc_xprt = bc_xprt;
xprt->bc_serv = serv;
INIT_LIST_HEAD(&serv->sv_cb_list);
spin_lock_init(&serv->sv_cb_lock);
init_waitqueue_head(&serv->sv_cb_waitq);
rqstp = svc_prepare_thread(serv, &serv->sv_pools[0]);
- if (IS_ERR(rqstp))
- svc_sock_destroy(bc_xprt);
+ if (IS_ERR(rqstp)) {
+ svc_xprt_put(serv->bc_xprt);
+ serv->bc_xprt = NULL;
+ }
out:
- dprintk("--> %s return %p\n", __func__, rqstp);
+ dprintk("--> %s return %ld\n", __func__, PTR_ERR(rqstp));
return rqstp;
}
--
1.6.6
next prev parent reply other threads:[~2010-12-17 18:20 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-12-17 18:20 [PATCH_V4 0/9] NFSv4 callback find client fix Version 4 andros
2010-12-17 18:20 ` [PATCH_V4 1/9] SUNRPC new transport for the NFSv4.1 shared back channel andros
2010-12-17 18:20 ` andros [this message]
2010-12-17 18:20 ` [PATCH_V4 3/9] NFS do not clear minor version at nfs_client free andros
2010-12-17 18:20 ` [PATCH_V4 4/9] NFS implement v4.0 callback_ident andros
2010-12-17 18:20 ` [PATCH_V4 5/9] NFS associate sessionid with callback connection andros
2010-12-17 18:20 ` [PATCH_V4 6/9] NFS reference nfs_client across cb_compound processing andros
2010-12-17 18:20 ` [PATCH_V4 7/9] NFS RPC_AUTH_GSS unsupported on v4.1 back channel andros
2010-12-17 18:20 ` [PATCH_V4 8/9] NFS add session back channel draining andros
2010-12-17 18:20 ` [PATCH_V4 9/9] NFS rename client back channel transport field andros
2010-12-20 14:05 ` [PATCH_V4 6/9] NFS reference nfs_client across cb_compound processing Benny Halevy
2010-12-20 15:44 ` Fred Isaman
2010-12-17 18:40 ` [PATCH_V4 1/9] SUNRPC new transport for the NFSv4.1 shared back channel J. Bruce Fields
2010-12-17 18:54 ` William A. (Andy) Adamson
2010-12-17 18:57 ` William A. (Andy) Adamson
2010-12-17 19:10 ` J. Bruce Fields
2010-12-17 19:16 ` William A. (Andy) Adamson
2010-12-17 19:33 ` J. Bruce Fields
2010-12-17 19:55 ` William A. (Andy) Adamson
2010-12-17 20:02 ` J. Bruce Fields
2010-12-17 20:11 ` William A. (Andy) Adamson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1292610010-19084-3-git-send-email-andros@netapp.com \
--to=andros@netapp.com \
--cc=bfields@redhat.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trond.myklebust@netapp.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).