drm/i915/bdw: Implement Wa4x4STCOptimizationDisable:bdw

Not implementing this W/A can lead to hangs.

Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Cc: Rafael Barbalho <rafael.barbalho@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

drivers/gpu/drm/i915/i915_reg.h

@@ -973,7 +973,8 @@ enum punit_power_well {
 #define CACHE_MODE_0_GEN7	0x7000 /* IVB+ */
 #define   HIZ_RAW_STALL_OPT_DISABLE (1<<2)
 #define CACHE_MODE_1		0x7004 /* IVB+ */
-#define   PIXEL_SUBSPAN_COLLECT_OPT_DISABLE (1<<6)
+#define   PIXEL_SUBSPAN_COLLECT_OPT_DISABLE	(1<<6)
+#define   GEN8_4x4_STC_OPTIMIZATION_DISABLE	(1<<6)
 
 #define GEN6_BLITTER_ECOSKPD	0x221d0
 #define   GEN6_BLITTER_LOCK_SHIFT			16

drivers/gpu/drm/i915/intel_pm.c

@@ -4882,6 +4882,10 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	/* WaDisableSDEUnitClockGating:bdw */
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
+	/* Wa4x4STCOptimizationDisable:bdw */
+	I915_WRITE(CACHE_MODE_1,
+		   _MASKED_BIT_ENABLE(GEN8_4x4_STC_OPTIMIZATION_DISABLE));
 }
 
 static void haswell_init_clock_gating(struct drm_device *dev)