On Mon, Oct 21, 2013 at 11:47:32AM +0100, Sebastian Hesselbarth wrote:

On 10/19/2013 01:37 PM, Andrew Lunn wrote:

quoted

Add a platform driver definition to instantiate the dove cpufreq
driver. Also indicate the ARCH has cpufreq support, so allowing the
cpufreq framework to be enabled.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---
 arch/arm/Kconfig                       |  1 +
 arch/arm/mach-dove/board-dt.c          |  2 ++
 arch/arm/mach-dove/common.c            | 36 ++++++++++++++++++++++++++++++++++
 arch/arm/mach-dove/common.h            |  1 +
 arch/arm/mach-dove/include/mach/dove.h |  1 +
 5 files changed, 41 insertions(+)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1ad6fb6..7744415 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig

@@ -495,6 +495,7 @@ config ARCH_IXP4XX

 config ARCH_DOVE
 	bool "Marvell Dove"
+	select ARCH_HAS_CPUFREQ
 	select ARCH_REQUIRE_GPIOLIB
 	select CPU_PJ4
 	select GENERIC_CLOCKEVENTS

diff --git a/arch/arm/mach-dove/board-dt.c b/arch/arm/mach-dove/board-dt.c
index 49f72a8..98202de 100644
--- a/arch/arm/mach-dove/board-dt.c
+++ b/arch/arm/mach-dove/board-dt.c

@@ -70,6 +70,8 @@ static void __init dove_dt_init(void)
 	/* Setup clocks for legacy devices */
 	dove_legacy_clk_init();

+	dove_cpufreq_init();
+

What ever the outcome of "DT: blessing or curse" discussion at ELCE will
be, are there any plans to probe the cpufreq/cpuidle drivers directly
from DT? Someday, we want to get rid of .machine_init.

Hi Sebastian

This has already been discussed once for cpufreq drivers. I triggered
it with the kirkwood cpufreq driver. The outcome was that not
everything needs to the DT, platform drivers are still OK for things
which are not really devices. But as you say, the outcome from
Edinburgh might repeal that decision.

quoted

+	[1] = {
+		.start  = DOVE_PMU_PHYS_BASE + 0x8000,
+		.end    = DOVE_PMU_PHYS_BASE + 0x8004,

BASE + 0x8004 - 1 ?

Ah, yes, was thinking DT.

See you soon,

    Andrew