[ofa-general] Re: [RFC][PATCH] last wqe event handler patch

[Shirley Ma]

Hello Eli,

>This means that if you receive
>such an event, you're guaranteed to see the CQE in the CQ (that is of of
>course if there is such a CQE; if there where no WQEs related to the QP
>at the time it entered an error state then the HW just generates the
>event).
Thanks for the clarification. Another question, after the QP entered an
error state, the HCA can generate a cqe for post_send drain WRs?

> Shirley, did you see this on Mellanox HCAs too or just ehca?
I saw some panic in Mellanox HCAs, I thought this senarios could cause a
panic. ehca doesn't support post_send() drain WRs after such an event. So
it can't be a problem for ehca.

Since there is no such a post_send() drain WRs for ehca, then we need a
timer value to release the resource from rx_flush_list. Otherwise the
resouces will remain there forever, for a long run, the node loses lots of
memory. We have seen lots of last WQE events when doing interoperability
test between MHCAs and ehcas.

thanks
Shirley
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20080626/95859006/attachment.html>