diff options
Diffstat (limited to 'net/rds')
-rw-r--r-- | net/rds/af_rds.c | 16 | ||||
-rw-r--r-- | net/rds/cong.c | 2 | ||||
-rw-r--r-- | net/rds/connection.c | 6 | ||||
-rw-r--r-- | net/rds/ib.h | 2 | ||||
-rw-r--r-- | net/rds/ib_cm.c | 6 | ||||
-rw-r--r-- | net/rds/ib_rdma.c | 9 | ||||
-rw-r--r-- | net/rds/ib_recv.c | 47 | ||||
-rw-r--r-- | net/rds/ib_send.c | 4 | ||||
-rw-r--r-- | net/rds/ib_sysctl.c | 26 | ||||
-rw-r--r-- | net/rds/iw.h | 2 | ||||
-rw-r--r-- | net/rds/iw_cm.c | 2 | ||||
-rw-r--r-- | net/rds/iw_rdma.c | 9 | ||||
-rw-r--r-- | net/rds/iw_recv.c | 47 | ||||
-rw-r--r-- | net/rds/iw_send.c | 7 | ||||
-rw-r--r-- | net/rds/iw_sysctl.c | 26 | ||||
-rw-r--r-- | net/rds/message.c | 3 | ||||
-rw-r--r-- | net/rds/rdma.c | 36 | ||||
-rw-r--r-- | net/rds/rdma.h | 1 | ||||
-rw-r--r-- | net/rds/recv.c | 11 | ||||
-rw-r--r-- | net/rds/send.c | 27 | ||||
-rw-r--r-- | net/rds/sysctl.c | 21 | ||||
-rw-r--r-- | net/rds/tcp_listen.c | 8 | ||||
-rw-r--r-- | net/rds/threads.c | 4 |
23 files changed, 183 insertions, 139 deletions
diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index 98e05382fd3c..853c52be781f 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -174,8 +174,8 @@ static unsigned int rds_poll(struct file *file, struct socket *sock, mask |= (POLLIN | POLLRDNORM); spin_unlock(&rs->rs_lock); } - if (!list_empty(&rs->rs_recv_queue) - || !list_empty(&rs->rs_notify_queue)) + if (!list_empty(&rs->rs_recv_queue) || + !list_empty(&rs->rs_notify_queue)) mask |= (POLLIN | POLLRDNORM); if (rs->rs_snd_bytes < rds_sk_sndbuf(rs)) mask |= (POLLOUT | POLLWRNORM); @@ -265,6 +265,9 @@ static int rds_setsockopt(struct socket *sock, int level, int optname, case RDS_GET_MR: ret = rds_get_mr(rs, optval, optlen); break; + case RDS_GET_MR_FOR_DEST: + ret = rds_get_mr_for_dest(rs, optval, optlen); + break; case RDS_FREE_MR: ret = rds_free_mr(rs, optval, optlen); break; @@ -305,8 +308,8 @@ static int rds_getsockopt(struct socket *sock, int level, int optname, if (len < sizeof(int)) ret = -EINVAL; else - if (put_user(rs->rs_recverr, (int __user *) optval) - || put_user(sizeof(int), optlen)) + if (put_user(rs->rs_recverr, (int __user *) optval) || + put_user(sizeof(int), optlen)) ret = -EFAULT; else ret = 0; @@ -407,7 +410,8 @@ static int __rds_create(struct socket *sock, struct sock *sk, int protocol) return 0; } -static int rds_create(struct net *net, struct socket *sock, int protocol) +static int rds_create(struct net *net, struct socket *sock, int protocol, + int kern) { struct sock *sk; @@ -431,7 +435,7 @@ void rds_sock_put(struct rds_sock *rs) sock_put(rds_rs_to_sk(rs)); } -static struct net_proto_family rds_family_ops = { +static const struct net_proto_family rds_family_ops = { .family = AF_RDS, .create = rds_create, .owner = THIS_MODULE, diff --git a/net/rds/cong.c b/net/rds/cong.c index dd2711df640b..6d06cac2649c 100644 --- a/net/rds/cong.c +++ b/net/rds/cong.c @@ -218,6 +218,8 @@ void rds_cong_queue_updates(struct rds_cong_map *map) spin_lock_irqsave(&rds_cong_lock, flags); list_for_each_entry(conn, &map->m_conn_list, c_map_item) { + if (conn->c_loopback) + continue; if (!test_and_set_bit(0, &conn->c_map_queued)) { rds_stats_inc(s_cong_update_queued); queue_delayed_work(rds_wq, &conn->c_send_w, 0); diff --git a/net/rds/connection.c b/net/rds/connection.c index cc8b568c0c84..278f607ab603 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -133,10 +133,8 @@ static struct rds_connection *__rds_conn_create(__be32 laddr, __be32 faddr, spin_lock_irqsave(&rds_conn_lock, flags); conn = rds_conn_lookup(head, laddr, faddr, trans); - if (conn - && conn->c_loopback - && conn->c_trans != &rds_loop_transport - && !is_outgoing) { + if (conn && conn->c_loopback && conn->c_trans != &rds_loop_transport && + !is_outgoing) { /* This is a looped back IB connection, and we're * called by the code handling the incoming connect. * We need a second connection