aboutsummaryrefslogtreecommitdiff
path: root/net/sunrpc/xprtrdma/transport.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/sunrpc/xprtrdma/transport.c')
-rw-r--r--net/sunrpc/xprtrdma/transport.c133
1 files changed, 49 insertions, 84 deletions
diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c
index 06cdbff79e4..66f91f0d071 100644
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -51,6 +51,7 @@
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/seq_file.h>
+#include <linux/sunrpc/addr.h>
#include "xprt_rdma.h"
@@ -85,7 +86,7 @@ static unsigned int max_memreg = RPCRDMA_LAST - 1;
static struct ctl_table_header *sunrpc_table_header;
-static ctl_table xr_tunables_table[] = {
+static struct ctl_table xr_tunables_table[] = {
{
.procname = "rdma_slot_table_entries",
.data = &xprt_rdma_slot_table_entries,
@@ -137,7 +138,7 @@ static ctl_table xr_tunables_table[] = {
{ },
};
-static ctl_table sunrpc_table[] = {
+static struct ctl_table sunrpc_table[] = {
{
.procname = "sunrpc",
.mode = 0555,
@@ -148,6 +149,11 @@ static ctl_table sunrpc_table[] = {
#endif
+#define RPCRDMA_BIND_TO (60U * HZ)
+#define RPCRDMA_INIT_REEST_TO (5U * HZ)
+#define RPCRDMA_MAX_REEST_TO (30U * HZ)
+#define RPCRDMA_IDLE_DISC_TO (5U * 60 * HZ)
+
static struct rpc_xprt_ops xprt_rdma_procs; /* forward reference */
static void
@@ -199,21 +205,15 @@ xprt_rdma_connect_worker(struct work_struct *work)
struct rpc_xprt *xprt = &r_xprt->xprt;
int rc = 0;
- if (!xprt->shutdown) {
- current->flags |= PF_FSTRANS;
- xprt_clear_connected(xprt);
+ current->flags |= PF_FSTRANS;
+ xprt_clear_connected(xprt);
- dprintk("RPC: %s: %sconnect\n", __func__,
- r_xprt->rx_ep.rep_connected != 0 ? "re" : "");
- rc = rpcrdma_ep_connect(&r_xprt->rx_ep, &r_xprt->rx_ia);
- if (rc)
- goto out;
- }
- goto out_clear;
+ dprintk("RPC: %s: %sconnect\n", __func__,
+ r_xprt->rx_ep.rep_connected != 0 ? "re" : "");
+ rc = rpcrdma_ep_connect(&r_xprt->rx_ep, &r_xprt->rx_ia);
+ if (rc)
+ xprt_wake_pending_tasks(xprt, rc);
-out:
- xprt_wake_pending_tasks(xprt, rc);
-out_clear:
dprintk("RPC: %s: exit\n", __func__);
xprt_clear_connecting(xprt);
current->flags &= ~PF_FSTRANS;
@@ -234,7 +234,6 @@ static void
xprt_rdma_destroy(struct rpc_xprt *xprt)
{
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
- int rc;
dprintk("RPC: %s: called\n", __func__);
@@ -243,10 +242,7 @@ xprt_rdma_destroy(struct rpc_xprt *xprt)
xprt_clear_connected(xprt);
rpcrdma_buffer_destroy(&r_xprt->rx_buf);
- rc = rpcrdma_ep_destroy(&r_xprt->rx_ep, &r_xprt->rx_ia);
- if (rc)
- dprintk("RPC: %s: rpcrdma_ep_destroy returned %i\n",
- __func__, rc);
+ rpcrdma_ep_destroy(&r_xprt->rx_ep, &r_xprt->rx_ia);
rpcrdma_ia_close(&r_xprt->rx_ia);
xprt_rdma_free_addresses(xprt);
@@ -294,9 +290,9 @@ xprt_setup_rdma(struct xprt_create *args)
/* 60 second timeout, no retries */
xprt->timeout = &xprt_rdma_default_timeout;
- xprt->bind_timeout = (60U * HZ);
- xprt->reestablish_timeout = (5U * HZ);
