> From: Peter J. Braam [mailto:[EMAIL PROTECTED]
> Sent: Sunday, October 09, 2005 2:18 PM
>
> Cluster File Systems, Inc and its customers have been wondering if the Lustre
> Network Driver (LND) for OpenIb gen2, which we will begin to develop during
> the coming months, should be based on kdapl or
What's the status on getting the ehca driver integrated into subversion?
If there's something holding it up, can we at least get a version that
can be dropped into drivers/infiniband/hw ?
Also, one final note, is it really appropriate to have ehca/ebus in the
infiniband directory? It's really a PP
Cluster File
Systems, Inc and its customers have been wondering if the Lustre Network
Driver (LND) for OpenIb gen2, which we will begin to develop during the
coming months, should be based on kdapl or verbs.
The driver we plan
to develop should strive to address several goals:
- high
re
Jack,
* On Oct,16 Jack Morgenstein<[EMAIL PROTECTED]> wrote :
> Sayantan,
> The Limit Event must be re-armed after an event has occurred (it is a
> "one-shot").
> (i.e., modify-srq/set-limit must be re-invoked).This is compliant with the
> IB Spec (see section 10.2.9.3, first paragraph). (Note t
On Sun, 2005-10-09 at 07:57 -0700, Sean Hefty wrote:
> >It is theoretically possible to support all this on an IPoIB based
> >network. Multiple subnets, multiple routes to remote peers, ICMP
> >redirect, multiple IP addresses for each physical interface, yada yada
> >yada. But IMHO, the only way to
Sayantan,
The Limit Event must be re-armed after an event has occurred (it is a
"one-shot").
(i.e., modify-srq/set-limit must be re-invoked).This is compliant with the
IB Spec (see section 10.2.9.3, first paragraph). (Note that after each SRQ LWM
event, the limit for the SRQ gets reset back to ze
Roland,
* On Oct,13 Roland Dreier<[EMAIL PROTECTED]> wrote :
> Sayantan> I noticed that the test re-posts buffers only when the
> Sayantan> outstanding recv count is <= 1. I set a SRQ limit as
> Sayantan> max_recv - 5. So, I should get the event when 5 WQEs are
> Sayantan> consumed
>It is theoretically possible to support all this on an IPoIB based
>network. Multiple subnets, multiple routes to remote peers, ICMP
>redirect, multiple IP addresses for each physical interface, yada yada
>yada. But IMHO, the only way to do this would be to tie directly into
>the existing routing,
>I think iWARP can be on top of TCP or SCTP. But why wouldn't it care ?
I'm referring to the case that iWarp is running over TCP. I know that it can
run over SCTP, but I'm not familiar with the details of that protocol. With
TCP, this is an end-to-end connection, so layering iWarp over it, only
Hi Hal,
During our tests on Windows we encountered an issue that is caused due
to some problem in the lower layer, but causes problem in the opensm.
If the osm_vendor_send call fails immediatly, we need to update
several counters (currently, only qp0_mads_sent is decremented), and
also all the dis
$B7k:'(B5$BG/L\(B28$B:P!#;R6!$,M_$7$/$F;EJ}$J$$$N$K=PMh$^$;$s!#(B
$BK\5$$G@:;R$r;d$N%*!{!{%3$K=P$7$F$/$l$^$;$s$+!)@dBP$K(B
$BLBOG$+$1$^$;$s!#(B
[EMAIL PROTECTED];v$G$9!#K\Ev$K=u$1$F$/[EMAIL PROTECTED]/(B
$B$*4j$$CW$7$^$9!#BT$C$F$^$9!#(B
http://awg.webchu.com/sweet-s/?gyakuen
$B!a
Hi,
I'm proposing a better fix. see below.
On Mon, Oct 03, 2005 at 06:13:51PM +0200, Roland Dreier wrote:
> Seems reasonable. However, looking back at the chip documentation, it
> seems that the max CQEs should really be 0x1 rather than 0x as
> I had it. Can you confirm?
>
> Thanks,
>
12 matches
Mail list logo