X-Git-Url: http://demsky.eecs.uci.edu/git/?a=blobdiff_plain;f=drivers%2Fgpu%2Fdrm%2Fi915%2Fintel_display.c;h=0d7c4f634bb92bbe1ee0fa3209f2591ff62a55fe;hb=7e7cb34f62dbb3471e4b1d3fae12a8b71e2224d9;hp=e600e1cfb6ea88b3e384ccfeff751e47b7474398;hpb=6aedd1f539f51b7b0c3d6be0088c3541f9d2c294;p=firefly-linux-kernel-4.4.55.git

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e600e1cfb6ea..0d7c4f634bb9 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2077,8 +2077,10 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
 	else
 		dspcntr &= ~DISPPLANE_TILED;
 
-	/* must disable */
-	dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
+	if (IS_HASWELL(dev))
+		dspcntr &= ~DISPPLANE_TRICKLE_FEED_DISABLE;
+	else
+		dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
 
 	I915_WRITE(reg, dspcntr);
 
@@ -2247,7 +2249,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		I915_WRITE(PIPESRC(intel_crtc->pipe),
 			   ((crtc->mode.hdisplay - 1) << 16) |
 			   (crtc->mode.vdisplay - 1));
-		if (!intel_crtc->config.pch_pfit.size &&
+		if (!intel_crtc->config.pch_pfit.enabled &&
 		    (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) ||
 		     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
 			I915_WRITE(PF_CTL(intel_crtc->pipe), 0);
@@ -2597,7 +2599,7 @@ static void ivb_manual_fdi_link_train(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	u32 reg, temp, i;
+	u32 reg, temp, i, j;
 
 	/* Train 1: umask FDI RX Interrupt symbol_lock and bit_lock bit
 	   for train result */
@@ -2613,97 +2615,99 @@ static void ivb_manual_fdi_link_train(struct drm_crtc *crtc)
 	DRM_DEBUG_KMS("FDI_RX_IIR before link train 0x%x\n",
 		      I915_READ(FDI_RX_IIR(pipe)));
 
-	/* enable CPU FDI TX and PCH FDI RX */
-	reg = FDI_TX_CTL(pipe);
-	temp = I915_READ(reg);
-	temp &= ~FDI_DP_PORT_WIDTH_MASK;
-	temp |= FDI_DP_PORT_WIDTH(intel_crtc->config.fdi_lanes);
-	temp &= ~(FDI_LINK_TRAIN_AUTO | FDI_LINK_TRAIN_NONE_IVB);
-	temp |= FDI_LINK_TRAIN_PATTERN_1_IVB;
-	temp &= ~FDI_LINK_TRAIN_VOL_EMP_MASK;
-	temp |= FDI_LINK_TRAIN_400MV_0DB_SNB_B;
-	temp |= FDI_COMPOSITE_SYNC;
-	I915_WRITE(reg, temp | FDI_TX_ENABLE);
-
-	I915_WRITE(FDI_RX_MISC(pipe),
-		   FDI_RX_TP1_TO_TP2_48 | FDI_RX_FDI_DELAY_90);
-
-	reg = FDI_RX_CTL(pipe);
-	temp = I915_READ(reg);
-	temp &= ~FDI_LINK_TRAIN_AUTO;
-	temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
-	temp |= FDI_LINK_TRAIN_PATTERN_1_CPT;
-	temp |= FDI_COMPOSITE_SYNC;
-	I915_WRITE(reg, temp | FDI_RX_ENABLE);
+	/* Try each vswing and preemphasis setting twice before moving on */
+	for (j = 0; j < ARRAY_SIZE(snb_b_fdi_train_param) * 2; j++) {
+		/* disable first in case we need to retry */
+		reg = FDI_TX_CTL(pipe);
+		temp = I915_READ(reg);
+		temp &= ~(FDI_LINK_TRAIN_AUTO | FDI_LINK_TRAIN_NONE_IVB);
+		temp &= ~FDI_TX_ENABLE;
+		I915_WRITE(reg, temp);
 
-	POSTING_READ(reg);
-	udelay(150);
+		reg = FDI_RX_CTL(pipe);
+		temp = I915_READ(reg);
+		temp &= ~FDI_LINK_TRAIN_AUTO;
+		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
+		temp &= ~FDI_RX_ENABLE;
+		I915_WRITE(reg, temp);
 