- xprt->idle_timeout = (5U * 60 * HZ);
+ xprt->bind_timeout = RPCRDMA_BIND_TO;
+ xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
+ xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO;
xprt->resvport = 0; /* privileged port not needed */
xprt->tsh_size = 0; /* RPC-RDMA handles framing */
@@ -396,7 +392,7 @@ out4:
xprt_rdma_free_addresses(xprt);
rc = -EINVAL;
out3:
- (void) rpcrdma_ep_destroy(new_ep, &new_xprt->rx_ia);
+ rpcrdma_ep_destroy(new_ep, &new_xprt->rx_ia);
out2:
rpcrdma_ia_close(&new_xprt->rx_ia);
out1:
@@ -432,9 +428,8 @@ xprt_rdma_set_port(struct rpc_xprt *xprt, u16 port)
}
static void
-xprt_rdma_connect(struct rpc_task *task)
+xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
{
- struct rpc_xprt *xprt = (struct rpc_xprt *)task->tk_xprt;
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
if (r_xprt->rx_ep.rep_connected != 0) {
@@ -442,10 +437,10 @@ xprt_rdma_connect(struct rpc_task *task)
schedule_delayed_work(&r_xprt->rdma_connect,
xprt->reestablish_timeout);
xprt->reestablish_timeout <<= 1;
- if (xprt->reestablish_timeout > (30 * HZ))
- xprt->reestablish_timeout = (30 * HZ);
- else if (xprt->reestablish_timeout < (5 * HZ))
- xprt->reestablish_timeout = (5 * HZ);
+ if (xprt->reestablish_timeout > RPCRDMA_MAX_REEST_TO)
+ xprt->reestablish_timeout = RPCRDMA_MAX_REEST_TO;
+ else if (xprt->reestablish_timeout < RPCRDMA_INIT_REEST_TO)
+ xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
} else {
schedule_delayed_work(&r_xprt->rdma_connect, 0);
if (!RPC_IS_ASYNC(task))
@@ -453,23 +448,6 @@ xprt_rdma_connect(struct rpc_task *task)
}
}
-static int
-xprt_rdma_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
-{
- struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
- int credits = atomic_read(&r_xprt->rx_buf.rb_credits);
-
- /* == RPC_CWNDSCALE @ init, but *after* setup */
- if (r_xprt->rx_buf.rb_cwndscale == 0UL) {
- r_xprt->rx_buf.rb_cwndscale = xprt->cwnd;
- dprintk("RPC: %s: cwndscale %lu\n", __func__,
- r_xprt->rx_buf.rb_cwndscale);
- BUG_ON(r_xprt->rx_buf.rb_cwndscale <= 0);
- }
- xprt->cwnd = credits * r_xprt->rx_buf.rb_cwndscale;
- return xprt_reserve_xprt_cong(xprt, task);
-}
-
/*
* The RDMA allocate/free functions need the task structure as a place
* to hide the struct rpcrdma_req, which is necessary for the actual send/recv
@@ -481,11 +459,12 @@ xprt_rdma_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
static void *
xprt_rdma_allocate(struct rpc_task *task, size_t size)
{
- struct rpc_xprt *xprt = task->tk_xprt;
+ struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt;
struct rpcrdma_req *req, *nreq;
req = rpcrdma_buffer_get(&rpcx_to_rdmax(xprt)->rx_buf);
- BUG_ON(NULL == req);
+ if (req == NULL)
+ return NULL;
if (size > req->rl_size) {
dprintk("RPC: %s: size %zd too large for buffer[%zd]: "
@@ -509,18 +488,6 @@ xprt_rdma_allocate(struct rpc_task *task, size_t size)
* If the allocation or registration fails, the RPC framework
* will (doggedly) retry.
