On 5/27/07, Michael S. Tsirkin <[EMAIL PROTECTED]> wrote:

> Quoting Or Gerlitz <[EMAIL PROTECTED]>:
> Subject: Re: Re: [PATCH for-2.6.22] IB/ipoib: fix SRQ WR leak
>
> Roland Dreier wrote:
> >OK, I crossed my fingers and merged this for 2.6.22
>
> Somehow it seems when applying this patch to OFED something goes wrong,
> please see https://bugs.openfabrics.org/show_bug.cgi?id=636

Yes, it seems that we shouldn't keep a QP in error state
for extended periods of time, since that moves hardware
to the slow path.


what actually is the "hardware slow path" ?

Or.
_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to