-	for (i = 0; i < 4; i++) {
+		/* enable CPU FDI TX and PCH FDI RX */
 		reg = FDI_TX_CTL(pipe);
 		temp = I915_READ(reg);
+		temp &= ~FDI_DP_PORT_WIDTH_MASK;
+		temp |= FDI_DP_PORT_WIDTH(intel_crtc->config.fdi_lanes);
+		temp |= FDI_LINK_TRAIN_PATTERN_1_IVB;
 		temp &= ~FDI_LINK_TRAIN_VOL_EMP_MASK;
-		temp |= snb_b_fdi_train_param[i];
-		I915_WRITE(reg, temp);
+		temp |= snb_b_fdi_train_param[j/2];
+		temp |= FDI_COMPOSITE_SYNC;
+		I915_WRITE(reg, temp | FDI_TX_ENABLE);
 
-		POSTING_READ(reg);
-		udelay(500);
+		I915_WRITE(FDI_RX_MISC(pipe),
+			   FDI_RX_TP1_TO_TP2_48 | FDI_RX_FDI_DELAY_90);
 
-		reg = FDI_RX_IIR(pipe);
+		reg = FDI_RX_CTL(pipe);
 		temp = I915_READ(reg);
-		DRM_DEBUG_KMS("FDI_RX_IIR 0x%x\n", temp);
-
-		if (temp & FDI_RX_BIT_LOCK ||
-		    (I915_READ(reg) & FDI_RX_BIT_LOCK)) {
-			I915_WRITE(reg, temp | FDI_RX_BIT_LOCK);
-			DRM_DEBUG_KMS("FDI train 1 done, level %i.\n", i);
-			break;
-		}
-	}
-	if (i == 4)
-		DRM_ERROR("FDI train 1 fail!\n");
+		temp |= FDI_LINK_TRAIN_PATTERN_1_CPT;
+		temp |= FDI_COMPOSITE_SYNC;
+		I915_WRITE(reg, temp | FDI_RX_ENABLE);
 
-	/* Train 2 */
-	reg = FDI_TX_CTL(pipe);
-	temp = I915_READ(reg);
-	temp &= ~FDI_LINK_TRAIN_NONE_IVB;
-	temp |= FDI_LINK_TRAIN_PATTERN_2_IVB;
-	temp &= ~FDI_LINK_TRAIN_VOL_EMP_MASK;
-	temp |= FDI_LINK_TRAIN_400MV_0DB_SNB_B;
-	I915_WRITE(reg, temp);
+		POSTING_READ(reg);
+		udelay(1); /* should be 0.5us */
 
-	reg = FDI_RX_CTL(pipe);
-	temp = I915_READ(reg);
-	temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
-	temp |= FDI_LINK_TRAIN_PATTERN_2_CPT;
-	I915_WRITE(reg, temp);
+		for (i = 0; i < 4; i++) {
+			reg = FDI_RX_IIR(pipe);
+			temp = I915_READ(reg);
+			DRM_DEBUG_KMS("FDI_RX_IIR 0x%x\n", temp);
 
-	POSTING_READ(reg);
-	udelay(150);
+			if (temp & FDI_RX_BIT_LOCK ||
+			    (I915_READ(reg) & FDI_RX_BIT_LOCK)) {
+				I915_WRITE(reg, temp | FDI_RX_BIT_LOCK);
+				DRM_DEBUG_KMS("FDI train 1 done, level %i.\n",
+					      i);
+				break;
+			}
+			udelay(1); /* should be 0.5us */
+		}
+		if (i == 4) {
+			DRM_DEBUG_KMS("FDI train 1 fail on vswing %d\n", j / 2);
+			continue;
+		}
 
-	for (i = 0; i < 4; i++) {
+		/* Train 2 */
 		reg = FDI_TX_CTL(pipe);
 		temp = I915_READ(reg);
-		temp &= ~FDI_LINK_TRAIN_VOL_EMP_MASK;
-		temp |= snb_b_fdi_train_param[i];
+		temp &= ~FDI_LINK_TRAIN_NONE_IVB;
+		temp |= FDI_LINK_TRAIN_PATTERN_2_IVB;
+		I915_WRITE(reg, temp);
+
+		reg = FDI_RX_CTL(pipe);
+		temp = I915_READ(reg);
+		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
+		temp |= FDI_LINK_TRAIN_PATTERN_2_CPT;
 		I915_WRITE(reg, temp);
 
 		POSTING_READ(reg);
-		udelay(500);
+		udelay(2); /* should be 1.5us */
 
-		reg = FDI_RX_IIR(pipe);
-		temp = I915_READ(reg);
-		DRM_DEBUG_KMS("FDI_RX_IIR 0x%x\n", temp);
+		for (i = 0; i < 4; i++) {
+			reg = FDI_RX_IIR(pipe);
+			temp = I915_READ(reg);
+			DRM_DEBUG_KMS("FDI_RX_IIR 0x%x\n", temp);
 
