static inline int pipelined_send(struct msg_queue *msq, struct msg_msg *msg)
{
struct list_head *tmp;
tmp = msq->q_receivers.next;
while (tmp != &msq->q_receivers) {
struct msg_receiver *msr;
msr = list_entry(tmp, struct msg_receiver, r_list);
tmp = tmp->next;
if (testmsg(msg, msr->r_msgtype, msr->r_mode) &&
!security_msg_queue_msgrcv(msq, msg, msr->r_tsk,
msr->r_msgtype, msr->r_mode)) {
list_del(&msr->r_list);
if (msr->r_maxsize < msg->m_ts) {
msr->r_msg = NULL;
wake_up_process(msr->r_tsk);
smp_mb();
msr->r_msg = ERR_PTR(-E2BIG);
} else {
msr->r_msg = NULL;
msq->q_lrpid = task_pid_vnr(msr->r_tsk);
msq->q_rtime = get_seconds();
wake_up_process(msr->r_tsk);
smp_mb();
msr->r_msg = msg;
return 1;
}
}
}
return 0;
}
static inline int pipelined_send(struct msg_queue *msq, struct msg_msg *msg)
{
struct list_head *tmp;
tmp = msq->q_receivers.next;
while (tmp != &msq->q_receivers) {
struct msg_receiver *msr;
msr = list_entry(tmp, struct msg_receiver, r_list);
tmp = tmp->next;
if (testmsg(msg, msr->r_msgtype, msr->r_mode) &&
!security_msg_queue_msgrcv(msq, msg, msr->r_tsk,
msr->r_msgtype, msr->r_mode)) {
/*
* Make sure that the wakeup doesnt preempt
* this CPU prematurely. (on PREEMPT_RT)
*/
preempt_disable_rt();
list_del(&msr->r_list);
if (msr->r_maxsize < msg->m_ts) {
msr->r_msg = NULL;
wake_up_process(msr->r_tsk);
smp_mb();
msr->r_msg = ERR_PTR(-E2BIG);
} else {
msr->r_msg = NULL;
msq->q_lrpid = task_pid_vnr(msr->r_tsk);
msq->q_rtime = get_seconds();
wake_up_process(msr->r_tsk);
smp_mb();
msr->r_msg = msg;
preempt_enable_rt();
return 1;
}
preempt_enable_rt();
}
}
return 0;
}
long do_msgrcv(int msqid, long *pmtype, void __user *mtext,
size_t msgsz, long msgtyp, int msgflg)
{
struct msg_queue *msq;
struct msg_msg *msg;
int mode;
struct ipc_namespace *ns;
if (msqid < 0 || (long) msgsz < 0)
return -EINVAL;
mode = convert_mode(&msgtyp, msgflg);
ns = current->nsproxy->ipc_ns;
msq = msg_lock_check(ns, msqid);
if (IS_ERR(msq))
return PTR_ERR(msq);
for (;;) {
struct msg_receiver msr_d;
struct list_head *tmp;
msg = ERR_PTR(-EACCES);
if (ipcperms(&msq->q_perm, S_IRUGO))
goto out_unlock;
msg = ERR_PTR(-EAGAIN);
tmp = msq->q_messages.next;
while (tmp != &msq->q_messages) {
struct msg_msg *walk_msg;
walk_msg = list_entry(tmp, struct msg_msg, m_list);
if (testmsg(walk_msg, msgtyp, mode) &&
!security_msg_queue_msgrcv(msq, walk_msg, current,
msgtyp, mode)) {
msg = walk_msg;
if (mode == SEARCH_LESSEQUAL &&
walk_msg->m_type != 1) {
msg = walk_msg;
msgtyp = walk_msg->m_type - 1;
} else {
msg = walk_msg;
break;
}
}
tmp = tmp->next;
}
if (!IS_ERR(msg)) {
/*
* Found a suitable message.
* Unlink it from the queue.
*/
if ((msgsz < msg->m_ts) && !(msgflg & MSG_NOERROR)) {
msg = ERR_PTR(-E2BIG);
goto out_unlock;
}
list_del(&msg->m_list);
msq->q_qnum--;
msq->q_rtime = get_seconds();
msq->q_lrpid = task_tgid_vnr(current);
msq->q_cbytes -= msg->m_ts;
atomic_sub(msg->m_ts, &ns->msg_bytes);
atomic_dec(&ns->msg_hdrs);
ss_wakeup(&msq->q_senders, 0);
msg_unlock(msq);
break;
}
/* No message waiting. Wait for a message */
if (msgflg & IPC_NOWAIT) {
msg = ERR_PTR(-ENOMSG);
goto out_unlock;
}
list_add_tail(&msr_d.r_list, &msq->q_receivers);
msr_d.r_tsk = current;
msr_d.r_msgtype = msgtyp;
msr_d.r_mode = mode;
if (msgflg & MSG_NOERROR)
msr_d.r_maxsize = INT_MAX;
else
msr_d.r_maxsize = msgsz;
msr_d.r_msg = ERR_PTR(-EAGAIN);
current->state = TASK_INTERRUPTIBLE;
msg_unlock(msq);
schedule();
/* Lockless receive, part 1:
* Disable preemption. We don't hold a reference to the queue
* and getting a reference would defeat the idea of a lockless
* operation, thus the code relies on rcu to guarantee the
* existance of msq:
* Prior to destruction, expunge_all(-EIRDM) changes r_msg.
* Thus if r_msg is -EAGAIN, then the queue not yet destroyed.
* rcu_read_lock() prevents preemption between reading r_msg
* and the spin_lock() inside ipc_lock_by_ptr().
*/
rcu_read_lock();
/* Lockless receive, part 2:
* Wait until pipelined_send or expunge_all are outside of
* wake_up_process(). There is a race with exit(), see
* ipc/mqueue.c for the details.
*/
msg = (struct msg_msg*)msr_d.r_msg;
while (msg == NULL) {
cpu_relax();
msg = (struct msg_msg *)msr_d.r_msg;
}
/* Lockless receive, part 3:
* If there is a message or an error then accept it without
* locking.
*/
if (msg != ERR_PTR(-EAGAIN)) {
rcu_read_unlock();
break;
}
/* Lockless receive, part 3:
* Acquire the queue spinlock.
*/
ipc_lock_by_ptr(&msq->q_perm);
rcu_read_unlock();
/* Lockless receive, part 4:
* Repeat test after acquiring the spinlock.
*/
msg = (struct msg_msg*)msr_d.r_msg;
if (msg != ERR_PTR(-EAGAIN))
goto out_unlock;
list_del(&msr_d.r_list);
if (signal_pending(current)) {
msg = ERR_PTR(-ERESTARTNOHAND);
out_unlock:
msg_unlock(msq);
break;
}
}
if (IS_ERR(msg))
return PTR_ERR(msg);
msgsz = (msgsz > msg->m_ts) ? msg->m_ts : msgsz;
*pmtype = msg->m_type;
if (store_msg(mtext, msg, msgsz))
msgsz = -EFAULT;
free_msg(msg);
return msgsz;
}
asmlinkage long sys_msgrcv (int msqid, struct msgbuf __user *msgp, size_t msgsz,
long msgtyp, int msgflg)
{
struct msg_queue *msq;
struct msg_receiver msr_d;
struct list_head* tmp;
struct msg_msg* msg, *found_msg;
int err;
int mode;
if (msqid < 0 || (long) msgsz < 0)
return -EINVAL;
mode = convert_mode(&msgtyp,msgflg);
msq = msg_lock(msqid);
if(msq==NULL)
return -EINVAL;
retry:
err = -EIDRM;
if (msg_checkid(msq,msqid))
goto out_unlock;
err=-EACCES;
if (ipcperms (&msq->q_perm, S_IRUGO))
goto out_unlock;
tmp = msq->q_messages.next;
found_msg=NULL;
while (tmp != &msq->q_messages) {
msg = list_entry(tmp,struct msg_msg,m_list);
if(testmsg(msg,msgtyp,mode) &&
!security_msg_queue_msgrcv(msq, msg, current, msgtyp, mode)) {
found_msg = msg;
if(mode == SEARCH_LESSEQUAL && msg->m_type != 1) {
found_msg=msg;
msgtyp=msg->m_type-1;
} else {
found_msg=msg;
break;
}
}
tmp = tmp->next;
}
if(found_msg) {
msg=found_msg;
if ((msgsz < msg->m_ts) && !(msgflg & MSG_NOERROR)) {
err=-E2BIG;
goto out_unlock;
}
list_del(&msg->m_list);
msq->q_qnum--;
msq->q_rtime = get_seconds();
msq->q_lrpid = current->tgid;
msq->q_cbytes -= msg->m_ts;
atomic_sub(msg->m_ts,&msg_bytes);
atomic_dec(&msg_hdrs);
ss_wakeup(&msq->q_senders,0);
msg_unlock(msq);
out_success:
msgsz = (msgsz > msg->m_ts) ? msg->m_ts : msgsz;
if (put_user (msg->m_type, &msgp->mtype) ||
store_msg(msgp->mtext, msg, msgsz)) {
msgsz = -EFAULT;
}
free_msg(msg);
return msgsz;
} else
{
/* no message waiting. Prepare for pipelined
* receive.
*/
if (msgflg & IPC_NOWAIT) {
err=-ENOMSG;
goto out_unlock;
}
list_add_tail(&msr_d.r_list,&msq->q_receivers);
msr_d.r_tsk = current;
msr_d.r_msgtype = msgtyp;
msr_d.r_mode = mode;
if(msgflg & MSG_NOERROR)
msr_d.r_maxsize = INT_MAX;
else
msr_d.r_maxsize = msgsz;
msr_d.r_msg = ERR_PTR(-EAGAIN);
current->state = TASK_INTERRUPTIBLE;
msg_unlock(msq);
schedule();
/*
* The below optimisation is buggy. A sleeping thread that is
* woken up checks if it got a message and if so, copies it to
* userspace and just returns without taking any locks.
* But this return to user space can be faster than the message
* send, and if the receiver immediately exits the
* wake_up_process performed by the sender will oops.
*/
#if 0
msg = (struct msg_msg*) msr_d.r_msg;
if(!IS_ERR(msg))
goto out_success;
#endif
msq = msg_lock(msqid);
msg = (struct msg_msg*)msr_d.r_msg;
if(!IS_ERR(msg)) {
/* our message arived while we waited for
* the spinlock. Process it.
*/
if(msq)
msg_unlock(msq);
goto out_success;
}
err = PTR_ERR(msg);
if(err == -EAGAIN) {
if(!msq)
BUG();
list_del(&msr_d.r_list);
if (signal_pending(current))
err=-EINTR;
else
goto retry;
}
}
out_unlock:
if(msq)
msg_unlock(msq);
return err;
}