cma.c

linux 内核源代码

out:
	cma_enable_remove(id_priv);
	cma_deref_id(id_priv);
	if (destroy)
		rdma_destroy_id(&id_priv->id);
	kfree(work);
}

static int cma_resolve_ib_route(struct rdma_id_private *id_priv, int timeout_ms)
{
	struct rdma_route *route = &id_priv->id.route;
	struct cma_work *work;
	int ret;

	work = kzalloc(sizeof *work, GFP_KERNEL);
	if (!work)
		return -ENOMEM;

	work->id = id_priv;
	INIT_WORK(&work->work, cma_work_handler);
	work->old_state = CMA_ROUTE_QUERY;
	work->new_state = CMA_ROUTE_RESOLVED;
	work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED;

	route->path_rec = kmalloc(sizeof *route->path_rec, GFP_KERNEL);
	if (!route->path_rec) {
		ret = -ENOMEM;
		goto err1;
	}

	ret = cma_query_ib_route(id_priv, timeout_ms, work);
	if (ret)
		goto err2;

	return 0;
err2:
	kfree(route->path_rec);
	route->path_rec = NULL;
err1:
	kfree(work);
	return ret;
}

int rdma_set_ib_paths(struct rdma_cm_id *id,
		      struct ib_sa_path_rec *path_rec, int num_paths)
{
	struct rdma_id_private *id_priv;
	int ret;

	id_priv = container_of(id, struct rdma_id_private, id);
	if (!cma_comp_exch(id_priv, CMA_ADDR_RESOLVED, CMA_ROUTE_RESOLVED))
		return -EINVAL;

	id->route.path_rec = kmalloc(sizeof *path_rec * num_paths, GFP_KERNEL);
	if (!id->route.path_rec) {
		ret = -ENOMEM;
		goto err;
	}

	memcpy(id->route.path_rec, path_rec, sizeof *path_rec * num_paths);
	return 0;
err:
	cma_comp_exch(id_priv, CMA_ROUTE_RESOLVED, CMA_ADDR_RESOLVED);
	return ret;
}
EXPORT_SYMBOL(rdma_set_ib_paths);

static int cma_resolve_iw_route(struct rdma_id_private *id_priv, int timeout_ms)
{
	struct cma_work *work;

	work = kzalloc(sizeof *work, GFP_KERNEL);
	if (!work)
		return -ENOMEM;

	work->id = id_priv;
	INIT_WORK(&work->work, cma_work_handler);
	work->old_state = CMA_ROUTE_QUERY;
	work->new_state = CMA_ROUTE_RESOLVED;
	work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED;
	queue_work(cma_wq, &work->work);
	return 0;
}

int rdma_resolve_route(struct rdma_cm_id *id, int timeout_ms)
{
	struct rdma_id_private *id_priv;
	int ret;

	id_priv = container_of(id, struct rdma_id_private, id);
	if (!cma_comp_exch(id_priv, CMA_ADDR_RESOLVED, CMA_ROUTE_QUERY))
		return -EINVAL;

	atomic_inc(&id_priv->refcount);
	switch (rdma_node_get_transport(id->device->node_type)) {
	case RDMA_TRANSPORT_IB:
		ret = cma_resolve_ib_route(id_priv, timeout_ms);
		break;
	case RDMA_TRANSPORT_IWARP:
		ret = cma_resolve_iw_route(id_priv, timeout_ms);
		break;
	default:
		ret = -ENOSYS;
		break;
	}
	if (ret)
		goto err;

	return 0;
err:
	cma_comp_exch(id_priv, CMA_ROUTE_QUERY, CMA_ADDR_RESOLVED);
	cma_deref_id(id_priv);
	return ret;
}
EXPORT_SYMBOL(rdma_resolve_route);

static int cma_bind_loopback(struct rdma_id_private *id_priv)
{
	struct cma_device *cma_dev;
	struct ib_port_attr port_attr;
	union ib_gid gid;
	u16 pkey;
	int ret;
	u8 p;

	mutex_lock(&lock);
	if (list_empty(&dev_list)) {
		ret = -ENODEV;
		goto out;
	}
	list_for_each_entry(cma_dev, &dev_list, list)
		for (p = 1; p <= cma_dev->device->phys_port_cnt; ++p)
			if (!ib_query_port(cma_dev->device, p, &port_attr) &&
			    port_attr.state == IB_PORT_ACTIVE)
				goto port_found;

