<html><body>
<p>Hello Roland,<br>
<br>
<tt>Roland Dreier <rdreier@cisco.com> wrote on 06/24/2008 08:11:56 PM:<br>
> > In case post_send drain WR failure, release connections in stale task.<br>
> What is "post_send drain WR failure"? When can it happen? <br>
> - R.<br>
</tt><br>
<tt>I saw it happened from below ipoib debug output on testing ehca2:</tt><br>
<br>
static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv *priv)<br>
{<br>
.... <br>
/*<br>
* QPs on flush list are error state. This way, a "flush<br>
* error" WC will be immediately generated for each WR we post.<br>
*/<br>
p = list_entry(priv->cm.rx_flush_list.next, typeof(*p), list);<br>
if (ib_post_send(p->qp, &ipoib_cm_rx_drain_wr, &bad_wr))<br>
ipoib_warn(priv, "failed to post drain wr\n");<br>
<tt>...</tt><br>
<tt>}</tt><br>
<br>
<tt>Thanks</tt><br>
<tt>Shirley</tt></body></html>