From patchwork Tue Oct 5 06:54:20 2010 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Bernard Metzler X-Patchwork-Id: 66760 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 2BA6BB70A8 for ; Tue, 5 Oct 2010 17:54:36 +1100 (EST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757690Ab0JEGy1 (ORCPT ); Tue, 5 Oct 2010 02:54:27 -0400 Received: from mtagate4.de.ibm.com ([195.212.17.164]:34646 "EHLO mtagate4.de.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757676Ab0JEGyZ (ORCPT ); Tue, 5 Oct 2010 02:54:25 -0400 Received: from d12nrmr1607.megacenter.de.ibm.com (d12nrmr1607.megacenter.de.ibm.com [9.149.167.49]) by mtagate4.de.ibm.com (8.13.1/8.13.1) with ESMTP id o956sLgv014484; Tue, 5 Oct 2010 06:54:21 GMT Received: from d12av01.megacenter.de.ibm.com (d12av01.megacenter.de.ibm.com [9.149.165.212]) by d12nrmr1607.megacenter.de.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id o956sLBj3948742; Tue, 5 Oct 2010 08:54:21 +0200 Received: from d12av01.megacenter.de.ibm.com (loopback [127.0.0.1]) by d12av01.megacenter.de.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id o956sKdr021702; Tue, 5 Oct 2010 08:54:21 +0200 Received: from inn.zurich.ibm.com (inn.zurich.ibm.com [9.4.4.229]) by d12av01.megacenter.de.ibm.com (8.12.11.20060308/8.12.11) with ESMTP id o956sKhl021693; Tue, 5 Oct 2010 08:54:20 +0200 Received: from localhost.localdomain (achilles.zurich.ibm.com [9.4.243.2]) by inn.zurich.ibm.com (AIX5.3/8.13.4/8.13.4) with ESMTP id o956sKD5679958; Tue, 5 Oct 2010 08:54:20 +0200 From: Bernard Metzler To: netdev@vger.kernel.org Cc: linux-rdma@vger.kernel.org, Bernard Metzler Subject: [PATCH] SIW: Connection management Date: Tue, 5 Oct 2010 08:54:20 +0200 Message-Id: <1286261660-5157-1-git-send-email-bmt@zurich.ibm.com> X-Mailer: git-send-email 1.5.4.3 Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org --- drivers/infiniband/hw/siw/siw_cm.c | 1939 ++++++++++++++++++++++++++++++++++++ drivers/infiniband/hw/siw/siw_cm.h | 155 +++ 2 files changed, 2094 insertions(+), 0 deletions(-) create mode 100644 drivers/infiniband/hw/siw/siw_cm.c create mode 100644 drivers/infiniband/hw/siw/siw_cm.h diff --git a/drivers/infiniband/hw/siw/siw_cm.c b/drivers/infiniband/hw/siw/siw_cm.c new file mode 100644 index 0000000..628b3b1 --- /dev/null +++ b/drivers/infiniband/hw/siw/siw_cm.c @@ -0,0 +1,1939 @@ +/* + * Software iWARP device driver for Linux + * + * Authors: Bernard Metzler + * Fredy Neeser + * + * Copyright (c) 2008-2010, IBM Corporation + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * - Neither the name of IBM nor the names of its contributors may be + * used to endorse or promote products derived from this software without + * specific prior written permission. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + + +#include +#include +#include +#include + +#include "siw.h" +#include "siw_cm.h" +#include "siw_obj.h" + +static int mpa_crc_enabled; +module_param(mpa_crc_enabled, int, 0644); +MODULE_PARM_DESC(mpa_crc_enabled, "MPA CRC enabled"); + +static int mpa_revision = 1; + + +/* + * siw_sock_nodelay() - Disable Nagle algorithm + * + * See also fs/ocfs2/cluster/tcp.c, o2net_set_nodelay() + */ +static int siw_sock_nodelay(struct socket *sock) +{ + int ret, val = 1; + mm_segment_t oldfs; + oldfs = get_fs(); + set_fs(KERNEL_DS); + + /* + * Don't use sock_setsockopt() for SOL_TCP. It doesn't check its level + * argument and assumes SOL_SOCKET so, say, your TCP_NODELAY will + * silently turn into SO_DEBUG. + */ + ret = sock->ops->setsockopt(sock, SOL_TCP, TCP_NODELAY, + (char __user *)&val, sizeof(val)); + set_fs(oldfs); + return ret; +} + +static void siw_cm_llp_state_change(struct sock *); +static void siw_cm_llp_data_ready(struct sock *, int); +static void siw_cm_llp_write_space(struct sock *); +static void siw_cm_llp_error_report(struct sock *); +static void siw_sk_assign_cm_upcalls(struct sock *sk) + +{ + write_lock_bh(&sk->sk_callback_lock); + sk->sk_state_change = siw_cm_llp_state_change; + sk->sk_data_ready = siw_cm_llp_data_ready; + sk->sk_write_space = siw_cm_llp_write_space; + sk->sk_error_report = siw_cm_llp_error_report; + write_unlock_bh(&sk->sk_callback_lock); +} + +static void siw_sk_save_upcalls(struct sock *sk) +{ + struct siw_cep *cep = sk_to_cep(sk); + BUG_ON(!cep); + + write_lock_bh(&sk->sk_callback_lock); + cep->sk_state_change = sk->sk_state_change; + cep->sk_data_ready = sk->sk_data_ready; + cep->sk_write_space = sk->sk_write_space; + cep->sk_error_report = sk->sk_error_report; + write_unlock_bh(&sk->sk_callback_lock); +} + +static void siw_sk_restore_upcalls(struct sock *sk, struct siw_cep *cep) +{ + sk->sk_state_change = cep->sk_state_change; + sk->sk_data_ready = cep->sk_data_ready; + sk->sk_write_space = cep->sk_write_space; + sk->sk_error_report = cep->sk_error_report; + sk->sk_user_data = NULL; + sk->sk_no_check = 0; +} + +static void siw_socket_disassoc(struct socket *s) +{ + struct sock *sk = s->sk; + struct siw_cep *cep; + + if (sk) { + write_lock_bh(&sk->sk_callback_lock); + cep = sk_to_cep(sk); + if (cep) { + siw_sk_restore_upcalls(sk, cep); + siw_cep_put(cep); + } + write_unlock_bh(&sk->sk_callback_lock); + } +} + + +static inline int kernel_peername(struct socket *s, struct sockaddr_in *addr) +{ + int unused; + return s->ops->getname(s, (struct sockaddr *)addr, &unused, 1); +} + +static inline int kernel_localname(struct socket *s, struct sockaddr_in *addr) +{ + int unused; + return s->ops->getname(s, (struct sockaddr *)addr, &unused, 0); +} + +static void siw_cep_socket_assoc(struct siw_cep *cep, struct socket *s) +{ + cep->llp.sock = s; + siw_cep_get(cep); + s->sk->sk_user_data = cep; + + siw_sk_save_upcalls(s->sk); + siw_sk_assign_cm_upcalls(s->sk); +} + + +static struct siw_cep *siw_cep_alloc(void) +{ + struct siw_cep *cep = kzalloc(sizeof *cep, GFP_KERNEL); + if (cep) { + INIT_LIST_HEAD(&cep->list); + INIT_LIST_HEAD(&cep->work_freelist); + + cep->mpa.hdr.params.c = mpa_crc_enabled ? 1 : 0; + cep->mpa.hdr.params.m = 0; + cep->mpa.hdr.params.rev = mpa_revision ? 