	p = 1;
	cma_dev = list_entry(dev_list.next, struct cma_device, list);

port_found:
	ret = ib_get_cached_gid(cma_dev->device, p, 0, &gid);
	if (ret)
		goto out;

	ret = ib_get_cached_pkey(cma_dev->device, p, 0, &pkey);
	if (ret)
		goto out;

	ib_addr_set_sgid(&id_priv->id.route.addr.dev_addr, &gid);
	ib_addr_set_pkey(&id_priv->id.route.addr.dev_addr, pkey);
	id_priv->id.port_num = p;
	cma_attach_to_dev(id_priv, cma_dev);
out:
	mutex_unlock(&lock);
	return ret;
}

static void addr_handler(int status, struct sockaddr *src_addr,
			 struct rdma_dev_addr *dev_addr, void *context)
{
	struct rdma_id_private *id_priv = context;
	struct rdma_cm_event event;

	memset(&event, 0, sizeof event);
	atomic_inc(&id_priv->dev_remove);

	/*
	 * Grab mutex to block rdma_destroy_id() from removing the device while
	 * we're trying to acquire it.
	 */
	mutex_lock(&lock);
	if (!cma_comp_exch(id_priv, CMA_ADDR_QUERY, CMA_ADDR_RESOLVED)) {
		mutex_unlock(&lock);
		goto out;
	}

	if (!status && !id_priv->cma_dev)
		status = cma_acquire_dev(id_priv);
	mutex_unlock(&lock);

	if (status) {
		if (!cma_comp_exch(id_priv, CMA_ADDR_RESOLVED, CMA_ADDR_BOUND))
			goto out;
		event.event = RDMA_CM_EVENT_ADDR_ERROR;
		event.status = status;
	} else {
		memcpy(&id_priv->id.route.addr.src_addr, src_addr,
		       ip_addr_size(src_addr));
		event.event = RDMA_CM_EVENT_ADDR_RESOLVED;
	}

	if (id_priv->id.event_handler(&id_priv->id, &event)) {
		cma_exch(id_priv, CMA_DESTROYING);
		cma_enable_remove(id_priv);
		cma_deref_id(id_priv);
		rdma_destroy_id(&id_priv->id);
		return;
	}
out:
	cma_enable_remove(id_priv);
	cma_deref_id(id_priv);
}

static int cma_resolve_loopback(struct rdma_id_private *id_priv)
{
	struct cma_work *work;
	struct sockaddr_in *src_in, *dst_in;
	union ib_gid gid;
	int ret;

	work = kzalloc(sizeof *work, GFP_KERNEL);
	if (!work)
		return -ENOMEM;

	if (!id_priv->cma_dev) {
		ret = cma_bind_loopback(id_priv);
		if (ret)
			goto err;
	}

	ib_addr_get_sgid(&id_priv->id.route.addr.dev_addr, &gid);
	ib_addr_set_dgid(&id_priv->id.route.addr.dev_addr, &gid);

	if (cma_zero_addr(&id_priv->id.route.addr.src_addr)) {
		src_in = (struct sockaddr_in *)&id_priv->id.route.addr.src_addr;
		dst_in = (struct sockaddr_in *)&id_priv->id.route.addr.dst_addr;
		src_in->sin_family = dst_in->sin_family;
		src_in->sin_addr.s_addr = dst_in->sin_addr.s_addr;
	}

	work->id = id_priv;
	INIT_WORK(&work->work, cma_work_handler);
	work->old_state = CMA_ADDR_QUERY;
	work->new_state = CMA_ADDR_RESOLVED;
	work->event.event = RDMA_CM_EVENT_ADDR_RESOLVED;
	queue_work(cma_wq, &work->work);
	return 0;
err:
	kfree(work);
	return ret;
}

static int cma_bind_addr(struct rdma_cm_id *id, struct sockaddr *src_addr,
			 struct sockaddr *dst_addr)
{
	if (src_addr && src_addr->sa_family)
		return rdma_bind_addr(id, src_addr);
	else
		return cma_bind_any(id, dst_addr->sa_family);
}

