Merge tag 'nfs-for-3.13-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client bugfixes:
- Stable fix for data corruption when retransmitting O_DIRECT writes
- Stable fix for a deep recursion/stack overflow bug in rpc_release_client
- Stable fix for infinite looping when mounting a NFSv4.x volume
- Fix a typo in the nfs mount option parser
- Allow pNFS layouts to be compiled into the kernel when NFSv4.1 is
* tag 'nfs-for-3.13-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
nfs: fix pnfs Kconfig defaults
NFS: correctly report misuse of "migration" mount option.
nfs: don't retry detect_trunking with RPC_AUTH_UNIX more than once
SUNRPC: Avoid deep recursion in rpc_release_client
SUNRPC: Fix a data corruption issue when retransmitting RPC calls
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 38c1768..3dece03 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -116,17 +116,17 @@
config PNFS_FILE_LAYOUT
tristate
depends on NFS_V4_1
- default m
+ default NFS_V4
config PNFS_BLOCK
tristate
depends on NFS_V4_1 && BLK_DEV_DM
- default m
+ default NFS_V4
config PNFS_OBJLAYOUT
tristate
depends on NFS_V4_1 && SCSI_OSD_ULD
- default m
+ default NFS_V4
config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN
string "NFSv4.1 Implementation ID Domain"
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 74a7e12..059c01b 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -2093,10 +2093,15 @@
nfs4_root_machine_cred(clp);
goto again;
}
- if (i > 2)
+ if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX)
break;
case -NFS4ERR_CLID_INUSE:
case -NFS4ERR_WRONGSEC:
+ /* No point in retrying if we already used RPC_AUTH_UNIX */
+ if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX) {
+ status = -EPERM;
+ break;
+ }
clnt = rpc_clone_client_set_auth(clnt, RPC_AUTH_UNIX);
if (IS_ERR(clnt)) {
status = PTR_ERR(clnt);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 317d6fc..910ed90 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1614,7 +1614,7 @@
goto out_minorversion_mismatch;
if (mnt->options & NFS_OPTION_MIGRATION &&
- mnt->version != 4 && mnt->minorversion != 0)
+ (mnt->version != 4 || mnt->minorversion != 0))
goto out_migration_misuse;
/*
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index dab09da..f09b7db 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -750,14 +750,16 @@
/*
* Free an RPC client
*/
-static void
+static struct rpc_clnt *
rpc_free_client(struct rpc_clnt *clnt)
{
+ struct rpc_clnt *parent = NULL;
+
dprintk_rcu("RPC: destroying %s client for %s\n",
clnt->cl_program->name,
rcu_dereference(clnt->cl_xprt)->servername);
if (clnt->cl_parent != clnt)
- rpc_release_client(clnt->cl_parent);
+ parent = clnt->cl_parent;
rpc_clnt_remove_pipedir(clnt);
rpc_unregister_client(clnt);
rpc_free_iostats(clnt->cl_metrics);
@@ -766,18 +768,17 @@
rpciod_down();
rpc_free_clid(clnt);
kfree(clnt);
+ return parent;
}
/*
* Free an RPC client
*/
-static void
+static struct rpc_clnt *
rpc_free_auth(struct rpc_clnt *clnt)
{
- if (clnt->cl_auth == NULL) {
- rpc_free_client(clnt);
- return;
- }
+ if (clnt->cl_auth == NULL)
+ return rpc_free_client(clnt);
/*
* Note: RPCSEC_GSS may need to send NULL RPC calls in order to
@@ -788,7 +789,8 @@
rpcauth_release(clnt->cl_auth);
clnt->cl_auth = NULL;
if (atomic_dec_and_test(&clnt->cl_count))
- rpc_free_client(clnt);
+ return rpc_free_client(clnt);
+ return NULL;
}
/*
@@ -799,10 +801,13 @@
{
dprintk("RPC: rpc_release_client(%p)\n", clnt);
- if (list_empty(&clnt->cl_tasks))
- wake_up(&destroy_wait);
- if (atomic_dec_and_test(&clnt->cl_count))
- rpc_free_auth(clnt);