object into which we diff --git a/net/rds/ib.h b/net/rds/ib.h index 1378b854cac0..64df4e79b29f 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -98,6 +98,7 @@ struct rds_ib_connection { struct rds_ib_send_work *i_sends; /* rx */ + struct tasklet_struct i_recv_tasklet; struct mutex i_recv_mutex; struct rds_ib_work_ring i_recv_ring; struct rds_ib_incoming *i_ibinc; @@ -303,6 +304,7 @@ void rds_ib_inc_free(struct rds_incoming *inc); int rds_ib_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov, size_t size); void rds_ib_recv_cq_comp_handler(struct ib_cq *cq, void *context); +void rds_ib_recv_tasklet_fn(unsigned long data); void rds_ib_recv_init_ring(struct rds_ib_connection *ic); void rds_ib_recv_clear_ring(struct rds_ib_connection *ic); void rds_ib_recv_init_ack(struct rds_ib_connection *ic); diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index c2d372f13dbb..647cb8ffc39b 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -377,8 +377,8 @@ static u32 rds_ib_protocol_compatible(struct rdma_cm_event *event) } /* Even if len is crap *now* I still want to check it. -ASG */ - if (event->param.conn.private_data_len < sizeof (*dp) - || dp->dp_protocol_major == 0) + if (event->param.conn.private_data_len < sizeof (*dp) || + dp->dp_protocol_major == 0) return RDS_PROTOCOL_3_0; common = be16_to_cpu(dp->dp_protocol_minor_mask) & RDS_IB_SUPPORTED_PROTOCOLS; @@ -694,6 +694,8 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp) return -ENOMEM; INIT_LIST_HEAD(&ic->ib_node); + tasklet_init(&ic->i_recv_tasklet, rds_ib_recv_tasklet_fn, + (unsigned long) ic); mutex_init(&ic->i_recv_mutex); #ifndef KERNEL_HAS_ATOMIC64 spin_lock_init(&ic->i_ack_lock); diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c index ef3ab5b7283e..4b0da865a72c 100644 --- a/net/rds/ib_rdma.c +++ b/net/rds/ib_rdma.c @@ -187,11 +187,8 @@ void __rds_ib_destroy_conns(struct list_head *list, spinlock_t *list_lock) INIT_LIST_HEAD(list); spin_unlock_irq(list_lock); - list_for_each_entry_safe(ic, _ic, &tmp_list, ib_node) { - if (ic->conn->c_passive) - rds_conn_destroy(ic->conn->c_passive); + list_for_each_entry_safe(ic, _ic, &tmp_list, ib_node) rds_conn_destroy(ic->conn); - } } struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev) @@ -573,8 +570,8 @@ void rds_ib_free_mr(void *trans_private, int invalidate) spin_unlock_irqrestore(&pool->list_lock, flags); /* If we've pinned too many pages, request a flush */ - if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned - || atomic_read(&pool->dirty_count) >= pool->max_items / 10) + if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned || + atomic_read(&pool->dirty_count) >= pool->max_items / 10) queue_work(rds_wq, &pool->flush_worker); if (invalidate) { diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index cd7a6cfcab03..04dc0d3f3c95 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -143,15 +143,16 @@ static int rds_ib_recv_refill_one(struct rds_connection *conn, int ret = -ENOMEM; if (recv->r_ibinc == NULL) { - if (atomic_read(&rds_ib_allocation) >= rds_ib_sysctl_max_recv_allocation) { + if (!atomic_add_unless(&rds_ib_allocation, 1, rds_ib_sysctl_max_recv_allocation)) { rds_ib_stats_inc(s_ib_rx_alloc_limit); goto out; } recv->r_ibinc = kmem_cache_alloc(rds_ib_incoming_slab, kptr_gfp); - if (recv->r_ibinc == NULL) + if (recv->r_ibinc == NULL) { + atomic_dec(&rds_ib_allocation); goto out; - atomic_inc(&rds_ib_allocation); + } INIT_LIST_HEAD(&recv->r_ibinc->ii_frags); rds_inc_init(&recv->r_ibinc->ii_inc, conn, conn->c_faddr); } @@ -229,8 +230,8 @@ int rds_ib_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, int ret = 0; u32 pos; - while ((prefill || rds_conn_up(conn)) - && rds_ib_ring_alloc(&ic->i_recv_ring, 1, &pos)) { + while ((prefill || rds_conn_up(conn)) && + rds_ib_ring_alloc(&ic->i_recv_ring, 1, &pos)) { if (pos >= ic->i_recv_ring.w_nr) { printk(KERN_NOTICE "Argh - ring alloc returned