summaryrefslogtreecommitdiff
path: root/net/sunrpc/svcsock.c
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2012-12-20 14:04:11 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2012-12-20 14:04:11 -0800
commit982197277c85018cc6eb77f1d3bef17933b0c5fd (patch)
tree805fcef9ec7c1e83867b89332fd37f751594fae3 /net/sunrpc/svcsock.c
parent40889e8d9fc6355980cf2bc94ef4356c10dec4ec (diff)
parent24ffb93872f7363a01ad639e3c8a9889b46c3f0a (diff)
Merge branch 'for-3.8' of git://linux-nfs.org/~bfields/linux
Pull nfsd update from Bruce Fields: "Included this time: - more nfsd containerization work from Stanislav Kinsbursky: we're not quite there yet, but should be by 3.9. - NFSv4.1 progress: implementation of basic backchannel security negotiation and the mandatory BACKCHANNEL_CTL operation. See http://wiki.linux-nfs.org/wiki/index.php/Server_4.0_and_4.1_issues for remaining TODO's - Fixes for some bugs that could be triggered by unusual compounds. Our xdr code wasn't designed with v4 compounds in mind, and it shows. A more thorough rewrite is still a todo. - If you've ever seen "RPC: multiple fragments per record not supported" logged while using some sort of odd userland NFS client, that should now be fixed. - Further work from Jeff Layton on our mechanism for storing information about NFSv4 clients across reboots. - Further work from Bryan Schumaker on his fault-injection mechanism (which allows us to discard selective NFSv4 state, to excercise rarely-taken recovery code paths in the client.) - The usual mix of miscellaneous bugs and cleanup. Thanks to everyone who tested or contributed this cycle." * 'for-3.8' of git://linux-nfs.org/~bfields/linux: (111 commits) nfsd4: don't leave freed stateid hashed nfsd4: free_stateid can use the current stateid nfsd4: cleanup: replace rq_resused count by rq_next_page pointer nfsd: warn on odd reply state in nfsd_vfs_read nfsd4: fix oops on unusual readlike compound nfsd4: disable zero-copy on non-final read ops svcrpc: fix some printks NFSD: Correct the size calculation in fault_inject_write NFSD: Pass correct buffer size to rpc_ntop nfsd: pass proper net to nfsd_destroy() from NFSd kthreads nfsd: simplify service shutdown nfsd: replace boolean nfsd_up flag by users counter nfsd: simplify NFSv4 state init and shutdown nfsd: introduce helpers for generic resources init and shutdown nfsd: make NFSd service structure allocated per net nfsd: make NFSd service boot time per-net nfsd: per-net NFSd up flag introduced nfsd: move per-net startup code to separated function nfsd: pass net to __write_ports() and down nfsd: pass net to nfsd_set_nrthreads() ...
Diffstat (limited to 'net/sunrpc/svcsock.c')
-rw-r--r--net/sunrpc/svcsock.c98
1 files changed, 50 insertions, 48 deletions
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index cc3020d16789..0a148c9d2a5c 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -605,6 +605,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
rqstp->rq_respages = rqstp->rq_pages + 1 +
DIV_ROUND_UP(rqstp->rq_arg.page_len, PAGE_SIZE);
}
+ rqstp->rq_next_page = rqstp->rq_respages+1;
if (serv->sv_stats)
serv->sv_stats->netudpcnt++;
@@ -878,9 +879,9 @@ static unsigned int svc_tcp_restore_pages(struct svc_sock *svsk, struct svc_rqst
{
unsigned int i, len, npages;
- if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+ if (svsk->sk_datalen == 0)
return 0;
- len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+ len = svsk->sk_datalen;
npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
for (i = 0; i < npages; i++) {
if (rqstp->rq_pages[i] != NULL)
@@ -897,9 +898,9 @@ static void svc_tcp_save_pages(struct svc_sock *svsk, struct svc_rqst *rqstp)
{
unsigned int i, len, npages;
- if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+ if (svsk->sk_datalen == 0)
return;
- len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+ len = svsk->sk_datalen;
npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
for (i = 0; i < npages; i++) {
svsk->sk_pages[i] = rqstp->rq_pages[i];
@@ -911,9 +912,9 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk)
{
unsigned int i, len, npages;
- if (svsk->sk_tcplen <= sizeof(rpc_fraghdr))
+ if (svsk->sk_datalen == 0)
goto out;
- len = svsk->sk_tcplen - sizeof(rpc_fraghdr);
+ len = svsk->sk_datalen;
npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
for (i = 0; i < npages; i++) {
BUG_ON(svsk->sk_pages[i] == NULL);
@@ -922,13 +923,12 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk)
}
out:
svsk->sk_tcplen = 0;
+ svsk->sk_datalen = 0;
}
/*
- * Receive data.
+ * Receive fragment record header.
* If we haven't gotten the record length yet, get the next four bytes.
- * Otherwise try to gobble up as much as possible up to the complete
- * record length.
*/
static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
{
@@ -954,32 +954,16 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp)
return -EAGAIN;
}
- svsk->sk_reclen = ntohl(svsk->sk_reclen);
- if (!(svsk->sk_reclen & RPC_LAST_STREAM_FRAGMENT)) {
- /* FIXME: technically, a record can be fragmented,
- * and non-terminal fragments will not have the top
- * bit set in the fragment length header.
