Merge tag 'drm-intel-next-2014-07-11' of git://anongit.freedesktop.org/drm-intel into drm-next

- fbc improvements when stolen memory is tight (Ben)
- cdclk handling improvements for vlv/chv (Ville)
- proper fix for stuck primary planes on gmch platforms with cxsr (Imre&Ebgert
  Eich)
- gen8 hw semaphore support (Ben)
- more execlist prep work from Oscar Mateo
- locking fixes for primary planes (Matt Roper)
- code rework to support runtime pm for dpms on hsw/bdw (Paulo, Imre & me), but
  not yet enabled because some fixes from Paulo haven't made the cut
- more gpu boost tuning from Chris
- as usual piles of little things all over

* tag 'drm-intel-next-2014-07-11' of git://anongit.freedesktop.org/drm-intel: (93 commits)
  drm/i915: Make the RPS interrupt generation mask handle the vlv wa
  drm/i915: Move RPS evaluation interval counters to i915->rps
  drm/i915: Don't cast a pointer to void* unnecessarily
  drm/i915: don't read LVDS regs at compute_config time
  drm/i915: check the power domains in intel_lvds_get_hw_state()
  drm/i915: check the power domains in ironlake_get_pipe_config()
  drm/i915: don't skip shared DPLL assertion on LPT
  drm/i915: Only touch WRPLL hw state in enable/disable hooks
  drm/i915: Switch to common shared dpll framework for WRPLLs
  drm/i915: ->enable hook for WRPLLs
  drm/i915: ->disable hook for WRPLLs
  drm/i915: State readout support for WRPLLs
  drm/i915: add POWER_DOMAIN_PLLS
  drm/i915: Document that the pll->mode_set hook is optional
  drm/i915: Basic shared dpll support for WRPLLs
  drm/i915: Precompute static ddi_pll_sel values in encoders
  drm/i915: BDW also has special-purpose DP DDI clocks
  drm/i915: State readout and cross-checking for ddi_pll_sel
  drm/i915: Move ddi_pll_sel into the pipe config
  drm/i915: Add a debugfs file for the shared dpll state
  ...

drivers/gpu/drm/i915/i915_debugfs.c

@@ -176,7 +176,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 
 static void describe_ctx(struct seq_file *m, struct intel_context *ctx)
 {
-	seq_putc(m, ctx->is_initialized ? 'I' : 'i');
+	seq_putc(m, ctx->legacy_hw_ctx.initialized ? 'I' : 'i');
 	seq_putc(m, ctx->remap_slice ? 'R' : 'r');
 	seq_putc(m, ' ');
 }
@@ -994,29 +994,6 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_next_seqno_fops,
 			i915_next_seqno_get, i915_next_seqno_set,
 			"0x%llx\n");
 
-static int i915_rstdby_delays(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u16 crstanddelay;
-	int ret;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
-	intel_runtime_pm_get(dev_priv);
-
-	crstanddelay = I915_READ16(CRSTANDVID);
-
-	intel_runtime_pm_put(dev_priv);
-	mutex_unlock(&dev->struct_mutex);
-
-	seq_printf(m, "w/ctx: %d, w/o ctx: %d\n", (crstanddelay >> 8) & 0x3f, (crstanddelay & 0x3f));
-
-	return 0;
-}
-
 static int i915_frequency_info(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = m->private;
@@ -1158,61 +1135,6 @@ out:
 	return ret;
 }
 
-static int i915_delayfreq_table(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 delayfreq;
-	int ret, i;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
-	intel_runtime_pm_get(dev_priv);
-
-	for (i = 0; i < 16; i++) {
-		delayfreq = I915_READ(PXVFREQ_BASE + i * 4);
-		seq_printf(m, "P%02dVIDFREQ: 0x%08x (VID: %d)\n", i, delayfreq,
-			   (delayfreq & PXVFREQ_PX_MASK) >> PXVFREQ_PX_SHIFT);
-	}
-
-	intel_runtime_pm_put(dev_priv);
-
-	mutex_unlock(&dev->struct_mutex);
-
-	return 0;
-}
-
-static inline int MAP_TO_MV(int map)
-{
-	return 1250 - (map * 25);
-}
-
-static int i915_inttoext_table(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 inttoext;
-	int ret, i;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
-	intel_runtime_pm_get(dev_priv);
-
-	for (i = 1; i <= 32; i++) {
-		inttoext = I915_READ(INTTOEXT_BASE_ILK + i * 4);
-		seq_printf(m, "INTTOEXT%02d: 0x%08x\n", i, inttoext);
-	}
-
-	intel_runtime_pm_put(dev_priv);
-	mutex_unlock(&dev->struct_mutex);
-
-	return 0;
-}
-
 static int ironlake_drpc_info(struct seq_file *m)
 {
 	struct drm_info_node *node = m->private;
@@ -1523,10 +1445,17 @@ static int i915_ips_status(struct seq_file *m, void *unused)
 
 	intel_runtime_pm_get(dev_priv);
 
-	if (IS_BROADWELL(dev) || I915_READ(IPS_CTL) & IPS_ENABLE)
-		seq_puts(m, "enabled\n");
-	else
-		seq_puts(m, "disabled\n");
+	seq_printf(m, "Enabled by kernel parameter: %s\n",
+		   yesno(i915.enable_ips));
+
+	if (INTEL_INFO(dev)->gen >= 8) {
+		seq_puts(m, "Currently: unknown\n");
+	} else {
+		if (I915_READ(IPS_CTL) & IPS_ENABLE)
+			seq_puts(m, "Currently: enabled\n");
+		else
+			seq_puts(m, "Currently: disabled\n");
+	}
 
 	intel_runtime_pm_put(dev_priv);
 
@@ -1630,26 +1559,6 @@ out:
 	return ret;
 }
 
-static int i915_gfxec(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
-
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
-	intel_runtime_pm_get(dev_priv);
-
-	seq_printf(m, "GFXEC: %ld\n", (unsigned long)I915_READ(0x112f4));
-	intel_runtime_pm_put(dev_priv);
-
-	mutex_unlock(&dev->struct_mutex);
-
-	return 0;
-}
-
 static int i915_opregion(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = m->private;
@@ -1746,7 +1655,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	}
 
 	list_for_each_entry(ctx, &dev_priv->context_list, link) {
-		if (ctx->obj == NULL)
+		if (ctx->legacy_hw_ctx.rcs_state == NULL)
 			continue;
 
 		seq_puts(m, "HW context ");
@@ -1755,7 +1664,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 			if (ring->default_context == ctx)
 				seq_printf(m, "(default context %s) ", ring->name);
 
-		describe_obj(m, ctx->obj);
+		describe_obj(m, ctx->legacy_hw_ctx.rcs_state);
 		seq_putc(m, '\n');
 	}
 
@@ -1869,7 +1778,7 @@ static int per_file_ctx(int id, void *ptr, void *data)
 	if (i915_gem_context_is_default(ctx))
 		seq_puts(m, "  default context:\n");
 	else
-		seq_printf(m, "  context %d:\n", ctx->id);
+		seq_printf(m, "  context %d:\n", ctx->user_handle);
 	ppgtt->debug_dump(ppgtt, m);
 
 	return 0;
@@ -2134,6 +2043,8 @@ static const char *power_domain_str(enum intel_display_power_domain domain)
 		return "VGA";
 	case POWER_DOMAIN_AUDIO:
 		return "AUDIO";
+	case POWER_DOMAIN_PLLS:
+		return "PLLS";
 	case POWER_DOMAIN_INIT:
 		return "INIT";
 	default:
@@ -2358,17 +2269,17 @@ static int i915_display_info(struct seq_file *m, void *unused)
 		bool active;
 		int x, y;
 
-		seq_printf(m, "CRTC %d: pipe: %c, active: %s\n",
+		seq_printf(m, "CRTC %d: pipe: %c, active=%s (size=%dx%d)\n",
 			   crtc->base.base.id, pipe_name(crtc->pipe),
-			   yesno(crtc->active));
+			   yesno(crtc->active), crtc->config.pipe_src_w, crtc->config.pipe_src_h);
 		if (crtc->active) {
 			intel_crtc_info(m, crtc);
 
 			active = cursor_position(dev, crtc->pipe, &x, &y);
-			seq_printf(m, "\tcursor visible? %s, position (%d, %d), addr 0x%08x, active? %s\n",
+			seq_printf(m, "\tcursor visible? %s, position (%d, %d), size %dx%d, addr 0x%08x, active? %s\n",
 				   yesno(crtc->cursor_base),
-				   x, y, crtc->cursor_addr,
-				   yesno(active));
+				   x, y, crtc->cursor_width, crtc->cursor_height,
+				   crtc->cursor_addr, yesno(active));
 		}
 
 		seq_printf(m, "\tunderrun reporting: cpu=%s pch=%s \n",
@@ -2388,6 +2299,104 @@ static int i915_display_info(struct seq_file *m, void *unused)
 	return 0;
 }
 
+static int i915_semaphore_status(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring;
+	int num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	int i, j, ret;
+
+	if (!i915_semaphore_is_enabled(dev)) {
+		seq_puts(m, "Semaphores are disabled\n");
+		return 0;
+	}
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+	intel_runtime_pm_get(dev_priv);
+
+	if (IS_BROADWELL(dev)) {
+		struct page *page;
+		uint64_t *seqno;
+
+		page = i915_gem_object_get_page(dev_priv->semaphore_obj, 0);
+
+		seqno = (uint64_t *)kmap_atomic(page);
+		for_each_ring(ring, dev_priv, i) {
+			uint64_t offset;
+
+			seq_printf(m, "%s\n", ring->name);
+
+			seq_puts(m, "  Last signal:");
+			for (j = 0; j < num_rings; j++) {
+				offset = i * I915_NUM_RINGS + j;
+				seq_printf(m, "0x%08llx (0x%02llx) ",
+					   seqno[offset], offset * 8);
+			}
+			seq_putc(m, '\n');
+
+			seq_puts(m, "  Last wait:  ");
+			for (j = 0; j < num_rings; j++) {
+				offset = i + (j * I915_NUM_RINGS);
+				seq_printf(m, "0x%08llx (0x%02llx) ",
+					   seqno[offset], offset * 8);
+			}
+			seq_putc(m, '\n');
+
+		}
+		kunmap_atomic(seqno);
+	} else {
+		seq_puts(m, "  Last signal:");
+		for_each_ring(ring, dev_priv, i)
+			for (j = 0; j < num_rings; j++)
+				seq_printf(m, "0x%08x\n",
+					   I915_READ(ring->semaphore.mbox.signal[j]));
+		seq_putc(m, '\n');
+	}
+
+	seq_puts(m, "\nSync seqno:\n");
+	for_each_ring(ring, dev_priv, i) {
+		for (j = 0; j < num_rings; j++) {
+			seq_printf(m, "  0x%08x ", ring->semaphore.sync_seqno[j]);
+		}
+		seq_putc(m, '\n');
+	}
+	seq_putc(m, '\n');
+
+	intel_runtime_pm_put(dev_priv);
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static int i915_shared_dplls_info(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int i;
+
+	drm_modeset_lock_all(dev);
+	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
+		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
+
+		seq_printf(m, "DPLL%i: %s, id: %i\n", i, pll->name, pll->id);
+		seq_printf(m, " refcount: %i, active: %i, on: %s\n", pll->refcount,
+			   pll->active, yesno(pll->on));
+		seq_printf(m, " tracked hardware state:\n");
+		seq_printf(m, " dpll:    0x%08x\n", pll->hw_state.dpll);
+		seq_printf(m, " dpll_md: 0x%08x\n", pll->hw_state.dpll_md);
+		seq_printf(m, " fp0:     0x%08x\n", pll->hw_state.fp0);
+		seq_printf(m, " fp1:     0x%08x\n", pll->hw_state.fp1);
+		seq_printf(m, " wrpll:   0x%08x\n", pll->hw_state.wrpll);
+	}
+	drm_modeset_unlock_all(dev);
+
+	return 0;
+}
+
 struct pipe_crc_info {
 	const char *name;
 	struct drm_device *dev;
@@ -2860,7 +2869,60 @@ static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 	return 0;
 }
 