1 : 0; + kref_init(&cep->ref); + cep->state = SIW_EPSTATE_IDLE; + init_waitqueue_head(&cep->waitq); + spin_lock_init(&cep->lock); + dprint(DBG_OBJ|DBG_CM, "(CEP 0x%p): New Object\n", cep); + } + return cep; +} + +static void siw_cm_free_work(struct siw_cep *cep) +{ + struct list_head *w, *tmp; + struct siw_cm_work *work; + + list_for_each_safe(w, tmp, &cep->work_freelist) { + work = list_entry(w, struct siw_cm_work, list); + list_del(&work->list); + kfree(work); + } +} + +static void siw_put_work(struct siw_cm_work *work) +{ + INIT_LIST_HEAD(&work->list); + spin_lock_bh(&work->cep->lock); + list_add(&work->list, &work->cep->work_freelist); + spin_unlock_bh(&work->cep->lock); +} + + +static void __siw_cep_dealloc(struct kref *ref) +{ + struct siw_cep *cep = container_of(ref, struct siw_cep, ref); + + dprint(DBG_OBJ|DBG_CM, "(CEP 0x%p): Free Object\n", cep); + + if (cep->listen_cep) + siw_cep_put(cep->listen_cep); + + /* kfree(NULL) is save */ + kfree(cep->mpa.pdata); + spin_lock_bh(&cep->lock); + if (!list_empty(&cep->work_freelist)) + siw_cm_free_work(cep); + spin_unlock_bh(&cep->lock); + + kfree(cep); +} + +static struct siw_cm_work *siw_get_work(struct siw_cep *cep) +{ + struct siw_cm_work *work = NULL; + + spin_lock_bh(&cep->lock); + if (!list_empty(&cep->work_freelist)) { + work = list_entry(cep->work_freelist.next, struct siw_cm_work, + list); + list_del_init(&work->list); + } + spin_unlock_bh(&cep->lock); + return work; +} + +static int siw_cm_alloc_work(struct siw_cep *cep, int num) +{ + struct siw_cm_work *work; + + BUG_ON(!list_empty(&cep->work_freelist)); + + while (num--) { + work = kmalloc(sizeof *work, GFP_KERNEL); + if (!work) { + if (!(list_empty(&cep->work_freelist))) + siw_cm_free_work(cep); + dprint(DBG_ON, " Failed\n"); + return -ENOMEM; + } + work->cep = cep; + INIT_LIST_HEAD(&work->list); + list_add(&work->list, &cep->work_freelist); + } + return 0; +} + +static void siw_cm_release(struct siw_cep *cep) +{ + if (cep->llp.sock) { + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + } + if (cep->qp) { + struct siw_qp *qp = cep->qp; + cep->qp = NULL; + siw_qp_put(qp); + } + if (cep->cm_id) { + cep->cm_id->rem_ref(cep->cm_id); + cep->cm_id = NULL; + siw_cep_put(cep); + } + cep->state = SIW_EPSTATE_CLOSED; +} + +/* + * Test and set CEP into CLOSE pending. After calling + * this function, the CEP conn_close flag is set. Returns: + * + * 1, if CEP is currently in use, + * 0, if CEP is not in use and not already in CLOSE, + * -1, if CEP is not in use and already in CLOSE. + */ +int siw_cep_in_close(struct siw_cep *cep) +{ + int rv; + + spin_lock_bh(&cep->lock); + + dprint(DBG_CM, " (CEP 0x%p): close %d, use %d\n", + cep, cep->conn_close, cep->in_use); + + rv = cep->in_use ? 1 : (cep->conn_close ? -1 : 0); + cep->conn_close = 1; /* may be redundant */ + + spin_unlock_bh(&cep->lock); + + return rv; +} + +/* + * siw_qp_cm_drop() + * + * Drops established LLP connection if present and not already + * scheduled for dropping. Called from user context, SQ workqueue + * or receive IRQ. Caller signals if socket can be immediately + * closed (basically, if not in IRQ) and if IWCM should get + * informed of LLP state change. + */ +void siw_qp_cm_drop(struct siw_qp *qp, int schedule) +{ + struct siw_cep *cep = qp->cep; + + qp->rx_ctx.rx_suspend = 1; + qp->tx_ctx.tx_suspend = 1; + + if (cep && !siw_cep_in_close(cep)) { + if (schedule) { + siw_cm_queue_work(cep, SIW_CM_WORK_CLOSE_LLP); + return; + } + /* + * Immediately close socket + */ + dprint(DBG_CM, "(): immediate close, cep->state=%d\n", + cep->state); + + if (cep->cm_id) { + switch (cep->state) { + + case SIW_EPSTATE_AWAIT_MPAREP: + siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, + IW_CM_EVENT_STATUS_EINVAL); + break; + + case SIW_EPSTATE_RDMA_MODE: + siw_cm_upcall(cep, IW_CM_EVENT_CLOSE, + IW_CM_EVENT_STATUS_OK); + + break; + + case SIW_EPSTATE_IDLE: + case SIW_EPSTATE_LISTENING: + case SIW_EPSTATE_CONNECTING: + case SIW_EPSTATE_AWAIT_MPAREQ: + case SIW_EPSTATE_RECVD_MPAREQ: + case SIW_EPSTATE_CLOSED: + default: + + break; + } + cep->cm_id->rem_ref(cep->cm_id); + cep->cm_id = NULL; + siw_cep_put(cep); + } + cep->state = SIW_EPSTATE_CLOSED; + + if (cep->llp.sock) { + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + } + cep->qp = NULL; + siw_qp_put(qp); + } +} + + +/* + * Set CEP in_use flag. Returns: + * + * 1, if CEP was not in use and not scheduled for closing, + * 0, if CEP was not in use but scheduled for closing, + * -1, if CEP is currently in use. + */ +static int siw_cep_set_inuse(struct siw_cep *cep) +{ + int rv; + + spin_lock_bh(&cep->lock); + + dprint(DBG_CM, " (CEP 0x%p): close %d, use %d\n", + cep, cep->conn_close, cep->in_use); + + rv = cep->in_use ? -1 : (cep->conn_close ? 0 : 1); + cep->in_use = 1; /* may be redundant */ + + spin_unlock_bh(&cep->lock); + + return rv; +} + +/* + * Clear CEP in_use flag. Returns: + * + * 1, if CEP is not scheduled for closing, + * 0, else. + */ +static int siw_cep_set_free(struct siw_cep *cep) +{ + int rv; + + spin_lock_bh(&cep->lock); + + dprint(DBG_CM, " (CEP 0x%p): close %d, use %d\n", + cep, cep->conn_close, cep->in_use); + + cep->in_use = 0; + rv = cep->conn_close ? 0 : 1; + + spin_unlock_bh(&cep->lock); + + wake_up(&cep->waitq); + + return rv; +} + + +void siw_cep_put(struct siw_cep *cep) +{ + dprint(DBG_OBJ|DBG_CM, "(CEP 0x%p): New refcount: %d\n", + cep, atomic_read(&cep->ref.refcount) - 1); + + if (!kref_put(&cep->ref, __siw_cep_dealloc)) + wake_up(&cep->waitq); +} + +void siw_cep_get(struct siw_cep *cep) +{ + kref_get(&cep->ref); + dprint(DBG_OBJ|DBG_CM, "(CEP 0x%p): New refcount: %d\n", + cep, atomic_read(&cep->ref.refcount)); +} + + + +static inline int ksock_recv(struct socket *sock, char *buf, size_t size, + int flags) +{ + struct kvec iov = {buf, size}; + struct msghdr msg = {.msg_name = NULL, .msg_flags = flags}; + + return kernel_recvmsg(sock, &msg, &iov, 1, size, flags); +} + +/* + * Receive MPA Request/Reply heder. + * + * Returns 0 if complete MPA Request/Reply haeder including + * eventual private data was received. Returns -EAGAIN if + * header was partially received or negative error code otherwise. + * + * Context: May be called in process context only + */ +static int siw_recv_mpa_rr(struct siw_cep *cep) +{ + struct mpa_rr *hdr = &cep->mpa.hdr; + struct socket *s = cep->llp.sock; + int rcvd, to_rcv; + + if (cep->mpa.bytes_rcvd < sizeof(struct mpa_rr)) { + + rcvd = ksock_recv(s, (char *)hdr + cep->mpa.bytes_rcvd, + sizeof(struct mpa_rr) - + cep->mpa.bytes_rcvd, 0); + + if (rcvd <= 0) + return -ECONNABORTED; + + cep->mpa.bytes_rcvd += rcvd; + + if (cep->mpa.bytes_rcvd < sizeof(struct mpa_rr)) + return -EAGAIN; + + hdr->params.pd_len = ntohs(hdr->params.pd_len); + + if (hdr->params.pd_len > MPA_MAX_PRIVDATA) + return -EPROTO; + } + + /* + * At least the MPA Request/Reply header (frame not including + * private data) has been received. + * Receive (or continue receiving) any private data. + */ + to_rcv = hdr->params.pd_len - + (cep->mpa.bytes_rcvd - sizeof(struct mpa_rr)); + + if (!to_rcv) { + /* + * We must have hdr->params.pd_len == 0 and thus received a + * complete MPA Request/Reply frame. + * Check against peer protocol violation. + */ + __u32 word; + + rcvd = ksock_recv(s, (char *)&word, sizeof word, MSG_DONTWAIT); + if (rcvd == -EAGAIN) + return 0; + + if (rcvd == 0) { + dprint(DBG_CM, " peer EOF\n"); + return -EPIPE; + } + if (rcvd < 0) { + dprint(DBG_CM, " ERROR: %d: \n", rcvd); + return rcvd; + } + dprint(DBG_CM, " peer sent extra data: %d\n", rcvd); + return -EPROTO; + } + + /* + * At this point, we must have hdr->params.pd_len != 0. + * A private data buffer gets allocated iff hdr->params.pd_len != 0. + * Ownership of this buffer will be transferred to the IWCM + * when calling siw_cm_upcall(). + */ + if (!cep->mpa.pdata && + !(cep->mpa.pdata = kmalloc(hdr->params.pd_len + 4, GFP_KERNEL))) + return -ENOMEM; + + rcvd = ksock_recv(s, cep->mpa.pdata + cep->mpa.bytes_rcvd + - sizeof(struct mpa_rr), to_rcv + 4, MSG_DONTWAIT); + + if (rcvd < 0) + return rcvd; + + if (rcvd > to_rcv) + return -EPROTO; + + cep->mpa.bytes_rcvd += rcvd; + + if (to_rcv == rcvd) { + dprint(DBG_CM, "%d bytes private_data received", + hdr->params.pd_len); + return 0; + } + return -EAGAIN; +} + + +static void siw_proc_mpareq(struct siw_cep *cep) +{ + int err = siw_recv_mpa_rr(cep); + + if (err) + goto out; + + if (cep->mpa.hdr.params.rev > MPA_REVISION_1) { + /* allow for 0 and 1 only */ + err = -EPROTO; + goto out; + } + + if (memcmp(cep->mpa.hdr.key, MPA_KEY_REQ, sizeof cep->mpa.hdr.key)) { + err = -EPROTO; + goto out; + } + cep->state = SIW_EPSTATE_RECVD_MPAREQ; + + if (cep->listen_cep->state == SIW_EPSTATE_LISTENING) { + /* + * Since siw_cm_upcall() called with success, iwcm must hold + * a reference to the CEP until the IW_CM_EVENT_CONNECT_REQUEST + * has been accepted or rejected. + * NOTE: If the iwcm never calls back with accept/reject, + * (e.g., the user types ^C instead), the CEP can never be + * free'd. It results in a memory hole which should be + * fixed by calling siw_reject() in case of application + * termination.. + */ + siw_cep_get(cep); + + err = siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REQUEST, + IW_CM_EVENT_STATUS_OK); + if (err) + siw_cep_put(cep); + } else { + /* + * listener lost: new connection cannot be signalled + */ + dprint(DBG_CM|DBG_ON, "(cep=0x%p): Listener lost:!\n", cep); + err = -EINVAL; + } +out: + if (err) { + dprint(DBG_CM|DBG_ON, "(cep=0x%p): error %d\n", cep, err); + + if (!siw_cep_in_close(cep)) { + /* + * remove reference from listening cep and clear + * information on related listener. + */ + siw_cep_put(cep->listen_cep); + cep->listen_cep = NULL; + + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + + cep->state = SIW_EPSTATE_CLOSED; + siw_cep_put(cep); + } + } +} + + +static void siw_proc_mpareply(struct siw_cep *cep) +{ + struct siw_qp_attrs qp_attrs; + struct siw_qp *qp = cep->qp; + int rv; + + rv = siw_recv_mpa_rr(cep); + if (rv == -EAGAIN) + /* incomplete mpa reply */ + return; + + if (rv) + goto error; + + if (cep->mpa.hdr.params.rev > MPA_REVISION_1) { + /* allow for 0 and 1 only */ + rv = -EPROTO; + goto error; + } + if (memcmp(cep->mpa.hdr.key, MPA_KEY_REP, sizeof cep->mpa.hdr.key)) { + rv = -EPROTO; + goto error; + } + /* + * TODO: 1. handle eventual MPA reject (upcall with ECONNREFUSED) + * 2. finish mpa parameter check/negotiation + */ + memset(&qp_attrs, 0, sizeof qp_attrs); + qp_attrs.mpa.marker_rcv = 0; + qp_attrs.mpa.marker_snd = 0; + qp_attrs.mpa.crc = CONFIG_RDMA_SIW_CRC_ENFORCED; + qp_attrs.mpa.version = 1; + qp_attrs.ird = cep->ird; + qp_attrs.ord = cep->ord; + qp_attrs.llp_stream_handle = cep->llp.sock; + qp_attrs.state = SIW_QP_STATE_RTS; + + /* Move socket RX/TX under QP control */ + down_write(&qp->state_lock); + if (qp->attrs.state > SIW_QP_STATE_RTR) { + rv = -EINVAL; + up_write(&qp->state_lock); + goto error; + } + rv = siw_qp_modify(qp, &qp_attrs, SIW_QP_ATTR_STATE| + SIW_QP_ATTR_LLP_HANDLE| + SIW_QP_ATTR_ORD| + SIW_QP_ATTR_IRD| + SIW_QP_ATTR_MPA); + + if (!rv) { + cep->state = SIW_EPSTATE_RDMA_MODE; + siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, + IW_CM_EVENT_STATUS_OK); + + up_write(&qp->state_lock); + return; + } + up_write(&qp->state_lock); +error: + /* + * failed socket handover returns responsibility: + * inform iwcm and drop connection + * TODO: 1. send MPA reject for MPA rev==1 + * if rv != ECONNREFUSED + */ + siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, rv); + + if (!siw_cep_in_close(cep)) { + + cep->cm_id->rem_ref(cep->cm_id); + cep->cm_id = NULL; + siw_cep_put(cep); + + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + + cep->qp = NULL; + siw_qp_put(cep->qp); + } + cep->state = SIW_EPSTATE_CLOSED; +} + +/* + * siw_accept_newconn - accept an incoming pending connection + * + */ +static void siw_accept_newconn(struct siw_cep *cep) +{ + struct socket *s = cep->llp.sock; + struct socket *new_s = NULL; + struct siw_cep *new_cep = NULL; + int rv = 0; /* debug only. should disappear */ + + new_cep = siw_cep_alloc(); + if (!new_cep) + goto error; + + if (siw_cm_alloc_work(new_cep, 4) != 0) + goto error; + + /* + * Copy saved socket callbacks from listening CEP + * and assign new socket with new CEP + */ + new_cep->sk_state_change = cep->sk_state_change; + new_cep->sk_data_ready = cep->sk_data_ready; + new_cep->sk_write_space = cep->sk_write_space; + new_cep->sk_error_report = cep->sk_error_report; + + rv = kernel_accept(s, &new_s, O_NONBLOCK); + if (rv != 0) { + /* + * TODO: Already aborted by peer? + * Is there anything we should do? + */ + dprint(DBG_CM|DBG_ON, "(cep=0x%p): ERROR: " + "kernel_accept(): rv=%d\n", cep, rv); + goto error; + } + new_cep->llp.sock = new_s; + siw_cep_get(new_cep); + new_s->sk->sk_user_data = new_cep; + + dprint(DBG_CM, "(cep=0x%p, s=0x%p, new_s=0x%p): " + "New LLP connection accepted\n", cep, s, new_s); + + rv = siw_sock_nodelay(new_s); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(cep=0x%p): ERROR: " + "siw_sock_nodelay(): rv=%d\n", cep, rv); + goto error; + } + + rv = kernel_peername(new_s, &new_cep->llp.raddr); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(cep=0x%p): ERROR: " + "kernel_peername(): rv=%d\n", cep, rv); + goto error; + } + rv = kernel_localname(new_s, &new_cep->llp.laddr); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(cep=0x%p): ERROR: " + "kernel_localname(): rv=%d\n", cep, rv); + goto error; + } + + /* + * See siw_proc_mpareq() etc. for the use of new_cep->listen_cep. + */ + new_cep->listen_cep = cep; + siw_cep_get(cep); + + new_cep->state = SIW_EPSTATE_AWAIT_MPAREQ; + + if (atomic_read(&new_s->sk->sk_rmem_alloc)) { + /* + * MPA REQ already queued + */ + dprint(DBG_CM, "(cep=0x%p): Immediate MPA req.\n", cep); + + siw_proc_mpareq(new_cep); + } + return; + +error: + if (new_cep) + siw_cep_put(new_cep); + + if (new_s) { + siw_socket_disassoc(new_s); + sock_release(new_s); + } + dprint(DBG_CM|DBG_ON, "(cep=0x%p): ERROR: rv=%d\n", cep, rv); +} + +/* + * Expects params->pd_len in host byte order + * + * TODO: We might want to combine the arguments params and pdata to a single + * pointer to a struct siw_mpa_info as defined in siw_cm.h. + * This way, all private data parameters would be in a common struct. + */ +static int siw_send_mpareqrep(struct socket *s, struct mpa_rr_params *params, + char *key, char *pdata) +{ + struct mpa_rr hdr; + struct kvec iov[2]; + struct msghdr msg; + + int rv; + unsigned short pd_len = params->pd_len; + + memset(&msg, 0, sizeof(msg)); + memset(&hdr, 0, sizeof hdr); + memcpy(hdr.key, key, 16); + + /* + * TODO: By adding a union to struct mpa_rr_params, it should be + * possible to replace the next 4 statements by one + */ + hdr.params.r = params->r; + hdr.params.c = params->c; + hdr.params.m = params->m; + hdr.params.rev = params->rev; + + if (pd_len > MPA_MAX_PRIVDATA) + return -EINVAL; + + hdr.params.pd_len = htons(pd_len); + + iov[0].iov_base = &hdr; + iov[0].iov_len = sizeof hdr; + + if (pd_len) { + iov[1].iov_base = pdata; + iov[1].iov_len = pd_len; + + rv = kernel_sendmsg(s, &msg, iov, 2, pd_len + sizeof hdr); + } else + rv = kernel_sendmsg(s, &msg, iov, 1, sizeof hdr); + + return rv < 0 ? rv : 0; +} + +/* + * siw_cm_upcall() + * + * Upcall to IWCM to inform about async connection events + */ +int siw_cm_upcall(struct siw_cep *cep, enum iw_cm_event_type reason, + enum iw_cm_event_status status) +{ + struct iw_cm_event event; + struct iw_cm_id *cm_id; + + memset(&event, 0, sizeof event); + event.status = status; + event.event = reason; + + if (cep->mpa.hdr.params.pd_len != 0) { + /* + * hand over MPA private data + */ + event.private_data_len = cep->mpa.hdr.params.pd_len; + event.private_data = cep->mpa.pdata; + cep->mpa.hdr.params.pd_len = 0; + } + if (reason == IW_CM_EVENT_CONNECT_REQUEST || + reason == IW_CM_EVENT_CONNECT_REPLY) { + event.local_addr = cep->llp.laddr; + event.remote_addr = cep->llp.raddr; + } + if (reason == IW_CM_EVENT_CONNECT_REQUEST) { + event.provider_data = cep; + cm_id = cep->listen_cep->cm_id; + } else + cm_id = cep->cm_id; + + dprint(DBG_CM, " (QP%d): cep=0x%p, id=0x%p, dev(id)=%s, " + "reason=%d, status=%d\n", + cep->qp ? QP_ID(cep->qp) : -1, cep, cm_id, + cm_id->device->name, reason, status); + + return cm_id->event_handler(cm_id, &event); +} + +static void siw_cm_work_handler(struct work_struct *w) +{ + struct siw_cm_work *work; + struct siw_cep *cep; + int rv; + + work = container_of(w, struct siw_cm_work, work); + cep = work->cep; + + dprint(DBG_CM, " (QP%d): WORK type: %d, CEP: 0x%p\n", + cep->qp ? QP_ID(cep->qp) : -1, work->type, cep); + + switch (work->type) { + + case SIW_CM_WORK_ACCEPT: + + rv = siw_cep_set_inuse(cep); + if (rv > 0) { + if (cep->state == SIW_EPSTATE_LISTENING) + siw_accept_newconn(cep); + + if (!siw_cep_set_free(cep)) { + siw_cm_release(cep); + siw_cep_put(cep); + } + break; + } + /* + * CEP already scheduled for closing + */ + if (!rv) { + siw_cm_release(cep); + (void) siw_cep_set_free(cep); + } + break; + + case SIW_CM_WORK_READ_MPAHDR: + + rv = siw_cep_set_inuse(cep); + if (rv > 0) { + switch (cep->state) { + + case SIW_EPSTATE_AWAIT_MPAREQ: + + siw_proc_mpareq(cep); + break; + + case SIW_EPSTATE_AWAIT_MPAREP: + + siw_proc_mpareply(cep); + break; + + default: + /* + * CEP already moved out of MPA handshake. + * any connection management already done. + * silently ignore the mpa packet. + */ + dprint(DBG_CM, "(): CEP not in MPA " + "handshake state: %d\n", cep->state); + } + if (!siw_cep_set_free(cep)) + siw_cm_release(cep); + + break; + } + /* + * CEP already scheduled for closing + */ + if (!rv) { + siw_cm_release(cep); + (void) siw_cep_set_free(cep); + } + break; + + case SIW_CM_WORK_CLOSE_LLP: + /* + * QP scheduled LLP close + */ + dprint(DBG_CM, "(): SIW_CM_WORK_CLOSE_LLP, cep->state=%d\n", + cep->state); + + cep->state = SIW_EPSTATE_CLOSED; + + if (cep->llp.sock) { + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + } + if (cep->qp) { + siw_qp_llp_close(cep->qp); + siw_qp_put(cep->qp); + cep->qp = NULL; + } + if (cep->cm_id) { + siw_cm_upcall(cep, IW_CM_EVENT_CLOSE, + IW_CM_EVENT_STATUS_OK); + + cep->cm_id->rem_ref(cep->cm_id); + cep->cm_id = NULL; + siw_cep_put(cep); + } + break; + + case SIW_CM_WORK_PEER_CLOSE: + + dprint(DBG_CM, "(): SIW_CM_WORK_PEER_CLOSE, " + "cep->state=%d\n", cep->state); + + if (cep->cm_id) { + switch (cep->state) { + + case SIW_EPSTATE_AWAIT_MPAREP: + /* + * MPA reply not received, but connection drop + */ + siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, + -ECONNRESET); + break; + + case SIW_EPSTATE_RDMA_MODE: + /* + * NOTE: IW_CM_EVENT_DISCONNECT is given just + * to transition IWCM into CLOSING. + * FIXME: is that needed? + */ + siw_cm_upcall(cep, IW_CM_EVENT_DISCONNECT, + IW_CM_EVENT_STATUS_OK); + siw_cm_upcall(cep, IW_CM_EVENT_CLOSE, + IW_CM_EVENT_STATUS_OK); + + break; + + default: + + break; + /* + * for these states there is no connection + * known to the IWCM. Even not for + * SIW_EPSTATE_RECVD_MPAREQ. + */ + } + cep->cm_id->rem_ref(cep->cm_id); + cep->cm_id = NULL; + siw_cep_put(cep); + } + if (cep->qp) { + siw_qp_llp_close(cep->qp); + siw_qp_put(cep->qp); + cep->qp = NULL; + } + if (cep->state != SIW_EPSTATE_CLOSED) { + cep->state = SIW_EPSTATE_CLOSED; + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + } + + break; + + default: + BUG(); + } + dprint(DBG_CM, " (Exit): WORK type: %d, CEP: 0x%p\n", work->type, cep); + siw_put_work(work); + siw_cep_put(cep); +} + +static struct workqueue_struct *siw_cm_wq; + +int siw_cm_queue_work(struct siw_cep *cep, enum siw_work_type type) +{ + struct siw_cm_work *work = siw_get_work(cep); + + dprint(DBG_CM, " (QP%d): WORK type: %d, CEP: 0x%p\n", + cep->qp ? QP_ID(cep->qp) : -1, type, cep); + + if (!work) { + dprint(DBG_ON, " Failed\n"); + return -ENOMEM; + } + work->type = type; + work->cep = cep; + + siw_cep_get(cep); + + INIT_WORK(&work->work, siw_cm_work_handler); + queue_work(siw_cm_wq, &work->work); + + return 0; +} + + +static void siw_cm_llp_data_ready(struct sock *sk, int flags) +{ + struct siw_cep *cep; + + read_lock(&sk->sk_callback_lock); + + cep = sk_to_cep(sk); + if (!cep) { + WARN_ON(1); + goto out; + } + + if (cep->conn_close) + goto out; + + dprint(DBG_CM, "(): cep 0x%p, state: %d, flags %x\n", cep, + cep->state, flags); + + switch (cep->state) { + + case SIW_EPSTATE_RDMA_MODE: + case SIW_EPSTATE_LISTENING: + + break; + + case SIW_EPSTATE_AWAIT_MPAREQ: + case SIW_EPSTATE_AWAIT_MPAREP: + + siw_cm_queue_work(cep, SIW_CM_WORK_READ_MPAHDR); + break; + + default: + dprint(DBG_CM, "(): Unexpected DATA, state %d\n", cep->state); + break; + } +out: + read_unlock(&sk->sk_callback_lock); +} + +static void siw_cm_llp_write_space(struct sock *sk) +{ + struct siw_cep *cep = sk_to_cep(sk); + + if (cep) + dprint(DBG_CM, "(): cep: 0x%p, state: %d\n", cep, cep->state); +} + +static void siw_cm_llp_error_report(struct sock *sk) +{ + struct siw_cep *cep = sk_to_cep(sk); + + dprint(DBG_CM, "(): error: %d, state: %d\n", sk->sk_err, sk->sk_state); + + if (cep) { + cep->sk_error = sk->sk_err; + dprint(DBG_CM, "(): cep->state: %d\n", cep->state); + cep->sk_error_report(sk); + } +} + +static void siw_cm_llp_state_change(struct sock *sk) +{ + struct siw_cep *cep; + struct socket *s; + void (*orig_state_change)(struct sock *); + + + read_lock(&sk->sk_callback_lock); + + cep = sk_to_cep(sk); + if (!cep) { + WARN_ON(1); + read_unlock(&sk->sk_callback_lock); + return; + } + orig_state_change = cep->sk_state_change; + + s = sk->sk_socket; + + dprint(DBG_CM, "(): cep: 0x%p, state: %d\n", cep, cep->state); + + switch (sk->sk_state) { + + case TCP_ESTABLISHED: + /* + * handle accepting socket as special case where only + * new connection is possible + */ + if (cep->conn_close) + break; + + if (cep->state == SIW_EPSTATE_LISTENING && + siw_cm_queue_work(cep, SIW_CM_WORK_ACCEPT) != 0) { + dprint(DBG_ON, "Cannot accept\n"); + } + break; + + case TCP_CLOSE: + case TCP_CLOSE_WAIT: + if (cep->state <= SIW_EPSTATE_LISTENING) { + dprint(DBG_CM, "() Close before accept()\n"); + break; + } + if (cep->qp) + cep->qp->tx_ctx.tx_suspend = 1; + + if (!siw_cep_in_close(cep)) + siw_cm_queue_work(cep, SIW_CM_WORK_PEER_CLOSE); + + break; + + default: + dprint(DBG_CM, "Unexpected sock state %d\n", sk->sk_state); + } + read_unlock(&sk->sk_callback_lock); + orig_state_change(sk); +} + + +static int kernel_bindconnect(struct socket *s, + struct sockaddr *laddr, int laddrlen, + struct sockaddr *raddr, int raddrlen, int flags) +{ + int err, s_val = 1; + /* + * XXX + * Tentative fix. Should not be needed but sometimes iwcm + * chooses ports in use + */ + err = kernel_setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char *)&s_val, + sizeof s_val); + if (err < 0) + goto done; + + err = s->ops->bind(s, laddr, laddrlen); + if (err < 0) + goto done; + + err = s->ops->connect(s, raddr, raddrlen, flags); + if (err < 0) + goto done; + + err = s->ops->getname(s, laddr, &s_val, 0); + +done: + return err; +} + + +int siw_connect(struct iw_cm_id *id, struct iw_cm_conn_param *params) +{ + struct siw_dev *dev = siw_dev_ofa2siw(id->device); + struct siw_qp *qp; + struct siw_cep *cep = NULL; + struct socket *s = NULL; + struct sockaddr *laddr, *raddr; + + u16 pd_len = params->private_data_len; + int rv, size; + + if (pd_len > MPA_MAX_PRIVDATA) + return -EINVAL; + + qp = siw_qp_id2obj(dev, params->qpn); + BUG_ON(!qp); + + dprint(DBG_CM, "(id=0x%p, QP%d): dev(id)=%s, l2dev=%s\n", + id, QP_ID(qp), dev->ofa_dev.name, dev->l2dev->name); + dprint(DBG_CM, "(id=0x%p, QP%d): laddr=(0x%x,%d), raddr=(0x%x,%d)\n", + id, QP_ID(qp), + ntohl(id->local_addr.sin_addr.s_addr), + ntohs(id->local_addr.sin_port), + ntohl(id->remote_addr.sin_addr.s_addr), + ntohs(id->remote_addr.sin_port)); + + down_write(&qp->state_lock); + if (qp->attrs.state > SIW_QP_STATE_RTR) { + rv = -EINVAL; + goto error; + } + + laddr = (struct sockaddr *)&id->local_addr; + raddr = (struct sockaddr *)&id->remote_addr; + + rv = sock_create(AF_INET, SOCK_STREAM, IPPROTO_TCP, &s); + if (rv < 0) + goto error; + + size = SOCKBUFSIZE; + rv = kernel_setsockopt(s, SOL_SOCKET, SO_SNDBUF, (char *)&size, + sizeof size); + if (rv < 0) + goto error; + + rv = kernel_setsockopt(s, SOL_SOCKET, SO_RCVBUF, (char *)&size, + sizeof size); + if (rv < 0) + goto error; + + /* + * NOTE: For simplification, connect() is called in blocking + * mode. Might be reconsidered for async connection setup at + * TCP level. + */ + rv = kernel_bindconnect(s, laddr, sizeof *laddr, raddr, + sizeof *raddr, 0); + if (rv != 0) { + dprint(DBG_CM, "(id=0x%p, QP%d): kernel_bindconnect: rv=%d\n", + id, QP_ID(qp), rv); + goto error; + } + rv = siw_sock_nodelay(s); + if (rv != 0) { + dprint(DBG_CM, "(id=0x%p, QP%d): siw_sock_nodelay(): rv=%d\n", + id, QP_ID(qp), rv); + goto error; + } + cep = siw_cep_alloc(); + if (!cep) { + rv = -ENOMEM; + goto error; + } + + /* Associate QP with CEP */ + siw_cep_get(cep); + qp->cep = cep; + + /* siw_qp_get(qp) already done by QP lookup */ + cep->qp = qp; + + id->add_ref(id); + cep->cm_id = id; + + rv = siw_cm_alloc_work(cep, 4); + if (rv != 0) { + rv = -ENOMEM; + goto error; + } + cep->mpa.hdr.params.pd_len = pd_len; + cep->ird = params->ird; + cep->ord = params->ord; + cep->state = SIW_EPSTATE_CONNECTING; + + rv = kernel_peername(s, &cep->llp.raddr); + if (rv) + goto error; + + rv = kernel_localname(s, &cep->llp.laddr); + if (rv) + goto error; + + dprint(DBG_CM, "(id=0x%p, QP%d): pd_len = %u\n", id, QP_ID(qp), pd_len); + if (pd_len) + dprint(DBG_CM, "%d bytes private_data\n", pd_len); + /* + * Associate CEP with socket + */ + siw_cep_socket_assoc(cep, s); + + cep->state = SIW_EPSTATE_AWAIT_MPAREP; + + rv = siw_send_mpareqrep(cep->llp.sock, &cep->mpa.hdr.params, + MPA_KEY_REQ, (char *)params->private_data); + + /* + * Reset private data len: in case connection drops w/o peer + * sending MPA reply we would report stale data pointer during + * IW_CM_EVENT_CONNECT_REPLY. + */ + cep->mpa.hdr.params.pd_len = 0; + + if (rv >= 0) { + dprint(DBG_CM, "(id=0x%p, QP%d): Exit\n", id, QP_ID(qp)); + up_write(&qp->state_lock); + return 0; + } +error: + up_write(&qp->state_lock); + + dprint(DBG_ON, " Failed: %d\n", rv); + + if (cep && !siw_cep_in_close(cep)) { + + siw_socket_disassoc(s); + sock_release(s); + cep->llp.sock = NULL; + + cep->qp = NULL; + + cep->cm_id = NULL; + id->rem_ref(id); + siw_cep_put(cep); + + qp->cep = NULL; + siw_cep_put(cep); + + cep->state = SIW_EPSTATE_CLOSED; + } else if (!cep && s) + sock_release(s); + + siw_qp_put(qp); + + return rv; +} + +/* + * siw_accept - Let SoftiWARP accept an RDMA connection request + * + * @id: New connection management id to be used for accepted + * connection request + * @params: Connection parameters provided by ULP for accepting connection + * + * Transition QP to RTS state, associate new CM id @id with accepted CEP + * and get prepared for TCP input by installing socket callbacks. + * Then send MPA Reply and generate the "connection established" event. + * Socket callbacks must be installed before sending MPA Reply, because + * the latter may cause a first RDMA message to arrive from the RDMA Initiator + * side very quickly, at which time the socket callbacks must be ready. + */ +int siw_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params) +{ + struct siw_dev *dev = siw_dev_ofa2siw(id->device); + struct siw_cep *cep = (struct siw_cep *)id->provider_data; + struct siw_qp *qp; + struct siw_qp_attrs qp_attrs; + char *pdata = NULL; + int rv; + +retry: + rv = siw_cep_set_inuse(cep); + if (rv < 0) { + dprint(DBG_CM, "(id=0x%p, cep=0x%p): CEP in use\n", id, cep); + wait_event(cep->waitq, !cep->in_use); + goto retry; + } + if (!rv) { + dprint(DBG_CM, "(id=0x%p, cep=0x%p): CEP in close\n", id, cep); + (void) siw_cep_set_free(cep); + return -EINVAL; + } + if (cep->state != SIW_EPSTATE_RECVD_MPAREQ) { + if (cep->state == SIW_EPSTATE_CLOSED) { + + dprint(DBG_CM, "(id=0x%p): Out of State\n", id); + (void) siw_cep_set_free(cep); + + siw_cep_put(cep); + return -ECONNRESET; + } + BUG(); + } + /* clear iwcm reference to CEP from IW_CM_EVENT_CONNECT_REQUEST */ + siw_cep_put(cep); + + qp = siw_qp_id2obj(dev, params->qpn); + BUG_ON(!qp); /* The OFA core should prevent this */ + + down_write(&qp->state_lock); + if (qp->attrs.state > SIW_QP_STATE_RTR) { + rv = -EINVAL; + goto unlock; + } + + dprint(DBG_CM, "(id=0x%p, QP%d): dev(id)=%s\n", + id, QP_ID(qp), dev->ofa_dev.name); + + if (params->ord > qp->attrs.ord || params->ird > qp->attrs.ird) { + dprint(DBG_CM|DBG_ON, "(id=0x%p, QP%d): " + "ORD: %d (max: %d), IRD: %d (max: %d)\n", + id, QP_ID(qp), + params->ord, qp->attrs.ord, + params->ird, qp->attrs.ird); + rv = -EINVAL; + goto unlock; + } + if (params->private_data_len > MPA_MAX_PRIVDATA) { + dprint(DBG_CM|DBG_ON, "(id=0x%p, QP%d): " + "Private data too long: %d (max: %d)\n", + id, QP_ID(qp), + params->private_data_len, MPA_MAX_PRIVDATA); + rv = -EINVAL; + goto unlock; + } + cep->cm_id = id; + id->add_ref(id); + + memset(&qp_attrs, 0, sizeof qp_attrs); + qp_attrs.ord = params->ord; + qp_attrs.ird = params->ird; + qp_attrs.llp_stream_handle = cep->llp.sock; + + /* + * TODO: Add MPA negotiation + */ + qp_attrs.mpa.marker_rcv = 0; + qp_attrs.mpa.marker_snd = 0; + qp_attrs.mpa.crc = CONFIG_RDMA_SIW_CRC_ENFORCED; + qp_attrs.mpa.version = 0; + qp_attrs.state = SIW_QP_STATE_RTS; + + dprint(DBG_CM, "(id=0x%p, QP%d): Moving to RTS\n", id, QP_ID(qp)); + + /* Associate QP with CEP */ + siw_cep_get(cep); + qp->cep = cep; + + /* siw_qp_get(qp) already done by QP lookup */ + cep->qp = qp; + + cep->state = SIW_EPSTATE_RDMA_MODE; + + /* Move socket RX/TX under QP control */ + rv = siw_qp_modify(qp, &qp_attrs, SIW_QP_ATTR_STATE| + SIW_QP_ATTR_LLP_HANDLE| + SIW_QP_ATTR_ORD| + SIW_QP_ATTR_IRD| + SIW_QP_ATTR_MPA); + up_write(&qp->state_lock); + + if (rv) + goto error; + + + /* + * TODO: It might be more elegant and concise to check the + * private data length cep->mpa.hdr.params.pd_len + * inside siw_send_mpareqrep(). + */ + if (params->private_data_len) { + pdata = (char *)params->private_data; + + dprint(DBG_CM, "(id=0x%p, QP%d): %d bytes private_data\n", + id, QP_ID(qp), params->private_data_len); + } + cep->mpa.hdr.params.pd_len = params->private_data_len; + + dprint(DBG_CM, "(id=0x%p, QP%d): Sending MPA Reply\n", id, QP_ID(qp)); + + rv = siw_send_mpareqrep(cep->llp.sock, &cep->mpa.hdr.params, + MPA_KEY_REP, pdata); + if (!rv) { + /* + * FIXME: In order to ensure that the first FPDU will be sent + * from the RDMA Initiator side, the "connection established" + * event should be delayed until Softiwarp has received the + * first FPDU from the RDMA Initiator side. + * Alternatively, Softiwarp could prevent this side to + * send a first FPDU until a first FPDU has been received. + * + * The two alternatives above will work if + * (1) the RDMA application is iWARP standards compliant + * by sending its first RDMA payload from the + * RDMA Initiator side, or + * (2) the RDMA Initiator side RNIC inserts an under-cover + * zero-length RDMA operation (negotiated through an + * extended MPA Request/Reply handshake) such as a + * zero-length RDMA Write or Read. + * Note that (2) would require an extension of the MPA RFC. + * + * A third alternative (which may be the easiest for now) is to + * return an error to an RDMA application that attempts to send + * the first RDMA payload from the RDMA Responder side. + */ + siw_cm_upcall(cep, IW_CM_EVENT_ESTABLISHED, + IW_CM_EVENT_STATUS_OK); + + if (!siw_cep_set_free(cep)) + siw_cm_release(cep); + + dprint(DBG_CM, "(id=0x%p, QP%d): Exit\n", id, QP_ID(qp)); + return 0; + } + +error: + if (siw_cep_set_free(cep)) { + + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + + cep->state = SIW_EPSTATE_CLOSED; + + cep->cm_id->rem_ref(id); + cep->cm_id = NULL; + + if (qp->cep) { + siw_cep_put(cep); + qp->cep = NULL; + } + cep->qp = NULL; + siw_qp_put(qp); + } + return rv; +unlock: + up_write(&qp->state_lock); + goto error; +} + +/* + * siw_reject() + * + * Local connection reject case. Send private data back to peer, + * close connection and dereference connection id. + */ +int siw_reject(struct iw_cm_id *id, const void *pdata, u8 plen) +{ + struct siw_cep *cep = (struct siw_cep *)id->provider_data; + + dprint(DBG_CM, "(id=0x%p): cep->state=%d\n", id, cep->state); + dprint(DBG_CM, " Reject: %s\n", plen ? (char *)pdata:"(no data)"); + + if (!siw_cep_in_close(cep)) { + + dprint(DBG_ON, " Sending REJECT not yet implemented\n"); + + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + + siw_cep_put(cep); + cep->state = SIW_EPSTATE_CLOSED; + } else { + dprint(DBG_CM, " (id=0x%p): Connection