On Thu, Dec 08, 2016 at 06:07:02PM +0100, Sylwester Nawrocki wrote: > On 12/08/2016 05:44 PM, Krzysztof Kozlowski wrote: > > On Thu, Dec 08, 2016 at 12:01:54PM +0100, Marek Szyprowski wrote: > >> Move assigned clocks properties from sound node to audio subsystem clock > >> controller node. This way clocks topology and rates are set just after > >> probing audio clocks controller. Leaving those properties under sound > >> node doesn't guarantee to configure them before they are being used > >> (for example i2s hardware module can be probed in parallel and it also > >> require proper audio clocks configuration). > >> > >> Signed-off-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx> > >> --- > >> arch/arm/boot/dts/exynos4412-odroid-common.dtsi | 20 ++++++++++---------- > >> 1 file changed, 10 insertions(+), 10 deletions(-) > > > > Looks correct, for the reference: > > Reviewed-by: Krzysztof Kozlowski <krzk@xxxxxxxxxx> > > > > Does this fixes any encountered issue (real one)? I wonder whether this > > should go to fixes or not. > > With current code there is no issues, and in fact there should be no > such timing dependencies in DT to rely on. However, the binding convention > is to put assigned-clock* properties in a node of a device they belong > to and in this case the 'clock_audss' is more appropriate than the 'sound' > node. This patch doesn't fix any bug with current code but it might save > us some trouble with future changes. > Thanks for explanation. I am not planning any pull requests for this cycle so the patch will have to wait till the end of 4.10-rc1 merge window. Best regards, Krzysztof -- To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html