aboutsummaryrefslogtreecommitdiff
path: root/net/rds
diff options
context:
space:
mode:
Diffstat (limited to 'net/rds')
-rw-r--r--net/rds/Kconfig4
-rw-r--r--net/rds/bind.c5
-rw-r--r--net/rds/connection.c21
-rw-r--r--net/rds/ib.c3
-rw-r--r--net/rds/ib.h5
-rw-r--r--net/rds/ib_cm.c13
-rw-r--r--net/rds/ib_recv.c47
-rw-r--r--net/rds/ib_send.c9
-rw-r--r--net/rds/ib_sysctl.c13
-rw-r--r--net/rds/info.c6
-rw-r--r--net/rds/iw.c3
-rw-r--r--net/rds/iw_cm.c2
-rw-r--r--net/rds/iw_recv.c13
-rw-r--r--net/rds/iw_send.c4
-rw-r--r--net/rds/iw_sysctl.c14
-rw-r--r--net/rds/loop.c4
-rw-r--r--net/rds/message.c8
-rw-r--r--net/rds/page.c9
-rw-r--r--net/rds/rdma_transport.c2
-rw-r--r--net/rds/rds.h4
-rw-r--r--net/rds/recv.c6
-rw-r--r--net/rds/send.c11
-rw-r--r--net/rds/stats.c1
-rw-r--r--net/rds/sysctl.c14
-rw-r--r--net/rds/tcp.h4
-rw-r--r--net/rds/tcp_connect.c4
-rw-r--r--net/rds/tcp_listen.c14
-rw-r--r--net/rds/tcp_recv.c23
-rw-r--r--net/rds/tcp_send.c6
29 files changed, 125 insertions, 147 deletions
diff --git a/net/rds/Kconfig b/net/rds/Kconfig
index ec753b3ae72..f2c670ba7b9 100644
--- a/net/rds/Kconfig
+++ b/net/rds/Kconfig
@@ -1,7 +1,7 @@
config RDS
- tristate "The RDS Protocol (EXPERIMENTAL)"
- depends on INET && EXPERIMENTAL
+ tristate "The RDS Protocol"
+ depends on INET
---help---
The RDS (Reliable Datagram Sockets) protocol provides reliable,
sequenced delivery of datagrams over Infiniband, iWARP,
diff --git a/net/rds/bind.c b/net/rds/bind.c
index 637bde56c9d..a2e6562da75 100644
--- a/net/rds/bind.c
+++ b/net/rds/bind.c
@@ -52,13 +52,12 @@ static struct rds_sock *rds_bind_lookup(__be32 addr, __be16 port,
struct rds_sock *insert)
{
struct rds_sock *rs;
- struct hlist_node *node;
struct hlist_head *head = hash_to_bucket(addr, port);
u64 cmp;
u64 needle = ((u64)be32_to_cpu(addr) << 32) | be16_to_cpu(port);
rcu_read_lock();
- hlist_for_each_entry_rcu(rs, node, head, rs_bound_node) {
+ hlist_for_each_entry_rcu(rs, head, rs_bound_node) {
cmp = ((u64)be32_to_cpu(rs->rs_bound_addr) << 32) |
be16_to_cpu(rs->rs_bound_port);
@@ -118,7 +117,7 @@ static int rds_add_bound(struct rds_sock *rs, __be32 addr, __be16 *port)
rover = be16_to_cpu(*port);
last = rover;
} else {
- rover = max_t(u16, net_random(), 2);
+ rover = max_t(u16, prandom_u32(), 2);
last = rover - 1;
}
diff --git a/net/rds/connection.c b/net/rds/connection.c
index 9e07c756d1f..378c3a6acf8 100644
--- a/net/rds/connection.c
+++ b/net/rds/connection.c
@@ -51,10 +51,16 @@ static struct kmem_cache *rds_conn_slab;
static struct hlist_head *rds_conn_bucket(__be32 laddr, __be32 faddr)
{
+ static u32 rds_hash_secret __read_mostly;
+
+ unsigned long hash;
+
+ net_get_random_once(&rds_hash_secret, sizeof(rds_hash_secret));
+
/* Pass NULL, don't need struct net for hash */
- unsigned long hash = inet_ehashfn(NULL,
- be32_to_cpu(laddr), 0,
- be32_to_cpu(faddr), 0);
+ hash = __inet_ehashfn(be32_to_cpu(laddr), 0,
+ be32_to_cpu(faddr), 0,
+ rds_hash_secret);
return &rds_conn_hash[hash & RDS_CONNECTION_HASH_MASK];
}
@@ -69,9 +75,8 @@ static struct rds_connection *rds_conn_lookup(struct hlist_head *head,
struct rds_transport *trans)
{
struct rds_connection *conn, *ret = NULL;
- struct hlist_node *pos;
- hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) {
+ hlist_for_each_entry_rcu(conn, head, c_hash_node) {
if (conn->c_faddr == faddr && conn->c_laddr == laddr &&
conn->c_trans == trans) {
ret = conn;
@@ -376,7 +381,6 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len,
int want_send)
{
struct hlist_head *head;
- struct hlist_node *pos;
struct list_head *list;
struct rds_connection *conn;
struct rds_message *rm;
@@ -390,7 +394,7 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len,
for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash);
i++, head++) {
- hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) {
+ hlist_for_each_entry_rcu(conn, head, c_hash_node) {
if (want_send)
list = &conn->c_send_queue;
else
@@ -439,7 +443,6 @@ void rds_for_each_conn_info(struct socket *sock, unsigned int len,
{
uint64_t buffer[(item_len + 7) / 8];
