[Xen-devel] [PATCH] linux/evtchn: Add memory barriers to evtchn ring accesses.

[Kouya Shimura <kouya@xxxxxxxxxxxxxx>]

Xenstore infrequently hangs up on IA64.
Actually the xenstored is still alive but no response from
xenstore-XXX commands.

After tracking down, I've found that evtchn_read() infrequently
returns a wrong evtchn port number and evtchn_write() never
unmask the exact port.

This patch fixes it.

Thanks,
Kouya

Signed-off-by: Kouya Shimura <kouya@xxxxxxxxxxxxxx>

diff -r 5644f68a7912 drivers/xen/evtchn/evtchn.c
--- a/drivers/xen/evtchn/evtchn.c       Wed Jul 16 13:51:36 2008 +0100
+++ b/drivers/xen/evtchn/evtchn.c       Thu Jul 17 16:21:09 2008 +0900
@@ -84,6 +84,7 @@
        if ((u = port_user[port]) != NULL) {
                if ((u->ring_prod - u->ring_cons) < EVTCHN_RING_SIZE) {
                        u->ring[EVTCHN_RING_MASK(u->ring_prod)] = port;
+                       wmb(); /* Ensure ring contents visible */
                        if (u->ring_cons == u->ring_prod++) {
                                wake_up_interruptible(&u->evtchn_wait);
                                kill_fasync(&u->evtchn_async_queue,
@@ -180,6 +181,7 @@
        }
 
        rc = -EFAULT;
+       rmb(); /* Ensure that we see the port before we copy it. */
        if (copy_to_user(buf, &u->ring[EVTCHN_RING_MASK(c)], bytes1) ||
            ((bytes2 != 0) &&
             copy_to_user(&buf[bytes1], &u->ring[0], bytes2)))

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel