Re: [PATCH v2 0/6] Port Tegra to generic clk framework

[Stephen Warren <swarren@wwwdotorg.org>]

On 07/03/2012 11:38 PM, Prashant Gaikwad wrote:
> On Tuesday 03 July 2012 09:59 PM, Stephen Warren wrote:
>> On 07/03/2012 05:09 AM, Prashant Gaikwad wrote:
>>> This patch set ports Tegra clock code to generic clock framework.
>>>
>>> Depends on
>>> [PATCH v3 0/9] Add clk_prepare/clk_unprepare series
>>> [PATCH] ARM: tegra: Remove second instance of uart clk
>>> [PATCH] dma: tegra: add clk_prepare/clk_unprepare
>>> [PATCH] ARM: tegra: dma: rename driver name for clock to "tegra-apbdma"
>>>
>>> Tested on Ventana and Cardhu
>>>
>>> v2:
>>>   - Rebased on Tegra's for-3.6/common-clk
>> OK, this applies to for-3.6/common-clk without issue now. However, there
>> are still runtime problems:
>>
>> On Tegra20(Ventana), running Tegra's for-next branch with these patches
>> merged in, if I play audio, the sound is much too low a pitch; the I2S
>> clock must be being set too low. Also, if I play the same file twice,
>> once encoded as 48KHz once encoded as 44.1KHz, the pitch is different.
>>
>> On Tegra20(Ventana), running next-20120703 plus a few local patches for
>> regulators plus these common clock patches, then I see the same
>> segfaults during boot that I mentioned before. It seems something in
>> linux-next not in Tegra's for-next branch is breaking the common clock
>> code for us.
> 
> Does these local patches include any changes in clock init table? Is it
> possible to share the changes?

My local dev branch is usually available at:
git://nv-tegra.nvidia.com/user/swarren/linux-2.6 linux-next_common

The local patches are mostly regulator support work-in-progress at the
moment, so nothing to do with clocks, except one minor warning fix.

I see you've posted v3; I assume that means you've found/fixed the
issues - I'll go test.