[PATCH] drm/msm: update adreno firmware path in MODULE_FIRMWARE

Rob Clark robdclark at gmail.com
Thu Nov 23 17:43:42 UTC 2017


On Thu, Nov 23, 2017 at 12:09 PM, Ben Hutchings <ben at decadent.org.uk> wrote:
> On Thu, 2017-11-23 at 16:09 +0100, Nicolas Dechesne wrote:
>> On Thu, Nov 23, 2017 at 3:54 PM, Rob Clark <robdclark at gmail.com> wrote:
>> > On Thu, Nov 23, 2017 at 5:32 AM, Nicolas Dechesne
>> > <nicolas.dechesne at linaro.org> wrote:
>> > > The preferred location for Adreno firmware files is now in qcom/ subfolder,
>> > > especially now that we are adding some of them in linux-firmware.
>> > >
>> > > Reported-by: Ben Hutchings <ben at decadent.org.uk>
>> > > Signed-off-by: Nicolas Dechesne <nicolas.dechesne at linaro.org>
>> >
>> > Thanks, I was wondering if we should perhaps list both old and new
>> > paths?  I'm not sure, maybe we don't need to care about dracut or
>> > initrd generation for the legacy case (since mostly there you are
>> > using fastboot).
>>
>> I've been going back and forth on that too. and i decided to ignore
>> the legacy paths... alternatively we could use the legacy paths for
>> a3xx and the new path for a5xx since we have links for a3xx files...
>> but i thought it was too much noise..
>>
>> >
>> > Also, I noticed we are missing a few a5xx fw files, but perhaps that
>> > should be fixed with a separate patch.
>>
>> as you want. I can include them if you prefer and resend.
>
> Whenever I've added MODULE_FIRMWARE information I've listed only the
> first-choice firmware paths.
>
> initramfs-tools will warn when including a module if some of the listed
> firmware is not available, so listing multiple paths for the same
> firmware will likely mean it always warns about one of them.
>

Ok, that sounds like good reason to not to list legacy paths.

BR,
-R


More information about the dri-devel mailing list