Quoting Aisheng Dong (2019-02-19 23:07:46) > > From: Rob Herring [mailto:robh@xxxxxxxxxx] > > Sent: Thursday, February 14, 2019 6:18 AM > > On Thu, Jan 24, 2019 at 01:22:45PM +0000, Aisheng Dong wrote: > > > Add a53 and a72 clock id, as there's still no users, we update > > > IMX_LSIO_MEM_CLK base to start from 6 to allow a53 and a72 clock id to > > > be continued with a35 clk. > > > > > > Cc: Stephen Boyd <sboyd@xxxxxxxxxx> > > > Cc: Rob Herring <robh+dt@xxxxxxxxxx> > > > Cc: devicetree@xxxxxxxxxxxxxxx > > > Cc: Shawn Guo <shawnguo@xxxxxxxxxx> > > > Cc: Sascha Hauer <kernel@xxxxxxxxxxxxxx> > > > Cc: Fabio Estevam <fabio.estevam@xxxxxxx> > > > Cc: Michael Turquette <mturquette@xxxxxxxxxxxx> > > > Signed-off-by: Dong Aisheng <aisheng.dong@xxxxxxx> > > > --- > > > v1->v2: > > > * change cpu clock to cpu cluster clock per Rob's suggestion > > > --- > > > include/dt-bindings/clock/imx8-clock.h | 6 ++++-- > > > 1 file changed, 4 insertions(+), 2 deletions(-) > > > > > > diff --git a/include/dt-bindings/clock/imx8-clock.h > > > b/include/dt-bindings/clock/imx8-clock.h > > > index b149e63..dcce744 100644 > > > --- a/include/dt-bindings/clock/imx8-clock.h > > > +++ b/include/dt-bindings/clock/imx8-clock.h > > > @@ -14,10 +14,12 @@ > > > /* CPU */ > > > #define IMX_A35_CLK 1 > > > #define IMX_CPU_CLUSTER_A35_CLK 1 > > > +#define IMX_CPU_CLUSTER_A53_CLK 2 > > > +#define IMX_CPU_CLUSTER_A72_CLK 3 > > > > I still don't get this. How many clock outputs does the clock controller have for > > CPUs? If 3, then this is correct. If it's the same clock controller bits across > > different SoCs, then just name it something like IMX_CPU_CLUSTER_CLK and > > reuse the same ID. > > > > For SCU firmware based platforms like mx8qxp/qm, the clocks are provided > by SCU firmware via SCU firmware call with unique IDs. So it's safe to use a common > Clock IDs file. > > But please ignore patch 3 and 4 first because we still met a few limitations with > current approach due to the device availability may vary a bit across CPUs and > Subsystems. > > We formerly planned to add all new IDs for each SS and dynamically check availability > in driver. That can be done but that may involve a lot effort and may result in more > changes In driver. Also hard to upstream device tree code due to dependency on Clock IDs. > > To relief this situation, we want to move the clock definition into device tree which > can fully decouple the dependency of Clock ID definition from device tree. > And no frequent changes required in clock driver. > > I will send a patch set to do it later. > Ok. I'll take this as a signal to drop these from the clk review queue. Thanks.