lost\n", id); + } + + /* + * clear iwcm reference to CEP from + * IW_CM_EVENT_CONNECT_REQUEST + */ + siw_cep_put(cep); + + return 0; +} + +int siw_listen_address(struct iw_cm_id *id, int backlog, struct sockaddr *laddr) +{ + struct socket *s; + struct siw_cep *cep = NULL; + int rv = 0, s_val; + + rv = sock_create(AF_INET, SOCK_STREAM, IPPROTO_TCP, &s); + if (rv < 0) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): ERROR: " + "sock_create(): rv=%d\n", id, rv); + return rv; + } + + s_val = SOCKBUFSIZE; + rv = kernel_setsockopt(s, SOL_SOCKET, SO_SNDBUF, (char *)&s_val, + sizeof s_val); + if (rv) + goto error; + + rv = kernel_setsockopt(s, SOL_SOCKET, SO_RCVBUF, (char *)&s_val, + sizeof s_val); + if (rv) + goto error; + + /* + * Probably to be removed later. Allows binding + * local port when still in TIME_WAIT from last close. + */ + s_val = 1; + rv = kernel_setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char *)&s_val, + sizeof s_val); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): ERROR: " + "kernel_setsockopt(): rv=%d\n", id, rv); + goto error; + } + + rv = s->ops->bind(s, laddr, sizeof *laddr); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): ERROR: bind(): rv=%d\n", + id, rv); + goto error; + } + + cep = siw_cep_alloc(); + if (!cep) { + rv = -ENOMEM; + goto error; + } + siw_cep_socket_assoc(cep, s); + + rv = siw_cm_alloc_work(cep, backlog); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): ERROR: " + "siw_cm_alloc_work(backlog=%d): rv=%d\n", + id, backlog, rv); + goto error; + } + + rv = s->ops->listen(s, backlog); + if (rv != 0) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): ERROR: listen() rv=%d\n", + id, rv); + goto error; + } + + /* + * TODO: Do we really need the copies of local_addr and remote_addr + * in CEP ??? + */ + memcpy(&cep->llp.laddr, &id->local_addr, sizeof cep->llp.laddr); + memcpy(&cep->llp.raddr, &id->remote_addr, sizeof cep->llp.raddr); + + cep->cm_id = id; + id->add_ref(id); + + /* + * In case of a wildcard rdma_listen on a multi-homed device, + * a listener's IWCM id is associated with more than one listening CEP. + * + * We currently use id->provider_data in three different ways: + * + * o For a listener's IWCM id, id->provider_data points to + * the list_head of the list of listening CEPs. + * Uses: siw_create_listen(), siw_destroy_listen() + * + * o For a passive-side IWCM id, id->provider_data points to + * the CEP itself. This is a consequence of + * - siw_cm_upcall() setting event.provider_data = cep and + * - the IWCM's cm_conn_req_handler() setting provider_data of the + * new passive-side IWCM id equal to event.provider_data + * Uses: siw_accept(), siw_reject() + * + * o For an active-side IWCM id, id->provider_data is not used at all. + * + */ + if (!id->provider_data) { + id->provider_data = kmalloc(sizeof(struct list_head), + GFP_KERNEL); + if (!id->provider_data) { + rv = -ENOMEM; + goto error; + } + INIT_LIST_HEAD((struct list_head *)id->provider_data); + } + + dprint(DBG_CM, "(id=0x%p): dev(id)=%s, l2dev=%s, " + "id->provider_data=0x%p, cep=0x%p\n", + id, id->device->name, + siw_dev_ofa2siw(id->device)->l2dev->name, + id->provider_data, cep); + + list_add_tail(&cep->list, (struct list_head *)id->provider_data); + cep->state = SIW_EPSTATE_LISTENING; + return 0; + +error: + dprint(DBG_ON, " Failed: %d\n", rv); + + if (cep) { + cep->llp.sock = NULL; + siw_socket_disassoc(s); + cep->state = SIW_EPSTATE_CLOSED; + siw_cep_put(cep); + } + sock_release(s); + return rv; +} + + +/* + * siw_create_listen - Create resources for a listener's IWCM ID @id + * + * Listens on the socket addresses id->local_addr and id->remote_addr. + * We support listening on multi-homed devices, i.e., Softiwarp devices + * whose underlying net_device is associated with multiple IP addresses. + * Wildcard listening (listening with zero IP address) is also supported. + * + * There are three design options for Softiwarp device management supporting + * - multiple physical Ethernet ports, i.e., multiple net_device instances, and + * - multiple IP addresses associated with net_device, + * as follows: + * + * Option 1: One Softiwarp device per net_device and + * IP address associated with the net_device + * Option 2: One Softiwarp device per net_device + * (and all IP addresses associated with the net_device) + * Option 3: Single Softiwarp device for all net_device instances + * (and all IP addresses associated with these instances) + * + * We currently use Option 2, registering a separate siw_dev for + * each net_device. Consequently, siw_create_listen() (called separately + * by the IWCM for each Softiwarp device) handles the associated IP address(es) + * as follows: + * + * - If the listener's @id provides a specific local IP address, at most one + * listening socket is created and associated with @id. + * + * - If the listener's @id provides the wildcard (zero) local IP address, + * a separate listen is performed for each local IP address of the device + * by creating a listening socket and binding to that local IP address. + * This avoids attempts to bind to the wildcard (zero) IP address + * on multiple devices, which fails with -EADDRINUSE on the second and + * all subsequent devices. + * + * For the given IWCM and Option 2 above, the alternative approach of doing + * a single wildcard listen by creating one listening socket and binding it + * to the wildcard IP address is not a good idea if + * - there is more than one Softiwarp device (e.g., for lo and eth0), or + * - there are non-Softiwarp iWARP devices that cannot cooperate. + */ +int siw_create_listen(struct iw_cm_id *id, int backlog) +{ + struct ib_device *ofa_dev = id->device; + struct siw_dev *dev = siw_dev_ofa2siw(ofa_dev); + int rv = 0; + + dprint(DBG_CM, "(id=0x%p): dev(id)=%s, l2dev=%s backlog=%d\n", + id, ofa_dev->name, dev->l2dev->name, backlog); + + /* + * IPv4/v6 design differences regarding multi-homing + * propagate up to iWARP: + * o For IPv4, use dev->l2dev->ip_ptr + * o For IPv6, use dev->l2dev->ipv6_ptr + */ + if (id->local_addr.sin_family == AF_INET) { + /* IPv4 */ + struct sockaddr_in laddr = id->local_addr; + u8 *l_ip, *r_ip; + struct in_device *in_dev; + + l_ip = (u8 *) &id->local_addr.sin_addr.s_addr; + r_ip = (u8 *) &id->remote_addr.sin_addr.s_addr; + dprint(DBG_CM, "(id=0x%p): " + "laddr(id) : ipv4=%d.%d.%d.%d, port=%d; " + "raddr(id) : ipv4=%d.%d.%d.