-static int ivb_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
+static void hsw_trans_edp_pipe_A_crc_wa(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *crtc =
+		to_intel_crtc(dev_priv->pipe_to_crtc_mapping[PIPE_A]);
+
+	drm_modeset_lock_all(dev);
+	/*
+	 * If we use the eDP transcoder we need to make sure that we don't
+	 * bypass the pfit, since otherwise the pipe CRC source won't work. Only
+	 * relevant on hsw with pipe A when using the always-on power well
+	 * routing.
+	 */
+	if (crtc->config.cpu_transcoder == TRANSCODER_EDP &&
+	    !crtc->config.pch_pfit.enabled) {
+		crtc->config.pch_pfit.force_thru = true;
+
+		intel_display_power_get(dev_priv,
+					POWER_DOMAIN_PIPE_PANEL_FITTER(PIPE_A));
+
+		dev_priv->display.crtc_disable(&crtc->base);
+		dev_priv->display.crtc_enable(&crtc->base);
+	}
+	drm_modeset_unlock_all(dev);
+}
+
+static void hsw_undo_trans_edp_pipe_A_crc_wa(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *crtc =
+		to_intel_crtc(dev_priv->pipe_to_crtc_mapping[PIPE_A]);
+
+	drm_modeset_lock_all(dev);
+	/*
+	 * If we use the eDP transcoder we need to make sure that we don't
+	 * bypass the pfit, since otherwise the pipe CRC source won't work. Only
+	 * relevant on hsw with pipe A when using the always-on power well
+	 * routing.
+	 */
+	if (crtc->config.pch_pfit.force_thru) {
+		crtc->config.pch_pfit.force_thru = false;
+
+		dev_priv->display.crtc_disable(&crtc->base);
+		dev_priv->display.crtc_enable(&crtc->base);
+
+		intel_display_power_put(dev_priv,
+					POWER_DOMAIN_PIPE_PANEL_FITTER(PIPE_A));
+	}
+	drm_modeset_unlock_all(dev);
+}
+
+static int ivb_pipe_crc_ctl_reg(struct drm_device *dev,
+				enum pipe pipe,
+				enum intel_pipe_crc_source *source,
 				uint32_t *val)
 {
 	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
@@ -2874,6 +2936,9 @@ static int ivb_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_SPRITE_IVB;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_PF:
+		if (IS_HASWELL(dev) && pipe == PIPE_A)
+			hsw_trans_edp_pipe_A_crc_wa(dev);
+
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PF_IVB;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_NONE:
@@ -2906,11 +2971,11 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 	else if (INTEL_INFO(dev)->gen < 5)
 		ret = i9xx_pipe_crc_ctl_reg(dev, pipe, &source, &val);
 	else if (IS_VALLEYVIEW(dev))
-		ret = vlv_pipe_crc_ctl_reg(dev,pipe, &source, &val);
+		ret = vlv_pipe_crc_ctl_reg(dev, pipe, &source, &val);
 	else if (IS_GEN5(dev) || IS_GEN6(dev))
 		ret = ilk_pipe_crc_ctl_reg(&source, &val);
 	else
-		ret = ivb_pipe_crc_ctl_reg(&source, &val);
+		ret = ivb_pipe_crc_ctl_reg(dev, pipe, &source, &val);
 
 	if (ret != 0)
 		return ret;
@@ -2962,6 +3027,8 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 			g4x_undo_pipe_scramble_reset(dev, pipe);
 		else if (IS_VALLEYVIEW(dev))
 			vlv_undo_pipe_scramble_reset(dev, pipe);
+		else if (IS_HASWELL(dev) && pipe == PIPE_A)
+			hsw_undo_trans_edp_pipe_A_crc_wa(dev);
 	}
 
 	return 0;
@@ -3815,14 +3882,10 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_blt", i915_hws_info, 0, (void *)BCS},
 	{"i915_gem_hws_bsd", i915_hws_info, 0, (void *)VCS},
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
-	{"i915_rstdby_delays", i915_rstdby_delays, 0},
 	{"i915_frequency_info", i915_frequency_info, 0},
-	{"i915_delayfreq_table", i915_delayfreq_table, 0},
-	{"i915_inttoext_table", i915_inttoext_table, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
 	{"i915_emon_status", i915_emon_status, 0},
 	{"i915_ring_freq_table", i915_ring_freq_table, 0},
-	{"i915_gfxec", i915_gfxec, 0},
 	{"i915_fbc_status", i915_fbc_status, 0},
 	{"i915_ips_status", i915_ips_status, 0},
 	{"i915_sr_status", i915_sr_status, 0},
@@ -3839,6 +3902,8 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_pc8_status", i915_pc8_status, 0},
 	{"i915_power_domain_info", i915_power_domain_info, 0},
 	{"i915_display_info", i915_display_info, 0},
+	{"i915_semaphore_status", i915_semaphore_status, 0},
+	{"i915_shared_dplls_info", i915_shared_dplls_info, 0},
 };
 #define I915_DEBUGFS_ENTRIES ARRAY_SIZE(i915_debugfs_list)
 

drivers/gpu/drm/i915/i915_dma.c

@@ -1593,7 +1593,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (dev_priv == NULL)
 		return -ENOMEM;
 
-	dev->dev_private = (void *)dev_priv;
+	dev->dev_private = dev_priv;
 	dev_priv->dev = dev;
 
 	/* copy initial configuration to dev_priv->info */
@@ -1954,11 +1954,11 @@ void i915_driver_lastclose(struct drm_device *dev)
 	i915_dma_cleanup(dev);
 }
 
-void i915_driver_preclose(struct drm_device *dev, struct drm_file *file_priv)
+void i915_driver_preclose(struct drm_device *dev, struct drm_file *file)
 {
 	mutex_lock(&dev->struct_mutex);
-	i915_gem_context_close(dev, file_priv);
-	i915_gem_release(dev, file_priv);
+	i915_gem_context_close(dev, file);
+	i915_gem_release(dev, file);
 	mutex_unlock(&dev->struct_mutex);
 }
 

drivers/gpu/drm/i915/i915_drv.c

@@ -477,10 +477,6 @@ bool i915_semaphore_is_enabled(struct drm_device *dev)
 	if (i915.semaphores >= 0)
 		return i915.semaphores;
 
-	/* Until we get further testing... */
-	if (IS_GEN8(dev))
-		return false;
-
 #ifdef CONFIG_INTEL_IOMMU
 	/* Enable semaphores on SNB when IO remapping is off */
 	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped)
@@ -520,6 +516,8 @@ static int i915_drm_freeze(struct drm_device *dev)
 			return error;
 		}
 
+		flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 		intel_runtime_pm_disable_interrupts(dev);
 
 		intel_suspend_gt_powersave(dev);
@@ -541,10 +539,11 @@ static int i915_drm_freeze(struct drm_device *dev)
 
 	i915_save_state(dev);
 
-	if (acpi_target_system_state() >= ACPI_STATE_S3)
-		opregion_target_state = PCI_D3cold;
-	else
+	opregion_target_state = PCI_D3cold;
+#if IS_ENABLED(CONFIG_ACPI_SLEEP)
+	if (acpi_target_system_state() < ACPI_STATE_S3)
 		opregion_target_state = PCI_D1;
+#endif
 	intel_opregion_notify_adapter(dev, opregion_target_state);
 
 	intel_uncore_forcewake_reset(dev, false);

drivers/gpu/drm/i915/i915_drv.h

@@ -129,6 +129,7 @@ enum intel_display_power_domain {
 	POWER_DOMAIN_PORT_OTHER,
 	POWER_DOMAIN_VGA,
 	POWER_DOMAIN_AUDIO,
+	POWER_DOMAIN_PLLS,
 	POWER_DOMAIN_INIT,
 
 	POWER_DOMAIN_NUM,
@@ -184,8 +185,10 @@ struct i915_mmu_object;
 enum intel_dpll_id {
 	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
 	/* real shared dpll ids must be >= 0 */
-	DPLL_ID_PCH_PLL_A,
-	DPLL_ID_PCH_PLL_B,
+	DPLL_ID_PCH_PLL_A = 0,
+	DPLL_ID_PCH_PLL_B = 1,
+	DPLL_ID_WRPLL1 = 0,
+	DPLL_ID_WRPLL2 = 1,
 };
 #define I915_NUM_PLLS 2
 
@@ -194,6 +197,7 @@ struct intel_dpll_hw_state {
 	uint32_t dpll_md;
 	uint32_t fp0;
 	uint32_t fp1;
+	uint32_t wrpll;
 };
 
 struct intel_shared_dpll {
@@ -204,6 +208,8 @@ struct intel_shared_dpll {
 	/* should match the index in the dev_priv->shared_dplls array */
 	enum intel_dpll_id id;
 	struct intel_dpll_hw_state hw_state;
+	/* The mode_set hook is optional and should be used together with the
+	 * intel_prepare_shared_dpll function. */
 	void (*mode_set)(struct drm_i915_private *dev_priv,
 			 struct intel_shared_dpll *pll);
 	void (*enable)(struct drm_i915_private *dev_priv,
@@ -228,12 +234,6 @@ void intel_link_compute_m_n(int bpp, int nlanes,
 			    int pixel_clock, int link_clock,
 			    struct intel_link_m_n *m_n);
 
-struct intel_ddi_plls {
-	int spll_refcount;
-	int wrpll1_refcount;
-	int wrpll2_refcount;
-};
-
 /* Interface history:
  *
  * 1.1: Original.
@@ -324,6 +324,7 @@ struct drm_i915_error_state {
 	u64 fence[I915_MAX_NUM_FENCES];
 	struct intel_overlay_error_state *overlay;
 	struct intel_display_error_state *display;
+	struct drm_i915_error_object *semaphore_obj;
 
 	struct drm_i915_error_ring {
 		bool valid;
@@ -584,27 +585,48 @@ struct i915_ctx_hang_stats {
 };
 
 /* This must match up with the value previously used for execbuf2.rsvd1. */
-#define DEFAULT_CONTEXT_ID 0
+#define DEFAULT_CONTEXT_HANDLE 0
+/**
+ * struct intel_context - as the name implies, represents a context.
+ * @ref: reference count.
+ * @user_handle: userspace tracking identity for this context.
+ * @remap_slice: l3 row remapping information.
+ * @file_priv: filp associated with this context (NULL for global default
+ *	       context).
+ * @hang_stats: information about the role of this context in possible GPU
+ *		hangs.
+ * @vm: virtual memory space used by this context.
+ * @legacy_hw_ctx: render context backing object and whether it is correctly
+ *                initialized (legacy ring submission mechanism only).
+ * @link: link in the global list of contexts.
+ *
+ * Contexts are memory images used by the hardware to store copies of their
+ * internal state.
+ */
 struct intel_context {
 	struct kref ref;
-	int id;
-	bool is_initialized;
+	int user_handle;
 	uint8_t remap_slice;
 	struct drm_i915_file_private *file_priv;
-	struct drm_i915_gem_object *obj;
 	struct i915_ctx_hang_stats hang_stats;
 	struct i915_address_space *vm;
 
+	struct {
+		struct drm_i915_gem_object *rcs_state;
+		bool initialized;
+	} legacy_hw_ctx;
+
 	struct list_head link;
 };
 
 struct i915_fbc {
 	unsigned long size;
+	unsigned threshold;
 	unsigned int fb_id;
 	enum plane plane;
 	int y;
 
-	struct drm_mm_node *compressed_fb;
+	struct drm_mm_node compressed_fb;
 	struct drm_mm_node *compressed_llb;
 
 	struct intel_fbc_work {
@@ -880,6 +902,12 @@ struct vlv_s0ix_state {
 	u32 clock_gate_dis2;
 };
 
+struct intel_rps_ei {
+	u32 cz_clock;
+	u32 render_c0;
+	u32 media_c0;
+};
+
 struct intel_gen6_power_mgmt {
 	/* work and pm_iir are protected by dev_priv->irq_lock */
 	struct work_struct work;
@@ -904,12 +932,17 @@ struct intel_gen6_power_mgmt {
 	u8 rp1_freq;		/* "less than" RP0 power/freqency */
 	u8 rp0_freq;		/* Non-overclocked max frequency. */
 
+	u32 ei_interrupt_count;
+
 	int last_adj;
 	enum { LOW_POWER, BETWEEN, HIGH_POWER } power;
 
 	bool enabled;
 	struct delayed_work delayed_resume_work;
 