- * But apparently no known nfs clients send fragmented
- * records. */
- net_notice_ratelimited("RPC: multiple fragments per record not supported\n");
- goto err_delete;
- }
-
- svsk->sk_reclen &= RPC_FRAGMENT_SIZE_MASK;
- dprintk("svc: TCP record, %d bytes\n", svsk->sk_reclen);
- if (svsk->sk_reclen > serv->sv_max_mesg) {
- net_notice_ratelimited("RPC: fragment too large: 0x%08lx\n",
- (unsigned long)svsk->sk_reclen);
+ dprintk("svc: TCP record, %d bytes\n", svc_sock_reclen(svsk));
+ if (svc_sock_reclen(svsk) + svsk->sk_datalen >
+ serv->sv_max_mesg) {
+ net_notice_ratelimited("RPC: fragment too large: %d\n",
+ svc_sock_reclen(svsk));
goto err_delete;
}
}
- if (svsk->sk_reclen < 8)
- goto err_delete; /* client is nuts. */
-
- len = svsk->sk_reclen;
-
- return len;
+ return svc_sock_reclen(svsk);
error:
dprintk("RPC: TCP recv_record got %d\n", len);
return len;
@@ -1023,7 +1007,7 @@ static int receive_cb_reply(struct svc_sock *svsk, struct svc_rqst *rqstp)
if (dst->iov_len < src->iov_len)
return -EAGAIN; /* whatever; just giving up. */
memcpy(dst->iov_base, src->iov_base, src->iov_len);
- xprt_complete_rqst(req->rq_task, svsk->sk_reclen);
+ xprt_complete_rqst(req->rq_task, rqstp->rq_arg.len);
rqstp->rq_arg.len = 0;
return 0;
}
@@ -1042,6 +1026,17 @@ static int copy_pages_to_kvecs(struct kvec *vec, struct page **pages, int len)
return i;
}
+static void svc_tcp_fragment_received(struct svc_sock *svsk)
+{
+ /* If we have more data, signal svc_xprt_enqueue() to try again */
+ if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
+ set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+ dprintk("svc: TCP %s record (%d bytes)\n",
+ svc_sock_final_rec(svsk) ? "final" : "nonfinal",
+ svc_sock_reclen(svsk));
+ svsk->sk_tcplen = 0;
+ svsk->sk_reclen = 0;
+}
/*
* Receive data from a TCP socket.
@@ -1068,29 +1063,39 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
goto error;
base = svc_tcp_restore_pages(svsk, rqstp);
- want = svsk->sk_reclen - base;
+ want = svc_sock_reclen(svsk) - (svsk->sk_tcplen - sizeof(rpc_fraghdr));
vec = rqstp->rq_vec;
pnum = copy_pages_to_kvecs(&vec[0], &rqstp->rq_pages[0],
- svsk->sk_reclen);
+ svsk->sk_datalen + want);
rqstp->rq_respages = &rqstp->rq_pages[pnum];
+ rqstp->rq_next_page = rqstp->rq_respages + 1;
/* Now receive data */
len = svc_partial_recvfrom(rqstp, vec, pnum, want, base);
- if (len >= 0)
+ if (len >= 0) {
svsk->sk_tcplen += len;
- if (len != want) {
+ svsk->sk_datalen += len;
+ }
+ if (len != want || !svc_sock_final_rec(svsk)) {
svc_tcp_save_pages(svsk, rqstp);
if (len < 0 && len != -EAGAIN)
- goto err_other;
- dprintk("svc: incomplete TCP record (%d of %d)\n",
- svsk->sk_tcplen, svsk->sk_reclen);
+ goto err_delete;
+ if (len == want)
+ svc_tcp_fragment_received(svsk);
+ else
+ dprintk("svc: incomplete TCP record (%d of %d)\n",
+ (int)(svsk->sk_tcplen - sizeof(rpc_fraghdr)),
+ svc_sock_reclen(svsk));
goto err_noclose;
}
- rqstp->rq_arg.len = svsk->sk_reclen;
+ if (svc_sock_reclen(svsk) < 8)
+ goto err_delete; /* client is nuts. */
+
+ rqstp->rq_arg.len = svsk->sk_datalen;
rqstp->rq_arg.page_base = 0;
if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) {
rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len;
@@ -1107,11 +1112,8 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
len = receive_cb_reply(svsk, rqstp);
/* Reset TCP read info */
- svsk->sk_reclen = 0;
- svsk->sk_tcplen = 0;
- /* If we have more data, signal svc_xprt_enqueue() to try again */
- if (svc_recv_available(svsk) > sizeof(rpc_fraghdr))
- set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
+ svsk->sk_datalen = 0;
+ svc_tcp_fragment_received(svsk);
if (len < 0)
goto error;
@@ -1120,15 +1122,14 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp)
if (serv->sv_stats)
serv->sv_stats->nettcpcnt++;
- dprintk("svc: TCP complete record (%d bytes)\n", rqstp->rq_arg.len);
return rqstp->rq_arg.len;
error:
if (len != -EAGAIN)
- goto err_other;
+ goto err_delete;
dprintk("RPC: TCP recvfrom got EAGAIN\n");
return 0;
-err_other:
+err_delete:
printk(KERN_NOTICE "%s: recvfrom returned errno %d\n",
svsk->sk_xprt.xpt_server->sv_name, -len);
set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
@@ -1305,6 +1306,7 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
svsk->sk_reclen = 0;
svsk->sk_tcplen = 0;
+ svsk->sk_datalen = 0;
memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;