struct hlist_head *head;
- struct hlist_node *pos;
struct rds_connection *conn;
size_t i;
@@ -450,7 +453,7 @@ void rds_for_each_conn_info(struct socket *sock, unsigned int len,
for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash);
i++, head++) {
- hlist_for_each_entry_rcu(conn, pos, head, c_hash_node) {
+ hlist_for_each_entry_rcu(conn, head, c_hash_node) {
/* XXX no c_lock usage.. */
if (!visitor(conn, buffer))
diff --git a/net/rds/ib.c b/net/rds/ib.c
index b4c8b0022fe..ba2dffeff60 100644
--- a/net/rds/ib.c
+++ b/net/rds/ib.c
@@ -338,7 +338,8 @@ static int rds_ib_laddr_check(__be32 addr)
ret = rdma_bind_addr(cm_id, (struct sockaddr *)&sin);
/* due to this, we will claim to support iWARP devices unless we
check node_type. */
- if (ret || cm_id->device->node_type != RDMA_NODE_IB_CA)
+ if (ret || !cm_id->device ||
+ cm_id->device->node_type != RDMA_NODE_IB_CA)
ret = -EADDRNOTAVAIL;
rdsdebug("addr %pI4 ret %d node type %d\n",
diff --git a/net/rds/ib.h b/net/rds/ib.h
index edfaaaf164e..7280ab8810c 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -50,7 +50,7 @@ struct rds_ib_cache_head {
};
struct rds_ib_refill_cache {
- struct rds_ib_cache_head *percpu;
+ struct rds_ib_cache_head __percpu *percpu;
struct list_head *xfer;
struct list_head *ready;
};
@@ -186,8 +186,7 @@ struct rds_ib_device {
struct work_struct free_work;
};
-#define pcidev_to_node(pcidev) pcibus_to_node(pcidev->bus)
-#define ibdev_to_node(ibdev) pcidev_to_node(to_pci_dev(ibdev->dma_device))
+#define ibdev_to_node(ibdev) dev_to_node(ibdev->dma_device)
#define rdsibdev_to_node(rdsibdev) ibdev_to_node(rdsibdev->dev)
/* bits for i_ack_flags */
diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index 51c868923f6..31b74f5e61a 100644
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -434,12 +434,11 @@ static u32 rds_ib_protocol_compatible(struct rdma_cm_event *event)
version = RDS_PROTOCOL_3_0;
while ((common >>= 1) != 0)
version++;
- }
- printk_ratelimited(KERN_NOTICE "RDS: Connection from %pI4 using "
- "incompatible protocol version %u.%u\n",
- &dp->dp_saddr,
- dp->dp_protocol_major,
- dp->dp_protocol_minor);
+ } else
+ printk_ratelimited(KERN_NOTICE "RDS: Connection from %pI4 using incompatible protocol version %u.%u\n",
+ &dp->dp_saddr,
+ dp->dp_protocol_major,
+ dp->dp_protocol_minor);
return version;
}
@@ -749,7 +748,7 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp)
int ret;
/* XXX too lazy? */
- ic = kzalloc(sizeof(struct rds_ib_connection), GFP_KERNEL);
+ ic = kzalloc(sizeof(struct rds_ib_connection), gfp);
if (!ic)
return -ENOMEM;
diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c
index e29e0ca32f7..d67de453c35 100644
--- a/net/rds/ib_recv.c
+++ b/net/rds/ib_recv.c
@@ -339,8 +339,8 @@ static int rds_ib_recv_refill_one(struct rds_connection *conn,
sge->length = sizeof(struct rds_header);
sge = &recv->r_sge[1];
- sge->addr = sg_dma_address(&recv->r_frag->f_sg);
- sge->length = sg_dma_len(&recv->r_frag->f_sg);
+ sge->addr = ib_sg_dma_address(ic->i_cm_id->device, &recv->r_frag->f_sg);
+ sge->length = ib_sg_dma_len(ic->i_cm_id->device, &recv->r_frag->f_sg);
ret = 0;
out:
@@ -381,7 +381,10 @@ void rds_ib_recv_refill(struct rds_connection *conn, int prefill)
ret = ib_post_recv(ic->i_cm_id->qp, &recv->r_wr, &failed_wr);
rdsdebug("recv %p ibinc %p page %p addr %lu ret %d\n", recv,
recv->r_ibinc, sg_page(&recv->r_frag->f_sg),
- (long) sg_dma_address(&recv->r_frag->f_sg), ret);
+ (long) ib_sg_dma_address(
+ ic->i_cm_id->device,
+ &recv->r_frag->f_sg),
+ ret);
if (ret) {
rds_ib_conn_error(conn, "recv post on "
"%pI4 returned %d, disconnecting and "
@@ -418,20 +421,20 @@ static void rds_ib_recv_cache_put(struct list_head *new_item,
struct rds_ib_refill_cache *cache)
{
unsigned long flags;
- struct rds_ib_cache_head *chp;
- struct list_head *old;
+ struct list_head *old, *chpfirst;
local_irq_save(flags);
- chp = per_cpu_ptr(cache->percpu, smp_processor_id());
- if (!chp->first)
+ chpfirst = __this_cpu_read(cache->percpu->first);
+ if (!chpfirst)
INIT_LIST_HEAD(new_item);
else /* put on front */