+	/* manual wa residency calculations */
+	struct intel_rps_ei up_ei, down_ei;
+
 	/*
 	 * Protects RPS/RC6 register access and PCU communication.
 	 * Must be taken after struct_mutex if nested.
@@ -1374,6 +1407,7 @@ struct drm_i915_private {
 
 	struct pci_dev *bridge_dev;
 	struct intel_engine_cs ring[I915_NUM_RINGS];
+	struct drm_i915_gem_object *semaphore_obj;
 	uint32_t last_seqno, next_seqno;
 
 	drm_dma_handle_t *status_page_dmah;
@@ -1480,7 +1514,6 @@ struct drm_i915_private {
 
 	int num_shared_dpll;
 	struct intel_shared_dpll shared_dplls[I915_NUM_PLLS];
-	struct intel_ddi_plls ddi_plls;
 	int dpio_phy_iosf_port[I915_NUM_PHYS_VLV];
 
 	/* Reclocking support */
@@ -1557,6 +1590,11 @@ struct drm_i915_private {
 
 	struct i915_runtime_pm pm;
 
+	struct intel_digital_port *hpd_irq_port[I915_MAX_PORTS];
+	u32 long_hpd_port_mask;
+	u32 short_hpd_port_mask;
+	struct work_struct dig_port_work;
+
 	/* Old dri1 support infrastructure, beware the dragons ya fools entering
 	 * here! */
 	struct i915_dri1_state dri1;
@@ -2097,12 +2135,12 @@ void i915_update_dri1_breadcrumb(struct drm_device *dev);
 extern void i915_kernel_lost_context(struct drm_device * dev);
 extern int i915_driver_load(struct drm_device *, unsigned long flags);
 extern int i915_driver_unload(struct drm_device *);
-extern int i915_driver_open(struct drm_device *dev, struct drm_file *file_priv);
+extern int i915_driver_open(struct drm_device *dev, struct drm_file *file);
 extern void i915_driver_lastclose(struct drm_device * dev);
 extern void i915_driver_preclose(struct drm_device *dev,
-				 struct drm_file *file_priv);
+				 struct drm_file *file);
 extern void i915_driver_postclose(struct drm_device *dev,
-				  struct drm_file *file_priv);
+				  struct drm_file *file);
 extern int i915_driver_device_is_agp(struct drm_device * dev);
 #ifdef CONFIG_COMPAT
 extern long i915_compat_ioctl(struct file *filp, unsigned int cmd,
@@ -2457,7 +2495,7 @@ static inline void i915_gem_context_unreference(struct intel_context *ctx)
 
 static inline bool i915_gem_context_is_default(const struct intel_context *c)
 {
-	return c->id == DEFAULT_CONTEXT_ID;
+	return c->user_handle == DEFAULT_CONTEXT_HANDLE;
 }
 
 int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
@@ -2488,7 +2526,7 @@ static inline void i915_gem_chipset_flush(struct drm_device *dev)
 
 /* i915_gem_stolen.c */
 int i915_gem_init_stolen(struct drm_device *dev);
-int i915_gem_stolen_setup_compression(struct drm_device *dev, int size);
+int i915_gem_stolen_setup_compression(struct drm_device *dev, int size, int fb_cpp);
 void i915_gem_stolen_cleanup_compression(struct drm_device *dev);
 void i915_gem_cleanup_stolen(struct drm_device *dev);
 struct drm_i915_gem_object *
@@ -2647,6 +2685,8 @@ extern void gen6_set_rps(struct drm_device *dev, u8 val);
 extern void valleyview_set_rps(struct drm_device *dev, u8 val);
 extern int valleyview_rps_max_freq(struct drm_i915_private *dev_priv);
 extern int valleyview_rps_min_freq(struct drm_i915_private *dev_priv);
+extern void intel_set_memory_cxsr(struct drm_i915_private *dev_priv,
+				  bool enable);
 extern void intel_detect_pch(struct drm_device *dev);
 extern int intel_trans_dp_port_sel(struct drm_crtc *crtc);
 extern int intel_enable_rc6(const struct drm_device *dev);

drivers/gpu/drm/i915/i915_gem.c

@@ -1168,7 +1168,7 @@ static int __wait_seqno(struct intel_engine_cs *ring, u32 seqno,
 
 	timeout_expire = timeout ? jiffies + timespec_to_jiffies_timeout(timeout) : 0;
 
-	if (INTEL_INFO(dev)->gen >= 6 && can_wait_boost(file_priv)) {
+	if (INTEL_INFO(dev)->gen >= 6 && ring->id == RCS && can_wait_boost(file_priv)) {
 		gen6_rps_boost(dev_priv);
 		if (file_priv)
 			mod_delayed_work(dev_priv->wq,
@@ -2330,7 +2330,7 @@ int __i915_add_request(struct intel_engine_cs *ring,
 	u32 request_ring_position, request_start;
 	int ret;
 
-	request_start = intel_ring_get_tail(ring);
+	request_start = intel_ring_get_tail(ring->buffer);
 	/*
 	 * Emit any outstanding flushes - execbuf can fail to emit the flush
 	 * after having emitted the batchbuffer command. Hence we need to fix
@@ -2351,7 +2351,7 @@ int __i915_add_request(struct intel_engine_cs *ring,
 	 * GPU processing the request, we never over-estimate the
 	 * position of the head.
 	 */
-	request_ring_position = intel_ring_get_tail(ring);
+	request_ring_position = intel_ring_get_tail(ring->buffer);
 
 	ret = ring->add_request(ring);
 	if (ret)
@@ -2842,6 +2842,8 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	idx = intel_ring_sync_index(from, to);
 
 	seqno = obj->last_read_seqno;
+	/* Optimization: Avoid semaphore sync when we are sure we already
+	 * waited for an object with higher seqno */
 	if (seqno <= from->semaphore.sync_seqno[idx])
 		return 0;
 

drivers/gpu/drm/i915/i915_gem_context.c

@@ -182,14 +182,14 @@ void i915_gem_context_free(struct kref *ctx_ref)
 						   typeof(*ctx), ref);
 	struct i915_hw_ppgtt *ppgtt = NULL;
 
-	if (ctx->obj) {
+	if (ctx->legacy_hw_ctx.rcs_state) {
 		/* We refcount even the aliasing PPGTT to keep the code symmetric */
-		if (USES_PPGTT(ctx->obj->base.dev))
+		if (USES_PPGTT(ctx->legacy_hw_ctx.rcs_state->base.dev))
 			ppgtt = ctx_to_ppgtt(ctx);
 
 		/* XXX: Free up the object before tearing down the address space, in
 		 * case we're bound in the PPGTT */
-		drm_gem_object_unreference(&ctx->obj->base);
+		drm_gem_object_unreference(&ctx->legacy_hw_ctx.rcs_state->base);
 	}
 
 	if (ppgtt)
@@ -198,6 +198,36 @@ void i915_gem_context_free(struct kref *ctx_ref)
 	kfree(ctx);
 }
 
+static struct drm_i915_gem_object *
+i915_gem_alloc_context_obj(struct drm_device *dev, size_t size)
+{
+	struct drm_i915_gem_object *obj;
+	int ret;
+
+	obj = i915_gem_alloc_object(dev, size);
+	if (obj == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	/*
+	 * Try to make the context utilize L3 as well as LLC.
+	 *
+	 * On VLV we don't have L3 controls in the PTEs so we
+	 * shouldn't touch the cache level, especially as that
+	 * would make the object snooped which might have a
+	 * negative performance impact.
+	 */
+	if (INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev)) {
+		ret = i915_gem_object_set_cache_level(obj, I915_CACHE_L3_LLC);
+		/* Failure shouldn't ever happen this early */
+		if (WARN_ON(ret)) {
+			drm_gem_object_unreference(&obj->base);
+			return ERR_PTR(ret);
+		}
+	}
+
+	return obj;
+}
+
 static struct i915_hw_ppgtt *
 create_vm_for_ctx(struct drm_device *dev, struct intel_context *ctx)
 {
@@ -234,40 +264,26 @@ __create_hw_context(struct drm_device *dev,
 	list_add_tail(&ctx->link, &dev_priv->context_list);
 
 	if (dev_priv->hw_context_size) {
-		ctx->obj = i915_gem_alloc_object(dev, dev_priv->hw_context_size);
-		if (ctx->obj == NULL) {
-			ret = -ENOMEM;
+		struct drm_i915_gem_object *obj =
+				i915_gem_alloc_context_obj(dev, dev_priv->hw_context_size);
+		if (IS_ERR(obj)) {
+			ret = PTR_ERR(obj);
 			goto err_out;
 		}
-
-		/*
-		 * Try to make the context utilize L3 as well as LLC.
-		 *
-		 * On VLV we don't have L3 controls in the PTEs so we
-		 * shouldn't touch the cache level, especially as that
-		 * would make the object snooped which might have a
-		 * negative performance impact.
-		 */
-		if (INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev)) {
-			ret = i915_gem_object_set_cache_level(ctx->obj,
-							      I915_CACHE_L3_LLC);
-			/* Failure shouldn't ever happen this early */
-			if (WARN_ON(ret))
-				goto err_out;
-		}
+		ctx->legacy_hw_ctx.rcs_state = obj;
 	}
 
 	/* Default context will never have a file_priv */
 	if (file_priv != NULL) {
 		ret = idr_alloc(&file_priv->context_idr, ctx,
-				DEFAULT_CONTEXT_ID, 0, GFP_KERNEL);
+				DEFAULT_CONTEXT_HANDLE, 0, GFP_KERNEL);
 		if (ret < 0)
 			goto err_out;
 	} else
-		ret = DEFAULT_CONTEXT_ID;
+		ret = DEFAULT_CONTEXT_HANDLE;
 
 	ctx->file_priv = file_priv;
-	ctx->id = ret;
+	ctx->user_handle = ret;
 	/* NB: Mark all slices as needing a remap so that when the context first
 	 * loads it will restore whatever remap state already exists. If there
 	 * is no remap info, it will be a NOP. */
@@ -301,7 +317,7 @@ i915_gem_create_context(struct drm_device *dev,
 	if (IS_ERR(ctx))
 		return ctx;
 
-	if (is_global_default_ctx && ctx->obj) {
+	if (is_global_default_ctx && ctx->legacy_hw_ctx.rcs_state) {
 		/* We may need to do things with the shrinker which
 		 * require us to immediately switch back to the default
 		 * context. This can cause a problem as pinning the
@@ -309,7 +325,7 @@ i915_gem_create_context(struct drm_device *dev,
 		 * be available. To avoid this we always pin the default
 		 * context.
 		 */
-		ret = i915_gem_obj_ggtt_pin(ctx->obj,
+		ret = i915_gem_obj_ggtt_pin(ctx->legacy_hw_ctx.rcs_state,
 					    get_context_alignment(dev), 0);
 		if (ret) {
 			DRM_DEBUG_DRIVER("Couldn't pin %d\n", ret);
@@ -349,8 +365,8 @@ i915_gem_create_context(struct drm_device *dev,
 	return ctx;
 
 err_unpin:
-	if (is_global_default_ctx && ctx->obj)
-		i915_gem_object_ggtt_unpin(ctx->obj);
+	if (is_global_default_ctx && ctx->legacy_hw_ctx.rcs_state)
+		i915_gem_object_ggtt_unpin(ctx->legacy_hw_ctx.rcs_state);
 err_destroy:
 	i915_gem_context_unreference(ctx);
 	return ERR_PTR(ret);
@@ -366,23 +382,27 @@ void i915_gem_context_reset(struct drm_device *dev)
 	for (i = 0; i < I915_NUM_RINGS; i++) {
 		struct intel_engine_cs *ring = &dev_priv->ring[i];
 		struct intel_context *dctx = ring->default_context;
+		struct intel_context *lctx = ring->last_context;
 
 		/* Do a fake switch to the default context */
-		if (ring->last_context == dctx)
+		if (lctx == dctx)
 			continue;
 
-		if (!ring->last_context)
+		if (!lctx)
 			continue;
 
