[PATCH] platform/x86: Export LPC attributes for the system SPI chip

[Mika Westerberg]

On Fri, May 08, 2020 at 05:27:12PM +0000, Mario.Limonciello at dell.com wrote:
> > -----Original Message-----
> > From: platform-driver-x86-owner at vger.kernel.org <platform-driver-x86-
> > owner at vger.kernel.org> On Behalf Of Mika Westerberg
> > Sent: Friday, May 8, 2020 3:20 AM
> > To: Limonciello, Mario
> > Cc: hughsient at gmail.com; platform-driver-x86 at vger.kernel.org; linux-
> > security-module at vger.kernel.org
> > Subject: Re: [PATCH] platform/x86: Export LPC attributes for the system
> > SPI chip
> > 
> > 
> > [EXTERNAL EMAIL]
> > 
> > On Thu, May 07, 2020 at 08:03:21PM +0000, Mario.Limonciello at dell.com
> > wrote:
> > > > -----Original Message-----
> > > > From: Richard Hughes <hughsient at gmail.com>
> > > > Sent: Thursday, May 7, 2020 2:49 PM
> > > > To: Limonciello, Mario
> > > > Cc: Platform Driver; linux-security-module;
> > mika.westerberg at linux.intel.com
> > > > Subject: Re: [PATCH] platform/x86: Export LPC attributes for the
> > system SPI
> > > > chip
> > > >
> > > >
> > > > [EXTERNAL EMAIL]
> > > >
> > > > On Thu, 7 May 2020 at 20:22, <Mario.Limonciello at dell.com> wrote:
> > > > > By default the driver exposes SPI serial flash contents as read-
> > only but it
> > > > can
> > > > > be changed from kernel command line, passing “intel-
> > spi.writeable=1”.
> > > >
> > > > Ahh, that was the bit I didn't know; having the SPI as readonly by
> > > > default is certainly a good idea, and probably sane enough to enable
> > > > for Fedora/RHEL as you still need to "do" something manual to enable
> > > > SPI writing. I guess I can add my securityfs additions to
> > > > intel-spi-pci.c with Mikas approval.
> > > >
> > > > Richard
> > >
> > > Mika,
> > >
> > > Since you're being joined into the thread late, here is the context:
> > > https://www.spinics.net/lists/platform-driver-x86/msg21646.html
> > 
> > Thanks for the information. I actually prefer that this would be in a
> > separate driver because I do not want distros to enable intel-spi just
> > for this. It is really only meant for special setups where firmware
> > upgrade/access flow has been thoroughly tested.
> 
> Mika,
> 
> Thanks for those comments and context on that driver.  Considering this,
> what do you think about as part of this new driver, moving the list of
> supported IDs in there to something that can be sourced by both drivers?
> I think it should help avoid having to keep the two lists fully in sync
> as new silicon comes out.

Hi,

Since this is binding to the LPC/eSPI device I think you may be adding
this functionality to drivers/mfd/lpc_ich.c. It is also filling the data
for intel-spi so it is natural place to re-use the code. I actually
think it already has some of these PCI IDs.