Skip to contentSkip to conversation

clocksource: time-efm32: Select CLKSRC_MMIO

PATCH In review Axel Lin 3 participants 7 messages Nov 20, 2013 – Dec 16, 2013 Clocksource, clockevent drivers

lore
Share & tools
lore

Axel Lin

author
#0 lore raw

PATCH 1/1 clocksource: time-efm32: Select CLKSRC_MMIO

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

---
This fixes build error when COMPILE_TEST=y.
drivers/built-in.o: In function `efm32_clocksource_init':
drivers/clocksource/time-efm32.c:162: undefined reference to clocksource_mmio_init' drivers/built-in.o: In function efm32_timer_init':
drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up'
make: *** [vmlinux] Error 1

 drivers/clocksource/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +1 -0
index bdb953e..99763eb 100644
@@ -75,6 +75,7 @@ config CLKSRC_DBX500_PRCMU_SCHED_CLOCK
 config CLKSRC_EFM32
 	bool "Clocksource for Energy Micro's EFM32 SoCs" if !ARCH_EFM32
 	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
+	select CLKSRC_MMIO
 	default ARCH_EFM32
 	help
 	  Support to use the timers of EFM32 SoCs as clock source and clock
Signature · 1 line
1.8.1.2
Acked-by on v1 1/1
#1 lore raw
Quoted 4 lines · Axel Lin

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

Good catch. I selected CLKSRC_MMIO from MACH_EFM32 (still out-of-tree) but of course this is wrong.

Thanks
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
#2 lore raw
In reply to Axel Lin
Quoted 20 lines · Axel Lin

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

---
This fixes build error when COMPILE_TEST=y.
drivers/built-in.o: In function `efm32_clocksource_init':
drivers/clocksource/time-efm32.c:162: undefined reference to clocksource_mmio_init' drivers/built-in.o: In function efm32_timer_init':
drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up'
make: *** [vmlinux] Error 1

 drivers/clocksource/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +0 -0
index bdb953e..99763eb 100644

drivers/clocksource/Kconfig config CLKSRC_DBX500_PRCMU_SCHED_CLOCK · line 79

Quoted 5 lines
drivers/clocksource/Kconfig +1 -0
@@ -75,6 +75,7 @@ config CLKSRC_DBX500_PRCMU_SCHED_CLOCK
 config CLKSRC_EFM32
 	bool "Clocksource for Energy Micro's EFM32 SoCs" if !ARCH_EFM32
 	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
+	select CLKSRC_MMIO
 	default ARCH_EFM32

I'd like to send support for arch/arm/mach-efm32 for 3.14. Formerly I had ARCH_EFM32 select CLKSRC_MMIO, but having it here is obviously the correct thing to do. So my possibilities are:

a) I take this patch and apply mach-efm32 on top of it; or
b) you take it early enough for me into 3.13-rc to base mach-efm32 on it; or
c) I send mach-efm32 with ARCH_EFM32 selecting CLKSRC_MMIO and drop this select later; or
d) we life with the build failure (i.e. drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up' and the same for clocksource_mmio_init in a different line) until this patch and mine come together.

I don't like d), but don't have a clear favorite between a), b) and c). What do you think? Probably a) is the easiest for all if there are no merge conflicts (I don't expect any). In this case I'd need Daniel's or Thomas' ack.

Thanks
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
#3 lore raw
Quoted 30 lines · Uwe Kleine-König
Quoted 20 lines · Axel Lin

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

---
This fixes build error when COMPILE_TEST=y.
drivers/built-in.o: In function `efm32_clocksource_init':
drivers/clocksource/time-efm32.c:162: undefined reference to clocksource_mmio_init' drivers/built-in.o: In function efm32_timer_init':
drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up'
make: *** [vmlinux] Error 1

  drivers/clocksource/Kconfig | 1 +
  1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +0 -0
index bdb953e..99763eb 100644

drivers/clocksource/Kconfig config CLKSRC_DBX500_PRCMU_SCHED_CLOCK · line 79

Quoted 5 lines
drivers/clocksource/Kconfig +1 -0
@@ -75,6 +75,7 @@ config CLKSRC_DBX500_PRCMU_SCHED_CLOCK
  config CLKSRC_EFM32
  	bool "Clocksource for Energy Micro's EFM32 SoCs" if !ARCH_EFM32
  	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
+	select CLKSRC_MMIO
  	default ARCH_EFM32

I'd like to send support for arch/arm/mach-efm32 for 3.14. Formerly I had ARCH_EFM32 select CLKSRC_MMIO, but having it here is obviously the correct thing to do. So my possibilities are:

Quoted 3 lines

a) I take this patch and apply mach-efm32 on top of it; or
b) you take it early enough for me into 3.13-rc to base mach-efm32 on it; or

I sent the PR with this patch. If everything is ok, it should be in the next -rc.

Quoted 15 lines

c) I send mach-efm32 with ARCH_EFM32 selecting CLKSRC_MMIO and drop this select later; or
d) we life with the build failure (i.e. drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up' and the same for clocksource_mmio_init in a different line) until this patch and mine come together.

I don't like d), but don't have a clear favorite between a), b) and c). What do you think? Probably a) is the easiest for all if there are no merge conflicts (I don't expect any). In this case I'd need Daniel's or Thomas' ack.

Thanks
Uwe

Signature · 5 lines
 <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
#4 lore raw

Hello Daniel,

Quoted 31 lines · Daniel Lezcano
Quoted 30 lines · Uwe Kleine-König
Quoted 20 lines · Axel Lin

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

---
This fixes build error when COMPILE_TEST=y.
drivers/built-in.o: In function `efm32_clocksource_init':
drivers/clocksource/time-efm32.c:162: undefined reference to clocksource_mmio_init' drivers/built-in.o: In function efm32_timer_init':
drivers/clocksource/time-efm32.c:274: undefined reference to `clocksource_mmio_readl_up'
make: *** [vmlinux] Error 1

 drivers/clocksource/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +0 -0
index bdb953e..99763eb 100644

drivers/clocksource/Kconfig config CLKSRC_DBX500_PRCMU_SCHED_CLOCK · line 79

Quoted 5 lines
drivers/clocksource/Kconfig +1 -0
@@ -75,6 +75,7 @@ config CLKSRC_DBX500_PRCMU_SCHED_CLOCK
 config CLKSRC_EFM32
 	bool "Clocksource for Energy Micro's EFM32 SoCs" if !ARCH_EFM32
 	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
+	select CLKSRC_MMIO
 	default ARCH_EFM32

I'd like to send support for arch/arm/mach-efm32 for 3.14. Formerly I had ARCH_EFM32 select CLKSRC_MMIO, but having it here is obviously the correct thing to do. So my possibilities are:

Quoted 6 lines
Quoted 3 lines

a) I take this patch and apply mach-efm32 on top of it; or
b) you take it early enough for me into 3.13-rc to base mach-efm32 on it; or

I sent the PR with this patch. If everything is ok, it should be in the next -rc.

ok. Thanks for your quick reply.

It would be great if your tree were included in linux-next. Don't you think this is feasible?

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
#5 lore raw
Quoted 7 lines · Uwe Kleine-König

Hello Daniel,

Quoted 4 lines · Daniel Lezcano
Quoted 3 lines · Uwe Kleine-König
Quoted 2 lines · Axel Lin

The time-efm32 driver uses the clocksource MMIO functions. Thus it needs to select CLKSRC_MMIO in Kconfig.

[ ... ]

Quoted 6 lines
Quoted 2 lines

I sent the PR with this patch. If everything is ok, it should be in the next -rc.

ok. Thanks for your quick reply.

It would be great if your tree were included in linux-next. Don't you think this is feasible?

Well, yes it could be a good idea. But if you are suggesting to use linux-next for the fixes, that should be already taken into account with the #auto-latest branch of tip.

Signature · 5 lines
 <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
#6 lore raw

Hello Daniel,

Quoted 5 lines · Daniel Lezcano
Quoted 4 lines · Uwe Kleine-König
Quoted 2 lines · Daniel Lezcano

I sent the PR with this patch. If everything is ok, it should be in the next -rc.

ok. Thanks for your quick reply.

It seems to have missed 3.13-rc4 btw :-(

Quoted 6 lines
Quoted 2 lines

It would be great if your tree were included in linux-next. Don't you think this is feasible?

Well, yes it could be a good idea. But if you are suggesting to use linux-next for the fixes, that should be already taken into account with the #auto-latest branch of tip.

I might be wrong, but it seems that the window when the patches are in your tree but not yet in tip is big. At least I hit that window I think three times already.

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |