From mboxrd@z Thu Jan  1 00:00:00 1970
From: hdegoede@redhat.com (Hans de Goede)
Date: Fri, 25 Sep 2015 10:00:23 +0200
Subject: [linux-sunxi] Re: [PATCH 1/3] Revert "ARM: dts: sun8i: Add new
 sun8i-a33-q8-tzx-723q4.dts file"
In-Reply-To: <20150925064258.GA4684@lukather>
References: <1443108234-14837-1-git-send-email-hdegoede@redhat.com>
 <20150925064258.GA4684@lukather>
Message-ID: <5604FF17.7000005@redhat.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi,

On 25-09-15 08:42, Maxime Ripard wrote:
> On Thu, Sep 24, 2015 at 05:23:52PM +0200, Hans de Goede wrote:
>> The tzx-723q[a]4 board variant of A33 based Q8 formfactor tablets was
>> given its own dts file because the mmc would not work with a drive
>> strength of 30mA, Android seemed to be using a lower drive-strength
>> and doing the same with the upstream kernel seemed to fix this issue.
>>
>> While working on a similar fix for u-boot I noticed that Android did
>> not use a lower drive strength for mmc0, but for mmc1, so mmc0 should
>> work fine with a drive strength of 30mA.
>>
>> Further investigation has shown that Allwinner is using slightly
>> tweaked clock sample phase settings in newer Android kernels, and
>> that updating the mainstream sunxi mmc driver to these settings fixes
>> this.
>>
>> With this fixed we no longer need a dts file specifically for these
>> pcb varants -> drop this dts file.
>
> I dropped the patch.
>
>> This reverts commit fb61c559a810f1487892b83d2d0606af1d020318.
>
> Checkpatch prints an error about this one. Make sure you have the
> right commit format.

<sigh> going off-topic here (very much so), but that line is actually
auto-generated by "git revert" it would be nice if the git and checkpatch
people got get together and agree on a common format here.

Regards,

Hans