- list_add_tail(new_item, chp->first);
- chp->first = new_item;
- chp->count++;
+ list_add_tail(new_item, chpfirst);
- if (chp->count < RDS_IB_RECYCLE_BATCH_COUNT)
+ __this_cpu_write(cache->percpu->first, new_item);
+ __this_cpu_inc(cache->percpu->count);
+
+ if (__this_cpu_read(cache->percpu->count) < RDS_IB_RECYCLE_BATCH_COUNT)
goto end;
/*
@@ -443,12 +446,13 @@ static void rds_ib_recv_cache_put(struct list_head *new_item,
do {
old = xchg(&cache->xfer, NULL);
if (old)
- list_splice_entire_tail(old, chp->first);
- old = cmpxchg(&cache->xfer, NULL, chp->first);
+ list_splice_entire_tail(old, chpfirst);
+ old = cmpxchg(&cache->xfer, NULL, chpfirst);
} while (old);
- chp->first = NULL;
- chp->count = 0;
+
+ __this_cpu_write(cache->percpu->first, NULL);
+ __this_cpu_write(cache->percpu->count, 0);
end:
local_irq_restore(flags);
}
@@ -594,7 +598,7 @@ static void rds_ib_set_ack(struct rds_ib_connection *ic, u64 seq,
{
atomic64_set(&ic->i_ack_next, seq);
if (ack_required) {
- smp_mb__before_clear_bit();
+ smp_mb__before_atomic();
set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
}
}
@@ -602,7 +606,7 @@ static void rds_ib_set_ack(struct rds_ib_connection *ic, u64 seq,
static u64 rds_ib_get_ack(struct rds_ib_connection *ic)
{
clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
return atomic64_read(&ic->i_ack_next);
}
@@ -763,7 +767,7 @@ static void rds_ib_cong_recv(struct rds_connection *conn,
to_copy = min(RDS_FRAG_SIZE - frag_off, PAGE_SIZE - map_off);
BUG_ON(to_copy & 7); /* Must be 64bit aligned. */
- addr = kmap_atomic(sg_page(&frag->f_sg), KM_SOFTIRQ0);
+ addr = kmap_atomic(sg_page(&frag->f_sg));
src = addr + frag_off;
dst = (void *)map->m_page_addrs[map_page] + map_off;
@@ -773,7 +777,7 @@ static void rds_ib_cong_recv(struct rds_connection *conn,
uncongested |= ~(*src) & *dst;
*dst++ = *src++;
}
- kunmap_atomic(addr, KM_SOFTIRQ0);
+ kunmap_atomic(addr);
copied += to_copy;
@@ -826,7 +830,7 @@ static void rds_ib_process_recv(struct rds_connection *conn,
if (data_len < sizeof(struct rds_header)) {
rds_ib_conn_error(conn, "incoming message "
- "from %pI4 didn't inclue a "
+ "from %pI4 didn't include a "
"header, disconnecting and "
"reconnecting\n",
&conn->c_faddr);
@@ -919,8 +923,7 @@ static void rds_ib_process_recv(struct rds_connection *conn,
rds_ib_cong_recv(conn, ibinc);
else {
rds_recv_incoming(conn, conn->c_faddr, conn->c_laddr,
- &ibinc->ii_inc, GFP_ATOMIC,
- KM_SOFTIRQ0);
+ &ibinc->ii_inc, GFP_ATOMIC);
state->ack_next = be64_to_cpu(hdr->h_sequence);
state->ack_next_valid = 1;
}
diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c
index e5909498117..1dde91e3dc7 100644
--- a/net/rds/ib_send.c
+++ b/net/rds/ib_send.c
@@ -298,7 +298,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context)
rds_ib_stats_inc(s_ib_tx_cq_event);
if (wc.wr_id == RDS_IB_ACK_WR_ID) {
- if (ic->i_ack_queued + HZ/2 < jiffies)
+ if (time_after(jiffies, ic->i_ack_queued + HZ/2))
rds_ib_stats_inc(s_ib_tx_stalled);
rds_ib_ack_send_complete(ic);
continue;
@@ -315,7 +315,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context)
rm = rds_ib_send_unmap_op(ic, send, wc.status);
- if (send->s_queued + HZ/2 < jiffies)
+ if (time_after(jiffies, send->s_queued + HZ/2))
rds_ib_stats_inc(s_ib_tx_stalled);
if (send->s_op) {
@@ -552,9 +552,8 @@ int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm,
&& rm->m_inc.i_hdr.h_flags & RDS_FLAG_CONG_BITMAP) {
rds_cong_map_updated(conn->c_fcong, ~(u64) 0);
scat = &rm->data.op_sg[sg];
- ret = sizeof(struct rds_header) + RDS_CONG_MAP_BYTES;
- ret = min_t(int, ret, scat->length - conn->c_xmit_data_off);
- return ret;
+ ret = max_t(int, RDS_CONG_MAP_BYTES, scat->length);
+ return sizeof(struct rds_header) + ret;
}
/* FIXME we may overallocate here */
diff --git a/net/rds/ib_sysctl.c b/net/rds/ib_sysctl.c
index 1253b006efd..e4e41b3afce 100644
--- a/net/rds/ib_sysctl.c
+++ b/net/rds/ib_sysctl.c
@@ -61,7 +61,7 @@ static unsigned long rds_ib_sysctl_max_unsig_wr_max = 64;
*/
unsigned int rds_ib_sysctl_flow_control = 0;
-static ctl_table rds_ib_sysctl_table[] = {
