Re: [PATCH 13/21] x86, acpi: Try to find SRAT in firmware earlier.

[Tang Chen]

On 07/24/2013 04:49 AM, Tejun Heo wrote:
> On Fri, Jul 19, 2013 at 03:59:26PM +0800, Tang Chen wrote:
......
> > +	for (pos = 0;
> > +	     pos<  acpi_gbl_root_table_list.current_table_count;
> > +	     pos++) {
> > +		if (!ACPI_COMPARE_NAME
> > +		    (&(acpi_gbl_root_table_list.tables[pos].signature),
> > +		    signature))
>
> Hohumm... creative formatting.  Can't you just cache the tables
> pointer in a local variable with short name and avoid the creativity?

OK, followed.

> > +			continue;
> > +
> > +		memcpy(out_desc,&acpi_gbl_root_table_list.tables[pos],
> > +		       sizeof(struct acpi_table_desc));
> > +
> > +		return_ACPI_STATUS(AE_OK);
> > +	}
> > +
> > +	return_ACPI_STATUS(AE_NOT_FOUND);
>
> Also, if we already know that SRAT is what we want, I wonder whether
> it'd be simpler to store the location of SRAT somewhere instead of
> trying to be generic with the early processing.

Do you mean get the SRAT's address without touching any ACPI global
variables, such as acpi_gbl_root_table_list ?

The physical addresses of all tables is stored in RSDT (Root System
Description Table), which is the root table. We need to parse RSDT
to get SRAT address.

Using acpi_gbl_root_table_list is very convenient. The initialization
of acpi_gbl_root_table_list is using acpi_os_map_memory(), so it can be
done before init_mem_mapping() and relocate_initrd().

With acpi_gbl_root_table_list initialized, we can iterate it and find
SRAT easily. Otherwise, we have to do the same procedure to parse RSDT,
and find SRAT, which I don't think could be any simpler. I think reuse
the existing acpi_gbl_root_table_list code is better.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/