| Message ID | b0bf70d911dffe170b0d2b2b3e680bb445c15fc4.1433768426.git.viresh.kumar@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 06/08/2015 03:40 PM, Viresh Kumar wrote: > Migrate bcm2835 driver to the new 'set-state' interface provided by > the clockevents core, the earlier 'set-mode' interface is marked > obsolete now. > > This also enables us to implement callbacks for new states of clockevent > devices, for example: ONESHOT_STOPPED. Please mention in the log you removed the RESUME because it does nothing and the core code will just ignore it. > Cc: Stephen Warren <swarren@wwwdotorg.org> > Cc: Lee Jones <lee@kernel.org> > Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> > --- > drivers/clocksource/bcm2835_timer.c | 17 ++++------------- > 1 file changed, 4 insertions(+), 13 deletions(-) > > diff --git a/drivers/clocksource/bcm2835_timer.c b/drivers/clocksource/bcm2835_timer.c > index 26ed331b1aad..b59b8e1ba503 100644 > --- a/drivers/clocksource/bcm2835_timer.c > +++ b/drivers/clocksource/bcm2835_timer.c > @@ -54,19 +54,9 @@ static u64 notrace bcm2835_sched_read(void) > return readl_relaxed(system_clock); > } > > -static void bcm2835_time_set_mode(enum clock_event_mode mode, > - struct clock_event_device *evt_dev) > +static int bcm2835_time_noop(struct clock_event_device *evt) > { > - switch (mode) { > - case CLOCK_EVT_MODE_ONESHOT: > - case CLOCK_EVT_MODE_UNUSED: > - case CLOCK_EVT_MODE_SHUTDOWN: > - case CLOCK_EVT_MODE_RESUME: > - break; > - default: > - WARN(1, "%s: unhandled event mode %d\n", __func__, mode); > - break; > - } > + return 0; > } > > static int bcm2835_time_set_next_event(unsigned long event, > @@ -129,7 +119,8 @@ static void __init bcm2835_timer_init(struct device_node *node) > timer->evt.name = node->name; > timer->evt.rating = 300; > timer->evt.features = CLOCK_EVT_FEAT_ONESHOT; > - timer->evt.set_mode = bcm2835_time_set_mode; > + timer->evt.set_state_shutdown = bcm2835_time_noop; > + timer->evt.set_state_oneshot = bcm2835_time_noop; The patch is ok but there is an inconsistency on how are handled the different callbacks. tick_resume can be omitted but not state_shutdown and state_oneshot forcing the drivers to define empty functions. > timer->evt.set_next_event = bcm2835_time_set_next_event; > timer->evt.cpumask = cpumask_of(0); > timer->act.name = node->name;
On 11-06-15, 17:25, Daniel Lezcano wrote: > Please mention in the log you removed the RESUME because it does > nothing and the core code will just ignore it. Okay. > The patch is ok but there is an inconsistency on how are handled the > different callbacks. tick_resume can be omitted but not > state_shutdown and state_oneshot forcing the drivers to define empty > functions. It was done because they are thought to be compulsory, but even I am not sure if we should make all callbacks as optional.
On 06/11/2015 06:00 PM, Viresh Kumar wrote: > On 11-06-15, 17:25, Daniel Lezcano wrote: >> Please mention in the log you removed the RESUME because it does >> nothing and the core code will just ignore it. > > Okay. > >> The patch is ok but there is an inconsistency on how are handled the >> different callbacks. tick_resume can be omitted but not >> state_shutdown and state_oneshot forcing the drivers to define empty >> functions. > > It was done because they are thought to be compulsory, but even I > am not sure if we should make all callbacks as optional. Yes, I think it would make sense to ignore NULL callback in __clockevents_set_state, otherwise we will end up to fill the different drivers with duplicated noop functions.
Hi Thomas, On 11-06-15, 18:13, Daniel Lezcano wrote: > Yes, I think it would make sense to ignore NULL callback in > __clockevents_set_state, otherwise we will end up to fill the > different drivers with duplicated noop functions. Do you suggest the same ? Thanks.
diff --git a/drivers/clocksource/bcm2835_timer.c b/drivers/clocksource/bcm2835_timer.c index 26ed331b1aad..b59b8e1ba503 100644 --- a/drivers/clocksource/bcm2835_timer.c +++ b/drivers/clocksource/bcm2835_timer.c @@ -54,19 +54,9 @@ static u64 notrace bcm2835_sched_read(void) return readl_relaxed(system_clock); } -static void bcm2835_time_set_mode(enum clock_event_mode mode, - struct clock_event_device *evt_dev) +static int bcm2835_time_noop(struct clock_event_device *evt) { - switch (mode) { - case CLOCK_EVT_MODE_ONESHOT: - case CLOCK_EVT_MODE_UNUSED: - case CLOCK_EVT_MODE_SHUTDOWN: - case CLOCK_EVT_MODE_RESUME: - break; - default: - WARN(1, "%s: unhandled event mode %d\n", __func__, mode); - break; - } + return 0; } static int bcm2835_time_set_next_event(unsigned long event, @@ -129,7 +119,8 @@ static void __init bcm2835_timer_init(struct device_node *node) timer->evt.name = node->name; timer->evt.rating = 300; timer->evt.features = CLOCK_EVT_FEAT_ONESHOT; - timer->evt.set_mode = bcm2835_time_set_mode; + timer->evt.set_state_shutdown = bcm2835_time_noop; + timer->evt.set_state_oneshot = bcm2835_time_noop; timer->evt.set_next_event = bcm2835_time_set_next_event; timer->evt.cpumask = cpumask_of(0); timer->act.name = node->name;
Migrate bcm2835 driver to the new 'set-state' interface provided by the clockevents core, the earlier 'set-mode' interface is marked obsolete now. This also enables us to implement callbacks for new states of clockevent devices, for example: ONESHOT_STOPPED. Cc: Stephen Warren <swarren@wwwdotorg.org> Cc: Lee Jones <lee@kernel.org> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> --- drivers/clocksource/bcm2835_timer.c | 17 ++++------------- 1 file changed, 4 insertions(+), 13 deletions(-)