Re: [PATCH] Add support of NVDIMM memory error notification in ACPI 6.2

[Kani, Toshimitsu]

On Thu, 2017-06-08 at 11:37 -0600, Toshi Kani wrote:
> On Thu, 2017-06-08 at 10:34 -0700, Dan Williams wrote:
> > On Thu, Jun 8, 2017 at 10:30 AM, Linda Knippers <linda.knippers@hpe
> > .c
> > om> wrote:
> > [..]
> > > Wasn't Dan concerned about how the OS can know whether the FW
> > > supports that bit in the Start ARS?
> > > 
> > > The Query ARS Capabilities DSM has a bit that tells the OS
> > > whether the platform supports the notification and the point of
> > > the notification was to tell the OS it could do a Start ARS with
> > > bit 1 set.ÂÂOf course, if you get the notification then that
> > > means the platform has the capability to deliver it, but it might
> > > not hurt to check the flag from the Query Capabilities bit.
> > 
> > Good point, yes, I think it is safe to assume that a BIOS that
> > claims to support un-correctable error notification also supports
> > this Start ARS flag.
> 
> Yes, ACPI 6.2, section 9.20.7.2, defines that:
> 
> ÂÂÂÂÂÂUpon receiving the notification, the OSPM may decide to issue
> ÂÂÂÂÂÂa Start ARS with Flags Bit [1] set to prepare for the retrieval
> ÂÂÂÂÂÂof existing records and issue the Query ARS Status function to
> ÂÂÂÂÂÂretrieve the records.
> 
> So,ÂI believe it is safe to assume that BIOS supporting 0x81 also
> supports flags Bit [1].ÂÂSorry, this is what I should have said in my
> previous email...

To reiterate my thinking, I believe the statement above clarifies that
the OS can assume BIOS support of Flags Bit[1] upon receiving a 0x81
notification.  Since BIOS may also support Flags Bit[1] without
supporting this 0x81 (in which case I do not know how to detect it, but
BIOS should simply ignore this bit when not supporting it), I am not
going to add a check/restriction that 0x81 support is necessary to set
Bit[1] in the scan function.

Thanks,
-Toshi