Re: [PATCH v2] ACPI: NUMA: debug invalid unused PXM value for CFMWs

[Jonathan Cameron]

On Fri, 14 Mar 2025 09:38:24 -0400
Gregory Price <gourry@xxxxxxxxxx> wrote:

> On Fri, Mar 14, 2025 at 10:12:26AM +0000, Jonathan Cameron wrote:
> > On Thu, 13 Mar 2025 11:02:37 -0400
> > Gregory Price <gourry@xxxxxxxxxx> wrote:
> >   
> > > On Thu, Mar 13, 2025 at 02:09:07PM +0800, Yuquan Wang wrote:  
> > > > @@ -441,6 +441,11 @@ static int __init acpi_parse_cfmws(union acpi_subtable_headers *header,
> > > >  	start = cfmws->base_hpa;
> > > >  	end = cfmws->base_hpa + cfmws->window_size;
> > > >  
> > > > +	if (srat_disabled()) {
> > > > +		pr_err("SRAT is missing or bad while processing CFMWS.\n");
> > > > +		return -EINVAL;
> > > > +	}
> > > > +    
> > > 
> > > I thought the srat was optional regardless of the presence of a CFMWS.
> > > Is this not the case?  
> > 
> > True in theory, but do we want to support it?
> > 
> > I'd vote no unless someone is shipping such a system and can't fix it up.
> > 
> > Jonathan
> >   
> 
> Well, this is really the patch trying to deal with that I suppose. The
> code here already states its creating 1 node per CFMWS in the absense of
> srat - but this patch just changes that and says "no nodes 4 u".  I
> don't think that's what we want either.

Under this specific set of circumstances, "no nodes 4 u" is
to me a perfectly valid answer.

Jonathan


> 
> ~Gregory