summaryrefslogtreecommitdiff
path: root/drivers/net/iseries_veth.c
diff options
context:
space:
mode:
authorMichael Ellerman <michael@ellerman.id.au>2005-09-01 11:29:00 +1000
committerJeff Garzik <jgarzik@pobox.com>2005-08-31 22:37:56 -0400
commit58c5900bdaffbf76afd7ad5e053410cb95eb3169 (patch)
tree1d6e6eba8392f496f9ac2f632fdda8aa48b0b732 /drivers/net/iseries_veth.c
parentabfda4719c61550be4efaf277d4a904a7930d410 (diff)
[PATCH] iseries_veth: Try to avoid pathological reset behaviour
The iseries_veth driver contains a state machine which is used to manage how connections are setup and neogotiated between LPARs. If one side of a connection resets for some reason, the two LPARs can get stuck in a race to re-setup the connection. This can lead to the connection being declared dead by one or both ends. In practice the connection is declared dead by one or both ends approximately 8/10 times a connection is reset, although it is rare for connections to be reset. (an example here: http://michael.ellerman.id.au/files/misc/veth-trace.html) The core of the problem is that the end that resets the connection doesn't wait for the other end to become aware of the reset. So the resetting end starts setting the connection back up, and then receives a reset from the other end (which is the response to the initial reset). And so on. We're severely limited in what we can do to fix this. The protocol between LPARs is essentially fixed, as we have to interoperate with both OS/400 and old Linux drivers. Which also means we need a fix that only changes the code on one end. The only fix I've found given that, is to just blindly sleep for a bit when resetting the connection, in the hope that the other end will get itself sorted. Needless to say I'd love it if someone has a better idea. This does work, I've so far been unable to get it to break, whereas without the fix a reset of one end will lead to a dead connection ~8/10 times. Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
Diffstat (limited to 'drivers/net/iseries_veth.c')
-rw-r--r--drivers/net/iseries_veth.c25
1 files changed, 23 insertions, 2 deletions
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index c19b32e0a5ad..db83b0d31327 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -324,8 +324,14 @@ static void veth_take_monitor_ack(struct veth_lpar_connection *cnx,
spin_lock_irqsave(&cnx->lock, flags);
veth_debug("cnx %d: lost connection.\n", cnx->remote_lp);
- cnx->state |= VETH_STATE_RESET;
- veth_kick_statemachine(cnx);
+
+ /* Avoid kicking the statemachine once we're shutdown.
+ * It's unnecessary and it could break veth_stop_connection(). */
+
+ if (! (cnx->state & VETH_STATE_SHUTDOWN)) {
+ cnx->state |= VETH_STATE_RESET;
+ veth_kick_statemachine(cnx);
+ }
spin_unlock_irqrestore(&cnx->lock, flags);
}
@@ -483,6 +489,12 @@ static void veth_statemachine(void *p)
if (cnx->state & VETH_STATE_RESET)
goto restart;
+
+ /* Hack, wait for the other end to reset itself. */
+ if (! (cnx->state & VETH_STATE_SHUTDOWN)) {
+ schedule_delayed_work(&cnx->statemachine_wq, 5 * HZ);
+ goto out;
+ }
}
if (cnx->state & VETH_STATE_SHUTDOWN)
@@ -667,6 +679,15 @@ static void veth_stop_connection(u8 rlp)
veth_kick_statemachine(cnx);
spin_unlock_irq(&cnx->lock);
+ /* There's a slim chance the reset code has just queued the
+ * statemachine to run in five seconds. If so we need to cancel
+ * that and requeue the work to run now. */
+ if (cancel_delayed_work(&cnx->statemachine_wq)) {
+ spin_lock_irq(&cnx->lock);
+ veth_kick_statemachine(cnx);
+ spin_unlock_irq(&cnx->lock);
+ }
+
/* Wait for the state machine to run. */
flush_scheduled_work();