From mboxrd@z Thu Jan  1 00:00:00 1970
From: ivo.g.dimitrov.75@gmail.com (Ivaylo Dimitrov)
Date: Thu, 28 Apr 2016 23:43:07 +0300
Subject: [PATCH 2/2] [media] ir-rx51: Fix build after multiarch changes
 broke it
In-Reply-To: <20160428202248.GG5995@atomide.com>
References: <1461714709-10455-1-git-send-email-tony@atomide.com>
 <1461714709-10455-3-git-send-email-tony@atomide.com>
 <572266AF.9020601@gmail.com> <20160428202248.GG5995@atomide.com>
Message-ID: <572275DB.8090300@gmail.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

??,

On 28.04.2016 23:22, Tony Lindgren wrote:
> * Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com> [160428 12:39]:
>> On 27.04.2016 02:51, Tony Lindgren wrote:
>>
>> omap_dm_timer_request_specific always fails with message "Please use
>> omap_dm_timer_request_by_cap/node()" with DT boot.
>>
>> I hacked the code to use omap_dm_timer_request_by_cap(OMAP_TIMER_HAS_PWM)
>> and it seems to use the correct timer (IR LED blinks, checked with the
>> camera of Samsung S4 mini), but it doesn't actually control either of the TV
>> sets here. The same SW(pierogi) controls them when device is booted to stock
>> kernel. However, this seems another problem not related to the patch.
>
> OK thanks for testing, I'll apply the pdata patch then.
>
> I assume you'll post a separate fix for the request_by_cap
> driver change?
>

Well, it was a hack, it just happens that the first matched timer is 
GPT9, I think we should aim for a proper solution (request_by_node()).

Shall I prepare a patch that gets the timer from the DT?

Ivo