+static struct ctl_table rds_ib_sysctl_table[] = {
{
.procname = "max_send_wr",
.data = &rds_ib_sysctl_max_send_wr,
@@ -106,22 +106,15 @@ static ctl_table rds_ib_sysctl_table[] = {
{ }
};
-static struct ctl_path rds_ib_sysctl_path[] = {
- { .procname = "net", },
- { .procname = "rds", },
- { .procname = "ib", },
- { }
-};
-
void rds_ib_sysctl_exit(void)
{
if (rds_ib_sysctl_hdr)
- unregister_sysctl_table(rds_ib_sysctl_hdr);
+ unregister_net_sysctl_table(rds_ib_sysctl_hdr);
}
int rds_ib_sysctl_init(void)
{
- rds_ib_sysctl_hdr = register_sysctl_paths(rds_ib_sysctl_path, rds_ib_sysctl_table);
+ rds_ib_sysctl_hdr = register_net_sysctl(&init_net, "net/rds/ib", rds_ib_sysctl_table);
if (!rds_ib_sysctl_hdr)
return -ENOMEM;
return 0;
diff --git a/net/rds/info.c b/net/rds/info.c
index f1c016c4146..9a6b4f66187 100644
--- a/net/rds/info.c
+++ b/net/rds/info.c
@@ -104,7 +104,7 @@ EXPORT_SYMBOL_GPL(rds_info_deregister_func);
void rds_info_iter_unmap(struct rds_info_iterator *iter)
{
if (iter->addr) {
- kunmap_atomic(iter->addr, KM_USER0);
+ kunmap_atomic(iter->addr);
iter->addr = NULL;
}
}
@@ -119,7 +119,7 @@ void rds_info_copy(struct rds_info_iterator *iter, void *data,
while (bytes) {
if (!iter->addr)
- iter->addr = kmap_atomic(*iter->pages, KM_USER0);
+ iter->addr = kmap_atomic(*iter->pages);
this = min(bytes, PAGE_SIZE - iter->offset);
@@ -134,7 +134,7 @@ void rds_info_copy(struct rds_info_iterator *iter, void *data,
iter->offset += this;
if (iter->offset == PAGE_SIZE) {
- kunmap_atomic(iter->addr, KM_USER0);
+ kunmap_atomic(iter->addr);
iter->addr = NULL;
iter->offset = 0;
iter->pages++;
diff --git a/net/rds/iw.c b/net/rds/iw.c
index 7826d46baa7..589935661d6 100644
--- a/net/rds/iw.c
+++ b/net/rds/iw.c
@@ -239,7 +239,8 @@ static int rds_iw_laddr_check(__be32 addr)
ret = rdma_bind_addr(cm_id, (struct sockaddr *)&sin);
/* due to this, we will claim to support IB devices unless we
check node_type. */
- if (ret || cm_id->device->node_type != RDMA_NODE_RNIC)
+ if (ret || !cm_id->device ||
+ cm_id->device->node_type != RDMA_NODE_RNIC)
ret = -EADDRNOTAVAIL;
rdsdebug("addr %pI4 ret %d node type %d\n",
diff --git a/net/rds/iw_cm.c b/net/rds/iw_cm.c
index 9556d2895f7..a91e1db62ee 100644
--- a/net/rds/iw_cm.c
+++ b/net/rds/iw_cm.c
@@ -694,7 +694,7 @@ int rds_iw_conn_alloc(struct rds_connection *conn, gfp_t gfp)
unsigned long flags;
/* XXX too lazy? */
- ic = kzalloc(sizeof(struct rds_iw_connection), GFP_KERNEL);
+ ic = kzalloc(sizeof(struct rds_iw_connection), gfp);
if (!ic)
return -ENOMEM;
diff --git a/net/rds/iw_recv.c b/net/rds/iw_recv.c
index 5e57347f49f..aa8bf678600 100644
--- a/net/rds/iw_recv.c
+++ b/net/rds/iw_recv.c
@@ -429,7 +429,7 @@ static void rds_iw_set_ack(struct rds_iw_connection *ic, u64 seq,
{
atomic64_set(&ic->i_ack_next, seq);
if (ack_required) {
- smp_mb__before_clear_bit();
+ smp_mb__before_atomic();
set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
}
}
@@ -437,7 +437,7 @@ static void rds_iw_set_ack(struct rds_iw_connection *ic, u64 seq,
static u64 rds_iw_get_ack(struct rds_iw_connection *ic)
{
clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
return atomic64_read(&ic->i_ack_next);
}
@@ -598,7 +598,7 @@ static void rds_iw_cong_recv(struct rds_connection *conn,
to_copy = min(RDS_FRAG_SIZE - frag_off, PAGE_SIZE - map_off);
BUG_ON(to_copy & 7); /* Must be 64bit aligned. */
- addr = kmap_atomic(frag->f_page, KM_SOFTIRQ0);
+ addr = kmap_atomic(frag->f_page);
src = addr + frag_off;
dst = (void *)map->m_page_addrs[map_page] + map_off;
@@ -608,7 +608,7 @@ static void rds_iw_cong_recv(struct rds_connection *conn,
uncongested |= ~(*src) & *dst;
*dst++ = *src++;
}
- kunmap_atomic(addr, KM_SOFTIRQ0);
+ kunmap_atomic(addr);
copied += to_copy;
@@ -661,7 +661,7 @@ static void rds_iw_process_recv(struct rds_connection *conn,
if (byte_len < sizeof(struct rds_header)) {
rds_iw_conn_error(conn, "incoming message "
- "from %pI4 didn't inclue a "
+ "from %pI4 didn't include a "
"header, disconnecting and "
"reconnecting\n",
&conn->c_faddr);
