Re: [PATCH v2] dell-wmi: process Dell Instant Launch hotkey on Dell Vostro V131

[MichaÅ KÄpieÅ]

> > > MichaÅ, can you prepare new (v3) version of this patch? Now required
> > > acpi video changes are included and so dell-wmi changes should go to...
> > > To finally fix this keypress bug on Dell Vostro V131 machine.
> > 
> > I keep this on my to-do list, but the updated patch will depend on the
> > final version of the SMBIOS API rework, so I guess there is little point
> > in posting it now as that API is subject to change.  But rest assured
> > that as soon as the final version of the API rework series (which I have
> > yet to prepare, of course) gets applied by Darren, I will post a v3 of
> > this patch - after all, it is the very reason I am working on the API
> > rework.
> 
> There is still need to patch dell-wmi.c? And this change does not depend
> on another SMBIOS change (in dell-laptop), right?

Well, back in December, you wrote [1]:

> This patch is not enough for enabling 0xe025 key on that Vostro machine.
> Some extra SMBIOS call is needed, without them ACPI will not send WMI
> keypress event.
> 
> (...)
> 
> Maybe now it could make sense to unify Dell SMBIOS API in kernel and
> move common functions to one place and let drivers to use just common
> functions. According to older Dell ACPI WMI documentation in DMI is bit
> which specify if BIOS support SMBIOS via WMI or not.
> 
> At least I think this one patch should not be included into kernel until
> there will be full support for 0xe025 key (adding that SMBIOS call).

>From the above I understood that first you want to unify the Dell SMBIOS
API used throughout the kernel (that's currently in progress), so that
it can then be used in dell-wmi as well to perform the SMBIOS call
needed on the Vostro V131.

If you want me to just rework the patch so that it doesn't introduce a
quirk structure, I recalled another reason to use it after all: there
are other Dell laptops which require the special SMI for enabling WMI
events, but report the Dell Instant Launch Hotkey using a different WMI
event code [2].  So I'd say that two separate issues should be fixed
using DMI matching:

  * whether the special SMBIOS call is required,
  * whether the 0x0e25 event should be translated into a keypress.

A quirk structure looks like an elegant way to deal with this.

Could you please advise how you would like me to proceed with this?

[1] http://www.spinics.net/lists/platform-driver-x86/msg07845.html
[2] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1205791/comments/12

-- 
Best regards,
MichaÅ KÄpieÅ