Re: [PATCH v3 5/6] OMAP: omap_device: Create a default omap_device_pm_latency

[Kevin Hilman <khilman@ti.com>]

Tony Lindgren <tony@atomide.com> writes:

> * Benoit Cousson <b-cousson@ti.com> [110927 09:45]:
>> Most devices are using the same default omap_device_pm_latency structure
>> during device built. In order to avoid the duplication of the same
>> structure everywhere, add a default structure that will be used if
>> the device does not have an explicit one.
>> 
>> Next patches will clean the duplicated structures.
>> 
>> Signed-off-by: Benoit Cousson <b-cousson@ti.com>
>> Cc: Kevin Hilman <khilman@ti.com>
>> Cc: Paul Walmsley <paul@pwsan.com>
>> ---
>>  arch/arm/plat-omap/omap_device.c |   19 ++++++++++++++++++-
>>  1 files changed, 18 insertions(+), 1 deletions(-)
>> 
>> diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c
>> index f832f92..cd8d977 100644
>> --- a/arch/arm/plat-omap/omap_device.c
>> +++ b/arch/arm/plat-omap/omap_device.c
>> @@ -97,6 +97,14 @@
>>  static int omap_device_register(struct platform_device *pdev);
>>  static int omap_early_device_register(struct platform_device *pdev);
>>  
>> +static struct omap_device_pm_latency omap_default_latency[] = {
>> +	{
>> +		.deactivate_func = omap_device_idle_hwmods,
>> +		.activate_func   = omap_device_enable_hwmods,
>> +		.flags = OMAP_DEVICE_LATENCY_AUTO_ADJUST,
>> +	}
>> +};
>> +
>>  /* Private functions */
>
> Isn't this racey between devices if the latency values get adjusted
> automatically for each device?

A copy of this is made for each device (uses kmemdup later in the patch.)

Kevin