[ofa-general] false warnings of multicast join failures

Yossi Etigin yosefe at Voltaire.COM
Mon Oct 27 11:47:22 PDT 2008


I'm referring to these:

ib0: multicast join failed for ff12:401b:ffff:0000:0000:0000:ffff:ffff, status -11

The patch in http://lists.openfabrics.org/pipermail/general/2008-May/050551.html 
is causing them.
The patch creates a state when there is no sm_ah, so all alloc_mad() calls
return -11 (-EAGAIN), this goes back to ipoib multicast join: ipoib asks the sa
to join, it queues work that calls send_join(), this calls ib_sa_mcmember_rec_query(),
this one calls alloc_mad() and gets -EAGAIN.

How about lowering the severity of this error in ipoib_mcast_join_complete() from warning
to debug?

Index: b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
===================================================================
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c	2008-10-22 20:28:06.000000000 +0200
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c	2008-10-27 20:13:59.000000000 +0200
@@ -443,7 +443,7 @@ static int ipoib_mcast_join_complete(int
 	}
 
 	if (mcast->logcount++ < 20) {
-		if (status == -ETIMEDOUT) {
+		if (status == -ETIMEDOUT || status == -EAGAIN) {
 			ipoib_dbg_mcast(priv, "multicast join failed for " IPOIB_GID_FMT
 					", status %d\n",
 					IPOIB_GID_ARG(mcast->mcmember.mgid),

-- 
--Yossi



More information about the general mailing list