linux-kernel - Re: [PATCH] mei: make hdcp and pxp depend on X86 && PCI

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230425115210.GA4151911@angband.pl>
Date:   Tue, 25 Apr 2023 13:52:10 +0200
From:   Adam Borowski <kilobyte@...band.pl>
To:     "Winkler, Tomas" <tomas.winkler@...el.com>
Cc:     Arnd Bergmann <arnd@...db.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "Usyskin, Alexander" <alexander.usyskin@...el.com>
Subject: Re: [PATCH] mei: make hdcp and pxp depend on X86 && PCI

On Tue, Apr 25, 2023 at 04:39:23AM +0000, Winkler, Tomas wrote:
> What is the exact issue you are experiencing, can you add the error message this fixes? 

The problem doesn't trigger in mainline, as DRM_I915 can't be enabled on
!X86; https://github.com/kilobyte/linux branch i915 has a WIP port.  Some
unrelated piece (I suspect ACPI changes, didn't bother to investigate) has
in the meantime grown enough bits to enable prerequsites that trigger:

WARNING: unmet direct dependencies detected for INTEL_MEI_ME
  Depends on [n]: X86 && PCI [=y]
  Selected by [y]:
  - INTEL_MEI_HDCP [=y] && DRM_I915 [=y]
  - INTEL_MEI_PXP [=y] && DRM_I915 [=y]

This doesn't cause an actual build failure, as drivers/misc/Makefile has
> obj-$(CONFIG_INTEL_MEI)         += mei/
which won't try to build mei/hdcp nor mei/pxp despite them being enabled
in the config.

Of all the settings in drivers/misc/mei/Kconfig, only these two (that
reside in their own subdirectories) lack this stanza.  You'd want to
deduplicate the selects one day -- but alas, "select" relations are not as
straightforward as "depends", thus I only did the same as for all other
settings in drivers/misc/mei/Kconfig.

Should I quote this warning in full in the commit message?  I thought that
"make ... depend on ..." already explains enough.

Also, as this lack of dependency is an obvious and self-contained change,
I decided to submit it directly via mei, rather than ask for an ACK then
spam you with every iteration of the patchset.  Should it go via drm/i915
instead?

(I'd also like to apologize for sitting on this patchset so long, my
brain's firmware is faulty :/ ).

> > From: Adam Borowski <kilobyte@...band.pl>
> > 
> > All other MEI configs do so already.  This fixes a Kconfig gripe if I915 gets
> > ported to other archs (such as RISC-V in Intel Horse Creek...).
> > 
> > Signed-off-by: Adam Borowski <kilobyte@...band.pl>
> > ---
> >  drivers/misc/mei/hdcp/Kconfig | 1 +
> >  drivers/misc/mei/pxp/Kconfig  | 1 +
> >  2 files changed, 2 insertions(+)
> > 
> > diff --git a/drivers/misc/mei/hdcp/Kconfig b/drivers/misc/mei/hdcp/Kconfig
> > index 54e1c9526909..83e67771ac6b 100644
> > --- a/drivers/misc/mei/hdcp/Kconfig
> > +++ b/drivers/misc/mei/hdcp/Kconfig
> > @@ -4,6 +4,7 @@
> >  config INTEL_MEI_HDCP
> >  	tristate "Intel HDCP2.2 services of ME Interface"
> >  	select INTEL_MEI_ME
> > +	depends on X86 && PCI
> >  	depends on DRM_I915
> >  	help
> >  	  MEI Support for HDCP2.2 Services on Intel platforms.
> > diff --git a/drivers/misc/mei/pxp/Kconfig b/drivers/misc/mei/pxp/Kconfig
> > index 4029b96afc04..95f2c2470d28 100644
> > --- a/drivers/misc/mei/pxp/Kconfig
> > +++ b/drivers/misc/mei/pxp/Kconfig
> > @@ -5,6 +5,7 @@
> >  config INTEL_MEI_PXP
> >  	tristate "Intel PXP services of ME Interface"
> >  	select INTEL_MEI_ME
> > +	depends on X86 && PCI
> >  	depends on DRM_I915
> >  	help
> >  	  MEI Support for PXP Services on Intel platforms.
> > --
> > 2.40.0

Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Q: Is it ok to combine wired, wifi, and/or bluetooth connections
⢿⡄⠘⠷⠚⠋⠀    in wearable computing?
⠈⠳⣄⠀⠀⠀⠀ A: No, that would be mixed fabric, which Lev19:19 forbids.