10 Apr
2013
10 Apr
'13
6:23 p.m.
On Wed, Apr 10, 2013 at 08:02:31PM +0200, Daniel Lezcano wrote:
On 04/10/2013 07:04 PM, Andrew Lunn wrote:
+int cpuidle_register(struct cpuidle_driver *drv,
const struct cpumask *const coupled_cpus)+{
- int ret, cpu;
- struct cpuidle_device *device;
- ret = cpuidle_register_driver(drv);
- if (ret) {
printk(KERN_ERR "failed to register cpuidle driver\n");return ret;- }
- for_each_possible_cpu(cpu) {
device = &per_cpu(cpuidle_dev, cpu);device->cpu = cpu;+#ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED
device->coupled_cpus = *coupled_cpus;+#endif
At least the kirkwood and the calxeda driver set
device->state_count
which you don't appear to do. cpuidle_add_state_sysfs() and cpuidle_remove_state_sysfs() use this. Is it now being set somewhere else?
Yes, in cpuidle_enable_device called from cpuidle_register_device:
O.K. It would be nice to add a comment in the change log message about this.
Ah, also, it would be good to update Documentation/cpuidle/drivers.txt with these new functions and update the text.
Thanks Andrew