drm/i915: Implement pipe CSC based limited range RGB output

HSW no longer has the PIPECONF bit for limited range RGB output.
Instead the pipe CSC unit must be used to perform that task.

The CSC pre offset are set to 0, since the incoming data is full
[0:255] range RGB, the coefficients are programmed to compress the
data into [0:219] range, and then we use either the CSC_MODE black
screen offset bit, or the CSC post offsets to shift the data to
the correct [16:235] range.

Also have to change the confiuration of all planes so that the
data is sent through the pipe CSC unit. For simplicity send the
plane data through the pipe CSC unit always, and in case full
range output is requested, the pipe CSC unit is set up with an
identity transform to pass the plane data through unchanged.

I've been told by some hardware people that the use of the pipe
CSC unit shouldn't result in any measurable increase in power
consumption numbers.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 7e77396..527b664 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2949,6 +2949,7 @@
 #define   CURSOR_ENABLE		0x80000000
 #define   CURSOR_GAMMA_ENABLE	0x40000000
 #define   CURSOR_STRIDE_MASK	0x30000000
+#define   CURSOR_PIPE_CSC_ENABLE (1<<24)
 #define   CURSOR_FORMAT_SHIFT	24
 #define   CURSOR_FORMAT_MASK	(0x07 << CURSOR_FORMAT_SHIFT)
 #define   CURSOR_FORMAT_2C	(0x00 << CURSOR_FORMAT_SHIFT)
@@ -3010,6 +3011,7 @@
 #define   DISPPLANE_RGBA888			(0xf<<26)
 #define   DISPPLANE_STEREO_ENABLE		(1<<25)
 #define   DISPPLANE_STEREO_DISABLE		0
+#define   DISPPLANE_PIPE_CSC_ENABLE		(1<<24)
 #define   DISPPLANE_SEL_PIPE_SHIFT		24
 #define   DISPPLANE_SEL_PIPE_MASK		(3<<DISPPLANE_SEL_PIPE_SHIFT)
 #define   DISPPLANE_SEL_PIPE_A			0
@@ -3098,6 +3100,7 @@
 #define   DVS_FORMAT_RGBX101010	(1<<25)
 #define   DVS_FORMAT_RGBX888	(2<<25)
 #define   DVS_FORMAT_RGBX161616	(3<<25)
+#define   DVS_PIPE_CSC_ENABLE   (1<<24)
 #define   DVS_SOURCE_KEY	(1<<22)
 #define   DVS_RGB_ORDER_XBGR	(1<<20)
 #define   DVS_YUV_BYTE_ORDER_MASK (3<<16)
@@ -3165,7 +3168,7 @@
 #define   SPRITE_FORMAT_RGBX161616	(3<<25)
 #define   SPRITE_FORMAT_YUV444		(4<<25)
 #define   SPRITE_FORMAT_XR_BGR101010	(5<<25) /* Extended range */
-#define   SPRITE_CSC_ENABLE		(1<<24)
+#define   SPRITE_PIPE_CSC_ENABLE	(1<<24)
 #define   SPRITE_SOURCE_KEY		(1<<22)
 #define   SPRITE_RGB_ORDER_RGBX		(1<<20) /* only for 888 and 161616 */
 #define   SPRITE_YUV_TO_RGB_CSC_DISABLE	(1<<19)
@@ -4645,4 +4648,51 @@
 #define  WM_DBG_DISALLOW_MAXFIFO	(1<<1)
 #define  WM_DBG_DISALLOW_SPRITE		(1<<2)
 
+/* pipe CSC */
+#define _PIPE_A_CSC_COEFF_RY_GY	0x49010
+#define _PIPE_A_CSC_COEFF_BY	0x49014
+#define _PIPE_A_CSC_COEFF_RU_GU	0x49018
+#define _PIPE_A_CSC_COEFF_BU	0x4901c
+#define _PIPE_A_CSC_COEFF_RV_GV	0x49020
+#define _PIPE_A_CSC_COEFF_BV	0x49024
+#define _PIPE_A_CSC_MODE	0x49028
+#define _PIPE_A_CSC_PREOFF_HI	0x49030
+#define _PIPE_A_CSC_PREOFF_ME	0x49034
+#define _PIPE_A_CSC_PREOFF_LO	0x49038
+#define _PIPE_A_CSC_POSTOFF_HI	0x49040
+#define _PIPE_A_CSC_POSTOFF_ME	0x49044
+#define _PIPE_A_CSC_POSTOFF_LO	0x49048
+
+#define _PIPE_B_CSC_COEFF_RY_GY	0x49110
+#define _PIPE_B_CSC_COEFF_BY	0x49114
+#define _PIPE_B_CSC_COEFF_RU_GU	0x49118
+#define _PIPE_B_CSC_COEFF_BU	0x4911c
+#define _PIPE_B_CSC_COEFF_RV_GV	0x49120
+#define _PIPE_B_CSC_COEFF_BV	0x49124
+#define _PIPE_B_CSC_MODE	0x49128
+#define _PIPE_B_CSC_PREOFF_HI	0x49130
+#define _PIPE_B_CSC_PREOFF_ME	0x49134
+#define _PIPE_B_CSC_PREOFF_LO	0x49138
+#define _PIPE_B_CSC_POSTOFF_HI	0x49140
+#define _PIPE_B_CSC_POSTOFF_ME	0x49144
+#define _PIPE_B_CSC_POSTOFF_LO	0x49148
+
+#define CSC_BLACK_SCREEN_OFFSET (1 << 2)
+#define CSC_POSITION_BEFORE_GAMMA (1 << 1)
+#define CSC_MODE_YUV_TO_RGB (1 << 0)
+
+#define PIPE_CSC_COEFF_RY_GY(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RY_GY, _PIPE_B_CSC_COEFF_RY_GY)
+#define PIPE_CSC_COEFF_BY(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BY, _PIPE_B_CSC_COEFF_BY)
+#define PIPE_CSC_COEFF_RU_GU(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RU_GU, _PIPE_B_CSC_COEFF_RU_GU)
+#define PIPE_CSC_COEFF_BU(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BU, _PIPE_B_CSC_COEFF_BU)
+#define PIPE_CSC_COEFF_RV_GV(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_RV_GV, _PIPE_B_CSC_COEFF_RV_GV)
+#define PIPE_CSC_COEFF_BV(pipe) _PIPE(pipe, _PIPE_A_CSC_COEFF_BV, _PIPE_B_CSC_COEFF_BV)
+#define PIPE_CSC_MODE(pipe) _PIPE(pipe, _PIPE_A_CSC_MODE, _PIPE_B_CSC_MODE)
+#define PIPE_CSC_PREOFF_HI(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_HI, _PIPE_B_CSC_PREOFF_HI)
+#define PIPE_CSC_PREOFF_ME(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_ME, _PIPE_B_CSC_PREOFF_ME)
+#define PIPE_CSC_PREOFF_LO(pipe) _PIPE(pipe, _PIPE_A_CSC_PREOFF_LO, _PIPE_B_CSC_PREOFF_LO)
+#define PIPE_CSC_POSTOFF_HI(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_HI, _PIPE_B_CSC_POSTOFF_HI)
+#define PIPE_CSC_POSTOFF_ME(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_ME, _PIPE_B_CSC_POSTOFF_ME)
+#define PIPE_CSC_POSTOFF_LO(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_LO, _PIPE_B_CSC_POSTOFF_LO)
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ba1d913..6eb3882 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5126,6 +5126,71 @@
 	POSTING_READ(PIPECONF(pipe));
 }
 
+/*
+ * Set up the pipe CSC unit.
+ *
+ * Currently only full range RGB to limited range RGB conversion
+ * is supported, but eventually this should handle various
+ * RGB<->YCbCr scenarios as well.
+ */
+static void intel_set_pipe_csc(struct drm_crtc *crtc,
+			       const struct drm_display_mode *adjusted_mode)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	uint16_t coeff = 0x7800; /* 1.0 */
+
+	/*
+	 * TODO: Check what kind of values actually come out of the pipe
+	 * with these coeff/postoff values and adjust to get the best
+	 * accuracy. Perhaps we even need to take the bpc value into
+	 * consideration.
+	 */
+
+	if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+		coeff = ((235 - 16) * (1 << 12) / 255) & 0xff8; /* 0.xxx... */
+
+	/*
+	 * GY/GU and RY/RU should be the other way around according
+	 * to BSpec, but reality doesn't agree. Just set them up in
+	 * a way that results in the correct picture.
+	 */
+	I915_WRITE(PIPE_CSC_COEFF_RY_GY(pipe), coeff << 16);
+	I915_WRITE(PIPE_CSC_COEFF_BY(pipe), 0);
+
+	I915_WRITE(PIPE_CSC_COEFF_RU_GU(pipe), coeff);
+	I915_WRITE(PIPE_CSC_COEFF_BU(pipe), 0);
+
+	I915_WRITE(PIPE_CSC_COEFF_RV_GV(pipe), 0);
+	I915_WRITE(PIPE_CSC_COEFF_BV(pipe), coeff << 16);
+
+	I915_WRITE(PIPE_CSC_PREOFF_HI(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_ME(pipe), 0);
+	I915_WRITE(PIPE_CSC_PREOFF_LO(pipe), 0);
+
+	if (INTEL_INFO(dev)->gen > 6) {
+		uint16_t postoff = 0;
+
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			postoff = (16 * (1 << 13) / 255) & 0x1fff;
+
+		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
+		I915_WRITE(PIPE_CSC_POSTOFF_LO(pipe), postoff);
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), 0);
+	} else {
+		uint32_t mode = CSC_MODE_YUV_TO_RGB;
+
+		if (adjusted_mode->private_flags & INTEL_MODE_LIMITED_COLOR_RANGE)
+			mode |= CSC_BLACK_SCREEN_OFFSET;
+
+		I915_WRITE(PIPE_CSC_MODE(pipe), mode);
+	}
+}
+
 static void haswell_set_pipeconf(struct drm_crtc *crtc,
 				 struct drm_display_mode *adjusted_mode,
 				 bool dither)
@@ -5714,8 +5779,10 @@
 
 	haswell_set_pipeconf(crtc, adjusted_mode, dither);
 
+	intel_set_pipe_csc(crtc, adjusted_mode);
+
 	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
 	POSTING_READ(DSPCNTR(plane));
 
 	ret = intel_pipe_set_base(crtc, x, y, fb);
@@ -6120,6 +6187,8 @@
 			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
 			cntl |= CURSOR_MODE_DISABLE;
 		}
+		if (IS_HASWELL(dev))
+			cntl |= CURSOR_PIPE_CSC_ENABLE;
 		I915_WRITE(CURCNTR_IVB(pipe), cntl);
 
 		intel_crtc->cursor_visible = visible;
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 03cfd62..d086e48 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -90,6 +90,9 @@
 	sprctl |= SPRITE_TRICKLE_FEED_DISABLE;
 	sprctl |= SPRITE_ENABLE;
 
+	if (IS_HASWELL(dev))
+		sprctl |= SPRITE_PIPE_CSC_ENABLE;
+
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;