Re: [PATCH v4 13/13] ARM: mvebu: register the cpuidle driver for the Armada XP SoCs
From: Gregory CLEMENT <hidden>
Date: 2014-02-19 16:52:31
Also in:
linux-arm-kernel
On 19/02/2014 17:46, Thomas Petazzoni wrote:
Dear Gregory CLEMENT, On Thu, 13 Feb 2014 18:33:36 +0100, Gregory CLEMENT wrote:quoted
The cpuidle is a platform driver so we have to register the device during the initialization of the boards. Signed-off-by: Gregory CLEMENT <redacted> --- arch/arm/mach-mvebu/armada-370-xp.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)diff --git a/arch/arm/mach-mvebu/armada-370-xp.c b/arch/arm/mach-mvebu/armada-370-xp.c index f6c9d1d85c14..81b42980311c 100644 --- a/arch/arm/mach-mvebu/armada-370-xp.c +++ b/arch/arm/mach-mvebu/armada-370-xp.c@@ -30,6 +30,11 @@ #include "common.h" #include "coherency.h" #include "mvebu-soc-id.h" +#include "pmsu.h" + +static struct platform_device armada_xp_cpuidle_device = { + .name = "cpuidle-armada-370-xp", +}; static void __init armada_370_xp_map_io(void) {@@ -80,6 +85,13 @@ static void __init armada_370_xp_dt_init(void) if (of_machine_is_compatible("plathome,openblocks-ax3-4")) i2c_quirk(); of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL); + if (of_find_compatible_node(NULL, NULL, "marvell,armada-370-xp-pmsu") + && of_find_compatible_node(NULL, NULL, "marvell,coherency-fabric") + && of_machine_is_compatible("marvell,armadaxp")) { + armada_370_xp_pmsu_enable_l2_powerdown_onidle(); + armada_370_xp_cpu_pm_init(); + platform_device_register(&armada_xp_cpuidle_device); + }What about putting this in pmsu.c, in an arch_initcall() (or some other initcall level) ? The cpuidle feature is really tied to the PMSU, so I believe it makes sense to have the cpuidle-armada-370-xp platform_device declared and registered in pmsu.c. As an added bonus, you don't need to expose armada_370_xp_pmsu_enable_l2_powerdown_onidle() and armada_370_xp_cpu_pm_init() in a header: they can remain static functions private to pmsu.c.
It sounds like a good idea, the tricky part will be to find the accurate initcall level: not too early and not too late. Thanks, Gregory
Thomas
-- Gregory Clement, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com