[openib-general] FW: patches to 2.6.19.1 kernel for switch Operation

[Suresh Shelvapille]

Just copying the list.

-----Original Message-----
From: Suresh Shelvapille [mailto:suri at baymicrosystems.com] 
Sent: Friday, February 09, 2007 4:33 PM
To: 'Hal Rosenstock'
Subject: RE: patches to 2.6.19.1 kernel for switch Operation



Hal:

Many thanks for your response,

Ref: comment on mad.c (ib_mad_recv_done_handler().

Even if I make the relevant changes to smi.c functions how do I get the
packet to get forwarded, without making additional changes in this function?

Meaning, when smi_handle_dr_smp_send(),smi_check_forward_dr_smp() are called
and you determine that the packet has to be forwarded instead of consuming
where do you actually do the send? I think this chain is missing!


Thanks,
Suri

> +                       if (!agent_send_response(&response->mad.mad,
> +                                                &response->grh, wc,
> +                                                port_priv->device,
> +                                                port_num,
> +                                                qp_info->qp->qp_num))
> +                               response = NULL;
> 
> Per the above change, it appears that smi_check_forward_dr_smp and
> smi_handle_dr_smp_send are no longer used at least here
> (smi_check_forward_dr_smp is not used at all with this change). Couldn't
> these be fixed to do the right thing for this case (as well as existing
> cases) ? I'm not sure your changes work for end ports (CA and router
> ports).
> 
> Also, based on smi comments below, there might also be changes to
> following:
> +                       if (!ib_get_smp_direction(&recv->mad.smp))
> +                               port_num =
> recv->mad.smp.initial_path[recv->mad.smp.hop_ptr+1];
> +                       else
> +                               port_num =
> recv->mad.smp.return_path[recv->mad.smp.hop_ptr-1];
> +
>