sunrpc: fix leak on error on socket xprt setup
Subsystems (3)
PATCH 1/1 sunrpc: fix leak on error on socket xprt setup
From: J. Bruce Fields <[email protected]>
Also collect exit code together while we're at it.
--- net/sunrpc/xprtsock.c | 29 ++++++++++++++++++----------- 1 files changed, 18 insertions(+), 11 deletions(-)
I think I sent this before, but it must have fallen through the cracks.
--b.
net/sunrpc/xprtsock.c +18 -11
@@ -2297,6 +2297,7 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)struct sockaddr *addr = args->dstaddr;struct rpc_xprt *xprt;struct sock_xprt *transport; +struct rpc_xprt *ret; xprt =xs_setup_xprt (args, xprt_udp_slot_table_entries);if (IS_ERR (xprt))
@@ -2334,8 +2335,8 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)xs_format_peer_addresses (xprt,"udp" ,RPCBIND_NETID_UDP6 );break ;default : -kfree (xprt); -return ERR_PTR (-EAFNOSUPPORT ); + ret =ERR_PTR (-EAFNOSUPPORT ); + goto out_err; }if (xprt_bound (xprt))
@@ -2350,10 +2351,11 @@ static struct rpc_xprt *xs_setup_udp(struct xprt_create *args)if (try_module_get (THIS_MODULE ))return xprt; - + ret =ERR_PTR (-EINVAL ); +out_err:kfree (xprt->slot);kfree (xprt); -return ERR_PTR (-EINVAL ); +return ret; }static const struct rpc_timeout xs_tcp_default_timeout = {
@@ -2372,6 +2374,7 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)struct sockaddr *addr = args->dstaddr;struct rpc_xprt *xprt;struct sock_xprt *transport; +struct rpc_xprt *ret; xprt =xs_setup_xprt (args, xprt_tcp_slot_table_entries);if (IS_ERR (xprt))
@@ -2407,8 +2410,8 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)xs_format_peer_addresses (xprt,"tcp" ,RPCBIND_NETID_TCP6 );break ;default : -kfree (xprt); -return ERR_PTR (-EAFNOSUPPORT ); + ret =ERR_PTR (-EAFNOSUPPORT ); + goto out_err; }if (xprt_bound (xprt))
@@ -2424,10 +2427,11 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args)if (try_module_get (THIS_MODULE ))return xprt; - + ret =ERR_PTR (-EINVAL ); +out_err:kfree (xprt->slot);kfree (xprt); -return ERR_PTR (-EINVAL ); +return ret; }/**
@@ -2441,6 +2445,7 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args)struct rpc_xprt *xprt;struct sock_xprt *transport;struct svc_sock *bc_sock; +struct rpc_xprt *ret; xprt =xs_setup_xprt (args, xprt_tcp_slot_table_entries);if (IS_ERR (xprt))
@@ -2480,8 +2485,8 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args) RPCBIND_NETID_TCP6);break ;default : -kfree (xprt); -return ERR_PTR (-EAFNOSUPPORT ); + ret =ERR_PTR (-EAFNOSUPPORT ); + goto out_err; }if (xprt_bound (xprt))
@@ -2503,9 +2508,11 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args)if (try_module_get (THIS_MODULE ))return xprt; + ret =ERR_PTR (-EINVAL ); +out_err:kfree (xprt->slot);kfree (xprt); -return ERR_PTR (-EINVAL ); +return ret; }static struct xprt_class xs_udp_transport = {
Signature · 1 line
1.7.0.4
Quoted 11 lines · J. Bruce Fields
From: J. Bruce Fields <[email protected]>
Also collect exit code together while we're at it.
Signed-off-by J. Bruce Fields <[email protected]>Cc Trond Myklebust <[email protected]>--- net/sunrpc/xprtsock.c | 29 ++++++++++++++++++----------- 1 files changed, 18 insertions(+), 11 deletions(-)I think I sent this before, but it must have fallen through the cracks.
The patch doesn't really ring a bell, but it looks correct. On the other hand, this doesn't look like a bug we'd ever hit in practice, so I'll probably delay merging it until I've got a few more critical bugfixes queued up.
Unless, of course, you'd prefer to merge it yourself?
Cheers
Trond
A message that is not in the archive (sent off-list or not yet received): 1274447779.3692.8.camel-rJ7iovZKK19ZJLDQqaL3InhyD016LWXt@public.gmane.org
Quoted 11 lines · Trond Myklebust
Quoted 10 lines · J. Bruce Fields
From: J. Bruce Fields <[email protected]>
Also collect exit code together while we're at it.
Signed-off-by J. Bruce Fields <[email protected]>Cc Trond Myklebust <[email protected]>--- net/sunrpc/xprtsock.c | 29 ++++++++++++++++++----------- 1 files changed, 18 insertions(+), 11 deletions(-)
Quoted 4 lines
Quoted 1 line
I think I sent this before, but it must have fallen through the cracks.
The patch doesn't really ring a bell, but it looks correct. On the other hand, this doesn't look like a bug we'd ever hit in practice,
... and even if we do, the consequences aren't much.
Quoted 3 lines
so I'll
probably delay merging it until I've got a few more critical bugfixes queued up.
I'd've probably left it to the next merge window.
Quoted 1 line
Unless, of course, you'd prefer to merge it yourself?
No preference; if you're in no hurry I'd just add it to my for-2.6.36 branch with your ack.
--b.