/*
- * linux/net/iucv/af_iucv.c
- *
* IUCV protocol stack for Linux on zSeries
*
- * Copyright 2006 IBM Corporation
+ * Copyright IBM Corp. 2006, 2009
*
* Author(s): Jennifer Hunt <jenhunt@us.ibm.com>
+ * Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
+ * PM functions:
+ * Ursula Braun <ursula.braun@de.ibm.com>
*/
+#define KMSG_COMPONENT "af_iucv"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
#include <linux/module.h>
#include <linux/types.h>
#include <linux/list.h>
#include <net/iucv/iucv.h>
#include <net/iucv/af_iucv.h>
-#define CONFIG_IUCV_SOCK_DEBUG 1
-
-#define IPRMDATA 0x80
-#define VERSION "1.0"
+#define VERSION "1.1"
static char iucv_userid[80];
.obj_size = sizeof(struct iucv_sock),
};
+/* special AF_IUCV IPRM messages */
+static const u8 iprm_shutdown[8] =
+ {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01};
+
+#define TRGCLS_SIZE (sizeof(((struct iucv_message *)0)->class))
+
+/* macros to set/get socket control buffer at correct offset */
+#define CB_TAG(skb) ((skb)->cb) /* iucv message tag */
+#define CB_TAG_LEN (sizeof(((struct iucv_message *) 0)->tag))
+#define CB_TRGCLS(skb) ((skb)->cb + CB_TAG_LEN) /* iucv msg target class */
+#define CB_TRGCLS_LEN (TRGCLS_SIZE)
+
+#define __iucv_sock_wait(sk, condition, timeo, ret) \
+do { \
+ DEFINE_WAIT(__wait); \
+ long __timeo = timeo; \
+ ret = 0; \
+ while (!(condition)) { \
+ prepare_to_wait(sk->sk_sleep, &__wait, TASK_INTERRUPTIBLE); \
+ if (!__timeo) { \
+ ret = -EAGAIN; \
+ break; \
+ } \
+ if (signal_pending(current)) { \
+ ret = sock_intr_errno(__timeo); \
+ break; \
+ } \
+ release_sock(sk); \
+ __timeo = schedule_timeout(__timeo); \
+ lock_sock(sk); \
+ ret = sock_error(sk); \
+ if (ret) \
+ break; \
+ } \
+ finish_wait(sk->sk_sleep, &__wait); \
+} while (0)
+
+#define iucv_sock_wait(sk, condition, timeo) \
+({ \
+ int __ret = 0; \
+ if (!(condition)) \
+ __iucv_sock_wait(sk, condition, timeo, __ret); \
+ __ret; \
+})
+
static void iucv_sock_kill(struct sock *sk);
static void iucv_sock_close(struct sock *sk);
static int iucv_callback_connreq(struct iucv_path *, u8 ipvmid[8],
u8 ipuser[16]);
static void iucv_callback_connrej(struct iucv_path *, u8 ipuser[16]);
+static void iucv_callback_shutdown(struct iucv_path *, u8 ipuser[16]);
static struct iucv_sock_list iucv_sk_list = {
- .lock = RW_LOCK_UNLOCKED,
+ .lock = __RW_LOCK_UNLOCKED(iucv_sk_list.lock),
.autobind_name = ATOMIC_INIT(0)
};
.path_complete = iucv_callback_connack,
.path_severed = iucv_callback_connrej,
.message_pending = iucv_callback_rx,
- .message_complete = iucv_callback_txdone
+ .message_complete = iucv_callback_txdone,
+ .path_quiesced = iucv_callback_shutdown,
};
static inline void high_nmcpy(unsigned char *dst, char *src)
memcpy(&dst[8], src, 8);
}
+static int afiucv_pm_prepare(struct device *dev)
+{
+#ifdef CONFIG_PM_DEBUG
+ printk(KERN_WARNING "afiucv_pm_prepare\n");
+#endif
+ return 0;
+}
+
+static void afiucv_pm_complete(struct device *dev)
+{
+#ifdef CONFIG_PM_DEBUG
+ printk(KERN_WARNING "afiucv_pm_complete\n");
+#endif
+ return;
+}
+
+/**
+ * afiucv_pm_freeze() - Freeze PM callback
+ * @dev: AFIUCV dummy device
+ *
+ * Sever all established IUCV communication pathes
+ */
+static int afiucv_pm_freeze(struct device *dev)
+{
+ struct iucv_sock *iucv;
+ struct sock *sk;
+ struct hlist_node *node;
+ int err = 0;
+
+#ifdef CONFIG_PM_DEBUG
+ printk(KERN_WARNING "afiucv_pm_freeze\n");
+#endif
+ read_lock(&iucv_sk_list.lock);
+ sk_for_each(sk, node, &iucv_sk_list.head) {
