[Intel-gfx] [PATCH 0/1] drm/i915: vlv sideband
Lucas De Marchi
lucas.demarchi at intel.com
Wed Oct 13 15:47:26 UTC 2021
On Wed, Oct 13, 2021 at 01:47:09PM +0300, Jani Nikula wrote:
>On Wed, 13 Oct 2021, Ville Syrjälä <ville.syrjala at linux.intel.com> wrote:
>> On Wed, Oct 13, 2021 at 01:11:58PM +0300, Jani Nikula wrote:
>>> Three main ideas here:
>>>
>>> - vlv sideband only has the name "sideband" in common with the rest of
>>> intel_sideband.[ch]
>>
>> I wouldn't put it like that. There are two actual sideband
>> implementtions in that file:
>> - vlv/chv iosf sideband (vlv_sideband)
>> - lpt/wpt iosf sideband (intel_sbi)
>>
>> And the third thing in that file is the snb+ pcode mailbox stuff,
>> which has nothing to do with sideband.
>
>Fair enough... but no opposition to the splitting out of vlv/chv iosf
>sideband? vlv_sideband.[ch] like here? I'm fine with renaming too.
>
>I can follow up with lpt/wpt iosf split out (intel_sbi.[ch]?) and snb+
>pcode (intel_pcode.[ch]?).
yeah, I think that if we move intel_pcode.[ch] out, then we probably
don't even have to worry about the iosf_* calls for other archs. The
common stuff would be in pcode and the others would be compiled out for
archs that don't have it (i.e. only x86 adds it).
+Siva, who was looking into this iosf abstraction.
Lucas De Marchi
>
>I think we've just put all of them together way back when this was all
>probably bundled in i915_drv.c or something...
>
>
>BR,
>Jani.
>
>
>
>--
>Jani Nikula, Intel Open Source Graphics Center
More information about the Intel-gfx
mailing list