-		if (temp & FDI_RX_SYMBOL_LOCK) {
-			I915_WRITE(reg, temp | FDI_RX_SYMBOL_LOCK);
-			DRM_DEBUG_KMS("FDI train 2 done, level %i.\n", i);
-			break;
+			if (temp & FDI_RX_SYMBOL_LOCK ||
+			    (I915_READ(reg) & FDI_RX_SYMBOL_LOCK)) {
+				I915_WRITE(reg, temp | FDI_RX_SYMBOL_LOCK);
+				DRM_DEBUG_KMS("FDI train 2 done, level %i.\n",
+					      i);
+				goto train_done;
+			}
+			udelay(2); /* should be 1.5us */
 		}
+		if (i == 4)
+			DRM_DEBUG_KMS("FDI train 2 fail on vswing %d\n", j / 2);
 	}
-	if (i == 4)
-		DRM_ERROR("FDI train 2 fail!\n");
 
+train_done:
 	DRM_DEBUG_KMS("FDI train done.\n");
 }
 
@@ -3199,7 +3203,7 @@ static void ironlake_pfit_enable(struct intel_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = crtc->pipe;
 
-	if (crtc->config.pch_pfit.size) {
+	if (crtc->config.pch_pfit.enabled) {
 		/* Force use of hard-coded filter coefficients
 		 * as some pre-programmed values are broken,
 		 * e.g. x201.
@@ -3424,7 +3428,7 @@ static void ironlake_pfit_disable(struct intel_crtc *crtc)
 
 	/* To avoid upsetting the power well on haswell only disable the pfit if
 	 * it's in use. The hw state code will make sure we get this right. */
-	if (crtc->config.pch_pfit.size) {
+	if (crtc->config.pch_pfit.enabled) {
 		I915_WRITE(PF_CTL(pipe), 0);
 		I915_WRITE(PF_WIN_POS(pipe), 0);
 		I915_WRITE(PF_WIN_SZ(pipe), 0);
@@ -5855,6 +5859,7 @@ static void ironlake_get_pfit_config(struct intel_crtc *crtc,
 	tmp = I915_READ(PF_CTL(crtc->pipe));
 
 	if (tmp & PF_ENABLE) {
+		pipe_config->pch_pfit.enabled = true;
 		pipe_config->pch_pfit.pos = I915_READ(PF_WIN_POS(crtc->pipe));
 		pipe_config->pch_pfit.size = I915_READ(PF_WIN_SZ(crtc->pipe));
 
@@ -5930,11 +5935,7 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
 	struct intel_crtc *crtc;
 	unsigned long irqflags;
-	uint32_t val, pch_hpd_mask;
-
-	pch_hpd_mask = SDE_PORTB_HOTPLUG_CPT | SDE_PORTC_HOTPLUG_CPT;
-	if (!(dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE))
-		pch_hpd_mask |= SDE_PORTD_HOTPLUG_CPT | SDE_CRT_HOTPLUG_CPT;
+	uint32_t val;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head)
 		WARN(crtc->base.enabled, "CRTC for pipe %c enabled\n",
@@ -5960,7 +5961,7 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	WARN((val & ~DE_PCH_EVENT_IVB) != val,
 	     "Unexpected DEIMR bits enabled: 0x%x\n", val);
 	val = I915_READ(SDEIMR);
-	WARN((val & ~pch_hpd_mask) != val,
+	WARN((val | SDE_HOTPLUG_MASK_CPT) != 0xffffffff,
 	     "Unexpected SDEIMR bits enabled: 0x%x\n", val);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
@@ -6031,16 +6032,21 @@ void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 		    LCPLL_POWER_DOWN_ALLOW)) == LCPLL_PLL_LOCK)
 		return;
 
+	/* Make sure we're not on PC8 state before disabling PC8, otherwise
+	 * we'll hang the machine! */
+	dev_priv->uncore.funcs.force_wake_get(dev_priv);
+
 	if (val & LCPLL_POWER_DOWN_ALLOW) {
 		val &= ~LCPLL_POWER_DOWN_ALLOW;
 		I915_WRITE(LCPLL_CTL, val);
+		POSTING_READ(LCPLL_CTL);
 	}
 
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_FORCE;
 	val &= ~D_COMP_COMP_DISABLE;
 	I915_WRITE(D_COMP, val);
-	I915_READ(D_COMP);
+	POSTING_READ(D_COMP);
 
 	val = I915_READ(LCPLL_CTL);
 	val &= ~LCPLL_PLL_DISABLE;
@@ -6058,6 +6064,168 @@ void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 					LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1))
 			DRM_ERROR("Switching back to LCPLL failed\n");
 	}
+
+	dev_priv->uncore.funcs.force_wake_put(dev_priv);
+}
+
+void hsw_enable_pc8_work(struct work_struct *__work)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(to_delayed_work(__work), struct drm_i915_private,
+			     pc8.enable_work);
+	struct drm_device *dev = dev_priv->dev;
+	uint32_t val;
+
+	if (dev_priv->pc8.enabled)
+		return;
+
+	DRM_DEBUG_KMS("Enabling package C8+\n");
+
+	dev_priv->pc8.enabled = true;
+
+	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
+		val = I915_READ(SOUTH_DSPCLK_GATE_D);
+		val &= ~PCH_LP_PARTITION_LEVEL_DISABLE;
+		I915_WRITE(SOUTH_DSPCLK_GATE_D, val);
+	}
+
+	lpt_disable_clkout_dp(dev);
+	hsw_pc8_disable_interrupts(dev);
+	hsw_disable_lcpll(dev_priv, true, true);
+}
+
+static void __hsw_enable_package_c8(struct drm_i915_private *dev_priv)
+{
+	WARN_ON(!mutex_is_locked(&dev_priv->pc8.lock));
+	WARN(dev_priv->pc8.disable_count < 1,
+	     "pc8.disable_count: %d\n", dev_priv->pc8.disable_count);
+
+	dev_priv->pc8.disable_count--;
+	if (dev_priv->pc8.disable_count != 0)
+		return;
+
+	schedule_delayed_work(&dev_priv->pc8.enable_work,
+			      msecs_to_jiffies(i915_pc8_timeout));
+}
+
+static void __hsw_disable_package_c8(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	uint32_t val;
+
+	WARN_ON(!mutex_is_locked(&dev_priv->pc8.lock));
+	WARN(dev_priv->pc8.disable_count < 0,
+	     "pc8.disable_count: %d\n", dev_priv->pc8.disable_count);
+
+	dev_priv->pc8.disable_count++;
+	if (dev_priv->pc8.disable_count != 1)
+		return;
+
+	cancel_delayed_work_sync(&dev_priv->pc8.enable_work);
+	if (!dev_priv->pc8.enabled)
+		return;
+
+	DRM_DEBUG_KMS("Disabling package C8+\n");
+
+	hsw_restore_lcpll(dev_priv);
+	hsw_pc8_restore_interrupts(dev);
+	lpt_init_pch_refclk(dev);
+
+	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
+		val = I915_READ(SOUTH_DSPCLK_GATE_D);
+		val |= PCH_LP_PARTITION_LEVEL_DISABLE;
+		I915_WRITE(SOUTH_DSPCLK_GATE_D, val);
+	}
+
+	intel_prepare_ddi(dev);
+	i915_gem_init_swizzling(dev);
+	mutex_lock(&dev_priv->rps.hw_lock);
+	gen6_update_ring_freq(dev);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+	dev_priv->pc8.enabled = false;
+}
+
+void hsw_enable_package_c8(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->pc8.lock);
+	__hsw_enable_package_c8(dev_priv);
+	mutex_unlock(&dev_priv->pc8.lock);
+}
+
+void hsw_disable_package_c8(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->pc8.lock);
+	__hsw_disable_package_c8(dev_priv);
+	mutex_unlock(&dev_priv->pc8.lock);
+}
+
+static bool hsw_can_enable_package_c8(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_crtc *crtc;
+	uint32_t val;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head)
+		if (crtc->base.enabled)
+			return false;
+
+	/* This case is still possible since we have the i915.disable_power_well
+	 * parameter and also the KVMr or something else might be requesting the
+	 * power well. */
+	val = I915_READ(HSW_PWR_WELL_DRIVER);
+	if (val != 0) {
+		DRM_DEBUG_KMS("Not enabling PC8: power well on\n");
+		return false;
+	}
+
+	return true;
+}
+
+/* Since we're called from modeset_global_resources there's no way to
+ * symmetrically increase and decrease the refcount, so we use
+ * dev_priv->pc8.requirements_met to track whether we already have the refcount
+ * or not.
+ */
+static void hsw_update_package_c8(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	bool allow;
+
+	if (!i915_enable_pc8)
+		return;
+
+	mutex_lock(&dev_priv->pc8.lock);
+
+	allow = hsw_can_enable_package_c8(dev_priv);
+
+	if (allow == dev_priv->pc8.requirements_met)
+		goto done;
+
+	dev_priv->pc8.requirements_met = allow;
+
+	if (allow)
+		__hsw_enable_package_c8(dev_priv);
+	else
+		__hsw_disable_package_c8(dev_priv);
+
+done:
+	mutex_unlock(&dev_priv->pc8.lock);
+}
+
+static void hsw_package_c8_gpu_idle(struct drm_i915_private *dev_priv)
+{
+	if (!dev_priv->pc8.gpu_idle) {
+		dev_priv->pc8.gpu_idle = true;
+		hsw_enable_package_c8(dev_priv);
+	}
+}
+
+static void hsw_package_c8_gpu_busy(struct drm_i915_private *dev_priv)
+{
+	if (dev_priv->pc8.gpu_idle) {
+		dev_priv->pc8.gpu_idle = false;
+		hsw_disable_package_c8(dev_priv);
+	}
 }
 
 static void haswell_modeset_global_resources(struct drm_device *dev)
@@ -6069,12 +6237,14 @@ static void haswell_modeset_global_resources(struct drm_device *dev)
 		if (!crtc->base.enabled)
 			continue;
 
-		if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.size ||
+		if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.enabled ||
 		    crtc->config.cpu_transcoder != TRANSCODER_EDP)
 			enable = true;
 	}
 
 	intel_set_power_well(dev, enable);
+
+	hsw_update_package_c8(dev);
 }
 
 static int haswell_crtc_mode_set(struct drm_crtc *crtc,
@@ -6325,15 +6495,15 @@ static void haswell_write_eld(struct drm_connector *connector,
 
 	/* Set ELD valid state */
 	tmp = I915_READ(aud_cntrl_st2);
-	DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%08x\n", tmp);
 	tmp |= (AUDIO_ELD_VALID_A << (pipe * 4));
 	I915_WRITE(aud_cntrl_st2, tmp);
 	tmp = I915_READ(aud_cntrl_st2);
-	DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%08x\n", tmp);
 
 	/* Enable HDMI mode */
 	tmp = I915_READ(aud_config);
-	DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%08x\n", tmp);
 	/* clear N_programing_enable and N_value_index */
 	tmp &= ~(AUD_CONFIG_N_VALUE_INDEX | AUD_CONFIG_N_PROG_ENABLE);
 	I915_WRITE(aud_config, tmp);
@@ -6595,8 +6765,10 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
 			cntl |= CURSOR_MODE_DISABLE;
 		}
-		if (IS_HASWELL(dev))
+		if (IS_HASWELL(dev)) {
 			cntl |= CURSOR_PIPE_CSC_ENABLE;
+			cntl &= ~CURSOR_TRICKLE_FEED_DISABLE;
+		}
 		I915_WRITE(CURCNTR_IVB(pipe), cntl);
 
 		intel_crtc->cursor_visible = visible;
@@ -7142,8 +7314,7 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 		}
 	}
 
-	pipe_config->adjusted_mode.clock = clock.dot *
-		pipe_config->pixel_multiplier;
+	pipe_config->adjusted_mode.clock = clock.dot;
 }
 
 static void ironlake_crtc_clock_get(struct intel_crtc *crtc,
@@ -7306,13 +7477,19 @@ static void intel_decrease_pllclock(struct drm_crtc *crtc)
 
 void intel_mark_busy(struct drm_device *dev)
 {
-	i915_update_gfx_val(dev->dev_private);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	hsw_package_c8_gpu_busy(dev_priv);
+	i915_update_gfx_val(dev_priv);
 }
 
 void intel_mark_idle(struct drm_device *dev)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
 
+	hsw_package_c8_gpu_idle(dev_priv);
+
 	if (!i915_powersave)
 		return;
 
@@ -7477,7 +7654,8 @@ inline static void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
 static int intel_gen2_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
-				 struct drm_i915_gem_object *obj)
+				 struct drm_i915_gem_object *obj,
+				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -7521,7 +7699,8 @@ err:
 static int intel_gen3_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
-				 struct drm_i915_gem_object *obj)
+				 struct drm_i915_gem_object *obj,
+				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -7562,7 +7741,8 @@ err:
 static int intel_gen4_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
-				 struct drm_i915_gem_object *obj)
+				 struct drm_i915_gem_object *obj,
+				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -7610,7 +7790,8 @@ err:
 static int intel_gen6_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
-				 struct drm_i915_gem_object *obj)
+				 struct drm_i915_gem_object *obj,
+				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -7651,22 +7832,21 @@ err:
 	return ret;
 }
 
-/*
- * On gen7 we currently use the blit ring because (in early silicon at least)
- * the render ring doesn't give us interrpts for page flip completion, which
- * means clients will hang after the first flip is queued.  Fortunately the
- * blit ring generates interrupts properly, so use it instead.
- */
 static int intel_gen7_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
-				 struct drm_i915_gem_object *obj)
+				 struct drm_i915_gem_object *obj,
+				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_ring_buffer *ring = &dev_priv->ring[BCS];
+	struct intel_ring_buffer *ring;
 	uint32_t plane_bit = 0;
-	int ret;
+	int len, ret;
+
+	ring = obj->ring;
+	if (IS_VALLEYVIEW(dev) || ring == NULL || ring->id != RCS)
+		ring = &dev_priv->ring[BCS];
 
 	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
 	if (ret)
@@ -7688,10 +7868,34 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		goto err_unpin;
 	}
 
-	ret = intel_ring_begin(ring, 4);
+	len = 4;
+	if (ring->id == RCS)
+		len += 6;
+
+	ret = intel_ring_begin(ring, len);
 	if (ret)
 		goto err_unpin;
 
+	/* Unmask the flip-done completion message. Note that the bspec says that
+	 * we should do this for both the BCS and RCS, and that we must not unmask
+	 * more than one flip event at any time (or ensure that one flip message
+	 * can be sent by waiting for flip-done prior to queueing new flips).
+	 * Experimentation says that BCS works despite DERRMR masking all
+	 * flip-done completion events and that unmasking all planes at once
+	 * for the RCS also doesn't appear to drop events. Setting the DERRMR
+	 * to zero does lead to lockups within MI_DISPLAY_FLIP.
+	 */
+	if (ring->id == RCS) {
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit(ring, DERRMR);
+		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
+					DERRMR_PIPEB_PRI_FLIP_DONE |
+					DERRMR_PIPEC_PRI_FLIP_DONE));
+		intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1));
+		intel_ring_emit(ring, DERRMR);
+		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
+	}
+
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
 	intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
 	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
@@ -7710,14 +7914,16 @@ err:
 static int intel_default_queue_flip(struct drm_device *dev,
 				    struct drm_crtc *crtc,
 				    struct drm_framebuffer *fb,
-				    struct drm_i915_gem_object *obj)
+				    struct drm_i915_gem_object *obj,
+				    uint32_t flags)
 {
 	return -ENODEV;
 }
 
 static int intel_crtc_page_flip(struct drm_crtc *crtc,
 				struct drm_framebuffer *fb,
-				struct drm_pending_vblank_event *event)
+				struct drm_pending_vblank_event *event,
+				uint32_t page_flip_flags)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -7787,7 +7993,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&intel_crtc->unpin_work_count);
 	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
-	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj);
+	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, page_flip_flags);
 	if (ret)
 		goto cleanup_pending;
 
@@ -8000,9 +8206,10 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->gmch_pfit.control,
 		      pipe_config->gmch_pfit.pgm_ratios,
 		      pipe_config->gmch_pfit.lvds_border_bits);
-	DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x\n",
+	DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x, %s\n",
 		      pipe_config->pch_pfit.pos,
-		      pipe_config->pch_pfit.size);
+		      pipe_config->pch_pfit.size,
+		      pipe_config->pch_pfit.enabled ? "enabled" : "disabled");
 	DRM_DEBUG_KMS("ips: %i\n", pipe_config->ips_enabled);
 }
 
@@ -8398,8 +8605,11 @@ intel_pipe_config_compare(struct drm_device *dev,
 	if (INTEL_INFO(dev)->gen < 4)
 		PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios);
 	PIPE_CONF_CHECK_I(gmch_pfit.lvds_border_bits);
-	PIPE_CONF_CHECK_I(pch_pfit.pos);
-	PIPE_CONF_CHECK_I(pch_pfit.size);
+	PIPE_CONF_CHECK_I(pch_pfit.enabled);
+	if (current_config->pch_pfit.enabled) {
+		PIPE_CONF_CHECK_I(pch_pfit.pos);
+		PIPE_CONF_CHECK_I(pch_pfit.size);
+	}
 
 	PIPE_CONF_CHECK_I(ips_enabled);
 
@@ -8546,9 +8756,11 @@ check_crtc_state(struct drm_device *dev)
 
 		list_for_each_entry(encoder, &dev->mode_config.encoder_list,
 				    base.head) {
+			enum pipe pipe;
 			if (encoder->base.crtc != &crtc->base)
 				continue;
-			if (encoder->get_config)
+			if (encoder->get_config &&
+			    encoder->get_hw_state(encoder, &pipe))
 				encoder->get_config(encoder, &pipe_config);
 		}
 
@@ -9840,6 +10052,33 @@ static void i915_disable_vga(struct drm_device *dev)
 	POSTING_READ(vga_reg);
 }
 
+static void i915_enable_vga_mem(struct drm_device *dev)
+{
+	/* Enable VGA memory on Intel HD */
+	if (HAS_PCH_SPLIT(dev)) {
+		vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO);
+		outb(inb(VGA_MSR_READ) | VGA_MSR_MEM_EN, VGA_MSR_WRITE);
+		vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO |
+						   VGA_RSRC_LEGACY_MEM |
+						   VGA_RSRC_NORMAL_IO |
+						   VGA_RSRC_NORMAL_MEM);
+		vga_put(dev->pdev, VGA_RSRC_LEGACY_IO);
+	}
+}
+
+void i915_disable_vga_mem(struct drm_device *dev)
+{
+	/* Disable VGA memory on Intel HD */
+	if (HAS_PCH_SPLIT(dev)) {
+		vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO);
+		outb(inb(VGA_MSR_READ) & ~VGA_MSR_MEM_EN, VGA_MSR_WRITE);
+		vga_set_legacy_decoding(dev->pdev, VGA_RSRC_LEGACY_IO |
+						   VGA_RSRC_NORMAL_IO |
+						   VGA_RSRC_NORMAL_MEM);
+		vga_put(dev->pdev, VGA_RSRC_LEGACY_IO);
+	}
+}
+
 void intel_modeset_init_hw(struct drm_device *dev)
 {
 	intel_init_power_well(dev);
@@ -10118,6 +10357,7 @@ void i915_redisable_vga(struct drm_device *dev)
 	if (I915_READ(vga_reg) != VGA_DISP_DISABLE) {
 		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
 		i915_disable_vga(dev);
+		i915_disable_vga_mem(dev);
 	}
 }
 
@@ -10331,6 +10571,8 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_disable_fbc(dev);
 
+	i915_enable_vga_mem(dev);
+
 	intel_disable_gt_powersave(dev);
 
 	ironlake_teardown_rc6(dev);
@@ -10385,6 +10627,8 @@ struct intel_display_error_state {
 
 	u32 power_well_driver;
 
+	int num_transcoders;
+
 	struct intel_cursor_error_state {
 		u32 control;
 		u32 position;
@@ -10393,16 +10637,7 @@ struct intel_display_error_state {
 	} cursor[I915_MAX_PIPES];
 
 	struct intel_pipe_error_state {
-		enum transcoder cpu_transcoder;
-		u32 conf;
 		u32 source;
-
-		u32 htotal;
-		u32 hblank;
-		u32 hsync;
-		u32 vtotal;
-		u32 vblank;
-		u32 vsync;
 	} pipe[I915_MAX_PIPES];
 
 	struct intel_plane_error_state {
@@ -10414,6 +10649,19 @@ struct intel_display_error_state {
 		u32 surface;
 		u32 tile_offset;
 	} plane[I915_MAX_PIPES];
+
+	struct intel_transcoder_error_state {
+		enum transcoder cpu_transcoder;
+
+		u32 conf;
+
+		u32 htotal;
+		u32 hblank;
+		u32 hsync;
+		u32 vtotal;
+		u32 vblank;
+		u32 vsync;
+	} transcoder[4];
 };
 
 struct intel_display_error_state *
@@ -10421,9 +10669,17 @@ intel_display_capture_error_state(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_display_error_state *error;
-	enum transcoder cpu_transcoder;
+	int transcoders[] = {
+		TRANSCODER_A,
+		TRANSCODER_B,
+		TRANSCODER_C,
+		TRANSCODER_EDP,
+	};
 	int i;
 
+	if (INTEL_INFO(dev)->num_pipes == 0)
+		return NULL;
+
 	error = kmalloc(sizeof(*error), GFP_ATOMIC);
 	if (error == NULL)
 		return NULL;
@@ -10432,9 +10688,6 @@ intel_display_capture_error_state(struct drm_device *dev)
 		error->power_well_driver = I915_READ(HSW_PWR_WELL_DRIVER);
 
 	for_each_pipe(i) {
-		cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv, i);
-		error->pipe[i].cpu_transcoder = cpu_transcoder;
-
 		if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev)) {
 			error->cursor[i].control = I915_READ(CURCNTR(i));
 			error->cursor[i].position = I915_READ(CURPOS(i));
@@ -10458,14 +10711,25 @@ intel_display_capture_error_state(struct drm_device *dev)
 			error->plane[i].tile_offset = I915_READ(DSPTILEOFF(i));
 		}
 
-		error->pipe[i].conf = I915_READ(PIPECONF(cpu_transcoder));
 		error->pipe[i].source = I915_READ(PIPESRC(i));
-		error->pipe[i].htotal = I915_READ(HTOTAL(cpu_transcoder));
-		error->pipe[i].hblank = I915_READ(HBLANK(cpu_transcoder));
-		error->pipe[i].hsync = I915_READ(HSYNC(cpu_transcoder));
-		error->pipe[i].vtotal = I915_READ(VTOTAL(cpu_transcoder));
-		error->pipe[i].vblank = I915_READ(VBLANK(cpu_transcoder));
-		error->pipe[i].vsync = I915_READ(VSYNC(cpu_transcoder));
+	}
+
+	error->num_transcoders = INTEL_INFO(dev)->num_pipes;
+	if (HAS_DDI(dev_priv->dev))
+		error->num_transcoders++; /* Account for eDP. */
+
+	for (i = 0; i < error->num_transcoders; i++) {
+		enum transcoder cpu_transcoder = transcoders[i];
+
+		error->transcoder[i].cpu_transcoder = cpu_transcoder;
+
+		error->transcoder[i].conf = I915_READ(PIPECONF(cpu_transcoder));
+		error->transcoder[i].htotal = I915_READ(HTOTAL(cpu_transcoder));
+		error->transcoder[i].hblank = I915_READ(HBLANK(cpu_transcoder));
+		error->transcoder[i].hsync = I915_READ(HSYNC(cpu_transcoder));
+		error->transcoder[i].vtotal = I915_READ(VTOTAL(cpu_transcoder));
+		error->transcoder[i].vblank = I915_READ(VBLANK(cpu_transcoder));
+		error->transcoder[i].vsync = I915_READ(VSYNC(cpu_transcoder));
 	}
 
 	/* In the code above we read the registers without checking if the power
@@ -10486,22 +10750,16 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 {
 	int i;
 
+	if (!error)
+		return;
+
 	err_printf(m, "Num Pipes: %d\n", INTEL_INFO(dev)->num_pipes);
 	if (HAS_POWER_WELL(dev))
 		err_printf(m, "PWR_WELL_CTL2: %08x\n",
 			   error->power_well_driver);
 	for_each_pipe(i) {
 		err_printf(m, "Pipe [%d]:\n", i);
-		err_printf(m, "  CPU transcoder: %c\n",
-			   transcoder_name(error->pipe[i].cpu_transcoder));
-		err_printf(m, "  CONF: %08x\n", error->pipe[i].conf);
 		err_printf(m, "  SRC: %08x\n", error->pipe[i].source);
-		err_printf(m, "  HTOTAL: %08x\n", error->pipe[i].htotal);
-		err_printf(m, "  HBLANK: %08x\n", error->pipe[i].hblank);
-		err_printf(m, "  HSYNC: %08x\n", error->pipe[i].hsync);
-		err_printf(m, "  VTOTAL: %08x\n", error->pipe[i].vtotal);
-		err_printf(m, "  VBLANK: %08x\n", error->pipe[i].vblank);
-		err_printf(m, "  VSYNC: %08x\n", error->pipe[i].vsync);
 
 		err_printf(m, "Plane [%d]:\n", i);
 		err_printf(m, "  CNTR: %08x\n", error->plane[i].control);
@@ -10522,4 +10780,16 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  POS: %08x\n", error->cursor[i].position);
 		err_printf(m, "  BASE: %08x\n", error->cursor[i].base);
 	}
+
+	for (i = 0; i < error->num_transcoders; i++) {
+		err_printf(m, "  CPU transcoder: %c\n",
+			   transcoder_name(error->transcoder[i].cpu_transcoder));
+		err_printf(m, "  CONF: %08x\n", error->transcoder[i].conf);
+		err_printf(m, "  HTOTAL: %08x\n", error->transcoder[i].htotal);
+		err_printf(m, "  HBLANK: %08x\n", error->transcoder[i].hblank);
+		err_printf(m, "  HSYNC: %08x\n", error->transcoder[i].hsync);
+		err_printf(m, "  VTOTAL: %08x\n", error->transcoder[i].vtotal);
+		err_printf(m, "  VBLANK: %08x\n", error->transcoder[i].vblank);
+		err_printf(m, "  VSYNC: %08x\n", error->transcoder[i].vsync);
+	}
 }