+ iucv = iucv_sk(sk);
+ skb_queue_purge(&iucv->send_skb_q);
+ skb_queue_purge(&iucv->backlog_skb_q);
+ switch (sk->sk_state) {
+ case IUCV_SEVERED:
+ case IUCV_DISCONN:
+ case IUCV_CLOSING:
+ case IUCV_CONNECTED:
+ if (iucv->path) {
+ err = iucv_path_sever(iucv->path, NULL);
+ iucv_path_free(iucv->path);
+ iucv->path = NULL;
+ }
+ break;
+ case IUCV_OPEN:
+ case IUCV_BOUND:
+ case IUCV_LISTEN:
+ case IUCV_CLOSED:
+ default:
+ break;
+ }
+ }
+ read_unlock(&iucv_sk_list.lock);
+ return err;
+}
+
+/**
+ * afiucv_pm_restore_thaw() - Thaw and restore PM callback
+ * @dev: AFIUCV dummy device
+ *
+ * socket clean up after freeze
+ */
+static int afiucv_pm_restore_thaw(struct device *dev)
+{
+ struct iucv_sock *iucv;
+ struct sock *sk;
+ struct hlist_node *node;
+
+#ifdef CONFIG_PM_DEBUG
+ printk(KERN_WARNING "afiucv_pm_restore_thaw\n");
+#endif
+ read_lock(&iucv_sk_list.lock);
+ sk_for_each(sk, node, &iucv_sk_list.head) {
+ iucv = iucv_sk(sk);
+ switch (sk->sk_state) {
+ case IUCV_CONNECTED:
+ sk->sk_err = EPIPE;
+ sk->sk_state = IUCV_DISCONN;
+ sk->sk_state_change(sk);
+ break;
+ case IUCV_DISCONN:
+ case IUCV_SEVERED:
+ case IUCV_CLOSING:
+ case IUCV_LISTEN:
+ case IUCV_BOUND:
+ case IUCV_OPEN:
+ default:
+ break;
+ }
+ }
+ read_unlock(&iucv_sk_list.lock);
+ return 0;
+}
+
+static struct dev_pm_ops afiucv_pm_ops = {
+ .prepare = afiucv_pm_prepare,
+ .complete = afiucv_pm_complete,
+ .freeze = afiucv_pm_freeze,
+ .thaw = afiucv_pm_restore_thaw,
+ .restore = afiucv_pm_restore_thaw,
+};
+
+static struct device_driver af_iucv_driver = {
+ .owner = THIS_MODULE,
+ .name = "afiucv",
+ .bus = &iucv_bus,
+ .pm = &afiucv_pm_ops,
+};
+
+/* dummy device used as trigger for PM functions */
+static struct device *af_iucv_dev;
+
+/**
+ * iucv_msg_length() - Returns the length of an iucv message.
+ * @msg: Pointer to struct iucv_message, MUST NOT be NULL
+ *
+ * The function returns the length of the specified iucv message @msg of data
+ * stored in a buffer and of data stored in the parameter list (PRMDATA).
+ *
+ * For IUCV_IPRMDATA, AF_IUCV uses the following convention to transport socket
+ * data:
+ * PRMDATA[0..6] socket data (max 7 bytes);
+ * PRMDATA[7] socket data length value (len is 0xff - PRMDATA[7])
+ *
+ * The socket data length is computed by substracting the socket data length
+ * value from 0xFF.
+ * If the socket data len is greater 7, then PRMDATA can be used for special
+ * notifications (see iucv_sock_shutdown); and further,
+ * if the socket data len is > 7, the function returns 8.
+ *
+ * Use this function to allocate socket buffers to store iucv message data.
+ */
+static inline size_t iucv_msg_length(struct iucv_message *msg)
+{
+ size_t datalen;
+
+ if (msg->flags & IUCV_IPRMDATA) {
+ datalen = 0xff - msg->rmmsg[7];
+ return (datalen < 8) ? datalen : 8;
+ }
+ return msg->length;
+}
+
+/**
+ * iucv_sock_in_state() - check for specific states
+ * @sk: sock structure
+ * @state: first iucv sk state
+ * @state: second iucv sk state
+ *
+ * Returns true if the socket in either in the first or second state.
+ */
+static int iucv_sock_in_state(struct sock *sk, int state, int state2)
+{
+ return (sk->sk_state == state || sk->sk_state == state2);
+}
+
+/**
+ * iucv_below_msglim() - function to check if messages can be sent
+ * @sk: sock structure
+ *
+ * Returns true if the send queue length is lower than the message limit.
+ * Always returns true if the socket is not connected (no iucv path for
+ * checking the message limit).