pos=%u\n", pos); @@ -770,10 +771,10 @@ static void rds_ib_process_recv(struct rds_connection *conn, hdr = &ibinc->ii_inc.i_hdr; /* We can't just use memcmp here; fragments of a * single message may carry different ACKs */ - if (hdr->h_sequence != ihdr->h_sequence - || hdr->h_len != ihdr->h_len - || hdr->h_sport != ihdr->h_sport - || hdr->h_dport != ihdr->h_dport) { + if (hdr->h_sequence != ihdr->h_sequence || + hdr->h_len != ihdr->h_len || + hdr->h_sport != ihdr->h_sport || + hdr->h_dport != ihdr->h_dport) { rds_ib_conn_error(conn, "fragment header mismatch; forcing reconnect\n"); return; @@ -824,17 +825,22 @@ void rds_ib_recv_cq_comp_handler(struct ib_cq *cq, void *context) { struct rds_connection *conn = context; struct rds_ib_connection *ic = conn->c_transport_data; - struct ib_wc wc; - struct rds_ib_ack_state state = { 0, }; - struct rds_ib_recv_work *recv; rdsdebug("conn %p cq %p\n", conn, cq); rds_ib_stats_inc(s_ib_rx_cq_call); - ib_req_notify_cq(cq, IB_CQ_SOLICITED); + tasklet_schedule(&ic->i_recv_tasklet); +} + +static inline void rds_poll_cq(struct rds_ib_connection *ic, + struct rds_ib_ack_state *state) +{ + struct rds_connection *conn = ic->conn; + struct ib_wc wc; + struct rds_ib_recv_work *recv; - while (ib_poll_cq(cq, 1, &wc) > 0) { + while (ib_poll_cq(ic->i_recv_cq, 1, &wc) > 0) { rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n", (unsigned long long)wc.wr_id, wc.status, wc.byte_len, be32_to_cpu(wc.ex.imm_data)); @@ -852,7 +858,7 @@ void rds_ib_recv_cq_comp_handler(struct ib_cq *cq, void *context) if (rds_conn_up(conn) || rds_conn_connecting(conn)) { /* We expect errors as the qp is drained during shutdown */ if (wc.status == IB_WC_SUCCESS) { - rds_ib_process_recv(conn, recv, wc.byte_len, &state); + rds_ib_process_recv(conn, recv, wc.byte_len, state); } else { rds_ib_conn_error(conn, "recv completion on " "%pI4 had status %u, disconnecting and " @@ -863,6 +869,17 @@ void rds_ib_recv_cq_comp_handler(struct ib_cq *cq, void *context) rds_ib_ring_free(&ic->i_recv_ring, 1); } +} + +void rds_ib_recv_tasklet_fn(unsigned long data) +{ + struct rds_ib_connection *ic = (struct rds_ib_connection *) data; + struct rds_connection *conn = ic->conn; + struct rds_ib_ack_state state = { 0, }; + + rds_poll_cq(ic, &state); + ib_req_notify_cq(ic->i_recv_cq, IB_CQ_SOLICITED); + rds_poll_cq(ic, &state); if (state.ack_next_valid) rds_ib_set_ack(ic, state.ack_next, state.ack_required); diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index 23bf830db2d5..a10fab6886d1 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -252,8 +252,8 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) rds_ib_ring_free(&ic->i_send_ring, completed); - if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) - || test_bit(0, &conn->c_map_queued)) + if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) || + test_bit(0, &conn->c_map_queued)) queue_delayed_work(rds_wq, &conn->c_send_w, 0); /* We expect errors as the qp is drained during shutdown */ diff --git a/net/rds/ib_sysctl.c b/net/rds/ib_sysctl.c index 84b5ffcb280f..03f01cb4e0fe 100644 --- a/net/rds/ib_sysctl.c +++ b/net/rds/ib_sysctl.c @@ -67,68 +67,62 @@ unsigned int rds_ib_sysctl_flow_control = 0; ctl_table rds_ib_sysctl_table[] = { { - .ctl_name = CTL_UNNUMBERED, .procname = "max_send_wr", .data = &rds_ib_sysctl_max_send_wr, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_ib_sysctl_max_wr_min, .extra2 = &rds_ib_sysctl_max_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_recv_wr", .data = &rds_ib_sysctl_max_recv_wr, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_ib_sysctl_max_wr_min, .extra2 = &rds_ib_sysctl_max_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unsignaled_wr", .data = &rds_ib_sysctl_max_unsig_wrs, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_ib_sysctl_max_unsig_wr_min, .extra2 = &rds_ib_sysctl_max_unsig_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unsignaled_bytes", .data = &rds_ib_sysctl_max_unsig_bytes, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_ib_sysctl_max_unsig_bytes_min, .extra2 = &rds_ib_sysctl_max_unsig_bytes_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_recv_allocation", .data = &rds_ib_sysctl_max_recv_allocation, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "flow_control", .data = &rds_ib_sysctl_flow_control, .maxlen = sizeof(rds_ib_sysctl_flow_control), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, - { .ctl_name = 0} + { } }; static struct ctl_path rds_ib_sysctl_path[] = { - { .procname = "net", .ctl_name = CTL_NET, }, - { .procname = "rds", .ctl_name = CTL_UNNUMBERED, }, - { .procname = "ib", .ctl_name = CTL_UNNUMBERED, }, + { .procname = "net", }, + { .procname = "rds", }, + { .procname = "ib", }, { } }; diff --git a/net/rds/iw.h b/net/rds/iw.h index dd72b62bd506..eef2f0c28476 100644 --- a/net/rds/iw.h +++ b/net/rds/iw.h @@ -119,6 +119,7 @@ struct rds_iw_connection { struct rds_iw_send_work *i_sends; /* rx */ + struct tasklet_struct i_recv_tasklet; struct mutex i_recv_mutex; struct rds_iw_work_ring i_recv_ring; struct rds_iw_incoming *i_iwinc; @@ -330,6 +331,7 @@ void rds_iw_inc_free(struct rds_incoming *inc); int rds_iw_inc_copy_to_user(struct rds_incoming *inc, struct iovec *iov, size_t size); void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context); +void rds_iw_recv_tasklet_fn(unsigned long data); void rds_iw_recv_init_ring(struct rds_iw_connection *ic); void rds_iw_recv_clear_ring(struct rds_iw_connection *ic); void rds_iw_recv_init_ack(struct rds_iw_connection *ic); diff --git a/net/rds/iw_cm.c b/net/rds/iw_cm.c index a416b0d492b1..394cf6b4d0aa 100644 --- a/net/rds/iw_cm.c +++ b/net/rds/iw_cm.c @@ -696,6 +696,8 @@ int rds_iw_conn_alloc(struct rds_connection *conn, gfp_t gfp) return -ENOMEM; INIT_LIST_HEAD(&ic->iw_node); + tasklet_init(&ic->i_recv_tasklet, rds_iw_recv_tasklet_fn, + (unsigned long) ic); mutex_init(&ic->i_recv_mutex); #ifndef KERNEL_HAS_ATOMIC64 spin_lock_init(&ic->i_ack_lock); diff --git a/net/rds/iw_rdma.c b/net/rds/iw_rdma.c index de4a1b16bf7b..9eda11cca956 100644 --- a/net/rds/iw_rdma.c +++ b/net/rds/iw_rdma.c @@ -245,11 +245,8 @@ void __rds_iw_destroy_conns(struct list_head *list, spinlock_t *list_lock) INIT_LIST_HEAD(list); spin_unlock_irq(list_lock); - list_for_each_entry_safe(ic, _ic, &tmp_list, iw_node) { - if (ic->conn->c_passive) - rds_conn_destroy(ic->conn->c_passive); + list_for_each_entry_safe(ic, _ic, &tmp_list, iw_node) rds_conn_destroy(ic->conn); - } } static void rds_iw_set_scatterlist(struct rds_iw_scatterlist *sg, @@ -576,8 +573,8 @@ void rds_iw_free_mr(void *trans_private, int invalidate) rds_iw_free_fastreg(pool, ibmr); /* If we've pinned too many pages, request a flush */ - if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned - || atomic_read(&pool->dirty_count) >= pool->max_items / 10) + if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned || + atomic_read(&pool->dirty_count) >= pool->max_items / 10) queue_work(rds_wq, &pool->flush_worker); if (invalidate) { diff --git a/net/rds/iw_recv.c b/net/rds/iw_recv.c index 8683f5f66c4b..54af7d6b92da 100644 --- a/net/rds/iw_recv.c +++ b/net/rds/iw_recv.c @@ -143,15 +143,16 @@ static int rds_iw_recv_refill_one(struct rds_connection *conn, int ret = -ENOMEM; if (recv->r_iwinc == NULL) { - if (atomic_read(&rds_iw_allocation) >= rds_iw_sysctl_max_recv_allocation) { + if (!atomic_add_unless(&rds_iw_allocation, 1, rds_iw_sysctl_max_recv_allocation)) { rds_iw_stats_inc(s_iw_rx_alloc_limit); goto out; } recv->r_iwinc = kmem_cache_alloc(rds_iw_incoming_slab, kptr_gfp); - if (recv->r_iwinc == NULL) + if (recv->r_iwinc == NULL) { + atomic_dec(&rds_iw_allocation); goto out; - atomic_inc(&rds_iw_allocation); + } INIT_LIST_HEAD(&recv->r_iwinc->ii_frags); rds_inc_init(&recv->r_iwinc->ii_inc, conn, conn->c_faddr); } @@ -229,8 +230,8 @@ int rds_iw_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp, int ret = 0; u32 pos; - while ((prefill || rds_conn_up(conn)) - && rds_iw_ring_alloc(&ic->i_recv_ring, 1, &pos)) { + while ((prefill || rds_conn_up(conn)) && + rds_iw_ring_alloc(&ic->i_recv_ring, 1, &pos)) { if (pos >= ic->i_recv_ring.w_nr) { printk(KERN_NOTICE "Argh - ring alloc returned pos=%u\n", pos); @@ -729,10 +730,10 @@ static void rds_iw_process_recv(struct rds_connection *conn, hdr = &iwinc->ii_inc.i_hdr; /* We can't just use memcmp here; fragments of a * single message may carry different ACKs */ - if (hdr->h_sequence != ihdr->h_sequence - || hdr->h_len != ihdr->h_len - || hdr->h_sport != ihdr->h_sport - || hdr->h_dport != ihdr->h_dport) { + if (hdr->h_sequence != ihdr->h_sequence || + hdr->h_len != ihdr->h_len || + hdr->h_sport != ihdr->h_sport || + hdr->h_dport != ihdr->h_dport) { rds_iw_conn_error(conn, "fragment header mismatch; forcing reconnect\n"); return; @@ -783,17 +784,22 @@ void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context) { struct rds_connection *conn = context; struct rds_iw_connection *ic = conn->c_transport_data; - struct ib_wc wc; - struct rds_iw_ack_state state = { 0, }; - struct rds_iw_recv_work *recv; rdsdebug("conn %p cq %p\n", conn, cq); rds_iw_stats_inc(s_iw_rx_cq_call); - ib_req_notify_cq(cq, IB_CQ_SOLICITED); + tasklet_schedule(&ic->i_recv_tasklet); +} + +static inline void rds_poll_cq(struct rds_iw_connection *ic, + struct rds_iw_ack_state *state) +{ + struct rds_connection *conn = ic->conn; + struct ib_wc wc; + struct rds_iw_recv_work *recv; - while (ib_poll_cq(cq, 1, &wc) > 0) { + while (ib_poll_cq(ic->i_recv_cq, 1, &wc) > 0) { rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n", (unsigned long long)wc.wr_id, wc.status, wc.byte_len, be32_to_cpu(wc.ex.imm_data)); @@ -811,7 +817,7 @@ void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context) if (rds_conn_up(conn) || rds_conn_connecting(conn)) { /* We expect errors as the qp is drained during shutdown */ if (wc.status == IB_WC_SUCCESS) { - rds_iw_process_recv(conn, recv, wc.byte_len, &state); + rds_iw_process_recv(conn, recv, wc.byte_len, state); } else { rds_iw_conn_error(conn, "recv completion on " "%pI4 had status %u, disconnecting and " @@ -822,6 +828,17 @@ void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context) rds_iw_ring_free(&ic->i_recv_ring, 1); } +} + +void rds_iw_recv_tasklet_fn(unsigned long data) +{ + struct rds_iw_connection *ic = (struct rds_iw_connection *) data; + struct rds_connection *conn = ic->conn; + struct rds_iw_ack_state state = { 0, }; + + rds_poll_cq(ic, &state); + ib_req_notify_cq(ic->i_recv_cq, IB_CQ_SOLICITED); + rds_poll_cq(ic, &state); if (state.ack_next_valid) rds_iw_set_ack(ic, state.ack_next, state.ack_required); diff --git a/net/rds/iw_send.c b/net/rds/iw_send.c index 1f5abe3cf2b4..1379e9d66a78 100644 --- a/net/rds/iw_send.c +++ b/net/rds/iw_send.c @@ -288,8 +288,8 @@ void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context) rds_iw_ring_free(&ic->i_send_ring, completed); - if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) - || test_bit(0, &conn->c_map_queued)) + if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) || + test_bit(0, &conn->c_map_queued)) queue_delayed_work(rds_wq, &conn->c_send_w, 0); /* We expect errors as the qp is drained during shutdown */ @@ -519,8 +519,7 @@ int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm, BUG_ON(hdr_off != 0 && hdr_off != sizeof(struct rds_header)); /* Fastreg support */ - if (rds_rdma_cookie_key(rm->m_rdma_cookie) - && !ic->i_fastreg_posted) { + if (rds_rdma_cookie_key(rm->m_rdma_cookie) && !ic->i_fastreg_posted) { ret = -EAGAIN; goto out; } diff --git a/net/rds/iw_sysctl.c b/net/rds/iw_sysctl.c index 9590678cd616..1c4428a61a02 100644 --- a/net/rds/iw_sysctl.c +++ b/net/rds/iw_sysctl.c @@ -57,68 +57,62 @@ unsigned int rds_iw_sysctl_flow_control = 1; ctl_table rds_iw_sysctl_table[] = { { - .ctl_name = CTL_UNNUMBERED, .procname = "max_send_wr", .data = &rds_iw_sysctl_max_send_wr, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_iw_sysctl_max_wr_min, .extra2 = &rds_iw_sysctl_max_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_recv_wr", .data = &rds_iw_sysctl_max_recv_wr, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_iw_sysctl_max_wr_min, .extra2 = &rds_iw_sysctl_max_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unsignaled_wr", .data = &rds_iw_sysctl_max_unsig_wrs, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_iw_sysctl_max_unsig_wr_min, .extra2 = &rds_iw_sysctl_max_unsig_wr_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unsignaled_bytes", .data = &rds_iw_sysctl_max_unsig_bytes, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = &rds_iw_sysctl_max_unsig_bytes_min, .extra2 = &rds_iw_sysctl_max_unsig_bytes_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_recv_allocation", .data = &rds_iw_sysctl_max_recv_allocation, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "flow_control", .data = &rds_iw_sysctl_flow_control, .maxlen = sizeof(rds_iw_sysctl_flow_control), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, - { .ctl_name = 0} + { } }; static struct ctl_path rds_iw_sysctl_path[] = { - { .procname = "net", .ctl_name = CTL_NET, }, - { .procname = "rds", .ctl_name = CTL_UNNUMBERED, }, - { .procname = "iw", .ctl_name = CTL_UNNUMBERED, }, + { .procname = "net", }, + { .procname = "rds", }, + { .procname = "iw", }, { } }; diff --git a/net/rds/message.c b/net/rds/message.c index ca50a8ec9742..73e600ffd87f 100644 --- a/net/rds/message.c +++ b/net/rds/message.c @@ -122,8 +122,7 @@ int rds_message_add_extension(struct rds_header *hdr, if (hdr->h_exthdr[0] != RDS_EXTHDR_NONE) return 0; - if (type >= __RDS_EXTHDR_MAX - || len != rds_exthdr_size[type]) + if (type >= __RDS_EXTHDR_MAX || len != rds_exthdr_size[type]) return 0; if (ext_len >= RDS_HEADER_EXT_SPACE) diff --git a/net/rds/rdma.c b/net/rds/rdma.c index 8dc83d2caa58..4c64daa1f5d5 100644 --- a/net/rds/rdma.c +++ b/net/rds/rdma.c @@ -317,6 +317,30 @@ int rds_get_mr(struct rds_sock *rs, char __user *optval, int optlen) return __rds_rdma_map(rs, &args, NULL, NULL); } +int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen) +{ + struct rds_get_mr_for_dest_args args; + struct rds_get_mr_args new_args; + + if (optlen != sizeof(struct rds_get_mr_for_dest_args)) + return -EINVAL; + + if (copy_from_user(&args, (struct rds_get_mr_for_dest_args __user *)optval, + sizeof(struct rds_get_mr_for_dest_args))) + return -EFAULT; + + /* + * Initially, just behave like get_mr(). + * TODO: Implement get_mr as wrapper around this + * and deprecate it. + */ + new_args.vec = args.vec; + new_args.cookie_addr = args.cookie_addr; + new_args.flags = args.flags; + + return __rds_rdma_map(rs, &new_args, NULL, NULL); +} + /* * Free the MR indicated by the given R_Key */ @@ -607,8 +631,8 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, { struct rds_rdma_op *op; - if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_rdma_args)) - || rm->m_rdma_op != NULL) + if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_rdma_args)) || + rm->m_rdma_op != NULL) return -EINVAL; op = rds_rdma_prepare(rs, CMSG_DATA(cmsg)); @@ -631,8 +655,8 @@ int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, u32 r_key; int err = 0; - if (cmsg->cmsg_len < CMSG_LEN(sizeof(rds_rdma_cookie_t)) - || rm->m_rdma_cookie != 0) + if (cmsg->cmsg_len < CMSG_LEN(sizeof(rds_rdma_cookie_t)) || + rm->m_rdma_cookie != 0) return -EINVAL; memcpy(&rm->m_rdma_cookie, CMSG_DATA(cmsg), sizeof(rm->m_rdma_cookie)); @@ -668,8 +692,8 @@ int rds_cmsg_rdma_dest(struct rds_sock *rs, struct rds_message *rm, int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm, struct cmsghdr *cmsg) { - if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_get_mr_args)) - || rm->m_rdma_cookie != 0) + if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_get_mr_args)) || + rm->m_rdma_cookie != 0) return -EINVAL; return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, &rm->m_rdma_mr); diff --git a/net/rds/rdma.h b/net/rds/rdma.h index 425512098b0b..909c39835a5d 100644 --- a/net/rds/rdma.h +++ b/net/rds/rdma.h @@ -61,6 +61,7 @@ static inline u32 rds_rdma_cookie_offset(rds_rdma_cookie_t cookie) } int rds_get_mr(struct rds_sock *rs, char __user *optval, int optlen); +int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen); int rds_free_mr(struct rds_sock *rs, char __user *optval, int optlen); void rds_rdma_drop_keys(struct rds_sock *rs); int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm, diff --git a/net/rds/recv.c b/net/rds/recv.c index fdff33c7b432..b426d67f760c 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -195,8 +195,8 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, * XXX we could spend more on the wire to get more robust failure * detection, arguably worth it to avoid data corruption. */ - if (be64_to_cpu(inc->i_hdr.h_sequence) < conn->c_next_rx_seq - && (inc->i_hdr.h_flags & RDS_FLAG_RETRANSMITTED)) { + if (be64_to_cpu(inc->i_hdr.h_sequence) < conn->c_next_rx_seq && + (inc->i_hdr.h_flags & RDS_FLAG_RETRANSMITTED)) { rds_stats_inc(s_recv_drop_old_seq); goto out; } @@ -432,10 +432,9 @@ int rds_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, } timeo = wait_event_interruptible_timeout(*sk->sk_sleep, - (!list_empty(&rs->rs_notify_queue) - || rs->rs_cong_notify - || rds_next_incoming(rs, &inc)), - timeo); + (!list_empty(&rs->rs_notify_queue) || + rs->rs_cong_notify || + rds_next_incoming(rs, &inc)), timeo); rdsdebug("recvmsg woke inc %p timeo %ld\n", inc, timeo); if (timeo > 0 || timeo == MAX_SCHEDULE_TIMEOUT) diff --git a/net/rds/send.c b/net/rds/send.c index 28c88ff3d038..b2fccfc20769 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -235,8 +235,8 @@ int rds_send_xmit(struct rds_connection *conn) * connection. * Therefore, we never retransmit messages with RDMA ops. */ - if (rm->m_rdma_op - && test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags)) { + if (rm->m_rdma_op && + test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags)) { spin_lock_irqsave(&conn->c_lock, flags); if (test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) list_move(&rm->m_conn_item, &to_be_dropped); @@ -247,8 +247,8 @@ int rds_send_xmit(struct rds_connection *conn) /* Require an ACK every once in a while */ len = ntohl(rm->m_inc.i_hdr.h_len); - if (conn->c_unacked_packets == 0 - || conn->c_unacked_bytes < len) { + if (conn->c_unacked_packets == 0 || + conn->c_unacked_bytes < len) { __set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); conn->c_unacked_packets = rds_sysctl_max_unacked_packets; @@ -418,8 +418,8 @@ void rds_rdma_send_complete(struct rds_message *rm, int status) spin_lock(&rm->m_rs_lock); ro = rm->m_rdma_op; - if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) - && ro && ro->r_notify && ro->r_notifier) { + if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) && + ro && ro->r_notify && ro->r_notifier) { notifier = ro->r_notifier; rs = rm->m_rs; sock_hold(rds_rs_to_sk(rs)); @@ -549,8 +549,7 @@ void rds_send_remove_from_sock(struct list_head *messages, int status) list_del_init(&rm->m_sock_item); rds_send_sndbuf_remove(rs, rm); - if (ro && ro->r_notifier - && (status || ro->r_notify)) { + if (ro && ro->r_notifier && (status || ro->r_notify)) { notifier = ro->r_notifier; list_add_tail(¬ifier->n_list, &rs->rs_notify_queue); @@ -877,8 +876,8 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, if (ret) goto out; - if ((rm->m_rdma_cookie || rm->m_rdma_op) - && conn->c_trans->xmit_rdma == NULL) { + if ((rm->m_rdma_cookie || rm->m_rdma_op) && + conn->c_trans->xmit_rdma == NULL) { if (printk_ratelimit()) printk(KERN_NOTICE "rdma_op %p conn xmit_rdma %p\n", rm->m_rdma_op, conn->c_trans->xmit_rdma); @@ -890,8 +889,8 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg, * have scheduled a delayed reconnect however - in this case * we should not interfere. */ - if (rds_conn_state(conn) == RDS_CONN_DOWN - && !