Re: [PATCH v5 3/4] amba: Don't unprepare the clocks if device driver wants IRQ safe runtime PM

[Russell King - ARM Linux <linux@arm.linux.org.uk>]

On Tue, Sep 16, 2014 at 08:25:25PM +0200, Ulf Hansson wrote:
> On 16 September 2014 14:59, Krzysztof Kozlowski <k.kozlowski@samsung.com> wrote:
> > The AMBA bus driver defines runtime Power Management functions which
> > disable and unprepare AMBA bus clock. This is problematic for runtime PM
> > because unpreparing a clock might sleep so it is not interrupt safe.
> >
> > However some drivers may want to implement runtime PM functions in
> > interrupt-safe way (see pm_runtime_irq_safe()). In such case the AMBA
> > bus driver should only disable/enable the clock in runtime suspend and
> > resume callbacks.
> >
> > Detect the device driver behavior after calling its probe function and
> > store it. During runtime suspend/resume deal with clocks according to
> > stored value.
> >
> > Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
> > ---
> >  drivers/amba/bus.c       | 29 +++++++++++++++++++++++++----
> >  include/linux/amba/bus.h |  1 +
> >  2 files changed, 26 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/amba/bus.c b/drivers/amba/bus.c
> > index 3cf61a127ee5..e8fd5706954f 100644
> > --- a/drivers/amba/bus.c
> > +++ b/drivers/amba/bus.c
> > @@ -94,8 +94,18 @@ static int amba_pm_runtime_suspend(struct device *dev)
> >         struct amba_device *pcdev = to_amba_device(dev);
> >         int ret = pm_generic_runtime_suspend(dev);
> >
> > -       if (ret == 0 && dev->driver)
> > -               clk_disable_unprepare(pcdev->pclk);
> > +       if (ret == 0 && dev->driver) {
> > +               /*
> > +                * Drivers should not change pm_runtime_irq_safe()
> > +                * after probe.
> > +                */
> > +               WARN_ON(pcdev->irq_safe != pm_runtime_is_irq_safe(dev));
> 
> Do we really need a WARN_ON here. Driver shouldn't update their
> irq_safe value dynamically, right!?

The driver shouldn't update it dynamically, and this makes sure *that*
is enforced since we end up depending on that property.  Hence the
check is sensible (and I even suggested it.)

> > +
> > +               if (pcdev->irq_safe)
> > +                       clk_disable(pcdev->pclk);
> 
> Since the irq_safe flag, could be considered as a special case, an
> option for these cases - could be to leave the clock to be entirely
> handled from the driver's runtime PM callback instead.

Too many sub-clauses to make much sense of that statement.

I don't want drivers messing around with this stuff.  This is the /bus/
clock, not a device specific clock.

-- 
FTTC broadband for 0.8mile line: currently at 9.5Mbps down 400kbps up
according to speedtest.net.