+ */
+static inline int iucv_below_msglim(struct sock *sk)
+{
+ struct iucv_sock *iucv = iucv_sk(sk);
+
+ if (sk->sk_state != IUCV_CONNECTED)
+ return 1;
+ return (skb_queue_len(&iucv->send_skb_q) < iucv->path->msglim);
+}
+
+/**
+ * iucv_sock_wake_msglim() - Wake up thread waiting on msg limit
+ */
+static void iucv_sock_wake_msglim(struct sock *sk)
+{
+ read_lock(&sk->sk_callback_lock);
+ if (sk_has_sleeper(sk))
+ wake_up_interruptible_all(sk->sk_sleep);
+ sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
+ read_unlock(&sk->sk_callback_lock);
+}
+
/* Timers */
static void iucv_sock_timeout(unsigned long arg)
{
sk_stop_timer(sk, &sk->sk_timer);
}
-static void iucv_sock_init_timer(struct sock *sk)
-{
- init_timer(&sk->sk_timer);
- sk->sk_timer.function = iucv_sock_timeout;
- sk->sk_timer.data = (unsigned long)sk;
-}
-
static struct sock *__iucv_get_sock_by_name(char *nm)
{
struct sock *sk;
timeo = sk->sk_lingertime;
else
timeo = IUCV_DISCONN_TIMEOUT;
- err = iucv_sock_wait_state(sk, IUCV_CLOSED, 0, timeo);
+ err = iucv_sock_wait(sk,
+ iucv_sock_in_state(sk, IUCV_CLOSED, 0),
+ timeo);
}
+ case IUCV_CLOSING: /* fall through */
sk->sk_state = IUCV_CLOSED;
sk->sk_state_change(sk);
{
struct sock *sk;
- sk = sk_alloc(&init_net, PF_IUCV, prio, &iucv_proto, 1);
+ sk = sk_alloc(&init_net, PF_IUCV, prio, &iucv_proto);
if (!sk)
return NULL;
spin_lock_init(&iucv_sk(sk)->message_q.lock);
skb_queue_head_init(&iucv_sk(sk)->backlog_skb_q);
iucv_sk(sk)->send_tag = 0;
+ iucv_sk(sk)->flags = 0;
+ iucv_sk(sk)->msglimit = IUCV_QUEUELEN_DEFAULT;
+ iucv_sk(sk)->path = NULL;
+ memset(&iucv_sk(sk)->src_user_id , 0, 32);
sk->sk_destruct = iucv_sock_destruct;
sk->sk_sndtimeo = IUCV_CONN_TIMEOUT;
sk->sk_protocol = proto;
sk->sk_state = IUCV_OPEN;
- iucv_sock_init_timer(sk);
+ setup_timer(&sk->sk_timer, iucv_sock_timeout, (unsigned long)sk);
iucv_sock_link(&iucv_sk_list, sk);
return sk;
{
struct sock *sk;
- if (sock->type != SOCK_STREAM)
- return -ESOCKTNOSUPPORT;
+ if (protocol && protocol != PF_IUCV)
+ return -EPROTONOSUPPORT;
sock->state = SS_UNCONNECTED;
- sock->ops = &iucv_sock_ops;
+
+ switch (sock->type) {
+ case SOCK_STREAM:
+ sock->ops = &iucv_sock_ops;
+ break;
+ case SOCK_SEQPACKET:
+ /* currently, proto ops can handle both sk types */
+ sock->ops = &iucv_sock_ops;
+ break;
+ default:
+ return -ESOCKTNOSUPPORT;
+ }
sk = iucv_sock_alloc(sock, protocol, GFP_KERNEL);
if (!sk)
return NULL;
}
-int iucv_sock_wait_state(struct sock *sk, int state, int state2,
- unsigned long timeo)
-{
- DECLARE_WAITQUEUE(wait, current);
- int err = 0;
-
- add_wait_queue(sk->sk_sleep, &wait);
- while (sk->sk_state != state && sk->sk_state != state2) {
- set_current_state(TASK_INTERRUPTIBLE);
-
- if (!timeo) {
- err = -EAGAIN;
- break;
- }
-
- if (signal_pending(current)) {
- err = sock_intr_errno(timeo);
- break;
- }
-
- release_sock(sk);
- timeo = schedule_timeout(timeo);
- lock_sock(sk);
-
- err = sock_error(sk);
- if (err)
- break;
- }
- set_current_state(TASK_RUNNING);
- remove_wait_queue(sk->sk_sleep, &wait);
- return err;
-}
-
/* Bind an unbound socket */
static int iucv_sock_bind(struct socket *sock, struct sockaddr *addr,
int addr_len)
if (sk->sk_state != IUCV_OPEN && sk->sk_state != IUCV_BOUND)
return -EBADFD;
- if (sk->sk_type != SOCK_STREAM)
+ if (sk->sk_type != SOCK_STREAM && sk->sk_type != SOCK_SEQPACKET)
return -EINVAL;
- iucv = iucv_sk(sk);
-
if (sk->sk_state == IUCV_OPEN) {
err = iucv_sock_autobind(sk);
if (unlikely(err))
iucv = iucv_sk(sk);
/* Create path. */
- iucv->path = iucv_path_alloc(IUCV_QUEUELEN_DEFAULT,
- IPRMDATA, GFP_KERNEL);
+ iucv->path = iucv_path_alloc(iucv->msglimit,