%d, port=%d\n", + id, + l_ip[0], l_ip[1], l_ip[2], l_ip[3], + ntohs(id->local_addr.sin_port), + r_ip[0], r_ip[1], r_ip[2], r_ip[3], + ntohs(id->remote_addr.sin_port)); + + in_dev = in_dev_get(dev->l2dev); + if (!in_dev) { + dprint(DBG_CM|DBG_ON, "(id=0x%p): " + "l2dev has no in_device\n", id); + return -ENODEV; + } + + /* + * If in_dev is not configured, in_dev->ifa_list may be empty + */ + for_ifa(in_dev) { + /* + * Create a listening socket if id->local_addr + * contains the wildcard IP address OR + * the IP address of the interface. + */ + if (ipv4_is_zeronet(id->local_addr.sin_addr.s_addr) || + id->local_addr.sin_addr.s_addr == + ifa->ifa_address) { + laddr.sin_addr.s_addr = ifa->ifa_address; + + l_ip = (u8 *) &laddr.sin_addr.s_addr; + dprint(DBG_CM, "(id=0x%p): " + "laddr(bind): ipv4=%d.%d.%d.%d," + " port=%d\n", id, + l_ip[0], l_ip[1], l_ip[2], + l_ip[3], ntohs(laddr.sin_port)); + + rv = siw_listen_address(id, backlog, + (struct sockaddr *)&laddr); + if (rv) + break; + } + } + endfor_ifa(in_dev); + in_dev_put(in_dev); + + if (rv) { + /* + * TODO: Cleanup resources already associated with + * id->provider_data + */ + dprint(DBG_CM|DBG_ON, "(id=0x%p): " + "TODO: Cleanup resources\n", id); + } + + } else { + /* IPv6 */ + dprint(DBG_CM|DBG_ON, "(id=0x%p): TODO: IPv6 support\n", id); + } + if (!rv) + dprint(DBG_CM, "(id=0x%p): Success\n", id); + + return rv; +} + + +int siw_destroy_listen(struct iw_cm_id *id) +{ + struct list_head *p, *tmp; + struct siw_cep *cep; + + dprint(DBG_CM, "(id=0x%p): dev(id)=%s, l2dev=%s\n", + id, id->device->name, + siw_dev_ofa2siw(id->device)->l2dev->name); + + if (!id->provider_data) { + /* + * TODO: See if there's a way to avoid getting any + * listener ids without a list of CEPs + */ + dprint(DBG_CM, "(id=0x%p): Listener id: no CEP(s)\n", id); + return 0; + } + + /* + * In case of a wildcard rdma_listen on a multi-homed device, + * a listener's IWCM id is associated with more than one listening CEP. + */ + list_for_each_safe(p, tmp, (struct list_head *)id->provider_data) { + + cep = list_entry(p, struct siw_cep, list); + list_del(p); + + if (siw_cep_set_inuse(cep) > 0) { + + cep->conn_close = 1; + + siw_socket_disassoc(cep->llp.sock); + sock_release(cep->llp.sock); + cep->llp.sock = NULL; + id->rem_ref(id); + + cep->state = SIW_EPSTATE_CLOSED; + /* + * Do not set the CEP free again. The CEP is dead. + * (void) siw_cep_set_free(cep); + */ + } else + cep->state = SIW_EPSTATE_CLOSED; + + siw_cep_put(cep); + } + kfree(id->provider_data); + id->provider_data = NULL; + + return 0; +} + +int __init siw_cm_init(void) +{ + /* + * create_single_workqueue for strict ordering + */ + siw_cm_wq = create_singlethread_workqueue("siw_cm_wq"); + if (!siw_cm_wq) + return -ENOMEM; + + return 0; +} + +void __exit siw_cm_exit(void) +{ + if (siw_cm_wq) { + flush_workqueue(siw_cm_wq); + destroy_workqueue(siw_cm_wq); + } +} diff --git a/drivers/infiniband/hw/siw/siw_cm.h b/drivers/infiniband/hw/siw/siw_cm.h new file mode 100644 index 0000000..fcb4544 --- /dev/null +++ b/drivers/infiniband/hw/siw/siw_cm.h @@ -0,0 +1,155 @@ +/* + * Software iWARP device driver for Linux + * + * Authors: Bernard Metzler + * + * Copyright (c) 2008-2010, IBM Corporation + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * - Neither the name of IBM nor the names of its contributors may be + * used to endorse or promote products derived from this software without + * specific prior written permission. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _SIW_CM_H +#define _SIW_CM_H + +#include +#include + +#include + + +enum siw_cep_state { + SIW_EPSTATE_IDLE = 1, + SIW_EPSTATE_LISTENING, + SIW_EPSTATE_CONNECTING, + SIW_EPSTATE_AWAIT_MPAREQ, + SIW_EPSTATE_RECVD_MPAREQ, + SIW_EPSTATE_AWAIT_MPAREP, + SIW_EPSTATE_RDMA_MODE, + SIW_EPSTATE_CLOSED +}; + +struct siw_mpa_info { + struct mpa_rr hdr; /* peer mpa hdr in host byte order */ + char *pdata; + int bytes_rcvd; +}; + +struct siw_llp_info { + struct socket *sock; + struct sockaddr_in laddr; /* redundant with socket info above */ + struct sockaddr_in raddr; /* dito, consider removal */ + struct siw_sk_upcalls sk_def_upcalls; +}; + +struct siw_cep { + struct iw_cm_id *cm_id; + + /* + * The provider_data element of a listener IWCM ID + * refers to a list of one or more listener CEPs + */ + struct list_head list; + + struct siw_cep *listen_cep; + struct siw_qp *qp; + spinlock_t lock; + wait_queue_head_t waitq; + struct kref ref; + enum siw_cep_state state; + short conn_close; /* sched. for closing or closed */ + short in_use; + struct siw_cm_work *mpa_timer; + struct list_head work_freelist; + struct siw_llp_info llp; + struct siw_mpa_info mpa; + int ord; + int ird; + int sk_error; /* not (yet) used XXX */ + + /* Saved upcalls of socket llp.sock */ + void (*sk_state_change)(struct sock *sk); + void (*sk_data_ready)(struct sock *sk, int bytes); + void (*sk_write_space)(struct sock *sk); + void (*sk_error_report)(struct sock *sk); +}; + +enum siw_work_type { + SIW_CM_WORK_ACCEPT = 1, + SIW_CM_WORK_READ_MPAHDR, + SIW_CM_WORK_CLOSE_LLP, /* close socket */ + SIW_CM_WORK_PEER_CLOSE, /* socket indicated peer close */ + SIW_CM_WORK_MPATIMEOUT /* to be done ! */ +}; + +struct siw_cm_work { + struct work_struct work; + struct list_head list; + enum siw_work_type type; + struct siw_cep *cep; +}; + +extern int siw_connect(struct iw_cm_id *, struct iw_cm_conn_param *); +extern int siw_accept(struct iw_cm_id *, struct iw_cm_conn_param *); +extern int siw_reject(struct iw_cm_id *, const void *, u8); +extern int siw_create_listen(struct iw_cm_id *, int); +extern int siw_destroy_listen(struct iw_cm_id *); + +extern int siw_cm_upcall(struct siw_cep *, enum iw_cm_event_type, + enum iw_cm_event_status); + +extern void siw_cep_upcall(struct siw_cep *, enum iw_cm_event_type); + +extern void siw_cep_put(struct siw_cep *); +extern void siw_cep_get(struct siw_cep *); +extern int siw_cep_in_close(struct siw_cep *); + +extern int siw_cm_queue_work(struct siw_cep *, enum siw_work_type); + +extern int siw_cm_init(void); +extern void siw_cm_exit(void); + +/* + * TCP socket interface + */ +#define sk_to_qp(sk) (((struct siw_cep *)((sk)->sk_user_data))->qp) +#define sk_to_cep(sk) ((struct siw_cep *)((sk)->sk_user_data)) + +/* + * Should we use tcp_current_mss()? + * But its not exported by kernel. + */ +static inline unsigned int get_tcp_mss(struct sock *sk) +{ + return ((struct tcp_sock *)sk)->xmit_size_goal_segs * + ((struct tcp_sock *)sk)->mss_cache; +} + +#endif