int rdma_resolve_addr(struct rdma_cm_id *id, struct sockaddr *src_addr,
		      struct sockaddr *dst_addr, int timeout_ms)
{
	struct rdma_id_private *id_priv;
	int ret;

	id_priv = container_of(id, struct rdma_id_private, id);
	if (id_priv->state == CMA_IDLE) {
		ret = cma_bind_addr(id, src_addr, dst_addr);
		if (ret)
			return ret;
	}

	if (!cma_comp_exch(id_priv, CMA_ADDR_BOUND, CMA_ADDR_QUERY))
		return -EINVAL;

	atomic_inc(&id_priv->refcount);
	memcpy(&id->route.addr.dst_addr, dst_addr, ip_addr_size(dst_addr));
	if (cma_any_addr(dst_addr))
		ret = cma_resolve_loopback(id_priv);
	else
		ret = rdma_resolve_ip(&addr_client, &id->route.addr.src_addr,
				      dst_addr, &id->route.addr.dev_addr,
				      timeout_ms, addr_handler, id_priv);
	if (ret)
		goto err;

	return 0;
err:
	cma_comp_exch(id_priv, CMA_ADDR_QUERY, CMA_ADDR_BOUND);
	cma_deref_id(id_priv);
	return ret;
}
EXPORT_SYMBOL(rdma_resolve_addr);

static void cma_bind_port(struct rdma_bind_list *bind_list,
			  struct rdma_id_private *id_priv)
{
	struct sockaddr_in *sin;

	sin = (struct sockaddr_in *) &id_priv->id.route.addr.src_addr;
	sin->sin_port = htons(bind_list->port);
	id_priv->bind_list = bind_list;
	hlist_add_head(&id_priv->node, &bind_list->owners);
}

static int cma_alloc_port(struct idr *ps, struct rdma_id_private *id_priv,
			  unsigned short snum)
{
	struct rdma_bind_list *bind_list;
	int port, ret;

	bind_list = kzalloc(sizeof *bind_list, GFP_KERNEL);
	if (!bind_list)
		return -ENOMEM;

	do {
		ret = idr_get_new_above(ps, bind_list, snum, &port);
	} while ((ret == -EAGAIN) && idr_pre_get(ps, GFP_KERNEL));

	if (ret)
		goto err1;

	if (port != snum) {
		ret = -EADDRNOTAVAIL;
		goto err2;
	}

	bind_list->ps = ps;
	bind_list->port = (unsigned short) port;
	cma_bind_port(bind_list, id_priv);
	return 0;
err2:
	idr_remove(ps, port);
err1:
	kfree(bind_list);
	return ret;
}

static int cma_alloc_any_port(struct idr *ps, struct rdma_id_private *id_priv)
{
	struct rdma_bind_list *bind_list;
	int port, ret, low, high;

	bind_list = kzalloc(sizeof *bind_list, GFP_KERNEL);
	if (!bind_list)
		return -ENOMEM;

retry:
	/* FIXME: add proper port randomization per like inet_csk_get_port */
	do {
		ret = idr_get_new_above(ps, bind_list, next_port, &port);
	} while ((ret == -EAGAIN) && idr_pre_get(ps, GFP_KERNEL));

	if (ret)
		goto err1;

	inet_get_local_port_range(&low, &high);
	if (port > high) {
		if (next_port != low) {
			idr_remove(ps, port);
			next_port = low;
			goto retry;
		}
		ret = -EADDRNOTAVAIL;
		goto err2;
	}

	if (port == high)
		next_port = low;
	else
		next_port = port + 1;

	bind_list->ps = ps;
	bind_list->port = (unsigned short) port;
	cma_bind_port(bind_list, id_priv);
	return 0;
err2:
	idr_remove(ps, port);
err1:
	kfree(bind_list);
	return ret;
}

static int cma_use_port(struct idr *ps, struct rdma_id_private *id_priv)
{
	struct rdma_id_private *cur_id;
	struct sockaddr_in *sin, *cur_sin;
	struct rdma_bind_list *bind_list;
	struct hlist_node *node;
	unsigned short snum;

	sin = (struct sockaddr_in *) &id_priv->id.route.addr.src_addr;
	snum = ntohs(sin->sin_port);
	if (snum < PROT_SOCK && !capable(CAP_NET_BIND_SERVICE))
		return -EACCES;

	bind_list = idr_find(ps, snum);
	if (!bind_list)
		return cma_alloc_port(ps, id_priv, snum);

