drm/i915: Optimize DIV_ROUND_CLOSEST call
authorJean Delvare <khali@linux-fr.org>
Mon, 12 Nov 2012 13:18:02 +0000 (14:18 +0100)
committerDaniel Vetter <daniel.vetter@ffwll.ch>
Wed, 21 Nov 2012 16:44:59 +0000 (17:44 +0100)
DIV_ROUND_CLOSEST is faster if the compiler knows it will only be
dealing with unsigned dividends. This optimization rips 32 bytes of
binary code on x86_64.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
drivers/gpu/drm/i915/intel_pm.c

index 0cbc0e6402b4ff49e2b843e0f74b871d8a1ecf69..0edb549d14343df0c83b6d215509faeec87d1fa3 100644 (file)
@@ -2552,7 +2552,8 @@ static void gen6_update_ring_freq(struct drm_device *dev)
 {
        struct drm_i915_private *dev_priv = dev->dev_private;
        int min_freq = 15;
-       int gpu_freq, ia_freq, max_ia_freq;
+       int gpu_freq;
+       unsigned int ia_freq, max_ia_freq;
        int scaling_factor = 180;
 
        WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));