+ IUCV_IPRMDATA, GFP_KERNEL);
+ if (!iucv->path) {
+ err = -ENOMEM;
+ goto done;
+ }
err = iucv_path_connect(iucv->path, &af_iucv_handler,
sa->siucv_user_id, NULL, user_data, sk);
if (err) {
iucv_path_free(iucv->path);
iucv->path = NULL;
- err = -ECONNREFUSED;
+ switch (err) {
+ case 0x0b: /* Target communicator is not logged on */
+ err = -ENETUNREACH;
+ break;
+ case 0x0d: /* Max connections for this guest exceeded */
+ case 0x0e: /* Max connections for target guest exceeded */
+ err = -EAGAIN;
+ break;
+ case 0x0f: /* Missing IUCV authorization */
+ err = -EACCES;
+ break;
+ default:
+ err = -ECONNREFUSED;
+ break;
+ }
goto done;
}
if (sk->sk_state != IUCV_CONNECTED) {
- err = iucv_sock_wait_state(sk, IUCV_CONNECTED, IUCV_DISCONN,
- sock_sndtimeo(sk, flags & O_NONBLOCK));
+ err = iucv_sock_wait(sk, iucv_sock_in_state(sk, IUCV_CONNECTED,
+ IUCV_DISCONN),
+ sock_sndtimeo(sk, flags & O_NONBLOCK));
}
if (sk->sk_state == IUCV_DISCONN) {
- release_sock(sk);
- return -ECONNREFUSED;
+ err = -ECONNREFUSED;
}
+
+ if (err) {
+ iucv_path_sever(iucv->path, NULL);
+ iucv_path_free(iucv->path);
+ iucv->path = NULL;
+ }
+
done:
release_sock(sk);
return err;
lock_sock(sk);
err = -EINVAL;
- if (sk->sk_state != IUCV_BOUND || sock->type != SOCK_STREAM)
+ if (sk->sk_state != IUCV_BOUND)
+ goto done;
+
+ if (sock->type != SOCK_STREAM && sock->type != SOCK_SEQPACKET)
goto done;
sk->sk_max_ack_backlog = backlog;
return 0;
}
+/**
+ * iucv_send_iprm() - Send socket data in parameter list of an iucv message.
+ * @path: IUCV path
+ * @msg: Pointer to a struct iucv_message
+ * @skb: The socket data to send, skb->len MUST BE <= 7
+ *
+ * Send the socket data in the parameter list in the iucv message
+ * (IUCV_IPRMDATA). The socket data is stored at index 0 to 6 in the parameter
+ * list and the socket data len at index 7 (last byte).
+ * See also iucv_msg_length().
+ *
+ * Returns the error code from the iucv_message_send() call.
+ */
+static int iucv_send_iprm(struct iucv_path *path, struct iucv_message *msg,
+ struct sk_buff *skb)
+{
+ u8 prmdata[8];
+
+ memcpy(prmdata, (void *) skb->data, skb->len);
+ prmdata[7] = 0xff - (u8) skb->len;
+ return iucv_message_send(path, msg, IUCV_IPRMDATA, 0,
+ (void *) prmdata, 8);
+}
+
static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
struct msghdr *msg, size_t len)
{
struct iucv_sock *iucv = iucv_sk(sk);
struct sk_buff *skb;
struct iucv_message txmsg;
+ struct cmsghdr *cmsg;
+ int cmsg_done;
+ long timeo;
+ char user_id[9];
+ char appl_id[9];
int err;
+ int noblock = msg->msg_flags & MSG_DONTWAIT;
err = sock_error(sk);
if (err)
if (msg->msg_flags & MSG_OOB)
return -EOPNOTSUPP;
+ /* SOCK_SEQPACKET: we do not support segmented records */
+ if (sk->sk_type == SOCK_SEQPACKET && !(msg->msg_flags & MSG_EOR))
+ return -EOPNOTSUPP;
+
lock_sock(sk);
if (sk->sk_shutdown & SEND_SHUTDOWN) {
goto out;
}
- if (sk->sk_state == IUCV_CONNECTED) {
- if (!(skb = sock_alloc_send_skb(sk, len,
- msg->msg_flags & MSG_DONTWAIT,
- &err)))
+ /* Return if the socket is not in connected state */
+ if (sk->sk_state != IUCV_CONNECTED) {
+ err = -ENOTCONN;
+ goto out;
+ }
+
+ /* initialize defaults */
+ cmsg_done = 0; /* check for duplicate headers */
+ txmsg.class = 0;
+
+ /* iterate over control messages */
+ for (cmsg = CMSG_FIRSTHDR(msg); cmsg;
+ cmsg = CMSG_NXTHDR(msg, cmsg)) {
+
+ if (!CMSG_OK(msg, cmsg)) {
+ err = -EINVAL;
goto out;
+ }
- if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
- err = -EFAULT;
- goto fail;
+ if (cmsg->cmsg_level != SOL_IUCV)
+ continue;
+
+ if (cmsg->cmsg_type & cmsg_done) {
+ err = -EINVAL;
+ goto out;
}