	/*
	 * We don't support binding to any address if anyone is bound to
	 * a specific address on the same port.
	 */
	if (cma_any_addr(&id_priv->id.route.addr.src_addr))
		return -EADDRNOTAVAIL;

	hlist_for_each_entry(cur_id, node, &bind_list->owners, node) {
		if (cma_any_addr(&cur_id->id.route.addr.src_addr))
			return -EADDRNOTAVAIL;

		cur_sin = (struct sockaddr_in *) &cur_id->id.route.addr.src_addr;
		if (sin->sin_addr.s_addr == cur_sin->sin_addr.s_addr)
			return -EADDRINUSE;
	}

	cma_bind_port(bind_list, id_priv);
	return 0;
}

static int cma_get_port(struct rdma_id_private *id_priv)
{
	struct idr *ps;
	int ret;

	switch (id_priv->id.ps) {
	case RDMA_PS_SDP:
		ps = &sdp_ps;
		break;
	case RDMA_PS_TCP:
		ps = &tcp_ps;
		break;
	case RDMA_PS_UDP:
		ps = &udp_ps;
		break;
	case RDMA_PS_IPOIB:
		ps = &ipoib_ps;
		break;
	default:
		return -EPROTONOSUPPORT;
	}

	mutex_lock(&lock);
	if (cma_any_port(&id_priv->id.route.addr.src_addr))
		ret = cma_alloc_any_port(ps, id_priv);
	else
		ret = cma_use_port(ps, id_priv);
	mutex_unlock(&lock);

	return ret;
}

int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr)
{
	struct rdma_id_private *id_priv;
	int ret;

	if (addr->sa_family != AF_INET)
		return -EAFNOSUPPORT;

	id_priv = container_of(id, struct rdma_id_private, id);
	if (!cma_comp_exch(id_priv, CMA_IDLE, CMA_ADDR_BOUND))
		return -EINVAL;

	if (!cma_any_addr(addr)) {
		ret = rdma_translate_ip(addr, &id->route.addr.dev_addr);
		if (ret)
			goto err1;

		mutex_lock(&lock);
		ret = cma_acquire_dev(id_priv);
		mutex_unlock(&lock);
		if (ret)
			goto err1;
	}

	memcpy(&id->route.addr.src_addr, addr, ip_addr_size(addr));
	ret = cma_get_port(id_priv);
	if (ret)
		goto err2;

	return 0;
err2:
	if (!cma_any_addr(addr)) {
		mutex_lock(&lock);
		cma_detach_from_dev(id_priv);
		mutex_unlock(&lock);
	}
err1:
	cma_comp_exch(id_priv, CMA_ADDR_BOUND, CMA_IDLE);
	return ret;
}
EXPORT_SYMBOL(rdma_bind_addr);

static int cma_format_hdr(void *hdr, enum rdma_port_space ps,
			  struct rdma_route *route)
{
	struct sockaddr_in *src4, *dst4;
	struct cma_hdr *cma_hdr;
	struct sdp_hh *sdp_hdr;

	src4 = (struct sockaddr_in *) &route->addr.src_addr;
	dst4 = (struct sockaddr_in *) &route->addr.dst_addr;

	switch (ps) {
	case RDMA_PS_SDP:
		sdp_hdr = hdr;
		if (sdp_get_majv(sdp_hdr->sdp_version) != SDP_MAJ_VERSION)
			return -EINVAL;
		sdp_set_ip_ver(sdp_hdr, 4);
		sdp_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
		sdp_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
		sdp_hdr->port = src4->sin_port;
		break;
	default:
		cma_hdr = hdr;
		cma_hdr->cma_version = CMA_VERSION;
		cma_set_ip_ver(cma_hdr, 4);
		cma_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
		cma_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
		cma_hdr->port = src4->sin_port;
		break;
	}
	return 0;
}

static int cma_sidr_rep_handler(struct ib_cm_id *cm_id,
				struct ib_cm_event *ib_event)
{
	struct rdma_id_private *id_priv = cm_id->context;
	struct rdma_cm_event event;