On Mon, Nov 14, 2016 at 06:35:08PM +0200, [email protected] wrote:
> From: Ville Syrjälä <[email protected]>
> 
> My little fix for populating intel_state->cdclk with
> dev_priv->atomic_cdclk_freq became a little meatier after I realized
> that we really need to protect it with what is essentially a crtc
> rwlock. That is, writers need to hold all the crtc locks, whereas
> readers need to hold at least one.
> 
> Ville Syrjälä (3):
>   drm/i915: Fix cdclk vs. dev_cdclk mess when not recomputing things
>   drm/i915: Protect dev_priv->atomic_cdclk_freq with all the crtc locks
>   drm/i915: Simplify error handling in intel_modeset_all_pipes()

Entire series pushed to dinq. Thanks for the reviews.

> 
>  drivers/gpu/drm/i915/i915_drv.h      |  9 +++++-
>  drivers/gpu/drm/i915/intel_display.c | 60 
> ++++++++++++++++++++++++++++--------
>  2 files changed, 55 insertions(+), 14 deletions(-)
> 
> -- 
> 2.7.4

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reply via email to