+ cmsg_done |= cmsg->cmsg_type;
- txmsg.class = 0;
- txmsg.tag = iucv->send_tag++;
- memcpy(skb->cb, &txmsg.tag, 4);
- skb_queue_tail(&iucv->send_skb_q, skb);
- err = iucv_message_send(iucv->path, &txmsg, 0, 0,
- (void *) skb->data, skb->len);
- if (err) {
- if (err == 3)
- printk(KERN_ERR "AF_IUCV msg limit exceeded\n");
+ switch (cmsg->cmsg_type) {
+ case SCM_IUCV_TRGCLS:
+ if (cmsg->cmsg_len != CMSG_LEN(TRGCLS_SIZE)) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ /* set iucv message target class */
+ memcpy(&txmsg.class,
+ (void *) CMSG_DATA(cmsg), TRGCLS_SIZE);
+
+ break;
+
+ default:
+ err = -EINVAL;
+ goto out;
+ break;
+ }
+ }
+
+ /* allocate one skb for each iucv message:
+ * this is fine for SOCK_SEQPACKET (unless we want to support
+ * segmented records using the MSG_EOR flag), but
+ * for SOCK_STREAM we might want to improve it in future */
+ skb = sock_alloc_send_skb(sk, len, noblock, &err);
+ if (!skb)
+ goto out;
+ if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
+ err = -EFAULT;
+ goto fail;
+ }
+
+ /* wait if outstanding messages for iucv path has reached */
+ timeo = sock_sndtimeo(sk, noblock);
+ err = iucv_sock_wait(sk, iucv_below_msglim(sk), timeo);
+ if (err)
+ goto fail;
+
+ /* return -ECONNRESET if the socket is no longer connected */
+ if (sk->sk_state != IUCV_CONNECTED) {
+ err = -ECONNRESET;
+ goto fail;
+ }
+
+ /* increment and save iucv message tag for msg_completion cbk */
+ txmsg.tag = iucv->send_tag++;
+ memcpy(CB_TAG(skb), &txmsg.tag, CB_TAG_LEN);
+ skb_queue_tail(&iucv->send_skb_q, skb);
+
+ if (((iucv->path->flags & IUCV_IPRMDATA) & iucv->flags)
+ && skb->len <= 7) {
+ err = iucv_send_iprm(iucv->path, &txmsg, skb);
+
+ /* on success: there is no message_complete callback
+ * for an IPRMDATA msg; remove skb from send queue */
+ if (err == 0) {
+ skb_unlink(skb, &iucv->send_skb_q);
+ kfree_skb(skb);
+ }
+
+ /* this error should never happen since the
+ * IUCV_IPRMDATA path flag is set... sever path */
+ if (err == 0x15) {
+ iucv_path_sever(iucv->path, NULL);
skb_unlink(skb, &iucv->send_skb_q);
err = -EPIPE;
goto fail;
}
-
- } else {
- err = -ENOTCONN;
- goto out;
+ } else
+ err = iucv_message_send(iucv->path, &txmsg, 0, 0,
+ (void *) skb->data, skb->len);
+ if (err) {
+ if (err == 3) {
+ user_id[8] = 0;
+ memcpy(user_id, iucv->dst_user_id, 8);
+ appl_id[8] = 0;
+ memcpy(appl_id, iucv->dst_name, 8);
+ pr_err("Application %s on z/VM guest %s"
+ " exceeds message limit\n",
+ appl_id, user_id);
+ err = -EAGAIN;
+ } else
+ err = -EPIPE;
+ skb_unlink(skb, &iucv->send_skb_q);
+ goto fail;
}
release_sock(sk);
if (!nskb)
return -ENOMEM;
+ /* copy target class to control buffer of new skb */
+ memcpy(CB_TRGCLS(nskb), CB_TRGCLS(skb), CB_TRGCLS_LEN);
+
+ /* copy data fragment */
memcpy(nskb->data, skb->data + copied, size);
copied += size;
dataleft -= size;
struct iucv_message *msg)
{
int rc;
+ unsigned int len;
+
+ len = iucv_msg_length(msg);
- if (msg->flags & IPRMDATA) {
- skb->data = NULL;
- skb->len = 0;
+ /* store msg target class in the second 4 bytes of skb ctrl buffer */
+ /* Note: the first 4 bytes are reserved for msg tag */
+ memcpy(CB_TRGCLS(skb), &msg->class, CB_TRGCLS_LEN);
+
+ /* check for special IPRM messages (e.g. iucv_sock_shutdown) */
+ if ((msg->flags & IUCV_IPRMDATA) && len > 7) {
+ if (memcmp(msg->rmmsg, iprm_shutdown, 8) == 0) {
+ skb->data = NULL;
+ skb->len = 0;
+ }
} else {
- rc = iucv_message_receive(path, msg, 0, skb->data,
- msg->length, NULL);
+ rc = iucv_message_receive(path, msg, msg->flags & IUCV_IPRMDATA,
+ skb->data, len, NULL);
if (rc) {
kfree_skb(skb);
return;
}
- if (skb->truesize >= sk->sk_rcvbuf / 4) {
