[ofw] (no subject)

[Smith, Stan]

Hello,
  Patch looks fine plus tests out OK.
Minor point about patches, when building a patch / patch file, it helps in applying the path, to make the patch relative to the trunk folder.

B:/users/irena/proj1/trunk/ulp/ipoib_NDIS6_CM/kernel/ipoib_port.cpp
vs
ulp/ipoib_NDIS6_CM/kernel/ipoib_port.cpp

thanks,

stan.

________________________________
From: ofw-bounces at lists.openfabrics.org [mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Irena Gannon
Sent: Sunday, November 28, 2010 2:27 AM
To: 'ofw at lists.openfabrics.org'
Subject: [ofw] (no subject)

Signed-off-by: Irena Gannon (irena at mellanox.co.il<mailto:irena at mellanox.co.il>)

This patch constructs the send_mgr.sg_pool in the __send_mgr_construct so that it has a valid state in cl_qcpool_destroy (in cl_is_state_valid assert) even if __port_init fails before __send_mgr_init

Index: B:/users/irena/proj1/trunk/ulp/ipoib_NDIS6_CM/kernel/ipoib_port.cpp
===================================================================
--- B:/users/irena/proj1/trunk/ulp/ipoib_NDIS6_CM/kernel/ipoib_port.cpp       (revision 6930)
+++ B:/users/irena/proj1/trunk/ulp/ipoib_NDIS6_CM/kernel/ipoib_port.cpp    (revision 6931)
@@ -3409,6 +3409,7 @@
                IPOIB_ENTER( IPOIB_DBG_SEND );
                p_port->send_mgr.depth = 0;
                cl_qlist_init( &p_port->send_mgr.pending_list );
+    cl_qpool_construct(&p_port->send_mgr.sg_pool);
                cl_qpool_construct( &p_port->send_mgr.send_pool );
                p_port->p_desc = NULL;
                IPOIB_EXIT( IPOIB_DBG_SEND );
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofw/attachments/20101129/6bd5170a/attachment.html>