-		if (dctx->obj && i == RCS) {
-			WARN_ON(i915_gem_obj_ggtt_pin(dctx->obj,
+		if (dctx->legacy_hw_ctx.rcs_state && i == RCS) {
+			WARN_ON(i915_gem_obj_ggtt_pin(dctx->legacy_hw_ctx.rcs_state,
 						      get_context_alignment(dev), 0));
 			/* Fake a finish/inactive */
-			dctx->obj->base.write_domain = 0;
-			dctx->obj->active = 0;
+			dctx->legacy_hw_ctx.rcs_state->base.write_domain = 0;
+			dctx->legacy_hw_ctx.rcs_state->active = 0;
 		}
 
-		i915_gem_context_unreference(ring->last_context);
+		if (lctx->legacy_hw_ctx.rcs_state && i == RCS)
+			i915_gem_object_ggtt_unpin(lctx->legacy_hw_ctx.rcs_state);
+
+		i915_gem_context_unreference(lctx);
 		i915_gem_context_reference(dctx);
 		ring->last_context = dctx;
 	}
@@ -429,7 +449,7 @@ void i915_gem_context_fini(struct drm_device *dev)
 	struct intel_context *dctx = dev_priv->ring[RCS].default_context;
 	int i;
 
-	if (dctx->obj) {
+	if (dctx->legacy_hw_ctx.rcs_state) {
 		/* The only known way to stop the gpu from accessing the hw context is
 		 * to reset it. Do this as the very last operation to avoid confusing
 		 * other code, leading to spurious errors. */
@@ -444,13 +464,13 @@ void i915_gem_context_fini(struct drm_device *dev)
 		WARN_ON(!dev_priv->ring[RCS].last_context);
 		if (dev_priv->ring[RCS].last_context == dctx) {
 			/* Fake switch to NULL context */
-			WARN_ON(dctx->obj->active);
-			i915_gem_object_ggtt_unpin(dctx->obj);
+			WARN_ON(dctx->legacy_hw_ctx.rcs_state->active);
+			i915_gem_object_ggtt_unpin(dctx->legacy_hw_ctx.rcs_state);
 			i915_gem_context_unreference(dctx);
 			dev_priv->ring[RCS].last_context = NULL;
 		}
 
-		i915_gem_object_ggtt_unpin(dctx->obj);
+		i915_gem_object_ggtt_unpin(dctx->legacy_hw_ctx.rcs_state);
 	}
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
@@ -570,7 +590,7 @@ mi_set_context(struct intel_engine_cs *ring,
 
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_emit(ring, MI_SET_CONTEXT);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(new_context->obj) |
+	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(new_context->legacy_hw_ctx.rcs_state) |
 			MI_MM_SPACE_GTT |
 			MI_SAVE_EXT_STATE_EN |
 			MI_RESTORE_EXT_STATE_EN |
@@ -602,8 +622,8 @@ static int do_switch(struct intel_engine_cs *ring,
 	int ret, i;
 
 	if (from != NULL && ring == &dev_priv->ring[RCS]) {
-		BUG_ON(from->obj == NULL);
-		BUG_ON(!i915_gem_obj_is_pinned(from->obj));
+		BUG_ON(from->legacy_hw_ctx.rcs_state == NULL);
+		BUG_ON(!i915_gem_obj_is_pinned(from->legacy_hw_ctx.rcs_state));
 	}
 
 	if (from == to && !to->remap_slice)
@@ -611,7 +631,7 @@ static int do_switch(struct intel_engine_cs *ring,
 
 	/* Trying to pin first makes error handling easier. */
 	if (ring == &dev_priv->ring[RCS]) {
-		ret = i915_gem_obj_ggtt_pin(to->obj,
+		ret = i915_gem_obj_ggtt_pin(to->legacy_hw_ctx.rcs_state,
 					    get_context_alignment(ring->dev), 0);
 		if (ret)
 			return ret;
@@ -644,17 +664,17 @@ static int do_switch(struct intel_engine_cs *ring,
 	 *
 	 * XXX: We need a real interface to do this instead of trickery.
 	 */
-	ret = i915_gem_object_set_to_gtt_domain(to->obj, false);
+	ret = i915_gem_object_set_to_gtt_domain(to->legacy_hw_ctx.rcs_state, false);
 	if (ret)
 		goto unpin_out;
 
-	if (!to->obj->has_global_gtt_mapping) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(to->obj,
+	if (!to->legacy_hw_ctx.rcs_state->has_global_gtt_mapping) {
+		struct i915_vma *vma = i915_gem_obj_to_vma(to->legacy_hw_ctx.rcs_state,
 							   &dev_priv->gtt.base);
-		vma->bind_vma(vma, to->obj->cache_level, GLOBAL_BIND);
+		vma->bind_vma(vma, to->legacy_hw_ctx.rcs_state->cache_level, GLOBAL_BIND);
 	}
 
-	if (!to->is_initialized || i915_gem_context_is_default(to))
+	if (!to->legacy_hw_ctx.initialized || i915_gem_context_is_default(to))
 		hw_flags |= MI_RESTORE_INHIBIT;
 
 	ret = mi_set_context(ring, to, hw_flags);
@@ -680,8 +700,8 @@ static int do_switch(struct intel_engine_cs *ring,
 	 * MI_SET_CONTEXT instead of when the next seqno has completed.
 	 */
 	if (from != NULL) {
-		from->obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
-		i915_vma_move_to_active(i915_gem_obj_to_ggtt(from->obj), ring);
+		from->legacy_hw_ctx.rcs_state->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
+		i915_vma_move_to_active(i915_gem_obj_to_ggtt(from->legacy_hw_ctx.rcs_state), ring);
 		/* As long as MI_SET_CONTEXT is serializing, ie. it flushes the
 		 * whole damn pipeline, we don't need to explicitly mark the
 		 * object dirty. The only exception is that the context must be
@@ -689,16 +709,16 @@ static int do_switch(struct intel_engine_cs *ring,
 		 * able to defer doing this until we know the object would be
 		 * swapped, but there is no way to do that yet.
 		 */
-		from->obj->dirty = 1;
-		BUG_ON(from->obj->ring != ring);
+		from->legacy_hw_ctx.rcs_state->dirty = 1;
+		BUG_ON(from->legacy_hw_ctx.rcs_state->ring != ring);
 
 		/* obj is kept alive until the next request by its active ref */
-		i915_gem_object_ggtt_unpin(from->obj);
+		i915_gem_object_ggtt_unpin(from->legacy_hw_ctx.rcs_state);
 		i915_gem_context_unreference(from);
 	}
 
-	uninitialized = !to->is_initialized && from == NULL;
-	to->is_initialized = true;
+	uninitialized = !to->legacy_hw_ctx.initialized && from == NULL;
+	to->legacy_hw_ctx.initialized = true;
 
 done:
 	i915_gem_context_reference(to);
@@ -714,7 +734,7 @@ done:
 
 unpin_out:
 	if (ring->id == RCS)
-		i915_gem_object_ggtt_unpin(to->obj);
+		i915_gem_object_ggtt_unpin(to->legacy_hw_ctx.rcs_state);
 	return ret;
 }
 
@@ -735,7 +755,7 @@ int i915_switch_context(struct intel_engine_cs *ring,
 
 	WARN_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
 
-	if (to->obj == NULL) { /* We have the fake context */
+	if (to->legacy_hw_ctx.rcs_state == NULL) { /* We have the fake context */
 		if (to != ring->last_context) {
 			i915_gem_context_reference(to);
 			if (ring->last_context)
@@ -773,7 +793,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);
 
-	args->ctx_id = ctx->id;
+	args->ctx_id = ctx->user_handle;
 	DRM_DEBUG_DRIVER("HW context %d created\n", args->ctx_id);
 
 	return 0;
@@ -787,7 +807,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 	struct intel_context *ctx;
 	int ret;
 
-	if (args->ctx_id == DEFAULT_CONTEXT_ID)
+	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE)
 		return -ENOENT;
 
 	ret = i915_mutex_lock_interruptible(dev);
@@ -800,7 +820,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 		return PTR_ERR(ctx);
 	}
 
-	idr_remove(&ctx->file_priv->context_idr, ctx->id);
+	idr_remove(&ctx->file_priv->context_idr, ctx->user_handle);
 	i915_gem_context_unreference(ctx);
 	mutex_unlock(&dev->struct_mutex);
 

drivers/gpu/drm/i915/i915_gem_execbuffer.c

@@ -938,7 +938,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 	struct intel_context *ctx = NULL;
 	struct i915_ctx_hang_stats *hs;
 
-	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_ID)
+	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_HANDLE)
 		return ERR_PTR(-EINVAL);
 
 	ctx = i915_gem_context_get(file->driver_priv, ctx_id);
@@ -1026,6 +1026,163 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 	return 0;
 }
 
+static int
+legacy_ringbuffer_submission(struct drm_device *dev, struct drm_file *file,
+			     struct intel_engine_cs *ring,
+			     struct intel_context *ctx,
+			     struct drm_i915_gem_execbuffer2 *args,
+			     struct list_head *vmas,
+			     struct drm_i915_gem_object *batch_obj,
+			     u64 exec_start, u32 flags)
+{
+	struct drm_clip_rect *cliprects = NULL;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u64 exec_len;
+	int instp_mode;
+	u32 instp_mask;
+	int i, ret = 0;
+
+	if (args->num_cliprects != 0) {
+		if (ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("clip rectangles are only valid with the render ring\n");
+			return -EINVAL;
+		}
+
+		if (INTEL_INFO(dev)->gen >= 5) {
+			DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
+			return -EINVAL;
+		}
+
+		if (args->num_cliprects > UINT_MAX / sizeof(*cliprects)) {
+			DRM_DEBUG("execbuf with %u cliprects\n",
+				  args->num_cliprects);
+			return -EINVAL;
+		}
+
+		cliprects = kcalloc(args->num_cliprects,
+				    sizeof(*cliprects),
+				    GFP_KERNEL);
+		if (cliprects == NULL) {
+			ret = -ENOMEM;
+			goto error;
+		}
+
+		if (copy_from_user(cliprects,
+				   to_user_ptr(args->cliprects_ptr),
+				   sizeof(*cliprects)*args->num_cliprects)) {
+			ret = -EFAULT;
+			goto error;
+		}
+	} else {
+		if (args->DR4 == 0xffffffff) {
+			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
+			args->DR4 = 0;
+		}
+
+		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
+			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
+			return -EINVAL;
+		}
+	}
+
+	ret = i915_gem_execbuffer_move_to_gpu(ring, vmas);
+	if (ret)
+		goto error;
+
+	ret = i915_switch_context(ring, ctx);
+	if (ret)
+		goto error;
+
+	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
+	instp_mask = I915_EXEC_CONSTANTS_MASK;
+	switch (instp_mode) {
+	case I915_EXEC_CONSTANTS_REL_GENERAL:
+	case I915_EXEC_CONSTANTS_ABSOLUTE:
+	case I915_EXEC_CONSTANTS_REL_SURFACE:
+		if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
+			ret = -EINVAL;
+			goto error;
+		}
+
+		if (instp_mode != dev_priv->relative_constants_mode) {
+			if (INTEL_INFO(dev)->gen < 4) {
+				DRM_DEBUG("no rel constants on pre-gen4\n");
+				ret = -EINVAL;
+				goto error;
+			}
+
+			if (INTEL_INFO(dev)->gen > 5 &&
+			    instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
+				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
+				ret = -EINVAL;
+				goto error;
+			}
+
+			/* The HW changed the meaning on this bit on gen6 */
+			if (INTEL_INFO(dev)->gen >= 6)
+				instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
+		}
+		break;
+	default:
+		DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	if (ring == &dev_priv->ring[RCS] &&
+			instp_mode != dev_priv->relative_constants_mode) {
+		ret = intel_ring_begin(ring, 4);
+		if (ret)
+			goto error;
+
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit(ring, INSTPM);
+		intel_ring_emit(ring, instp_mask << 16 | instp_mode);
+		intel_ring_advance(ring);
+
+		dev_priv->relative_constants_mode = instp_mode;
+	}
+
+	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
+		ret = i915_reset_gen7_sol_offsets(dev, ring);
+		if (ret)
+			goto error;
+	}
+
+	exec_len = args->batch_len;
+	if (cliprects) {
+		for (i = 0; i < args->num_cliprects; i++) {
+			ret = i915_emit_box(dev, &cliprects[i],
+					    args->DR1, args->DR4);
+			if (ret)
+				goto error;
+
+			ret = ring->dispatch_execbuffer(ring,
+							exec_start, exec_len,
+							flags);
+			if (ret)
+				goto error;
+		}
+	} else {
+		ret = ring->dispatch_execbuffer(ring,
+						exec_start, exec_len,
+						flags);
+		if (ret)
+			return ret;
+	}
+
+	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
+
+	i915_gem_execbuffer_move_to_active(vmas, ring);
+	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
+
+error:
+	kfree(cliprects);
+	return ret;
+}
+
 /**
  * Find one BSD ring to dispatch the corresponding BSD command.
  * The Ring ID is returned.
@@ -1085,14 +1242,13 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
-	struct drm_clip_rect *cliprects = NULL;
 	struct intel_engine_cs *ring;
 	struct intel_context *ctx;
 	struct i915_address_space *vm;
 	const u32 ctx_id = i915_execbuffer2_get_context_id(*args);
-	u64 exec_start = args->batch_start_offset, exec_len;
-	u32 mask, flags;
-	int ret, mode, i;
+	u64 exec_start = args->batch_start_offset;
+	u32 flags;
+	int ret;
 	bool need_relocs;
 