- rc = iucv_fragment_skb(sk, skb, msg->length);
+ /* we need to fragment iucv messages for SOCK_STREAM only;
+ * for SOCK_SEQPACKET, it is only relevant if we support
+ * record segmentation using MSG_EOR (see also recvmsg()) */
+ if (sk->sk_type == SOCK_STREAM &&
+ skb->truesize >= sk->sk_rcvbuf / 4) {
+ rc = iucv_fragment_skb(sk, skb, len);
kfree_skb(skb);
skb = NULL;
if (rc) {
} else {
skb_reset_transport_header(skb);
skb_reset_network_header(skb);
- skb->len = msg->length;
+ skb->len = len;
}
}
struct sock_msg_q *p, *n;
list_for_each_entry_safe(p, n, &iucv->message_q.list, list) {
- skb = alloc_skb(p->msg.length, GFP_ATOMIC | GFP_DMA);
+ skb = alloc_skb(iucv_msg_length(&p->msg), GFP_ATOMIC | GFP_DMA);
if (!skb)
break;
iucv_process_message(sk, skb, p->path, &p->msg);
int noblock = flags & MSG_DONTWAIT;
struct sock *sk = sock->sk;
struct iucv_sock *iucv = iucv_sk(sk);
- int target, copied = 0;
+ unsigned int copied, rlen;
struct sk_buff *skb, *rskb, *cskb;
int err = 0;
if (flags & (MSG_OOB))
return -EOPNOTSUPP;
- target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
-
+ /* receive/dequeue next skb:
+ * the function understands MSG_PEEK and, thus, does not dequeue skb */
skb = skb_recv_datagram(sk, flags, noblock, &err);
if (!skb) {
if (sk->sk_shutdown & RCV_SHUTDOWN)
return err;
}
- copied = min_t(unsigned int, skb->len, len);
+ rlen = skb->len; /* real length of skb */
+ copied = min_t(unsigned int, rlen, len);
cskb = skb;
if (memcpy_toiovec(msg->msg_iov, cskb->data, copied)) {
- skb_queue_head(&sk->sk_receive_queue, skb);
- if (copied == 0)
- return -EFAULT;
- goto done;
+ if (!(flags & MSG_PEEK))
+ skb_queue_head(&sk->sk_receive_queue, skb);
+ return -EFAULT;
+ }
+
+ /* SOCK_SEQPACKET: set MSG_TRUNC if recv buf size is too small */
+ if (sk->sk_type == SOCK_SEQPACKET) {
+ if (copied < rlen)
+ msg->msg_flags |= MSG_TRUNC;
+ /* each iucv message contains a complete record */
+ msg->msg_flags |= MSG_EOR;
}
- len -= copied;
+ /* create control message to store iucv msg target class:
+ * get the trgcls from the control buffer of the skb due to
+ * fragmentation of original iucv message. */
+ err = put_cmsg(msg, SOL_IUCV, SCM_IUCV_TRGCLS,
+ CB_TRGCLS_LEN, CB_TRGCLS(skb));
+ if (err) {
+ if (!(flags & MSG_PEEK))
+ skb_queue_head(&sk->sk_receive_queue, skb);
+ return err;
+ }
/* Mark read part of skb as used */
if (!(flags & MSG_PEEK)) {
- skb_pull(skb, copied);
- if (skb->len) {
- skb_queue_head(&sk->sk_receive_queue, skb);
- goto done;
+ /* SOCK_STREAM: re-queue skb if it contains unreceived data */
+ if (sk->sk_type == SOCK_STREAM) {
+ skb_pull(skb, copied);
+ if (skb->len) {
+ skb_queue_head(&sk->sk_receive_queue, skb);
+ goto done;
+ }
}
kfree_skb(skb);
iucv_process_message_q(sk);
spin_unlock_bh(&iucv->message_q.lock);
}
-
- } else
- skb_queue_head(&sk->sk_receive_queue, skb);
+ }
done:
- return err ? : copied;
+ /* SOCK_SEQPACKET: return real length if MSG_TRUNC is set */
+ if (sk->sk_type == SOCK_SEQPACKET && (flags & MSG_TRUNC))
+ copied = rlen;
+
+ return copied;
}
static inline unsigned int iucv_accept_poll(struct sock *parent)
struct sock *sk = sock->sk;
unsigned int mask = 0;
- poll_wait(file, sk->sk_sleep, wait);
+ sock_poll_wait(file, sk->sk_sleep, wait);
if (sk->sk_state == IUCV_LISTEN)
return iucv_accept_poll(sk);
struct iucv_sock *iucv = iucv_sk(sk);
struct iucv_message txmsg;
int err = 0;
- u8 prmmsg[8] = {0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01};
how++;
lock_sock(sk);
switch (sk->sk_state) {
+ case IUCV_DISCONN:
+ case IUCV_CLOSING:
+ case IUCV_SEVERED:
case IUCV_CLOSED:
err = -ENOTCONN;
goto fail;
txmsg.class = 0;