@@ -754,8 +754,7 @@ static void rds_iw_process_recv(struct rds_connection *conn,
rds_iw_cong_recv(conn, iwinc);
else {
rds_recv_incoming(conn, conn->c_faddr, conn->c_laddr,
- &iwinc->ii_inc, GFP_ATOMIC,
- KM_SOFTIRQ0);
+ &iwinc->ii_inc, GFP_ATOMIC);
state->ack_next = be64_to_cpu(hdr->h_sequence);
state->ack_next_valid = 1;
}
diff --git a/net/rds/iw_send.c b/net/rds/iw_send.c
index e40c3c5db2c..9105ea03aec 100644
--- a/net/rds/iw_send.c
+++ b/net/rds/iw_send.c
@@ -232,7 +232,7 @@ void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context)
}
if (wc.wr_id == RDS_IW_ACK_WR_ID) {
- if (ic->i_ack_queued + HZ/2 < jiffies)
+ if (time_after(jiffies, ic->i_ack_queued + HZ/2))
rds_iw_stats_inc(s_iw_tx_stalled);
rds_iw_ack_send_complete(ic);
continue;
@@ -267,7 +267,7 @@ void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context)
send->s_wr.opcode = 0xdead;
send->s_wr.num_sge = 1;
- if (send->s_queued + HZ/2 < jiffies)
+ if (time_after(jiffies, send->s_queued + HZ/2))
rds_iw_stats_inc(s_iw_tx_stalled);
/* If a RDMA operation produced an error, signal this right
diff --git a/net/rds/iw_sysctl.c b/net/rds/iw_sysctl.c
index e2e47176e72..139239d2cb2 100644
--- a/net/rds/iw_sysctl.c
+++ b/net/rds/iw_sysctl.c
@@ -55,7 +55,7 @@ static unsigned long rds_iw_sysctl_max_unsig_bytes_max = ~0UL;
unsigned int rds_iw_sysctl_flow_control = 1;
-static ctl_table rds_iw_sysctl_table[] = {
+static struct ctl_table rds_iw_sysctl_table[] = {
{
.procname = "max_send_wr",
.data = &rds_iw_sysctl_max_send_wr,
@@ -109,22 +109,14 @@ static ctl_table rds_iw_sysctl_table[] = {
{ }
};
-static struct ctl_path rds_iw_sysctl_path[] = {
- { .procname = "net", },
- { .procname = "rds", },
- { .procname = "iw", },
- { }
-};
-
void rds_iw_sysctl_exit(void)
{
- if (rds_iw_sysctl_hdr)
- unregister_sysctl_table(rds_iw_sysctl_hdr);
+ unregister_net_sysctl_table(rds_iw_sysctl_hdr);
}
int rds_iw_sysctl_init(void)
{
- rds_iw_sysctl_hdr = register_sysctl_paths(rds_iw_sysctl_path, rds_iw_sysctl_table);
+ rds_iw_sysctl_hdr = register_net_sysctl(&init_net, "net/rds/iw", rds_iw_sysctl_table);
if (!rds_iw_sysctl_hdr)
return -ENOMEM;
return 0;
diff --git a/net/rds/loop.c b/net/rds/loop.c
index bca6761a3ca..6b12b68541a 100644
--- a/net/rds/loop.c
+++ b/net/rds/loop.c
@@ -79,7 +79,7 @@ static int rds_loop_xmit(struct rds_connection *conn, struct rds_message *rm,
rds_message_addref(rm);
rds_recv_incoming(conn, conn->c_laddr, conn->c_faddr, &rm->m_inc,
- GFP_KERNEL, KM_USER0);
+ GFP_KERNEL);
rds_send_drop_acked(conn, be64_to_cpu(rm->m_inc.i_hdr.h_sequence),
NULL);
@@ -121,7 +121,7 @@ static int rds_loop_conn_alloc(struct rds_connection *conn, gfp_t gfp)
struct rds_loop_connection *lc;
unsigned long flags;
- lc = kzalloc(sizeof(struct rds_loop_connection), GFP_KERNEL);
+ lc = kzalloc(sizeof(struct rds_loop_connection), gfp);
if (!lc)
return -ENOMEM;
diff --git a/net/rds/message.c b/net/rds/message.c
index f0a4658f327..aba232f9f30 100644
--- a/net/rds/message.c
+++ b/net/rds/message.c
@@ -82,10 +82,7 @@ static void rds_message_purge(struct rds_message *rm)
void rds_message_put(struct rds_message *rm)
{
rdsdebug("put rm %p ref %d\n", rm, atomic_read(&rm->m_refcount));
- if (atomic_read(&rm->m_refcount) == 0) {
-printk(KERN_CRIT "danger refcount zero on %p\n", rm);
-WARN_ON(1);
- }
+ WARN(!atomic_read(&rm->m_refcount), "danger refcount zero on %p\n", rm);
if (atomic_dec_and_test(&rm->m_refcount)) {
BUG_ON(!list_empty(&rm->m_sock_item));
BUG_ON(!list_empty(&rm->m_conn_item));
@@ -197,6 +194,9 @@ struct rds_message *rds_message_alloc(unsigned int extra_len, gfp_t gfp)
{
struct rds_message *rm;
+ if (extra_len > KMALLOC_MAX_SIZE - sizeof(struct rds_message))
+ return NULL;
+
rm = kzalloc(sizeof(struct rds_message) + extra_len, gfp);
if (!rm)
goto out;
diff --git a/net/rds/page.c b/net/rds/page.c
index 2499cd10842..9005a2c920e 100644
--- a/net/rds/page.c
+++ b/net/rds/page.c
@@ -74,11 +74,12 @@ int rds_page_copy_user(struct page *page, unsigned long offset,
}
EXPORT_SYMBOL_GPL(rds_page_copy_user);
-/*
- * Message allocation uses this to build up regions of a message.