 	if (!i915_gem_check_execbuffer(args))
@@ -1136,87 +1292,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	mode = args->flags & I915_EXEC_CONSTANTS_MASK;
-	mask = I915_EXEC_CONSTANTS_MASK;
-	switch (mode) {
-	case I915_EXEC_CONSTANTS_REL_GENERAL:
-	case I915_EXEC_CONSTANTS_ABSOLUTE:
-	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (mode != 0 && ring != &dev_priv->ring[RCS]) {
-			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
-			return -EINVAL;
-		}
-
-		if (mode != dev_priv->relative_constants_mode) {
-			if (INTEL_INFO(dev)->gen < 4) {
-				DRM_DEBUG("no rel constants on pre-gen4\n");
-				return -EINVAL;
-			}
-
-			if (INTEL_INFO(dev)->gen > 5 &&
-			    mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
-				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
-				return -EINVAL;
-			}
-
-			/* The HW changed the meaning on this bit on gen6 */
-			if (INTEL_INFO(dev)->gen >= 6)
-				mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
-		}
-		break;
-	default:
-		DRM_DEBUG("execbuf with unknown constants: %d\n", mode);
-		return -EINVAL;
-	}
-
 	if (args->buffer_count < 1) {
 		DRM_DEBUG("execbuf with %d buffers\n", args->buffer_count);
 		return -EINVAL;
 	}
 
-	if (args->num_cliprects != 0) {
-		if (ring != &dev_priv->ring[RCS]) {
-			DRM_DEBUG("clip rectangles are only valid with the render ring\n");
-			return -EINVAL;
-		}
-
-		if (INTEL_INFO(dev)->gen >= 5) {
-			DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
-			return -EINVAL;
-		}
-
-		if (args->num_cliprects > UINT_MAX / sizeof(*cliprects)) {
-			DRM_DEBUG("execbuf with %u cliprects\n",
-				  args->num_cliprects);
-			return -EINVAL;
-		}
-
-		cliprects = kcalloc(args->num_cliprects,
-				    sizeof(*cliprects),
-				    GFP_KERNEL);
-		if (cliprects == NULL) {
-			ret = -ENOMEM;
-			goto pre_mutex_err;
-		}
-
-		if (copy_from_user(cliprects,
-				   to_user_ptr(args->cliprects_ptr),
-				   sizeof(*cliprects)*args->num_cliprects)) {
-			ret = -EFAULT;
-			goto pre_mutex_err;
-		}
-	} else {
-		if (args->DR4 == 0xffffffff) {
-			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
-			args->DR4 = 0;
-		}
-
-		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
-			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
-			return -EINVAL;
-		}
-	}
-
 	intel_runtime_pm_get(dev_priv);
 
 	ret = i915_mutex_lock_interruptible(dev);
@@ -1320,63 +1400,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	else
 		exec_start += i915_gem_obj_offset(batch_obj, vm);
 
-	ret = i915_gem_execbuffer_move_to_gpu(ring, &eb->vmas);
+	ret = legacy_ringbuffer_submission(dev, file, ring, ctx,
+			args, &eb->vmas, batch_obj, exec_start, flags);
 	if (ret)
 		goto err;
 
-	ret = i915_switch_context(ring, ctx);
-	if (ret)
-		goto err;
-
-	if (ring == &dev_priv->ring[RCS] &&
-	    mode != dev_priv->relative_constants_mode) {
-		ret = intel_ring_begin(ring, 4);
-		if (ret)
-				goto err;
-
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit(ring, INSTPM);
-		intel_ring_emit(ring, mask << 16 | mode);
-		intel_ring_advance(ring);
-
-		dev_priv->relative_constants_mode = mode;
-	}
-
-	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
-		ret = i915_reset_gen7_sol_offsets(dev, ring);
-		if (ret)
-			goto err;
-	}
-
-
-	exec_len = args->batch_len;
-	if (cliprects) {
-		for (i = 0; i < args->num_cliprects; i++) {
-			ret = i915_emit_box(dev, &cliprects[i],
-					    args->DR1, args->DR4);
-			if (ret)
-				goto err;
-
-			ret = ring->dispatch_execbuffer(ring,
-							exec_start, exec_len,
-							flags);
-			if (ret)
-				goto err;
-		}
-	} else {
-		ret = ring->dispatch_execbuffer(ring,
-						exec_start, exec_len,
-						flags);
-		if (ret)
-			goto err;
-	}
-
-	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
-
-	i915_gem_execbuffer_move_to_active(&eb->vmas, ring);
-	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
-
 err:
 	/* the request owns the ref now */
 	i915_gem_context_unreference(ctx);
@@ -1385,8 +1413,6 @@ err:
 	mutex_unlock(&dev->struct_mutex);
 
 pre_mutex_err:
-	kfree(cliprects);
-
 	/* intel_gpu_busy should also get a ref, so it will free when the device
 	 * is really idle. */
 	intel_runtime_pm_put(dev_priv);

drivers/gpu/drm/i915/i915_gem_stolen.c

@@ -103,30 +103,68 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 	return base;
 }
 
-static int i915_setup_compression(struct drm_device *dev, int size)
+static int find_compression_threshold(struct drm_device *dev,
+				      struct drm_mm_node *node,
+				      int size,
+				      int fb_cpp)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_mm_node *compressed_fb, *uninitialized_var(compressed_llb);
+	int compression_threshold = 1;
 	int ret;
 
-	compressed_fb = kzalloc(sizeof(*compressed_fb), GFP_KERNEL);
-	if (!compressed_fb)
-		goto err_llb;
+	/* HACK: This code depends on what we will do in *_enable_fbc. If that
+	 * code changes, this code needs to change as well.
+	 *
+	 * The enable_fbc code will attempt to use one of our 2 compression
+	 * thresholds, therefore, in that case, we only have 1 resort.
+	 */
 
-	/* Try to over-allocate to reduce reallocations and fragmentation */
-	ret = drm_mm_insert_node(&dev_priv->mm.stolen, compressed_fb,
+	/* Try to over-allocate to reduce reallocations and fragmentation. */
+	ret = drm_mm_insert_node(&dev_priv->mm.stolen, node,
 				 size <<= 1, 4096, DRM_MM_SEARCH_DEFAULT);
-	if (ret)
-		ret = drm_mm_insert_node(&dev_priv->mm.stolen, compressed_fb,
-					 size >>= 1, 4096,
-					 DRM_MM_SEARCH_DEFAULT);
-	if (ret)
+	if (ret == 0)
+		return compression_threshold;
+
+again:
+	/* HW's ability to limit the CFB is 1:4 */
+	if (compression_threshold > 4 ||
+	    (fb_cpp == 2 && compression_threshold == 2))
+		return 0;
+
+	ret = drm_mm_insert_node(&dev_priv->mm.stolen, node,
+				 size >>= 1, 4096,
+				 DRM_MM_SEARCH_DEFAULT);
+	if (ret && INTEL_INFO(dev)->gen <= 4) {
+		return 0;
+	} else if (ret) {
+		compression_threshold <<= 1;
+		goto again;
+	} else {
+		return compression_threshold;
+	}
+}
+
+static int i915_setup_compression(struct drm_device *dev, int size, int fb_cpp)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_mm_node *uninitialized_var(compressed_llb);
+	int ret;
+
+	ret = find_compression_threshold(dev, &dev_priv->fbc.compressed_fb,
+					 size, fb_cpp);
+	if (!ret)
 		goto err_llb;
+	else if (ret > 1) {
+		DRM_INFO("Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.\n");
+
+	}
+
+	dev_priv->fbc.threshold = ret;
 
 	if (HAS_PCH_SPLIT(dev))
-		I915_WRITE(ILK_DPFC_CB_BASE, compressed_fb->start);
+		I915_WRITE(ILK_DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
 	else if (IS_GM45(dev)) {
-		I915_WRITE(DPFC_CB_BASE, compressed_fb->start);
+		I915_WRITE(DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
 	} else {
 		compressed_llb = kzalloc(sizeof(*compressed_llb), GFP_KERNEL);
 		if (!compressed_llb)
@@ -140,13 +178,12 @@ static int i915_setup_compression(struct drm_device *dev, int size)
 		dev_priv->fbc.compressed_llb = compressed_llb;
 
 		I915_WRITE(FBC_CFB_BASE,
-			   dev_priv->mm.stolen_base + compressed_fb->start);
+			   dev_priv->mm.stolen_base + dev_priv->fbc.compressed_fb.start);
 		I915_WRITE(FBC_LL_BASE,
 			   dev_priv->mm.stolen_base + compressed_llb->start);
 	}
 
-	dev_priv->fbc.compressed_fb = compressed_fb;
-	dev_priv->fbc.size = size;
+	dev_priv->fbc.size = size / dev_priv->fbc.threshold;
 
 	DRM_DEBUG_KMS("reserved %d bytes of contiguous stolen space for FBC\n",
 		      size);
@@ -155,14 +192,13 @@ static int i915_setup_compression(struct drm_device *dev, int size)
 
 err_fb:
 	kfree(compressed_llb);
-	drm_mm_remove_node(compressed_fb);
+	drm_mm_remove_node(&dev_priv->fbc.compressed_fb);
 err_llb:
-	kfree(compressed_fb);
 	pr_info_once("drm: not enough stolen space for compressed buffer (need %d more bytes), disabling. Hint: you may be able to increase stolen memory size in the BIOS to avoid this.\n", size);
 	return -ENOSPC;
 }
 
-int i915_gem_stolen_setup_compression(struct drm_device *dev, int size)
+int i915_gem_stolen_setup_compression(struct drm_device *dev, int size, int fb_cpp)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -175,7 +211,7 @@ int i915_gem_stolen_setup_compression(struct drm_device *dev, int size)
 	/* Release any current block */
 	i915_gem_stolen_cleanup_compression(dev);
 
-	return i915_setup_compression(dev, size);
+	return i915_setup_compression(dev, size, fb_cpp);
 }
 
 void i915_gem_stolen_cleanup_compression(struct drm_device *dev)
@@ -185,10 +221,7 @@ void i915_gem_stolen_cleanup_compression(struct drm_device *dev)
 	if (dev_priv->fbc.size == 0)
 		return;
 
-	if (dev_priv->fbc.compressed_fb) {
-		drm_mm_remove_node(dev_priv->fbc.compressed_fb);
-		kfree(dev_priv->fbc.compressed_fb);
-	}
+	drm_mm_remove_node(&dev_priv->fbc.compressed_fb);
 
 	if (dev_priv->fbc.compressed_llb) {
 		drm_mm_remove_node(dev_priv->fbc.compressed_llb);

drivers/gpu/drm/i915/i915_gpu_error.c

@@ -327,6 +327,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	struct drm_device *dev = error_priv->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_error_state *error = error_priv->error;
+	struct drm_i915_error_object *obj;
 	int i, j, offset, elt;
 	int max_hangcheck_score;
 
@@ -395,8 +396,6 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 				    error->pinned_bo_count[0]);
 
 	for (i = 0; i < ARRAY_SIZE(error->ring); i++) {
-		struct drm_i915_error_object *obj;
-
 		obj = error->ring[i].batchbuffer;
 		if (obj) {
 			err_puts(m, dev_priv->ring[i].name);
@@ -459,6 +458,18 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		}
 	}
 