txmsg.tag = 0;
err = iucv_message_send(iucv->path, &txmsg, IUCV_IPRMDATA, 0,
- (void *) prmmsg, 8);
+ (void *) iprm_shutdown, 8);
if (err) {
switch (err) {
case 1:
return err;
}
+/* getsockopt and setsockopt */
+static int iucv_sock_setsockopt(struct socket *sock, int level, int optname,
+ char __user *optval, int optlen)
+{
+ struct sock *sk = sock->sk;
+ struct iucv_sock *iucv = iucv_sk(sk);
+ int val;
+ int rc;
+
+ if (level != SOL_IUCV)
+ return -ENOPROTOOPT;
+
+ if (optlen < sizeof(int))
+ return -EINVAL;
+
+ if (get_user(val, (int __user *) optval))
+ return -EFAULT;
+
+ rc = 0;
+
+ lock_sock(sk);
+ switch (optname) {
+ case SO_IPRMDATA_MSG:
+ if (val)
+ iucv->flags |= IUCV_IPRMDATA;
+ else
+ iucv->flags &= ~IUCV_IPRMDATA;
+ break;
+ case SO_MSGLIMIT:
+ switch (sk->sk_state) {
+ case IUCV_OPEN:
+ case IUCV_BOUND:
+ if (val < 1 || val > (u16)(~0))
+ rc = -EINVAL;
+ else
+ iucv->msglimit = val;
+ break;
+ default:
+ rc = -EINVAL;
+ break;
+ }
+ break;
+ default:
+ rc = -ENOPROTOOPT;
+ break;
+ }
+ release_sock(sk);
+
+ return rc;
+}
+
+static int iucv_sock_getsockopt(struct socket *sock, int level, int optname,
+ char __user *optval, int __user *optlen)
+{
+ struct sock *sk = sock->sk;
+ struct iucv_sock *iucv = iucv_sk(sk);
+ int val, len;
+
+ if (level != SOL_IUCV)
+ return -ENOPROTOOPT;
+
+ if (get_user(len, optlen))
+ return -EFAULT;
+
+ if (len < 0)
+ return -EINVAL;
+
+ len = min_t(unsigned int, len, sizeof(int));
+
+ switch (optname) {
+ case SO_IPRMDATA_MSG:
+ val = (iucv->flags & IUCV_IPRMDATA) ? 1 : 0;
+ break;
+ case SO_MSGLIMIT:
+ lock_sock(sk);
+ val = (iucv->path != NULL) ? iucv->path->msglim /* connected */
+ : iucv->msglimit; /* default */
+ release_sock(sk);
+ break;
+ default:
+ return -ENOPROTOOPT;
+ }
+
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, &val, len))
+ return -EFAULT;
+
+ return 0;
+}
+
+
/* Callback wrappers - called from iucv base support */
static int iucv_callback_connreq(struct iucv_path *path,
u8 ipvmid[8], u8 ipuser[16])
ASCEBC(user_data, sizeof(user_data));
if (sk->sk_state != IUCV_LISTEN) {
err = iucv_path_sever(path, user_data);
+ iucv_path_free(path);
goto fail;
}
/* Check for backlog size */
if (sk_acceptq_is_full(sk)) {
err = iucv_path_sever(path, user_data);
+ iucv_path_free(path);
goto fail;
}
/* Create the new socket */
- nsk = iucv_sock_alloc(NULL, SOCK_STREAM, GFP_ATOMIC);
+ nsk = iucv_sock_alloc(NULL, sk->sk_type, GFP_ATOMIC);
if (!nsk) {
err = iucv_path_sever(path, user_data);
+ iucv_path_free(path);
goto fail;
}
memcpy(nuser_data + 8, niucv->src_name, 8);
ASCEBC(nuser_data + 8, 8);
- path->msglim = IUCV_QUEUELEN_DEFAULT;
+ /* set message limit for path based on msglimit of accepting socket */
+ niucv->msglimit = iucv->msglimit;
+ path->msglim = iucv->msglimit;
err = iucv_path_accept(path, &af_iucv_handler, nuser_data, nsk);
if (err) {
err = iucv_path_sever(path, user_data);
+ iucv_path_free(path);
+ iucv_sock_kill(nsk);
goto fail;
}
struct sock_msg_q *save_msg;
int len;
- if (sk->sk_shutdown & RCV_SHUTDOWN)
+ if (sk->sk_shutdown & RCV_SHUTDOWN) {
+ iucv_message_reject(path, msg);
return;
+ }
+
+ spin_lock(&iucv->message_q.lock);
if (!list_empty(&iucv->message_q.list) ||
!skb_queue_empty(&iucv->backlog_skb_q))
goto save_message;
len = atomic_read(&sk->sk_rmem_alloc);
- len += msg->length + sizeof(struct sk_buff);
+ len += iucv_msg_length(msg) + sizeof(struct sk_buff);
if (len > sk->sk_rcvbuf)
goto save_message;
- skb = alloc_skb(msg->length, GFP_ATOMIC | GFP_DMA);
+ skb = alloc_skb(iucv_msg_length(msg), GFP_ATOMIC | GFP_DMA);
if (!skb)
goto save_message;
- spin_lock(&iucv->message_q.lock);