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) + if (rds_conn_state(conn) == RDS_CONN_DOWN && + !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) queue_delayed_work(rds_wq, &conn->c_conn_w, 0); ret = rds_cong_wait(conn->c_fcong, dport, nonblock, rs); @@ -973,8 +972,8 @@ rds_send_pong(struct rds_connection *conn, __be16 dport) * have scheduled a delayed reconnect however - in this case * we should not interfere. */ - if (rds_conn_state(conn) == RDS_CONN_DOWN - && !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) + if (rds_conn_state(conn) == RDS_CONN_DOWN && + !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) queue_delayed_work(rds_wq, &conn->c_conn_w, 0); ret = rds_cong_wait(conn->c_fcong, dport, 1, NULL); diff --git a/net/rds/sysctl.c b/net/rds/sysctl.c index 307dc5c1be15..7829a20325d3 100644 --- a/net/rds/sysctl.c +++ b/net/rds/sysctl.c @@ -51,55 +51,50 @@ unsigned int rds_sysctl_ping_enable = 1; static ctl_table rds_sysctl_rds_table[] = { { - .ctl_name = CTL_UNNUMBERED, .procname = "reconnect_min_delay_ms", .data = &rds_sysctl_reconnect_min_jiffies, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .proc_handler = proc_doulongvec_ms_jiffies_minmax, .extra1 = &rds_sysctl_reconnect_min, .extra2 = &rds_sysctl_reconnect_max_jiffies, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "reconnect_max_delay_ms", .data = &rds_sysctl_reconnect_max_jiffies, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .proc_handler = proc_doulongvec_ms_jiffies_minmax, .extra1 = &rds_sysctl_reconnect_min_jiffies, .extra2 = &rds_sysctl_reconnect_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unacked_packets", .data = &rds_sysctl_max_unacked_packets, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "max_unacked_bytes", .data = &rds_sysctl_max_unacked_bytes, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "ping_enable", .data = &rds_sysctl_ping_enable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, - { .ctl_name = 0} + { } }; static struct ctl_path rds_sysctl_path[] = { - { .procname = "net", .ctl_name = CTL_NET, }, - { .procname = "rds", .ctl_name = CTL_UNNUMBERED, }, + { .procname = "net", }, + { .procname = "rds", }, { } }; diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 24b743eb0b1b..45474a436862 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -67,11 +67,11 @@ static int rds_tcp_accept_one(struct socket *sock) inet = inet_sk(new_sock->sk); rdsdebug("accepted tcp %u.%u.%u.%u:%u -> %u.%u.%u.%u:%u\n", - NIPQUAD(inet->saddr), ntohs(inet->sport), - NIPQUAD(inet->daddr), ntohs(inet->dport)); + NIPQUAD(inet->inet_saddr), ntohs(inet->inet_sport), + NIPQUAD(inet->inet_daddr), ntohs(inet->inet_dport)); - conn = rds_conn_create(inet->saddr, inet->daddr, &rds_tcp_transport, - GFP_KERNEL); + conn = rds_conn_create(inet->inet_saddr, inet->inet_daddr, + &rds_tcp_transport, GFP_KERNEL); if (IS_ERR(conn)) { ret = PTR_ERR(conn); goto out; diff --git a/net/rds/threads.c b/net/rds/threads.c index dd7e0cad1e7c..00fa10e59af8 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -170,8 +170,8 @@ void rds_shutdown_worker(struct work_struct *work) * handler is supposed to check for state DISCONNECTING */ mutex_lock(&conn->c_cm_lock); - if (!rds_conn_transition(conn, RDS_CONN_UP, RDS_CONN_DISCONNECTING) - && !rds_conn_transition(conn, RDS_CONN_ERROR, RDS_CONN_DISCONNECTING)) { + if (!rds_conn_transition(conn, RDS_CONN_UP, RDS_CONN_DISCONNECTING) && + !rds_conn_transition(conn, RDS_CONN_ERROR, RDS_CONN_DISCONNECTING)) { rds_conn_error(conn, "shutdown called in state %d\n", atomic_read(&conn->c_state)); mutex_unlock(&conn->c_cm_lock); |