+ do {
+ if (list_empty(&clnt->cl_tasks))
+ wake_up(&destroy_wait);
+ if (!atomic_dec_and_test(&clnt->cl_count))
+ break;
+ clnt = rpc_free_auth(clnt);
+ } while (clnt != NULL);
}
EXPORT_SYMBOL_GPL(rpc_release_client);
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 17c8892..dd9d295 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -393,8 +393,10 @@
return kernel_sendmsg(sock, &msg, NULL, 0, 0);
}
-static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned int base, int more)
+static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned int base, int more, bool zerocopy)
{
+ ssize_t (*do_sendpage)(struct socket *sock, struct page *page,
+ int offset, size_t size, int flags);
struct page **ppage;
unsigned int remainder;
int err, sent = 0;
@@ -403,6 +405,9 @@
base += xdr->page_base;
ppage = xdr->pages + (base >> PAGE_SHIFT);
base &= ~PAGE_MASK;
+ do_sendpage = sock->ops->sendpage;
+ if (!zerocopy)
+ do_sendpage = sock_no_sendpage;
for(;;) {
unsigned int len = min_t(unsigned int, PAGE_SIZE - base, remainder);
int flags = XS_SENDMSG_FLAGS;
@@ -410,7 +415,7 @@
remainder -= len;
if (remainder != 0 || more)
flags |= MSG_MORE;
- err = sock->ops->sendpage(sock, *ppage, base, len, flags);
+ err = do_sendpage(sock, *ppage, base, len, flags);
if (remainder == 0 || err != len)
break;
sent += err;
@@ -431,9 +436,10 @@
* @addrlen: UDP only -- length of destination address
* @xdr: buffer containing this request
* @base: starting position in the buffer
+ * @zerocopy: true if it is safe to use sendpage()
*
*/
-static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen, struct xdr_buf *xdr, unsigned int base)
+static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen, struct xdr_buf *xdr, unsigned int base, bool zerocopy)
{
unsigned int remainder = xdr->len - base;
int err, sent = 0;
@@ -461,7 +467,7 @@
if (base < xdr->page_len) {
unsigned int len = xdr->page_len - base;
remainder -= len;
- err = xs_send_pagedata(sock, xdr, base, remainder != 0);
+ err = xs_send_pagedata(sock, xdr, base, remainder != 0, zerocopy);
if (remainder == 0 || err != len)
goto out;
sent += err;
@@ -564,7 +570,7 @@
req->rq_svec->iov_base, req->rq_svec->iov_len);
status = xs_sendpages(transport->sock, NULL, 0,
- xdr, req->rq_bytes_sent);
+ xdr, req->rq_bytes_sent, true);
dprintk("RPC: %s(%u) = %d\n",
__func__, xdr->len - req->rq_bytes_sent, status);
if (likely(status >= 0)) {
@@ -620,7 +626,7 @@
status = xs_sendpages(transport->sock,
xs_addr(xprt),
xprt->addrlen, xdr,
- req->rq_bytes_sent);
+ req->rq_bytes_sent, true);
dprintk("RPC: xs_udp_send_request(%u) = %d\n",
xdr->len - req->rq_bytes_sent, status);
@@ -693,6 +699,7 @@
struct rpc_xprt *xprt = req->rq_xprt;
struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
struct xdr_buf *xdr = &req->rq_snd_buf;
+ bool zerocopy = true;
int status;
xs_encode_stream_record_marker(&req->rq_snd_buf);
@@ -700,13 +707,20 @@
xs_pktdump("packet data:",
req->rq_svec->iov_base,
req->rq_svec->iov_len);
+ /* Don't use zero copy if this is a resend. If the RPC call
+ * completes while the socket holds a reference to the pages,
+ * then we may end up resending corrupted data.
+ */
+ if (task->tk_flags & RPC_TASK_SENT)
+ zerocopy = false;
/* Continue transmitting the packet/record. We must be careful
* to cope with writespace callbacks arriving _after_ we have
* called sendmsg(). */
while (1) {
status = xs_sendpages(transport->sock,
- NULL, 0, xdr, req->rq_bytes_sent);
+ NULL, 0, xdr, req->rq_bytes_sent,
+ zerocopy);
dprintk("RPC: xs_tcp_send_request(%u) = %d\n",
xdr->len - req->rq_bytes_sent, status);
