[openib-general] Re: openib and mellanox hca problem
Michael Di Domenico
mdidomenico at gmail.com
Thu Feb 9 14:01:36 PST 2006
On 2/9/06, Grant Grundler <iod00d at hp.com> wrote:
> On Thu, Feb 09, 2006 at 07:32:31AM -0800, Roland Dreier wrote:
> > Michael> My problems if you recall with the mellanox cards seem to
> > Michael> be related to Linux Kernel 2.6.15.3... Installed RHEL4
> > Michael> instead of Fedora Core and get the same results using
> > Michael> kernel 2.6.15.3 and RHEL4...
> >
> > No, I think the problem is that something about your system is causing
> > problems when the driver resets the HCA. Presumably the silverstorm
> > stack does not reset the HCA.
> >
> > I had one thought: do you know the speed (100 MHz, 133 MHz, ...) of
> > the PCI slot that your HCA is in?
>
> Or can you move the card to a different slot and try again?
> Sometimes cards will work in one slot but not another due
> to timing differences or signal quality differences between slots.
> >
> > In any case you should be able to comment out the call to
> > mthca_reset() in mthca_init_one() as a workaround for now.
Commenting out the mthca_reset seems to have fixed the problem... My
apologies... So i'm now running RHEL4 w/ Kernel 2.6.15.3 with the
openib stack... here's some outputs if you guys want it, from lspci
and dmesg
I'm still curious why the reset caused an issue and whether it is
thought that this is a random occurance because of the hardware i
have?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dmesg.out
Type: application/octet-stream
Size: 15421 bytes
Desc: not available
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20060209/65e94646/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lspci.out
Type: application/octet-stream
Size: 15442 bytes
Desc: not available
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20060209/65e94646/attachment-0001.obj>
More information about the general
mailing list