2015-05-19 11:59 GMT+02:00 Daniel Lezcano [off-list ref]:

On 05/19/2015 11:44 AM, Maxime Coquelin wrote:

quoted

2015-05-19 11:06 GMT+02:00 Daniel Lezcano [off-list ref]:

quoted

On 05/19/2015 10:55 AM, Maxime Coquelin wrote:

quoted

2015-05-19 10:16 GMT+02:00 Daniel Lezcano [off-list ref]:

quoted

On 05/18/2015 04:03 PM, Maxime Coquelin wrote:

quoted

2015-05-18 15:10 GMT+02:00 Daniel Lezcano [off-list ref]:

quoted

On 05/09/2015 09:53 AM, Maxime Coquelin wrote:

quoted

STM32 MCUs feature 16 and 32 bits general purpose timers with
prescalers.
The drivers detects whether the time is 16 or 32 bits, and applies a
1024 prescaler value if it is 16 bits.

Reviewed-by: Linus Walleij <redacted>
Tested-by: Chanwoo Choi <redacted>
Signed-off-by: Maxime Coquelin <redacted>
---
     drivers/clocksource/Kconfig       |   8 ++
     drivers/clocksource/Makefile      |   1 +
     drivers/clocksource/timer-stm32.c | 184
++++++++++++++++++++++++++++++++++++++
     3 files changed, 193 insertions(+)
     create mode 100644 drivers/clocksource/timer-stm32.c

diff --git a/drivers/clocksource/Kconfig

b/drivers/clocksource/Kconfig
index bf9364c..2443520 100644

--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig

@@ -106,6 +106,14 @@ config CLKSRC_EFM32
             Support to use the timers of EFM32 SoCs as clock source

and
clock
             event device.

+config CLKSRC_STM32
+       bool "Clocksource for STM32 SoCs" if !ARCH_STM32
+       depends on OF && ARM && (ARCH_STM32 || COMPILE_TEST)

Are the interactive bool and the 'COMPILE_TEST' necessary ?

The interactive bool is necessary if we want to be able to
select/deselect it in COMPILE_TEST configuration.
And personnaly, I think COMPILE_TEST use makes sense.

Note that other timer drivers are doing the same thing today
(CLKSRC_EFM32, SH_TIMER_CMT, EM_TIMER_STI...).

Do you have a specific concern regarding COMPILE_TEST?

Actually, we try to keep the timer selection non-interactive and let
the
platform's Kconfig to select the timer.

Ok.

quoted

I like when the code is consistent. The COMPILE_TEST was introduced and
created a precedence. I would like to get rid of the interactive timer
selection but I did not have time to go through this yet.

Indeed, consistency is important.
On my side, I don't have a strong opinion regarding the COMPILE_TEST
thing.
IMHO, it is more a subsystem's maintainer choice.

So, if as a maintainer you don't use it and prefer not supporting it,
I'm fine to provide you a new version without COMPILE_TEST.
Doing that, the interactive selection will disappear too.

I can provide you a new version this evenning.

Ok, great.

Is the below Kconfig entry fine for you?

config CLKSRC_STM32
     def_bool y if ARCH_STM32
     select CLKSRC_MMIO

config CLKSRC_STM32
        bool
        select CLKSRC_MMIO

and in the arch/arm/mach-stm32/Kconfig add select CLKSRC_STM32

Ok, I will send a patch for arch/arm/Kconfig, as Arnd already applied
the one intruducing ARCH_STM32.

Thanks,
Maxime

quoted

Best regards,
Maxime

quoted

Thanks
   -- Daniel

quoted

quoted

--
   <http://www.linaro.org/> Linaro.org │ Open source software for ARM
SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

--
  <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

--
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog