Re: [PATCH] OMAP3 PM export chip IDCODE, Production ID and Die ID

[Kevin Hilman <khilman@deeprootsystems.com>]

Eduardo Valentin <eduardo.valentin@nokia.com> writes:

> Hello Tony and Kevin,
>
> On Wed, Oct 14, 2009 at 11:44:07AM +0200, De-Schrijver Peter (Nokia-D/Helsinki) wrote:
>> On Tue, Oct 13, 2009 at 09:51:30PM +0200, ext Kevin Hilman wrote:
>> > On Tue, Oct 13, 2009 at 12:48 PM, Tony Lindgren <tony@atomide.com> wrote:
>> > > * Kevin Hilman <khilman@deeprootsystems.com> [091013 12:09]:
>> > >> "Peter 'p2' De Schrijver" <peter.de-schrijver@nokia.com> writes:
>> > >>
>> > >> > From: De-Schrijver Peter (Nokia-D/Helsinki) <Peter.De-Schrijver@nokia.com>
>> > >> >
>> > >> > This patch exports the OMAP3 IDCODE, Production ID and Die ID to userspace via
>> > >> > sysfs. This can be used to track down silicon specific issues. The info is
>> > >> > exported via sysfs because it should be possible to include this in corematic
>> > >> > dumps.
>> > >> >
>> > >> > Signed-off-by: Peter 'p2' De Schrijver <peter.de-schrijver@nokia.com>
>> > >>
>> > >> Please export these via debugfs.
>> > >
>> > > I don't think we want to export unique chip identifiers by default.
>> > >
>> > 
>> > Right, which is why I suggested using debugfs, which is something that
>> > probably wouldn't be enabled/exported on default production kernels.
>> > 
>> 
>> Which is why I do not want it in debugfs as we log this info in
>> crash reports on devices which might not have debugfs enabled.
>
> As Peter said here, this patch is to identify the chip from user land.
> So it is not really a debug tool, but a identification tool.
>
> Any good reason why we should not have it?
>
> Another thing is, it could go into /proc/cpuinfo. Currently I don't
> see any OMAP related info there, only ARM. Do you guys know why we don't
> put anything in /proc/cpuinfo?

I don't know any reason not to.  This sounds like a good candiate
for OMAP specific details in /proc/cpuinfo.

Kevin