[PATCH 1/2] clk: rockchip: add binding ID for DMC (memory controller) clocks on rk3288

[PATCH 1/2] clk: rockchip: add binding ID for DMC (memory controller) clocks on rk3288

[Doug Anderson]

From: Jeff Chen <cym@rock-chips.com>

The DMC clocks need to be turned off at runtime, so we should have IDs
so we can export them.

Signed-off-by: Jeff Chen <cym@rock-chips.com>
[dianders: split into two patches; adjusted commit msg]
Signed-off-by: Doug Anderson <dianders@chromium.org>
---
 include/dt-bindings/clock/rk3288-cru.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/dt-bindings/clock/rk3288-cru.h b/include/dt-bindings/clock/rk3288-cru.h
index 100a08c..8e19ed5 100644
--- a/include/dt-bindings/clock/rk3288-cru.h
+++ b/include/dt-bindings/clock/rk3288-cru.h
@@ -141,6 +141,10 @@
 #define PCLK_VIO2_H2P		361
 #define PCLK_CPU		362
 #define PCLK_PERI		363
+#define PCLK_DDRUPCTL0		364
+#define PCLK_PUBL0		365
+#define PCLK_DDRUPCTL1		366
+#define PCLK_PUBL1		367
 
 /* hclk gates */
 #define HCLK_GPS		448
-- 
2.1.0.rc2.206.gedb03e5

[PATCH 2/2] clk: rockchip: use clock ID for DMC (memory controller) on rk3288

[Doug Anderson]

From: Jeff Chen <cym@rock-chips.com>

The DMC clocks need to be turned off at runtime.  Use the newly
assigned clock IDs to export them.

Signed-off-by: Jeff Chen <cym@rock-chips.com>
[dianders: split into two patches; adjusted commit msg]
Signed-off-by: Doug Anderson <dianders@chromium.org>
---
 drivers/clk/rockchip/clk-rk3288.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/clk/rockchip/clk-rk3288.c b/drivers/clk/rockchip/clk-rk3288.c
index ad8a27a..2328097 100644
--- a/drivers/clk/rockchip/clk-rk3288.c
+++ b/drivers/clk/rockchip/clk-rk3288.c
@@ -625,10 +625,10 @@ static struct rockchip_clk_branch rk3288_clk_branches[] __initdata = {
 	GATE(PCLK_TIMER, "pclk_timer", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 1, GFLAGS),
 	GATE(PCLK_I2C0, "pclk_i2c0", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 2, GFLAGS),
 	GATE(PCLK_I2C2, "pclk_i2c2", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 3, GFLAGS),
-	GATE(0, "pclk_ddrupctl0", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 14, GFLAGS),
-	GATE(0, "pclk_publ0", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 15, GFLAGS),
-	GATE(0, "pclk_ddrupctl1", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 0, GFLAGS),
-	GATE(0, "pclk_publ1", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 1, GFLAGS),
+	GATE(PCLK_DDRUPCTL0, "pclk_ddrupctl0", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 14, GFLAGS),
+	GATE(PCLK_PUBL0, "pclk_publ0", "pclk_cpu", 0, RK3288_CLKGATE_CON(10), 15, GFLAGS),
+	GATE(PCLK_DDRUPCTL1, "pclk_ddrupctl1", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 0, GFLAGS),
+	GATE(PCLK_PUBL1, "pclk_publ1", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 1, GFLAGS),
 	GATE(0, "pclk_efuse_1024", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 2, GFLAGS),
 	GATE(PCLK_TZPC, "pclk_tzpc", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 3, GFLAGS),
 	GATE(PCLK_UART2, "pclk_uart2", "pclk_cpu", 0, RK3288_CLKGATE_CON(11), 9, GFLAGS),
-- 
2.1.0.rc2.206.gedb03e5

Re: [PATCH 1/2] clk: rockchip: add binding ID for DMC (memory controller) clocks on rk3288

[Heiko Stübner]

Am Dienstag, 25. November 2014, 16:13:02 schrieb Doug Anderson:
> From: Jeff Chen <cym@rock-chips.com>
> 
> The DMC clocks need to be turned off at runtime, so we should have IDs
> so we can export them.
> 
> Signed-off-by: Jeff Chen <cym@rock-chips.com>
> [dianders: split into two patches; adjusted commit msg]
> Signed-off-by: Doug Anderson <dianders@chromium.org>

both patches look ok to me and I'll probably pick them up tomorrow.