iucv_process_message(sk, skb, path, msg);
- spin_unlock(&iucv->message_q.lock);
-
- return;
+ goto out_unlock;
save_message:
save_msg = kzalloc(sizeof(struct sock_msg_q), GFP_ATOMIC | GFP_DMA);
+ if (!save_msg)
+ return;
save_msg->path = path;
save_msg->msg = *msg;
- spin_lock(&iucv->message_q.lock);
list_add_tail(&save_msg->list, &iucv->message_q.list);
+
+out_unlock:
spin_unlock(&iucv->message_q.lock);
}
struct iucv_message *msg)
{
struct sock *sk = path->private;
- struct sk_buff *this;
+ struct sk_buff *this = NULL;
struct sk_buff_head *list = &iucv_sk(sk)->send_skb_q;
struct sk_buff *list_skb = list->next;
unsigned long flags;
- if (list_skb) {
+ if (!skb_queue_empty(list)) {
spin_lock_irqsave(&list->lock, flags);
- do {
- this = list_skb;
+ while (list_skb != (struct sk_buff *)list) {
+ if (!memcmp(&msg->tag, CB_TAG(list_skb), CB_TAG_LEN)) {
+ this = list_skb;
+ break;
+ }
list_skb = list_skb->next;
- } while (memcmp(&msg->tag, this->cb, 4) && list_skb);
+ }
+ if (this)
+ __skb_unlink(this, list);
spin_unlock_irqrestore(&list->lock, flags);
- skb_unlink(this, &iucv_sk(sk)->send_skb_q);
- kfree_skb(this);
+ if (this) {
+ kfree_skb(this);
+ /* wake up any process waiting for sending */
+ iucv_sock_wake_msglim(sk);
+ }
}
+ BUG_ON(!this);
if (sk->sk_state == IUCV_CLOSING) {
if (skb_queue_empty(&iucv_sk(sk)->send_skb_q)) {
sk->sk_state_change(sk);
}
+/* called if the other communication side shuts down its RECV direction;
+ * in turn, the callback sets SEND_SHUTDOWN to disable sending of data.
+ */
+static void iucv_callback_shutdown(struct iucv_path *path, u8 ipuser[16])
+{
+ struct sock *sk = path->private;
+
+ bh_lock_sock(sk);
+ if (sk->sk_state != IUCV_CLOSED) {
+ sk->sk_shutdown |= SEND_SHUTDOWN;
+ sk->sk_state_change(sk);
+ }
+ bh_unlock_sock(sk);
+}
+
static struct proto_ops iucv_sock_ops = {
.family = PF_IUCV,
.owner = THIS_MODULE,
.mmap = sock_no_mmap,
.socketpair = sock_no_socketpair,
.shutdown = iucv_sock_shutdown,
- .setsockopt = sock_no_setsockopt,
- .getsockopt = sock_no_getsockopt
+ .setsockopt = iucv_sock_setsockopt,
+ .getsockopt = iucv_sock_getsockopt,
};
static struct net_proto_family iucv_sock_family_ops = {
int err;
if (!MACHINE_IS_VM) {
- printk(KERN_ERR "AF_IUCV connection needs VM as base\n");
+ pr_err("The af_iucv module cannot be loaded"
+ " without z/VM\n");
err = -EPROTONOSUPPORT;
goto out;
}
cpcmd("QUERY USERID", iucv_userid, sizeof(iucv_userid), &err);
if (unlikely(err)) {
- printk(KERN_ERR "AF_IUCV needs the VM userid\n");
+ WARN_ON(err);
err = -EPROTONOSUPPORT;
goto out;
}
err = sock_register(&iucv_sock_family_ops);
if (err)
goto out_proto;
- printk(KERN_INFO "AF_IUCV lowlevel driver initialized\n");
+ /* establish dummy device */
+ err = driver_register(&af_iucv_driver);
+ if (err)
+ goto out_sock;
+ af_iucv_dev = kzalloc(sizeof(struct device), GFP_KERNEL);
+ if (!af_iucv_dev) {
+ err = -ENOMEM;
+ goto out_driver;
+ }
+ dev_set_name(af_iucv_dev, "af_iucv");
+ af_iucv_dev->bus = &iucv_bus;
+ af_iucv_dev->parent = iucv_root;
+ af_iucv_dev->release = (void (*)(struct device *))kfree;
+ af_iucv_dev->driver = &af_iucv_driver;
+ err = device_register(af_iucv_dev);
+ if (err)
+ goto out_driver;
+
return 0;
+out_driver:
+ driver_unregister(&af_iucv_driver);
+out_sock:
+ sock_unregister(PF_IUCV);
out_proto:
proto_unregister(&iucv_proto);
out_iucv:
static void __exit afiucv_exit(void)
{
+ device_unregister(af_iucv_dev);
+ driver_unregister(&af_iucv_driver);
sock_unregister(PF_IUCV);
proto_unregister(&iucv_proto);
iucv_unregister(&af_iucv_handler, 0);
-
- printk(KERN_INFO "AF_IUCV lowlevel driver unloaded\n");
}
module_init(afiucv_init);