| Message ID | 1425364906-31769-1-git-send-email-ramalingam.c@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Looks enough for me... Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> On Mon, Mar 2, 2015 at 10:41 PM, Ramalingam C <ramalingam.c@intel.com> wrote: > In invalidate and flush functions of eDP DRRS, if deferred downclock > work starts execution at a time window between acquiring the drrs > mutex and cancellation of the deferred work > (intel_edp_drrs_downclock_work), then deferred work will find > drrs mutex locked and wait for the same. > > Meanwhile the function that acquired mutex drrs invalidate/flush will > wait for the completion of the deferred work before releasing the mutex. > Thats a deadlock. > > To avoid such deadlock scenario, this change cancels the deferred work > before acquiring the mutex at invalidate and flush functions. > > Signed-off-by: Ramalingam C <ramalingam.c@intel.com> > --- > drivers/gpu/drm/i915/intel_dp.c | 7 ++++--- > 1 file changed, 4 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c > index d1141d3..0a57763 100644 > --- a/drivers/gpu/drm/i915/intel_dp.c > +++ b/drivers/gpu/drm/i915/intel_dp.c > @@ -4966,12 +4966,13 @@ void intel_edp_drrs_invalidate(struct drm_device *dev, > if (!dev_priv->drrs.dp) > return; > > + cancel_delayed_work_sync(&dev_priv->drrs.work); > + > mutex_lock(&dev_priv->drrs.mutex); > crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; > pipe = to_intel_crtc(crtc)->pipe; > > if (dev_priv->drrs.refresh_rate_type == DRRS_LOW_RR) { > - cancel_delayed_work_sync(&dev_priv->drrs.work); > intel_dp_set_drrs_state(dev_priv->dev, > dev_priv->drrs.dp->attached_connector->panel. > fixed_mode->vrefresh); > @@ -5004,13 +5005,13 @@ void intel_edp_drrs_flush(struct drm_device *dev, > if (!dev_priv->drrs.dp) > return; > > + cancel_delayed_work_sync(&dev_priv->drrs.work); > + > mutex_lock(&dev_priv->drrs.mutex); > crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; > pipe = to_intel_crtc(crtc)->pipe; > dev_priv->drrs.busy_frontbuffer_bits &= ~frontbuffer_bits; > > - cancel_delayed_work_sync(&dev_priv->drrs.work); > - > if (dev_priv->drrs.refresh_rate_type != DRRS_LOW_RR && > !dev_priv->drrs.busy_frontbuffer_bits) > schedule_delayed_work(&dev_priv->drrs.work, > -- > 1.7.9.5 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
On Wed, Mar 04, 2015 at 02:55:23PM -0800, Rodrigo Vivi wrote: > Looks enough for me... > > Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> > > On Mon, Mar 2, 2015 at 10:41 PM, Ramalingam C <ramalingam.c@intel.com> wrote: > > In invalidate and flush functions of eDP DRRS, if deferred downclock > > work starts execution at a time window between acquiring the drrs > > mutex and cancellation of the deferred work > > (intel_edp_drrs_downclock_work), then deferred work will find > > drrs mutex locked and wait for the same. > > > > Meanwhile the function that acquired mutex drrs invalidate/flush will > > wait for the completion of the deferred work before releasing the mutex. > > Thats a deadlock. > > > > To avoid such deadlock scenario, this change cancels the deferred work > > before acquiring the mutex at invalidate and flush functions. > > > > Signed-off-by: Ramalingam C <ramalingam.c@intel.com> Queued for -next, thanks for the patch. -Daniel > > --- > > drivers/gpu/drm/i915/intel_dp.c | 7 ++++--- > > 1 file changed, 4 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c > > index d1141d3..0a57763 100644 > > --- a/drivers/gpu/drm/i915/intel_dp.c > > +++ b/drivers/gpu/drm/i915/intel_dp.c > > @@ -4966,12 +4966,13 @@ void intel_edp_drrs_invalidate(struct drm_device *dev, > > if (!dev_priv->drrs.dp) > > return; > > > > + cancel_delayed_work_sync(&dev_priv->drrs.work); > > + > > mutex_lock(&dev_priv->drrs.mutex); > > crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; > > pipe = to_intel_crtc(crtc)->pipe; > > > > if (dev_priv->drrs.refresh_rate_type == DRRS_LOW_RR) { > > - cancel_delayed_work_sync(&dev_priv->drrs.work); > > intel_dp_set_drrs_state(dev_priv->dev, > > dev_priv->drrs.dp->attached_connector->panel. > > fixed_mode->vrefresh); > > @@ -5004,13 +5005,13 @@ void intel_edp_drrs_flush(struct drm_device *dev, > > if (!dev_priv->drrs.dp) > > return; > > > > + cancel_delayed_work_sync(&dev_priv->drrs.work); > > + > > mutex_lock(&dev_priv->drrs.mutex); > > crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; > > pipe = to_intel_crtc(crtc)->pipe; > > dev_priv->drrs.busy_frontbuffer_bits &= ~frontbuffer_bits; > > > > - cancel_delayed_work_sync(&dev_priv->drrs.work); > > - > > if (dev_priv->drrs.refresh_rate_type != DRRS_LOW_RR && > > !dev_priv->drrs.busy_frontbuffer_bits) > > schedule_delayed_work(&dev_priv->drrs.work, > > -- > > 1.7.9.5 > > > > _______________________________________________ > > Intel-gfx mailing list > > Intel-gfx@lists.freedesktop.org > > http://lists.freedesktop.org/mailman/listinfo/intel-gfx > > > > -- > Rodrigo Vivi > Blog: http://blog.vivi.eng.br
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index d1141d3..0a57763 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -4966,12 +4966,13 @@ void intel_edp_drrs_invalidate(struct drm_device *dev, if (!dev_priv->drrs.dp) return; + cancel_delayed_work_sync(&dev_priv->drrs.work); + mutex_lock(&dev_priv->drrs.mutex); crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; pipe = to_intel_crtc(crtc)->pipe; if (dev_priv->drrs.refresh_rate_type == DRRS_LOW_RR) { - cancel_delayed_work_sync(&dev_priv->drrs.work); intel_dp_set_drrs_state(dev_priv->dev, dev_priv->drrs.dp->attached_connector->panel. fixed_mode->vrefresh); @@ -5004,13 +5005,13 @@ void intel_edp_drrs_flush(struct drm_device *dev, if (!dev_priv->drrs.dp) return; + cancel_delayed_work_sync(&dev_priv->drrs.work); + mutex_lock(&dev_priv->drrs.mutex); crtc = dp_to_dig_port(dev_priv->drrs.dp)->base.base.crtc; pipe = to_intel_crtc(crtc)->pipe; dev_priv->drrs.busy_frontbuffer_bits &= ~frontbuffer_bits; - cancel_delayed_work_sync(&dev_priv->drrs.work); - if (dev_priv->drrs.refresh_rate_type != DRRS_LOW_RR && !dev_priv->drrs.busy_frontbuffer_bits) schedule_delayed_work(&dev_priv->drrs.work,
In invalidate and flush functions of eDP DRRS, if deferred downclock work starts execution at a time window between acquiring the drrs mutex and cancellation of the deferred work (intel_edp_drrs_downclock_work), then deferred work will find drrs mutex locked and wait for the same. Meanwhile the function that acquired mutex drrs invalidate/flush will wait for the completion of the deferred work before releasing the mutex. Thats a deadlock. To avoid such deadlock scenario, this change cancels the deferred work before acquiring the mutex at invalidate and flush functions. Signed-off-by: Ramalingam C <ramalingam.c@intel.com> --- drivers/gpu/drm/i915/intel_dp.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)