*/
- if (rpcx_to_rdmax(xprt)->rx_ia.ri_memreg_strategy ==
- RPCRDMA_BOUNCEBUFFERS) {
- /* forced to "pure inline" */
- dprintk("RPC: %s: too much data (%zd) for inline "
- "(r/w max %d/%d)\n", __func__, size,
- rpcx_to_rdmad(xprt).inline_rsize,
- rpcx_to_rdmad(xprt).inline_wsize);
- size = req->rl_size;
- rpc_exit(task, -EIO); /* fail the operation */
- rpcx_to_rdmax(xprt)->rx_stats.failed_marshal_count++;
- goto out;
- }
if (task->tk_flags & RPC_TASK_SWAPPER)
nreq = kmalloc(sizeof *req + size, GFP_ATOMIC);
else
@@ -549,7 +516,6 @@ xprt_rdma_allocate(struct rpc_task *task, size_t size)
req = nreq;
}
dprintk("RPC: %s: size %zd, request 0x%p\n", __func__, size, req);
-out:
req->rl_connect_cookie = 0; /* our reserved value */
return req->rl_xdr_buf;
@@ -585,9 +551,7 @@ xprt_rdma_free(void *buffer)
__func__, rep, (rep && rep->rr_func) ? " (with waiter)" : "");
/*
- * Finish the deregistration. When using mw bind, this was
- * begun in rpcrdma_reply_handler(). In all other modes, we
- * do it here, in thread context. The process is considered
+ * Finish the deregistration. The process is considered
* complete when the rr_func vector becomes NULL - this
* was put in place during rpcrdma_reply_handler() - the wait
* call below will not block if the dereg is "done". If
@@ -596,12 +560,7 @@ xprt_rdma_free(void *buffer)
for (i = 0; req->rl_nchunks;) {
--req->rl_nchunks;
i += rpcrdma_deregister_external(
- &req->rl_segments[i], r_xprt, NULL);
- }
-
- if (rep && wait_event_interruptible(rep->rr_unbind, !rep->rr_func)) {
- rep->rr_func = NULL; /* abandon the callback */
- req->rl_reply = NULL;
+ &req->rl_segments[i], r_xprt);
}
if (req->rl_iov.length == 0) { /* see allocate above */
@@ -633,16 +592,15 @@ static int
xprt_rdma_send_request(struct rpc_task *task)
{
struct rpc_rqst *rqst = task->tk_rqstp;
- struct rpc_xprt *xprt = task->tk_xprt;
+ struct rpc_xprt *xprt = rqst->rq_xprt;
struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+ int rc;
- /* marshal the send itself */
- if (req->rl_niovs == 0 && rpcrdma_marshal_req(rqst) != 0) {
- r_xprt->rx_stats.failed_marshal_count++;
- dprintk("RPC: %s: rpcrdma_marshal_req failed\n",
- __func__);
- return -EIO;
+ if (req->rl_niovs == 0) {
+ rc = rpcrdma_marshal_req(rqst);
+ if (rc < 0)
+ goto failed_marshal;
}
if (req->rl_reply == NULL) /* e.g. reconnection */
@@ -666,6 +624,12 @@ xprt_rdma_send_request(struct rpc_task *task)
rqst->rq_bytes_sent = 0;
return 0;
+failed_marshal:
+ r_xprt->rx_stats.failed_marshal_count++;
+ dprintk("RPC: %s: rpcrdma_marshal_req failed, status %i\n",
+ __func__, rc);
+ if (rc == -EIO)
+ return -EIO;
drop_connection:
xprt_disconnect_done(xprt);
return -ENOTCONN; /* implies disconnect */
@@ -711,8 +675,9 @@ static void xprt_rdma_print_stats(struct rpc_xprt *xprt, struct seq_file *seq)
*/
static struct rpc_xprt_ops xprt_rdma_procs = {
- .reserve_xprt = xprt_rdma_reserve_xprt,
+ .reserve_xprt = xprt_reserve_xprt_cong,
.release_xprt = xprt_release_xprt_cong, /* sunrpc/xprt.c */
+ .alloc_slot = xprt_alloc_slot,
.release_request = xprt_release_rqst_cong, /* ditto */
.set_retrans_timeout = xprt_set_retrans_timeout_def, /* ditto */
.rpcbind = rpcb_getport_async, /* sunrpc/rpcb_clnt.c */
@@ -738,7 +703,7 @@ static void __exit xprt_rdma_cleanup(void)
{
int rc;
- dprintk(KERN_INFO "RPCRDMA Module Removed, deregister RPC RDMA transport\n");
+ dprintk("RPCRDMA Module Removed, deregister RPC RDMA transport\n");
#ifdef RPC_DEBUG
if (sunrpc_table_header) {
unregister_sysctl_table(sunrpc_table_header);
@@ -760,14 +725,14 @@ static int __init xprt_rdma_init(void)
if (rc)
return rc;
- dprintk(KERN_INFO "RPCRDMA Module Init, register RPC RDMA transport\n");
+ dprintk("RPCRDMA Module Init, register RPC RDMA transport\n");
- dprintk(KERN_INFO "Defaults:\n");
- dprintk(KERN_INFO "\tSlots %d\n"
+ dprintk("Defaults:\n");
+ dprintk("\tSlots %d\n"
"\tMaxInlineRead %d\n\tMaxInlineWrite %d\n",
xprt_rdma_slot_table_entries,
xprt_rdma_max_inline_read, xprt_rdma_max_inline_write);
- dprintk(KERN_INFO "\tPadding %d\n\tMemreg %d\n",
+ dprintk("\tPadding %d\n\tMemreg %d\n",
xprt_rdma_inline_write_padding, xprt_rdma_memreg_strategy);
#ifdef RPC_DEBUG