On Tue, Sep 28, 2010 at 11:36 AM, Thomson, Patrick S
<patrick.s.thomson@xxxxxxxxx> wrote:
> Alrighty...
>
> Based on Dan's feedback and some from Nate, here's another go
> at the changes. I removed the spinlock and renamed the elements in
> enum cfg. The default behavior remains the same however. I'm
> still leery on changing existing behavior, unless of course people
> think it's a good idea
I guess I'm not understanding the usage model for this change. When
would you want two phys attached to the same address to not form a
wide port, and would this be a per-phy policy rather than
per-controller?
General question about the code change below:
> +void notify_libsas_config(struct sas_ha_struct *sas_ha, enum cfg_state cfg)
> +{
> + unsigned long flags;
> +
> + switch (cfg) {
> + case CFGE_USE_LOCAL_AND_ATTACHED_SAS:
> + sas_ha->use_host_phy_addr = 1;
> + break;
> + case CFGE_USE_ONLY_ATTACHED_SAS:
> + sas_ha->use_host_phy_addr = 0;
> + break;
> + }
> +}
Is it not sufficient to just set this flag before calling
sas_register_ha()? If this was truly a dynamic notification I would
expect you would want to re-validate existing ports against this newly
set policy.
--
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