+	if ((obj = error->semaphore_obj)) {
+		err_printf(m, "Semaphore page = 0x%08x\n", obj->gtt_offset);
+		for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
+			err_printf(m, "[%04x] %08x %08x %08x %08x\n",
+				   elt * 4,
+				   obj->pages[0][elt],
+				   obj->pages[0][elt+1],
+				   obj->pages[0][elt+2],
+				   obj->pages[0][elt+3]);
+		}
+	}
+
 	if (error->overlay)
 		intel_overlay_print_error_state(m, error->overlay);
 
@@ -529,6 +540,7 @@ static void i915_error_state_free(struct kref *error_ref)
 		kfree(error->ring[i].requests);
 	}
 
+	i915_error_object_free(error->semaphore_obj);
 	kfree(error->active_bo);
 	kfree(error->overlay);
 	kfree(error->display);
@@ -746,7 +758,52 @@ static void i915_gem_record_fences(struct drm_device *dev,
 	}
 }
 
+
+static void gen8_record_semaphore_state(struct drm_i915_private *dev_priv,
+					struct drm_i915_error_state *error,
+					struct intel_engine_cs *ring,
+					struct drm_i915_error_ring *ering)
+{
+	struct intel_engine_cs *useless;
+	int i;
+
+	if (!i915_semaphore_is_enabled(dev_priv->dev))
+		return;
+
+	if (!error->semaphore_obj)
+		error->semaphore_obj =
+			i915_error_object_create(dev_priv,
+						 dev_priv->semaphore_obj,
+						 &dev_priv->gtt.base);
+
+	for_each_ring(useless, dev_priv, i) {
+		u16 signal_offset =
+			(GEN8_SIGNAL_OFFSET(ring, i) & PAGE_MASK) / 4;
+		u32 *tmp = error->semaphore_obj->pages[0];
+
+		ering->semaphore_mboxes[i] = tmp[signal_offset];
+		ering->semaphore_seqno[i] = ring->semaphore.sync_seqno[i];
+	}
+}
+
+static void gen6_record_semaphore_state(struct drm_i915_private *dev_priv,
+					struct intel_engine_cs *ring,
+					struct drm_i915_error_ring *ering)
+{
+	ering->semaphore_mboxes[0] = I915_READ(RING_SYNC_0(ring->mmio_base));
+	ering->semaphore_mboxes[1] = I915_READ(RING_SYNC_1(ring->mmio_base));
+	ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0];
+	ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1];
+
+	if (HAS_VEBOX(dev_priv->dev)) {
+		ering->semaphore_mboxes[2] =
+			I915_READ(RING_SYNC_2(ring->mmio_base));
+		ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2];
+	}
+}
+
 static void i915_record_ring_state(struct drm_device *dev,
+				   struct drm_i915_error_state *error,
 				   struct intel_engine_cs *ring,
 				   struct drm_i915_error_ring *ering)
 {
@@ -755,18 +812,10 @@ static void i915_record_ring_state(struct drm_device *dev,
 	if (INTEL_INFO(dev)->gen >= 6) {
 		ering->rc_psmi = I915_READ(ring->mmio_base + 0x50);
 		ering->fault_reg = I915_READ(RING_FAULT_REG(ring));
-		ering->semaphore_mboxes[0]
-			= I915_READ(RING_SYNC_0(ring->mmio_base));
-		ering->semaphore_mboxes[1]
-			= I915_READ(RING_SYNC_1(ring->mmio_base));
-		ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0];
-		ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1];
-	}
-
-	if (HAS_VEBOX(dev)) {
-		ering->semaphore_mboxes[2] =
-			I915_READ(RING_SYNC_2(ring->mmio_base));
-		ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2];
+		if (INTEL_INFO(dev)->gen >= 8)
+			gen8_record_semaphore_state(dev_priv, error, ring, ering);
+		else
+			gen6_record_semaphore_state(dev_priv, ring, ering);
 	}
 
 	if (INTEL_INFO(dev)->gen >= 4) {
@@ -895,7 +944,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 
 		error->ring[i].valid = true;
 
-		i915_record_ring_state(dev, ring, &error->ring[i]);
+		i915_record_ring_state(dev, error, ring, &error->ring[i]);
 
 		request = i915_gem_find_active_request(ring);
 		if (request) {

drivers/gpu/drm/i915/i915_irq.c

@@ -1090,6 +1090,53 @@ static bool intel_hpd_irq_event(struct drm_device *dev,
 	return true;
 }
 
+static void i915_digport_work_func(struct work_struct *work)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(work, struct drm_i915_private, dig_port_work);
+	unsigned long irqflags;
+	u32 long_port_mask, short_port_mask;
+	struct intel_digital_port *intel_dig_port;
+	int i, ret;
+	u32 old_bits = 0;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	long_port_mask = dev_priv->long_hpd_port_mask;
+	dev_priv->long_hpd_port_mask = 0;
+	short_port_mask = dev_priv->short_hpd_port_mask;
+	dev_priv->short_hpd_port_mask = 0;
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	for (i = 0; i < I915_MAX_PORTS; i++) {
+		bool valid = false;
+		bool long_hpd = false;
+		intel_dig_port = dev_priv->hpd_irq_port[i];
+		if (!intel_dig_port || !intel_dig_port->hpd_pulse)
+			continue;
+
+		if (long_port_mask & (1 << i))  {
+			valid = true;
+			long_hpd = true;
+		} else if (short_port_mask & (1 << i))
+			valid = true;
+
+		if (valid) {
+			ret = intel_dig_port->hpd_pulse(intel_dig_port, long_hpd);
+			if (ret == true) {
+				/* if we get true fallback to old school hpd */
+				old_bits |= (1 << intel_dig_port->base.hpd_pin);
+			}
+		}
+	}
+
+	if (old_bits) {
+		spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+		dev_priv->hpd_event_bits |= old_bits;
+		spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+		schedule_work(&dev_priv->hotplug_work);
+	}
+}
+
 /*
  * Handle hotplug events outside the interrupt handler proper.
  */
@@ -1221,6 +1268,131 @@ static void notify_ring(struct drm_device *dev,
 	i915_queue_hangcheck(dev);
 }
 
+static u32 vlv_c0_residency(struct drm_i915_private *dev_priv,
+			    struct intel_rps_ei *rps_ei)
+{
+	u32 cz_ts, cz_freq_khz;
+	u32 render_count, media_count;
+	u32 elapsed_render, elapsed_media, elapsed_time;
+	u32 residency = 0;
+
+	cz_ts = vlv_punit_read(dev_priv, PUNIT_REG_CZ_TIMESTAMP);
+	cz_freq_khz = DIV_ROUND_CLOSEST(dev_priv->mem_freq * 1000, 4);
+
+	render_count = I915_READ(VLV_RENDER_C0_COUNT_REG);
+	media_count = I915_READ(VLV_MEDIA_C0_COUNT_REG);
+
+	if (rps_ei->cz_clock == 0) {
+		rps_ei->cz_clock = cz_ts;
+		rps_ei->render_c0 = render_count;
+		rps_ei->media_c0 = media_count;
+
+		return dev_priv->rps.cur_freq;
+	}
+
+	elapsed_time = cz_ts - rps_ei->cz_clock;
+	rps_ei->cz_clock = cz_ts;
+
+	elapsed_render = render_count - rps_ei->render_c0;
+	rps_ei->render_c0 = render_count;
+
+	elapsed_media = media_count - rps_ei->media_c0;
+	rps_ei->media_c0 = media_count;
+
+	/* Convert all the counters into common unit of milli sec */
+	elapsed_time /= VLV_CZ_CLOCK_TO_MILLI_SEC;
+	elapsed_render /=  cz_freq_khz;
+	elapsed_media /= cz_freq_khz;
+
+	/*
+	 * Calculate overall C0 residency percentage
+	 * only if elapsed time is non zero
+	 */
+	if (elapsed_time) {
+		residency =
+			((max(elapsed_render, elapsed_media) * 100)
+				/ elapsed_time);
+	}
+
+	return residency;
+}
+
+/**
+ * vlv_calc_delay_from_C0_counters - Increase/Decrease freq based on GPU
+ * busy-ness calculated from C0 counters of render & media power wells
+ * @dev_priv: DRM device private
+ *
+ */
+static u32 vlv_calc_delay_from_C0_counters(struct drm_i915_private *dev_priv)
+{
+	u32 residency_C0_up = 0, residency_C0_down = 0;
+	u8 new_delay, adj;
+
+	dev_priv->rps.ei_interrupt_count++;
+
+	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
+
+
+	if (dev_priv->rps.up_ei.cz_clock == 0) {
+		vlv_c0_residency(dev_priv, &dev_priv->rps.up_ei);
+		vlv_c0_residency(dev_priv, &dev_priv->rps.down_ei);
+		return dev_priv->rps.cur_freq;
+	}
+
+
+	/*
+	 * To down throttle, C0 residency should be less than down threshold
+	 * for continous EI intervals. So calculate down EI counters
+	 * once in VLV_INT_COUNT_FOR_DOWN_EI
+	 */
+	if (dev_priv->rps.ei_interrupt_count == VLV_INT_COUNT_FOR_DOWN_EI) {
+
+		dev_priv->rps.ei_interrupt_count = 0;
+
+		residency_C0_down = vlv_c0_residency(dev_priv,
+						     &dev_priv->rps.down_ei);
+	} else {
+		residency_C0_up = vlv_c0_residency(dev_priv,
+						   &dev_priv->rps.up_ei);
+	}
+
+	new_delay = dev_priv->rps.cur_freq;
+
+	adj = dev_priv->rps.last_adj;
+	/* C0 residency is greater than UP threshold. Increase Frequency */
+	if (residency_C0_up >= VLV_RP_UP_EI_THRESHOLD) {
+		if (adj > 0)
+			adj *= 2;
+		else
+			adj = 1;
+
+		if (dev_priv->rps.cur_freq < dev_priv->rps.max_freq_softlimit)
+			new_delay = dev_priv->rps.cur_freq + adj;
+
+		/*
+		 * For better performance, jump directly
+		 * to RPe if we're below it.
+		 */
+		if (new_delay < dev_priv->rps.efficient_freq)
+			new_delay = dev_priv->rps.efficient_freq;
+
+	} else if (!dev_priv->rps.ei_interrupt_count &&
+			(residency_C0_down < VLV_RP_DOWN_EI_THRESHOLD)) {
+		if (adj < 0)
+			adj *= 2;
+		else
+			adj = -1;
+		/*
+		 * This means, C0 residency is less than down threshold over
+		 * a period of VLV_INT_COUNT_FOR_DOWN_EI. So, reduce the freq
+		 */
+		if (dev_priv->rps.cur_freq > dev_priv->rps.min_freq_softlimit)
+			new_delay = dev_priv->rps.cur_freq + adj;
+	}
+
+	return new_delay;
+}
+
 static void gen6_pm_rps_work(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
@@ -1269,6 +1441,8 @@ static void gen6_pm_rps_work(struct work_struct *work)
 		else
 			new_delay = dev_priv->rps.min_freq_softlimit;
 		adj = 0;
+	} else if (pm_iir & GEN6_PM_RP_UP_EI_EXPIRED) {
+		new_delay = vlv_calc_delay_from_C0_counters(dev_priv);
 	} else if (pm_iir & GEN6_PM_RP_DOWN_THRESHOLD) {
 		if (adj < 0)
 			adj *= 2;
@@ -1517,23 +1691,104 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 #define HPD_STORM_DETECT_PERIOD 1000
 #define HPD_STORM_THRESHOLD 5
 