+/**
+ * rds_page_remainder_alloc - build up regions of a message.
*
- * @bytes - the number of bytes needed.
- * @gfp - the waiting behaviour of the allocation
+ * @scat: Scatter list for message
+ * @bytes: the number of bytes needed.
+ * @gfp: the waiting behaviour of the allocation
*
* @gfp is always ored with __GFP_HIGHMEM. Callers must be prepared to
* kmap the pages, etc.
diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c
index c2be901d19e..6cd9d1deafc 100644
--- a/net/rds/rdma_transport.c
+++ b/net/rds/rdma_transport.c
@@ -168,7 +168,7 @@ static int rds_rdma_listen_init(void)
return ret;
}
- sin.sin_family = AF_INET,
+ sin.sin_family = AF_INET;
sin.sin_addr.s_addr = (__force u32)htonl(INADDR_ANY);
sin.sin_port = (__force u16)htons(RDS_PORT);
diff --git a/net/rds/rds.h b/net/rds/rds.h
index 7eaba1831f0..48f8ffc60f8 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -704,7 +704,7 @@ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn,
__be32 saddr);
void rds_inc_put(struct rds_incoming *inc);
void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr,
- struct rds_incoming *inc, gfp_t gfp, enum km_type km);
+ struct rds_incoming *inc, gfp_t gfp);
int rds_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
size_t size, int msg_flags);
void rds_clear_recv_queue(struct rds_sock *rs);
@@ -749,7 +749,7 @@ void rds_atomic_send_complete(struct rds_message *rm, int wc_status);
int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm,
struct cmsghdr *cmsg);
-extern void __rds_put_mr_final(struct rds_mr *mr);
+void __rds_put_mr_final(struct rds_mr *mr);
static inline void rds_mr_put(struct rds_mr *mr)
{
if (atomic_dec_and_test(&mr->r_refcount))
diff --git a/net/rds/recv.c b/net/rds/recv.c
index bc3f8cd6d07..bd82522534f 100644
--- a/net/rds/recv.c
+++ b/net/rds/recv.c
@@ -155,7 +155,7 @@ static void rds_recv_incoming_exthdrs(struct rds_incoming *inc, struct rds_sock
* tell us which roles the addrs in the conn are playing for this message.
*/
void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr,
- struct rds_incoming *inc, gfp_t gfp, enum km_type km)
+ struct rds_incoming *inc, gfp_t gfp)
{
struct rds_sock *rs = NULL;
struct sock *sk;
@@ -402,7 +402,7 @@ int rds_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
struct rds_sock *rs = rds_sk_to_rs(sk);
long timeo;
int ret = 0, nonblock = msg_flags & MSG_DONTWAIT;
- struct sockaddr_in *sin;
+ DECLARE_SOCKADDR(struct sockaddr_in *, sin, msg->msg_name);
struct rds_incoming *inc = NULL;
/* udp_recvmsg()->sock_recvtimeo() gets away without locking too.. */
@@ -479,12 +479,12 @@ int rds_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
rds_stats_inc(s_recv_delivered);
- sin = (struct sockaddr_in *)msg->msg_name;
if (sin) {
sin->sin_family = AF_INET;
sin->sin_port = inc->i_hdr.h_sport;
sin->sin_addr.s_addr = inc->i_saddr;
memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+ msg->msg_namelen = sizeof(*sin);
}
break;
}
diff --git a/net/rds/send.c b/net/rds/send.c
index e2d63c59e7c..23718160d71 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -107,7 +107,7 @@ static int acquire_in_xmit(struct rds_connection *conn)
static void release_in_xmit(struct rds_connection *conn)
{
clear_bit(RDS_IN_XMIT, &conn->c_flags);
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
/*
* We don't use wait_on_bit()/wake_up_bit() because our waking is in a
* hot path and finding waiters is very rare. We don't want to walk
@@ -661,7 +661,7 @@ void rds_send_drop_acked(struct rds_connection *conn, u64 ack,
/* order flag updates with spin locks */
if (!list_empty(&list))
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
spin_unlock_irqrestore(&conn->c_lock, flags);
@@ -691,7 +691,7 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest)
}
/* order flag updates with the rs lock */
- smp_mb__after_clear_bit();
+ smp_mb__after_atomic();
spin_unlock_irqrestore(&rs->rs_lock, flags);
@@ -922,7 +922,7 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
{
struct sock *sk = sock->sk;
struct rds_sock *rs = rds_sk_to_rs(sk);
- struct sockaddr_in *usin = (struct sockaddr_in *)msg->msg_name;
+ DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
__be32 daddr;
__be16 dport;
struct rds_message *rm = NULL;
