Re: [PATCH 3/4] acpi/ghes, efi/cper: Recognize and process CXL Protocol Errors.

From: Ira Weiny
Date: Wed Jan 03 2024 - 17:32:43 EST


Smita Koralahalli wrote:
> On 1/2/2024 9:58 AM, Ira Weiny wrote:
> > Smita Koralahalli wrote:

[snip]

> >> +
> >> + return 0;
> >> +}
> >> diff --git a/include/linux/cxl-event.h b/include/linux/cxl-event.h
> >> index 90d8390a73cb..7ba2dfd6619e 100644
> >> --- a/include/linux/cxl-event.h
> >> +++ b/include/linux/cxl-event.h
> >> @@ -154,11 +154,17 @@ struct cxl_ras_capability_regs {
> >>
> >> struct cxl_cper_rec_data {
> >> struct cxl_cper_event_rec rec;
> >> + struct cxl_ras_capability_regs *cxl_ras;
> >> + int severity;
> >
> > NIT: When I saw this without any addition to event type I wondered if the
> > series would bisect. I see it does because the record is not sent until
> > the next patch. But I wonder if the 2 patches would be best reversed.
>
> You mean to say patch 4 to be 3 and 3 to be 4?
>
> And since I haven't used severity yet, fix it by declaring severity to
> where it is used..

Yes. What you have is not technically wrong but it threw me in review.

>
> >
> > Also, should cxl_ras + severity be put in a protocol error sub-struct?
> > Does severity ever apply to event records?
>
> No, not in any of the component event records. Only place is in "Event
> Record Flags" in Common Event Record Format (Table 8-42).
>
> Addressed in patch 1 to make a sub-struct.

Thanks!
Ira