[Intel-gfx] [PATCH] drm/i915: Add might_sleep() check to wait_for()

[Chris Wilson]

Quoting Tvrtko Ursulin (2017-11-15 09:11:13)
> 
> On 14/11/2017 21:56, Chris Wilson wrote:
> > We should long past the time of trying to use wait_for() from inside
> > atomic contexts, so add a might_sleep() check to prevent misuse.
> > 
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > ---
> >   drivers/gpu/drm/i915/intel_drv.h | 11 ++---------
> >   1 file changed, 2 insertions(+), 9 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
> > index fd64a5e8ea12..a898ded7efe9 100644
> > --- a/drivers/gpu/drm/i915/intel_drv.h
> > +++ b/drivers/gpu/drm/i915/intel_drv.h
> > @@ -47,14 +47,11 @@
> >    * contexts. Note that it's important that we check the condition again after
> >    * having timed out, since the timeout could be due to preemption or similar and
> >    * we've never had a chance to check the condition before the timeout.
> > - *
> > - * TODO: When modesetting has fully transitioned to atomic, the below
> > - * drm_can_sleep() can be removed and in_atomic()/!in_atomic() asserts
> > - * added.
> >    */
> >   #define _wait_for(COND, US, W) ({ \
> >       unsigned long timeout__ = jiffies + usecs_to_jiffies(US) + 1;   \
> >       int ret__;                                                      \
> > +     might_sleep();                                                  \
> >       for (;;) {                                                      \
> >               bool expired__ = time_after(jiffies, timeout__);        \
> >               if (COND) {                                             \
> > @@ -65,11 +62,7 @@
> >                       ret__ = -ETIMEDOUT;                             \
> >                       break;                                          \
> >               }                                                       \
> > -             if ((W) && drm_can_sleep()) {                           \
> > -                     usleep_range((W), (W)*2);                       \
> > -             } else {                                                \
> > -                     cpu_relax();                                    \
> > -             }                                                       \
> > +             usleep_range((W), (W)*2);                               \
> >       }                                                               \
> >       ret__;                                                          \
> >   })
> > 
> 
> Seems OK under my understanding how things work at least.
> 
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> 

Thanks for the review. Maarten also gave his r-b on irc, and with an all
green pw run, pushed.

We now have no more drm_can_sleep() in i915! No more in_atomic() policy
violations!
-Chris