@@ -935,7 +935,6 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
/* Mirror Linux UDP mirror of BSD error message compatibility */
/* XXX: Perhaps MSG_MORE someday */
if (msg->msg_flags & ~(MSG_DONTWAIT | MSG_CMSG_COMPAT)) {
- printk(KERN_INFO "msg_flags 0x%08X\n", msg->msg_flags);
ret = -EOPNOTSUPP;
goto out;
}
@@ -1123,7 +1122,7 @@ rds_send_pong(struct rds_connection *conn, __be16 dport)
rds_stats_inc(s_send_pong);
if (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags))
- rds_send_xmit(conn);
+ queue_delayed_work(rds_wq, &conn->c_send_w, 0);
rds_message_put(rm);
return 0;
diff --git a/net/rds/stats.c b/net/rds/stats.c
index 7be790d60b9..73be187d389 100644
--- a/net/rds/stats.c
+++ b/net/rds/stats.c
@@ -87,6 +87,7 @@ void rds_stats_info_copy(struct rds_info_iterator *iter,
for (i = 0; i < nr; i++) {
BUG_ON(strlen(names[i]) >= sizeof(ctr.name));
strncpy(ctr.name, names[i], sizeof(ctr.name) - 1);
+ ctr.name[sizeof(ctr.name) - 1] = '\0';
ctr.value = values[i];
rds_info_copy(iter, &ctr, sizeof(ctr));
diff --git a/net/rds/sysctl.c b/net/rds/sysctl.c
index 25ad0c77a26..c3b0cd43eb5 100644
--- a/net/rds/sysctl.c
+++ b/net/rds/sysctl.c
@@ -49,7 +49,7 @@ unsigned int rds_sysctl_max_unacked_bytes = (16 << 20);
unsigned int rds_sysctl_ping_enable = 1;
-static ctl_table rds_sysctl_rds_table[] = {
+static struct ctl_table rds_sysctl_rds_table[] = {
{
.procname = "reconnect_min_delay_ms",
.data = &rds_sysctl_reconnect_min_jiffies,
@@ -92,17 +92,9 @@ static ctl_table rds_sysctl_rds_table[] = {
{ }
};
-static struct ctl_path rds_sysctl_path[] = {
- { .procname = "net", },
- { .procname = "rds", },
- { }
-};
-
-
void rds_sysctl_exit(void)
{
- if (rds_sysctl_reg_table)
- unregister_sysctl_table(rds_sysctl_reg_table);
+ unregister_net_sysctl_table(rds_sysctl_reg_table);
}
int rds_sysctl_init(void)
@@ -110,7 +102,7 @@ int rds_sysctl_init(void)
rds_sysctl_reconnect_min = msecs_to_jiffies(1);
rds_sysctl_reconnect_min_jiffies = rds_sysctl_reconnect_min;
- rds_sysctl_reg_table = register_sysctl_paths(rds_sysctl_path, rds_sysctl_rds_table);
+ rds_sysctl_reg_table = register_net_sysctl(&init_net,"net/rds", rds_sysctl_rds_table);
if (!rds_sysctl_reg_table)
return -ENOMEM;
return 0;
diff --git a/net/rds/tcp.h b/net/rds/tcp.h
index 9cf2927d002..65637491f72 100644
--- a/net/rds/tcp.h
+++ b/net/rds/tcp.h
@@ -61,12 +61,12 @@ void rds_tcp_state_change(struct sock *sk);
/* tcp_listen.c */
int rds_tcp_listen_init(void);
void rds_tcp_listen_stop(void);
-void rds_tcp_listen_data_ready(struct sock *sk, int bytes);
+void rds_tcp_listen_data_ready(struct sock *sk);
/* tcp_recv.c */
int rds_tcp_recv_init(void);
void rds_tcp_recv_exit(void);
-void rds_tcp_data_ready(struct sock *sk, int bytes);
+void rds_tcp_data_ready(struct sock *sk);
int rds_tcp_recv(struct rds_connection *conn);
void rds_tcp_inc_free(struct rds_incoming *inc);
int rds_tcp_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov,
diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c
index af95c8e058f..a65ee78db0c 100644
--- a/net/rds/tcp_connect.c
+++ b/net/rds/tcp_connect.c
@@ -43,7 +43,7 @@ void rds_tcp_state_change(struct sock *sk)
struct rds_connection *conn;
struct rds_tcp_connection *tc;
- read_lock_bh(&sk->sk_callback_lock);
+ read_lock(&sk->sk_callback_lock);
conn = sk->sk_user_data;
if (!conn) {
state_change = sk->sk_state_change;
@@ -68,7 +68,7 @@ void rds_tcp_state_change(struct sock *sk)
break;
}
out:
- read_unlock_bh(&sk->sk_callback_lock);
+ read_unlock(&sk->sk_callback_lock);
state_change(sk);
}
diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c
index 8b5cc4aa886..23ab4dcd1d9 100644
--- a/net/rds/tcp_listen.c
+++ b/net/rds/tcp_listen.c
@@ -108,13 +108,13 @@ static void rds_tcp_accept_worker(struct work_struct *work)
cond_resched();
}
-void rds_tcp_listen_data_ready(struct sock *sk, int bytes)
+void rds_tcp_listen_data_ready(struct sock *sk)
{
- void (*ready)(struct sock *sk, int bytes);
+ void (*ready)(struct sock *sk);
rdsdebug("listen data ready sk %p\n", sk);
- read_lock_bh(&sk->sk_callback_lock);
+ read_lock(&sk->sk_callback_lock);
ready = sk->sk_user_data;
if (!ready) { /* check for teardown race */
ready = sk->sk_data_ready;
@@ -131,8 +131,8 @@ void rds_tcp_listen_data_ready(struct sock *sk, int bytes)
