On Tue, 2017-08-15 at 17:50 +0200, Borislav Petkov wrote:
> On Tue, Aug 15, 2017 at 03:35:51PM +0000, Kani, Toshimitsu wrote:
> > ghes_edac instantiates an mci as a pseudo device representing a
> > GHES error source. Each error source associates with all DIMMs,
> > and may report errors independently. As ghes_edac is an GHES
> > error-reporting wrapper to edac, this abstraction makes sense.
>
> Bullshit.
>
> An MCI is a memory controller descriptor. That doesn't fit the GHES
> platform devices that get probed. GHES platform device != MCI. How
> many times do I need to say this for it to get through to you?
Right, but it has to be a "pseudo" device for ghes_edac. There is no
memory controller info available. A single mci does not make it a real
memory controller, either.
> > I do not see a problem in having counters for each GHES error
> > source.
>
> And the error counters of that "simulated" mci get incremented
> depending on which pointer gets passed in from GHES? More bullshit.
>
> > This is just statistics info, and ghes_edac does not expect any OS
> > action from the counters.
>
> So let me know if you don't want to do it and rather would prefer to
> pointlessly debate. I certainly don't want to waste my time debating.
Yes, ghes_edac refactoring like this should be considered a separate
item. My patchset is aimed to introduce a platform-check to attach
ghes_edac on supported platforms.
Thanks,
-Toshi
��.n��������+%������w��{.n�����{�����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