Just as handling question, do you plan on sending dts changes using these 
clocks in time for 3.19 [i.e. in the next couple of days]?

We have currently three clock patches/series adding new clock ids [this one, 
the mmc phases and Sonny's i2s_clkout] and I'm trying to dertermine if I need 
a shared branch for those.

Currently it doesn't look like the others will provide dts patches using the 
clocks in time, so it looks like I can skip the shared branch and we can use 
the clock ids regularly after the merge-window. Some true for your clocks?


Thanks
Heiko

Re: [PATCH 1/2] clk: rockchip: add binding ID for DMC (memory controller) clocks on rk3288

[Doug Anderson]

Heiko,

On Tue, Nov 25, 2014 at 4:45 PM, Heiko Stübner <heiko@sntech.de> wrote:
> Am Dienstag, 25. November 2014, 16:13:02 schrieb Doug Anderson:
>> From: Jeff Chen <cym@rock-chips.com>
>>
>> The DMC clocks need to be turned off at runtime, so we should have IDs
>> so we can export them.
>>
>> Signed-off-by: Jeff Chen <cym@rock-chips.com>
>> [dianders: split into two patches; adjusted commit msg]
>> Signed-off-by: Doug Anderson <dianders@chromium.org>
>
> both patches look ok to me and I'll probably pick them up tomorrow.

OK, thanks!


> Just as handling question, do you plan on sending dts changes using these
> clocks in time for 3.19 [i.e. in the next couple of days]?

There's pretty much zero chance.  Maybe negative.


> We have currently three clock patches/series adding new clock ids [this one,
> the mmc phases and Sonny's i2s_clkout] and I'm trying to dertermine if I need
> a shared branch for those.
>
> Currently it doesn't look like the others will provide dts patches using the
> clocks in time, so it looks like I can skip the shared branch and we can use
> the clock ids regularly after the merge-window. Some true for your clocks?

Yup, no problem for me.  I provided this as two patches to you just in
case it was convenient, since I saw that as feedback in previous
submissions.

Thanks!

-Doug

Re: [PATCH 1/2] clk: rockchip: add binding ID for DMC (memory controller) clocks on rk3288

[Heiko Stübner]

Am Dienstag, 25. November 2014, 16:49:08 schrieb Doug Anderson:
> Heiko,
> 
> On Tue, Nov 25, 2014 at 4:45 PM, Heiko Stübner <heiko@sntech.de> wrote:
> > Am Dienstag, 25. November 2014, 16:13:02 schrieb Doug Anderson:
> >> From: Jeff Chen <cym@rock-chips.com>
> >> 
> >> The DMC clocks need to be turned off at runtime, so we should have IDs
> >> so we can export them.
> >> 
> >> Signed-off-by: Jeff Chen <cym@rock-chips.com>
> >> [dianders: split into two patches; adjusted commit msg]
> >> Signed-off-by: Doug Anderson <dianders@chromium.org>
> > 
> > both patches look ok to me and I'll probably pick them up tomorrow.
> 
> OK, thanks!

I've now added both patches to my clk branch.


> > We have currently three clock patches/series adding new clock ids [this
> > one, the mmc phases and Sonny's i2s_clkout] and I'm trying to dertermine
> > if I need a shared branch for those.
> > 
> > Currently it doesn't look like the others will provide dts patches using
> > the clocks in time, so it looks like I can skip the shared branch and we
> > can use the clock ids regularly after the merge-window. Some true for
> > your clocks?
> Yup, no problem for me.  I provided this as two patches to you just in
> case it was convenient, since I saw that as feedback in previous
> submissions.

Yep, I think it is the sanest of doing it in one way all the time, so we have 
the option of easily creating shared branches depending on what is needed.


Heiko