queue_work(rds_wq, &rds_tcp_listen_work);
out:
- read_unlock_bh(&sk->sk_callback_lock);
- ready(sk, bytes);
+ read_unlock(&sk->sk_callback_lock);
+ ready(sk);
}
int rds_tcp_listen_init(void)
@@ -145,7 +145,7 @@ int rds_tcp_listen_init(void)
if (ret < 0)
goto out;
- sock->sk->sk_reuse = 1;
+ sock->sk->sk_reuse = SK_CAN_REUSE;
rds_tcp_nonagle(sock);
write_lock_bh(&sock->sk->sk_callback_lock);
@@ -153,7 +153,7 @@ int rds_tcp_listen_init(void)
sock->sk->sk_data_ready = rds_tcp_listen_data_ready;
write_unlock_bh(&sock->sk->sk_callback_lock);
- sin.sin_family = PF_INET,
+ sin.sin_family = PF_INET;
sin.sin_addr.s_addr = (__force u32)htonl(INADDR_ANY);
sin.sin_port = (__force u16)htons(RDS_TCP_PORT);
diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c
index 78205e25500..9ae6e0a264e 100644
--- a/net/rds/tcp_recv.c
+++ b/net/rds/tcp_recv.c
@@ -169,7 +169,6 @@ static void rds_tcp_cong_recv(struct rds_connection *conn,
struct rds_tcp_desc_arg {
struct rds_connection *conn;
gfp_t gfp;
- enum km_type km;
};
static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb,
@@ -255,7 +254,7 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb,
else
rds_recv_incoming(conn, conn->c_faddr,
conn->c_laddr, &tinc->ti_inc,
- arg->gfp, arg->km);
+ arg->gfp);
tc->t_tinc_hdr_rem = sizeof(struct rds_header);
tc->t_tinc_data_rem = 0;
@@ -272,8 +271,7 @@ out:
}
/* the caller has to hold the sock lock */
-static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp,
- enum km_type km)
+static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp)
{
struct rds_tcp_connection *tc = conn->c_transport_data;
struct socket *sock = tc->t_sock;
@@ -283,7 +281,6 @@ static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp,
/* It's like glib in the kernel! */
arg.conn = conn;
arg.gfp = gfp;
- arg.km = km;
desc.arg.data = &arg;
desc.error = 0;
desc.count = 1; /* give more than one skb per call */
@@ -311,21 +308,21 @@ int rds_tcp_recv(struct rds_connection *conn)
rdsdebug("recv worker conn %p tc %p sock %p\n", conn, tc, sock);
lock_sock(sock->sk);
- ret = rds_tcp_read_sock(conn, GFP_KERNEL, KM_USER0);
+ ret = rds_tcp_read_sock(conn, GFP_KERNEL);
release_sock(sock->sk);
return ret;
}
-void rds_tcp_data_ready(struct sock *sk, int bytes)
+void rds_tcp_data_ready(struct sock *sk)
{
- void (*ready)(struct sock *sk, int bytes);
+ void (*ready)(struct sock *sk);
struct rds_connection *conn;
struct rds_tcp_connection *tc;
- rdsdebug("data ready sk %p bytes %d\n", sk, bytes);
+ rdsdebug("data ready sk %p\n", sk);
- read_lock_bh(&sk->sk_callback_lock);
+ read_lock(&sk->sk_callback_lock);
conn = sk->sk_user_data;
if (!conn) { /* check for teardown race */
ready = sk->sk_data_ready;
@@ -336,11 +333,11 @@ void rds_tcp_data_ready(struct sock *sk, int bytes)
ready = tc->t_orig_data_ready;
rds_tcp_stats_inc(s_tcp_data_ready_calls);
- if (rds_tcp_read_sock(conn, GFP_ATOMIC, KM_SOFTIRQ0) == -ENOMEM)
+ if (rds_tcp_read_sock(conn, GFP_ATOMIC) == -ENOMEM)
queue_delayed_work(rds_wq, &conn->c_recv_w, 0);
out:
- read_unlock_bh(&sk->sk_callback_lock);
- ready(sk, bytes);
+ read_unlock(&sk->sk_callback_lock);
+ ready(sk);
}
int rds_tcp_recv_init(void)
diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c
index 1b4fd68f0c7..53b17ca0dff 100644
--- a/net/rds/tcp_send.c
+++ b/net/rds/tcp_send.c
@@ -93,7 +93,7 @@ int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm,
rm->m_ack_seq = tc->t_last_sent_nxt +
sizeof(struct rds_header) +
be32_to_cpu(rm->m_inc.i_hdr.h_len) - 1;
- smp_mb__before_clear_bit();
+ smp_mb__before_atomic();
set_bit(RDS_MSG_HAS_ACK_SEQ, &rm->m_flags);
tc->t_last_expected_una = rm->m_ack_seq + 1;
@@ -174,7 +174,7 @@ void rds_tcp_write_space(struct sock *sk)
struct rds_connection *conn;
struct rds_tcp_connection *tc;
- read_lock_bh(&sk->sk_callback_lock);
+ read_lock(&sk->sk_callback_lock);
conn = sk->sk_user_data;
if (!conn) {
write_space = sk->sk_write_space;
@@ -194,7 +194,7 @@ void rds_tcp_write_space(struct sock *sk)
queue_delayed_work(rds_wq, &conn->c_send_w, 0);
out:
- read_unlock_bh(&sk->sk_callback_lock);
+ read_unlock(&sk->sk_callback_lock);
/*
* write_space is only called when data leaves tcp's send queue if
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 11:48:33 +0000