[PATCH 3.16.y-ckt 059/126] thermal: exynos: Fix unbalanced regulator disable on probe failure

[Krzysztof Kozlowski]

On 09.12.2015 22:59, Luis Henriques wrote:
> On Wed, Dec 09, 2015 at 10:22:37PM +0900, Krzysztof Kozlowski wrote:
>> W dniu 09.12.2015 o 18:37, Luis Henriques pisze:
>>> 3.16.7-ckt21 -stable review patch.  If anyone has any objections, please let me know.
>>>
>>> ------------------
>>>
>>> From: Krzysztof Kozlowski <k.kozlowski at samsung.com>
>>>
>>> commit 824ead03b78403a21449cb7eb153a4344cd3b4c8 upstream.
>>>
>>> During probe if the regulator could not be enabled, the error exit path
>>> would still disable it. This could lead to unbalanced counter of
>>> regulator enable/disable.
>>>
>>> The patch moves code for getting and enabling the regulator from
>>> exynos_map_dt_data() to probe function because it is really not a part
>>> of getting Device Tree properties.
>>>
>>> Acked-by: Lukasz Majewski <l.majewski at samsung.com>
>>> Tested-by: Lukasz Majewski <l.majewski at samsung.com>
>>> Reviewed-by: Alim Akhtar <alim.akhtar at samsung.com>
>>> Signed-off-by: Krzysztof Kozlowski <k.kozlowski at samsung.com>
>>> Fixes: 5f09a5cbd14a ("thermal: exynos: Disable the regulator on probe failure")
>>> Signed-off-by: Eduardo Valentin <edubezval at gmail.com>
>>> [ luis: backported to 3.16: adjusted context ]
>>> Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
>>
>> This patch shouldn't be backported before 4.2. It fixes a bug introduced
>> in 4.2: 5f09a5cbd14a.
>>
>> For kernels <4.2 this bug does not exist but fortunately the patch looks
>> harmless. Anyway still for <4.2 I think it shouldn't be ported.
>>
>> Best regards,
>> Krzysztof
>>
> 
> Thank you Krzysztof.
> 
> I understand that this fixes an issue with commit 5f09a5cbd14a ("thermal:
> exynos: Disable the regulator on probe failure"), which was included in
> 4.2.  However, since this commit was also tagged for stable, it has also
> been backported to some stable kernels (at least 3.16.y-ckt and
> 3.19.y-ckt).
> 
> Would you agree that, for this reason, this patch is in fact applicable to
> the 3.16 (and 3.19) kernel?

Ahh, you are right! The 5f09a5cbd14a was backported so this applies to
backporting as well.

Sorry for the noise.

Best regards,
Krzysztof