[ofw] correcting a side-effect of a previous bugfix

[Tzachi Dar]

Just to clarify, all /Mellanox patches are in.


________________________________

	From: ofw-bounces at lists.openfabrics.org
[mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Leonid Keller
	Sent: Thursday, November 13, 2008 7:42 PM
	To: Leonid Keller; ofw at lists.openfabrics.org
	Subject: RE: [ofw] correcting a side-effect of a previous bugfix
	
	
	Applied in 1760,1761.


________________________________

		From: ofw-bounces at lists.openfabrics.org
[mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Leonid Keller
		Sent: Thursday, November 13, 2008 5:54 PM
		To: ofw at lists.openfabrics.org
		Subject: [ofw] correcting a side-effect of a previous
bugfix
		
		

		1361 fix added a bug: now all application, creating QP
with zero receive QP size, will fail with send completion error.

		This patch fixes that.

		 
		 
		Index: V:/svn/winib/trunk/hw/mthca/kernel/mthca_qp.c
	
===================================================================
		--- V:/svn/winib/trunk/hw/mthca/kernel/mthca_qp.c
(revision 3466)
		+++ V:/svn/winib/trunk/hw/mthca/kernel/mthca_qp.c
(revision 3467)
		@@ -1374,6 +1374,12 @@
		  if (qp->transport == MLX && cap->max_recv_sge + 2 >
(u32)dev->limits.max_sg)
		   return -EINVAL;
		 
		+ /* Enable creating zero-sized QPs */
		+ if (!cap->max_recv_wr)
		+  cap->max_recv_wr = 1;
		+ if (!cap->max_send_wr)
		+  cap->max_send_wr = 1;
		+ 
		  if (mthca_is_memfree(dev)) {
		   qp->rq.max = cap->max_recv_wr ?
		    roundup_pow_of_two(cap->max_recv_wr) : 0;
		

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofw/attachments/20081113/2cc8154c/attachment.html>