On Thu, 2008-03-27 at 10:04 -0700, Grant Grundler wrote:
> On Wed, Mar 26, 2008 at 11:55 PM, Ke Wei <kewei@xxxxxxxxxxx> wrote:
> > removed unused code and attached SATA address makes use of port id.
> > enable HBA interrupt after calling sas_register_ha();
>
> just one small nit and one question.
>
> ...
> > @@ -2837,20 +2812,34 @@ static void mvs_update_phyinfo(struct mvs_info *mvi, int i,
> > } else {
> > dev_printk(KERN_DEBUG, &pdev->dev,
> > "No sig fis\n");
> > + phy->phy_type &= ~(PORT_TYPE_SATA);
>
> shouldn't need '()'s around the constant.
>
> > + goto out_done;
> > }
> > }
> > + tmp64 = cpu_to_be64(phy->att_dev_sas_addr);
> > + memcpy(sas_phy->attached_sas_addr, &tmp64, SAS_ADDR_SIZE);
>
> Why does this need to be a memcpy and not just a simple store operation?
> Is it misaligned? (even if it is, the kernel will deal with it. It
> just won't be fast.)
Because if you look in include/scsi/libsas.h, you find the definition:
struct asd_sas_phy {
[...]
u8 attached_sas_addr[SAS_ADDR_SIZE]; /* class:RO, driver: R/W */
We're annoyingly schizophrenic on this ... most of the time we have a
sas_addr defined as u64, but not in a lot of the libsas code.
James
--
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
