On Thu, Oct 10, 2013 at 05:21:37PM -0300, Paulo Zanoni wrote: > 2013/10/10 Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>: > > As part of the device quiesceing we need to disable all active timers > > and delayed workers so that they do not execute after the module is > > unloaded. > > > > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > > Cc: Paulo Zanoni <paulo.r.zanoni@xxxxxxxxx> > > Cc: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx> > > --- > > drivers/gpu/drm/i915/intel_display.c | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c > > index 683b68e..5ce1558 100644 > > --- a/drivers/gpu/drm/i915/intel_display.c > > +++ b/drivers/gpu/drm/i915/intel_display.c > > @@ -11065,6 +11065,7 @@ void intel_modeset_quiesce(struct drm_device *dev) > > I don't see this "intel_modeset_quiesce" on my drm-intel-nightly > branch from today :( > > Besides, we already cancel the work at i915_driver_unload(). Well that at least allays my immediate fears that the timer is still running then. However, it probably wants to be hsw_disable_package_c8() instead. Oh well, we know a timer is still running somewhere. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