+static int ilk_port_to_hotplug_shift(enum port port)
+{
+	switch (port) {
+	case PORT_A:
+	case PORT_E:
+	default:
+		return -1;
+	case PORT_B:
+		return 0;
+	case PORT_C:
+		return 8;
+	case PORT_D:
+		return 16;
+	}
+}
+
+static int g4x_port_to_hotplug_shift(enum port port)
+{
+	switch (port) {
+	case PORT_A:
+	case PORT_E:
+	default:
+		return -1;
+	case PORT_B:
+		return 17;
+	case PORT_C:
+		return 19;
+	case PORT_D:
+		return 21;
+	}
+}
+
+static inline enum port get_port_from_pin(enum hpd_pin pin)
+{
+	switch (pin) {
+	case HPD_PORT_B:
+		return PORT_B;
+	case HPD_PORT_C:
+		return PORT_C;
+	case HPD_PORT_D:
+		return PORT_D;
+	default:
+		return PORT_A; /* no hpd */
+	}
+}
+
 static inline void intel_hpd_irq_handler(struct drm_device *dev,
 					 u32 hotplug_trigger,
+					 u32 dig_hotplug_reg,
 					 const u32 *hpd)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
+	enum port port;
 	bool storm_detected = false;
+	bool queue_dig = false, queue_hp = false;
+	u32 dig_shift;
+	u32 dig_port_mask = 0;
 
 	if (!hotplug_trigger)
 		return;
 
-	DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x\n",
-			  hotplug_trigger);
+	DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x, dig 0x%08x\n",
+			 hotplug_trigger, dig_hotplug_reg);
 
 	spin_lock(&dev_priv->irq_lock);
 	for (i = 1; i < HPD_NUM_PINS; i++) {
+		if (!(hpd[i] & hotplug_trigger))
+			continue;
+
+		port = get_port_from_pin(i);
+		if (port && dev_priv->hpd_irq_port[port]) {
+			bool long_hpd;
+
+			if (IS_G4X(dev)) {
+				dig_shift = g4x_port_to_hotplug_shift(port);
+				long_hpd = (hotplug_trigger >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
+			} else {
+				dig_shift = ilk_port_to_hotplug_shift(port);
+				long_hpd = (dig_hotplug_reg >> dig_shift) & PORTB_HOTPLUG_LONG_DETECT;
+			}
 
+			DRM_DEBUG_DRIVER("digital hpd port %d %d\n", port, long_hpd);
+			/* for long HPD pulses we want to have the digital queue happen,
+			   but we still want HPD storm detection to function. */
+			if (long_hpd) {
+				dev_priv->long_hpd_port_mask |= (1 << port);
+				dig_port_mask |= hpd[i];
+			} else {
+				/* for short HPD just trigger the digital queue */
+				dev_priv->short_hpd_port_mask |= (1 << port);
+				hotplug_trigger &= ~hpd[i];
+			}
+			queue_dig = true;
+		}
+	}
+
+	for (i = 1; i < HPD_NUM_PINS; i++) {
 		if (hpd[i] & hotplug_trigger &&
 		    dev_priv->hpd_stats[i].hpd_mark == HPD_DISABLED) {
 			/*
@@ -1553,7 +1808,11 @@ static inline void intel_hpd_irq_handler(struct drm_device *dev,
 		    dev_priv->hpd_stats[i].hpd_mark != HPD_ENABLED)
 			continue;
 
-		dev_priv->hpd_event_bits |= (1 << i);
+		if (!(dig_port_mask & hpd[i])) {
+			dev_priv->hpd_event_bits |= (1 << i);
+			queue_hp = true;
+		}
+
 		if (!time_in_range(jiffies, dev_priv->hpd_stats[i].hpd_last_jiffies,
 				   dev_priv->hpd_stats[i].hpd_last_jiffies
 				   + msecs_to_jiffies(HPD_STORM_DETECT_PERIOD))) {
@@ -1582,7 +1841,10 @@ static inline void intel_hpd_irq_handler(struct drm_device *dev,
 	 * queue for otherwise the flush_work in the pageflip code will
 	 * deadlock.
 	 */
-	schedule_work(&dev_priv->hotplug_work);
+	if (queue_dig)
+		schedule_work(&dev_priv->dig_port_work);
+	if (queue_hp)
+		schedule_work(&dev_priv->hotplug_work);
 }
 
 static void gmbus_irq_handler(struct drm_device *dev)
@@ -1823,11 +2085,11 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev)
 		if (IS_G4X(dev)) {
 			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
 
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_g4x);
+			intel_hpd_irq_handler(dev, hotplug_trigger, 0, hpd_status_g4x);
 		} else {
 			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
 
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
+			intel_hpd_irq_handler(dev, hotplug_trigger, 0, hpd_status_i915);
 		}
 
 		if ((IS_G4X(dev) || IS_VALLEYVIEW(dev)) &&
@@ -1925,8 +2187,12 @@ static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
+	u32 dig_hotplug_reg;
+
+	dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
+	I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
 
-	intel_hpd_irq_handler(dev, hotplug_trigger, hpd_ibx);
+	intel_hpd_irq_handler(dev, hotplug_trigger, dig_hotplug_reg, hpd_ibx);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >>
@@ -2032,8 +2298,12 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
+	u32 dig_hotplug_reg;
 
-	intel_hpd_irq_handler(dev, hotplug_trigger, hpd_cpt);
+	dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
+	I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
+
+	intel_hpd_irq_handler(dev, hotplug_trigger, dig_hotplug_reg, hpd_cpt);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >>
@@ -2780,12 +3050,7 @@ static bool
 ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr)
 {
 	if (INTEL_INFO(dev)->gen >= 8) {
-		/*
-		 * FIXME: gen8 semaphore support - currently we don't emit
-		 * semaphores on bdw anyway, but this needs to be addressed when
-		 * we merge that code.
-		 */
-		return false;
+		return (ipehr >> 23) == 0x1c;
 	} else {
 		ipehr &= ~MI_SEMAPHORE_SYNC_MASK;
 		return ipehr == (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE |
@@ -2794,19 +3059,20 @@ ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr)
 }
 
 static struct intel_engine_cs *
-semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr)
+semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr, u64 offset)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	struct intel_engine_cs *signaller;
 	int i;
 
 	if (INTEL_INFO(dev_priv->dev)->gen >= 8) {
-		/*
-		 * FIXME: gen8 semaphore support - currently we don't emit
-		 * semaphores on bdw anyway, but this needs to be addressed when
-		 * we merge that code.
-		 */
-		return NULL;
+		for_each_ring(signaller, dev_priv, i) {
+			if (ring == signaller)
+				continue;
+
+			if (offset == signaller->semaphore.signal_ggtt[ring->id])
+				return signaller;
+		}
 	} else {
 		u32 sync_bits = ipehr & MI_SEMAPHORE_SYNC_MASK;
 
@@ -2819,8 +3085,8 @@ semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr)
 		}
 	}
 
-	DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x\n",
-		  ring->id, ipehr);
+	DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x, offset 0x%016llx\n",
+		  ring->id, ipehr, offset);
 
 	return NULL;
 }
@@ -2830,7 +3096,8 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 cmd, ipehr, head;
-	int i;
+	u64 offset = 0;
+	int i, backwards;
 
 	ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
 	if (!ipehr_is_semaphore_wait(ring->dev, ipehr))
@@ -2839,13 +3106,15 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 	/*
 	 * HEAD is likely pointing to the dword after the actual command,
 	 * so scan backwards until we find the MBOX. But limit it to just 3
-	 * dwords. Note that we don't care about ACTHD here since that might
+	 * or 4 dwords depending on the semaphore wait command size.
+	 * Note that we don't care about ACTHD here since that might
 	 * point at at batch, and semaphores are always emitted into the
 	 * ringbuffer itself.
 	 */
 	head = I915_READ_HEAD(ring) & HEAD_ADDR;
+	backwards = (INTEL_INFO(ring->dev)->gen >= 8) ? 5 : 4;
 
-	for (i = 4; i; --i) {
+	for (i = backwards; i; --i) {
 		/*
 		 * Be paranoid and presume the hw has gone off into the wild -
 		 * our ring is smaller than what the hardware (and hence
@@ -2865,7 +3134,12 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 		return NULL;
 
 	*seqno = ioread32(ring->buffer->virtual_start + head + 4) + 1;
-	return semaphore_wait_to_signaller_ring(ring, ipehr);
+	if (INTEL_INFO(ring->dev)->gen >= 8) {
+		offset = ioread32(ring->buffer->virtual_start + head + 12);
+		offset <<= 32;
+		offset = ioread32(ring->buffer->virtual_start + head + 8);
+	}
+	return semaphore_wait_to_signaller_ring(ring, ipehr, offset);
 }
 
 static int semaphore_passed(struct intel_engine_cs *ring)
@@ -4354,12 +4628,17 @@ void intel_irq_init(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func);
+	INIT_WORK(&dev_priv->dig_port_work, i915_digport_work_func);
 	INIT_WORK(&dev_priv->gpu_error.work, i915_error_work_func);
 	INIT_WORK(&dev_priv->rps.work, gen6_pm_rps_work);
 	INIT_WORK(&dev_priv->l3_parity.error_work, ivybridge_parity_work);
 
 	/* Let's track the enabled rps events */
-	dev_priv->pm_rps_events = GEN6_PM_RPS_EVENTS;
+	if (IS_VALLEYVIEW(dev))
+		/* WaGsvRC0ResidenncyMethod:VLV */
+		dev_priv->pm_rps_events = GEN6_PM_RP_UP_EI_EXPIRED;
+	else
+		dev_priv->pm_rps_events = GEN6_PM_RPS_EVENTS;
 
 	setup_timer(&dev_priv->gpu_error.hangcheck_timer,
 		    i915_hangcheck_elapsed,

drivers/gpu/drm/i915/i915_reg.h

@@ -240,7 +240,7 @@
 #define   MI_DISPLAY_FLIP_IVB_SPRITE_B (3 << 19)
 #define   MI_DISPLAY_FLIP_IVB_PLANE_C  (4 << 19)
 #define   MI_DISPLAY_FLIP_IVB_SPRITE_C (5 << 19)
-#define MI_SEMAPHORE_MBOX	MI_INSTR(0x16, 1) /* gen6+ */
+#define MI_SEMAPHORE_MBOX	MI_INSTR(0x16, 1) /* gen6, gen7 */
 #define   MI_SEMAPHORE_GLOBAL_GTT    (1<<22)
 #define   MI_SEMAPHORE_UPDATE	    (1<<21)
 #define   MI_SEMAPHORE_COMPARE	    (1<<20)
@@ -266,6 +266,11 @@
 #define   MI_RESTORE_EXT_STATE_EN	(1<<2)
 #define   MI_FORCE_RESTORE		(1<<1)
 #define   MI_RESTORE_INHIBIT		(1<<0)
+#define MI_SEMAPHORE_SIGNAL	MI_INSTR(0x1b, 0) /* GEN8+ */
+#define   MI_SEMAPHORE_TARGET(engine)	((engine)<<15)
+#define MI_SEMAPHORE_WAIT	MI_INSTR(0x1c, 2) /* GEN8+ */
+#define   MI_SEMAPHORE_POLL		(1<<15)
+#define   MI_SEMAPHORE_SAD_GTE_SDD	(1<<12)
 #define MI_STORE_DWORD_IMM	MI_INSTR(0x20, 1)
 #define   MI_MEM_VIRTUAL	(1 << 22) /* 965+ only */
 #define MI_STORE_DWORD_INDEX	MI_INSTR(0x21, 1)
@@ -360,6 +365,7 @@
 #define   PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE		(1<<10) /* GM45+ only */
 #define   PIPE_CONTROL_INDIRECT_STATE_DISABLE		(1<<9)
 #define   PIPE_CONTROL_NOTIFY				(1<<8)
+#define   PIPE_CONTROL_FLUSH_ENABLE			(1<<7) /* gen7+ */
 #define   PIPE_CONTROL_VF_CACHE_INVALIDATE		(1<<4)
 #define   PIPE_CONTROL_CONST_CACHE_INVALIDATE		(1<<3)
 #define   PIPE_CONTROL_STATE_CACHE_INVALIDATE		(1<<2)
@@ -525,6 +531,7 @@ enum punit_power_well {
 #define PUNIT_REG_GPU_FREQ_STS			0xd8
 #define   GENFREQSTATUS				(1<<0)
 #define PUNIT_REG_MEDIA_TURBO_FREQ_REQ		0xdc
+#define PUNIT_REG_CZ_TIMESTAMP			0xce
 
 #define PUNIT_FUSE_BUS2				0xf6 /* bits 47:40 */
 #define PUNIT_FUSE_BUS1				0xf5 /* bits 55:48 */
@@ -550,6 +557,11 @@ enum punit_power_well {
 #define   FB_FMAX_VMIN_FREQ_LO_SHIFT		27
 #define   FB_FMAX_VMIN_FREQ_LO_MASK		0xf8000000
 
+#define VLV_CZ_CLOCK_TO_MILLI_SEC		100000
+#define VLV_RP_UP_EI_THRESHOLD			90
+#define VLV_RP_DOWN_EI_THRESHOLD		70
+#define VLV_INT_COUNT_FOR_DOWN_EI		5
+
 /* vlv2 north clock has */
 #define CCK_FUSE_REG				0x8
 #define  CCK_FUSE_HPLL_FREQ_MASK		0x3
@@ -584,6 +596,11 @@ enum punit_power_well {
 #define  DSI_PLL_M1_DIV_SHIFT			0
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
+#define  DISPLAY_TRUNK_FORCE_ON			(1 << 17)
+#define  DISPLAY_TRUNK_FORCE_OFF		(1 << 16)
+#define  DISPLAY_FREQUENCY_STATUS		(0x1f << 8)
+#define  DISPLAY_FREQUENCY_STATUS_SHIFT		8
+#define  DISPLAY_FREQUENCY_VALUES		(0x1f << 0)
 
 /**
  * DOC: DPIO
@@ -5383,6 +5400,7 @@ enum punit_power_well {
 #define   VLV_GTLC_ALLOWWAKEERR			(1 << 1)
 #define   VLV_GTLC_PW_MEDIA_STATUS_MASK		(1 << 5)
 #define   VLV_GTLC_PW_RENDER_STATUS_MASK	(1 << 7)
+#define VLV_GTLC_SURVIVABILITY_REG              0x130098
 #define  FORCEWAKE_MT				0xa188 /* multi-threaded */
 #define   FORCEWAKE_KERNEL			0x1
 #define   FORCEWAKE_USER			0x2
@@ -5530,6 +5548,8 @@ enum punit_power_well {
 #define GEN6_GT_GFX_RC6_LOCKED			0x138104
 #define VLV_COUNTER_CONTROL			0x138104
 #define   VLV_COUNT_RANGE_HIGH			(1<<15)
+#define   VLV_MEDIA_RC0_COUNT_EN		(1<<5)
+#define   VLV_RENDER_RC0_COUNT_EN		(1<<4)
 #define   VLV_MEDIA_RC6_COUNT_EN		(1<<1)
 #define   VLV_RENDER_RC6_COUNT_EN		(1<<0)
 #define GEN6_GT_GFX_RC6				0x138108
@@ -5538,6 +5558,8 @@ enum punit_power_well {
 
 #define GEN6_GT_GFX_RC6p			0x13810C
 #define GEN6_GT_GFX_RC6pp			0x138110
+#define VLV_RENDER_C0_COUNT_REG		0x138118
+#define VLV_MEDIA_C0_COUNT_REG			0x13811C
 
 #define GEN6_PCODE_MAILBOX			0x138124
 #define   GEN6_PCODE_READY			(1<<31)
@@ -5772,6 +5794,7 @@ enum punit_power_well {
 #define  TRANS_DDI_FUNC_ENABLE		(1<<31)
 /* Those bits are ignored by pipe EDP since it can only connect to DDI A */
 #define  TRANS_DDI_PORT_MASK		(7<<28)
+#define  TRANS_DDI_PORT_SHIFT		28
 #define  TRANS_DDI_SELECT_PORT(x)	((x)<<28)
 #define  TRANS_DDI_PORT_NONE		(0<<28)
 #define  TRANS_DDI_MODE_SELECT_MASK	(7<<24)
@@ -5899,10 +5922,12 @@ enum punit_power_well {
 /* WRPLL */
 #define WRPLL_CTL1			0x46040
 #define WRPLL_CTL2			0x46060
+#define WRPLL_CTL(pll)			(pll == 0 ? WRPLL_CTL1 : WRPLL_CTL2)
 #define  WRPLL_PLL_ENABLE		(1<<31)
-#define  WRPLL_PLL_SELECT_SSC		(0x01<<28)
-#define  WRPLL_PLL_SELECT_NON_SSC	(0x02<<28)
-#define  WRPLL_PLL_SELECT_LCPLL_2700	(0x03<<28)
+#define  WRPLL_PLL_SSC			(1<<28)
+#define  WRPLL_PLL_NON_SSC		(2<<28)
+#define  WRPLL_PLL_LCPLL		(3<<28)
+#define  WRPLL_PLL_REF_MASK		(3<<28)
 /* WRPLL divider programming */
 #define  WRPLL_DIVIDER_REFERENCE(x)	((x)<<0)
 #define  WRPLL_DIVIDER_REF_MASK		(0xff)
@@ -5921,6 +5946,7 @@ enum punit_power_well {
 #define  PORT_CLK_SEL_LCPLL_1350	(1<<29)
 #define  PORT_CLK_SEL_LCPLL_810		(2<<29)
 #define  PORT_CLK_SEL_SPLL		(3<<29)
+#define  PORT_CLK_SEL_WRPLL(pll)	(((pll)+4)<<29)
 #define  PORT_CLK_SEL_WRPLL1		(4<<29)
 #define  PORT_CLK_SEL_WRPLL2		(5<<29)
 #define  PORT_CLK_SEL_NONE		(7<<29)
@@ -5962,7 +5988,10 @@ enum punit_power_well {
 #define  LCPLL_CD_SOURCE_FCLK		(1<<21)
 #define  LCPLL_CD_SOURCE_FCLK_DONE	(1<<19)
 
-#define D_COMP				(MCHBAR_MIRROR_BASE_SNB + 0x5F0C)
+/* Please see hsw_read_dcomp() and hsw_write_dcomp() before using this register,
+ * since on HSW we can't write to it using I915_WRITE. */
+#define D_COMP_HSW			(MCHBAR_MIRROR_BASE_SNB + 0x5F0C)
+#define D_COMP_BDW			0x138144
 #define  D_COMP_RCOMP_IN_PROGRESS	(1<<9)
 #define  D_COMP_COMP_FORCE		(1<<8)
 #define  D_COMP_COMP_DISABLE		(1<<0)

drivers/gpu/drm/i915/intel_crt.c

@@ -137,6 +137,18 @@ static void hsw_crt_get_config(struct intel_encoder *encoder,
 	pipe_config->adjusted_mode.flags |= intel_crt_get_flags(encoder);
 }
 
+static void hsw_crt_pre_enable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	WARN(I915_READ(SPLL_CTL) & SPLL_PLL_ENABLE, "SPLL already enabled\n");
+	I915_WRITE(SPLL_CTL,
+		   SPLL_PLL_ENABLE | SPLL_PLL_FREQ_1350MHz | SPLL_PLL_SSC);
+	POSTING_READ(SPLL_CTL);
+	udelay(20);
+}
+
 /* Note: The caller is required to filter out dpms modes not supported by the
  * platform. */
 static void intel_crt_set_dpms(struct intel_encoder *encoder, int mode)
@@ -194,6 +206,20 @@ static void intel_disable_crt(struct intel_encoder *encoder)
 	intel_crt_set_dpms(encoder, DRM_MODE_DPMS_OFF);
 }
 
+
+static void hsw_crt_post_disable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t val;
+
+	DRM_DEBUG_KMS("Disabling SPLL\n");
+	val = I915_READ(SPLL_CTL);
+	WARN_ON(!(val & SPLL_PLL_ENABLE));
+	I915_WRITE(SPLL_CTL, val & ~SPLL_PLL_ENABLE);
+	POSTING_READ(SPLL_CTL);
+}
+
 static void intel_enable_crt(struct intel_encoder *encoder)
 {
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
@@ -289,8 +315,10 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder,
 		pipe_config->pipe_bpp = 24;
 
 	/* FDI must always be 2.7 GHz */
-	if (HAS_DDI(dev))
+	if (HAS_DDI(dev)) {
+		pipe_config->ddi_pll_sel = PORT_CLK_SEL_SPLL;
 		pipe_config->port_clock = 135000 * 2;
+	}
 
 	return true;
 }
@@ -860,6 +888,8 @@ void intel_crt_init(struct drm_device *dev)
 	if (HAS_DDI(dev)) {
 		crt->base.get_config = hsw_crt_get_config;
 		crt->base.get_hw_state = intel_ddi_get_hw_state;
+		crt->base.pre_enable = hsw_crt_pre_enable;
+		crt->base.post_disable = hsw_crt_post_disable;
 	} else {
 		crt->base.get_config = intel_crt_get_config;
 		crt->base.get_hw_state = intel_crt_get_hw_state;

drivers/gpu/drm/i915/intel_ddi.c

@@ -277,7 +277,8 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 	I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
 
 	/* Configure Port Clock Select */
-	I915_WRITE(PORT_CLK_SEL(PORT_E), intel_crtc->ddi_pll_sel);
+	I915_WRITE(PORT_CLK_SEL(PORT_E), intel_crtc->config.ddi_pll_sel);
+	WARN_ON(intel_crtc->config.ddi_pll_sel != PORT_CLK_SEL_SPLL);
 
 	/* Start the training iterating through available voltages and emphasis,
 	 * testing each value twice. */
@@ -385,53 +386,6 @@ intel_ddi_get_crtc_encoder(struct drm_crtc *crtc)
 	return ret;
 }
 
-void intel_ddi_put_crtc_pll(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
-	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	uint32_t val;
-
-	switch (intel_crtc->ddi_pll_sel) {
-	case PORT_CLK_SEL_SPLL:
-		plls->spll_refcount--;
-		if (plls->spll_refcount == 0) {
-			DRM_DEBUG_KMS("Disabling SPLL\n");
-			val = I915_READ(SPLL_CTL);
-			WARN_ON(!(val & SPLL_PLL_ENABLE));
-			I915_WRITE(SPLL_CTL, val & ~SPLL_PLL_ENABLE);
-			POSTING_READ(SPLL_CTL);
-		}
-		break;
-	case PORT_CLK_SEL_WRPLL1:
-		plls->wrpll1_refcount--;
-		if (plls->wrpll1_refcount == 0) {
-			DRM_DEBUG_KMS("Disabling WRPLL 1\n");
-			val = I915_READ(WRPLL_CTL1);
-			WARN_ON(!(val & WRPLL_PLL_ENABLE));
-			I915_WRITE(WRPLL_CTL1, val & ~WRPLL_PLL_ENABLE);
-			POSTING_READ(WRPLL_CTL1);
-		}
-		break;
-	case PORT_CLK_SEL_WRPLL2:
-		plls->wrpll2_refcount--;
-		if (plls->wrpll2_refcount == 0) {
-			DRM_DEBUG_KMS("Disabling WRPLL 2\n");
-			val = I915_READ(WRPLL_CTL2);
-			WARN_ON(!(val & WRPLL_PLL_ENABLE));
-			I915_WRITE(WRPLL_CTL2, val & ~WRPLL_PLL_ENABLE);
-			POSTING_READ(WRPLL_CTL2);
-		}
-		break;
-	}
-
-	WARN(plls->spll_refcount < 0, "Invalid SPLL refcount\n");
-	WARN(plls->wrpll1_refcount < 0, "Invalid WRPLL1 refcount\n");
-	WARN(plls->wrpll2_refcount < 0, "Invalid WRPLL2 refcount\n");
-
-	intel_crtc->ddi_pll_sel = PORT_CLK_SEL_NONE;
-}
-
 #define LC_FREQ 2700
 #define LC_FREQ_2K (LC_FREQ * 2000)
 
@@ -592,9 +546,9 @@ static int intel_ddi_calc_wrpll_link(struct drm_i915_private *dev_priv,
 	u32 wrpll;
 
 	wrpll = I915_READ(reg);
-	switch (wrpll & SPLL_PLL_REF_MASK) {
-	case SPLL_PLL_SSC:
-	case SPLL_PLL_NON_SSC:
+	switch (wrpll & WRPLL_PLL_REF_MASK) {
+	case WRPLL_PLL_SSC:
+	case WRPLL_PLL_NON_SSC:
 		/*
 		 * We could calculate spread here, but our checking
 		 * code only cares about 5% accuracy, and spread is a max of
@@ -602,7 +556,7 @@ static int intel_ddi_calc_wrpll_link(struct drm_i915_private *dev_priv,
 		 */
 		refclk = 135;
 		break;
-	case SPLL_PLL_LCPLL:
+	case WRPLL_PLL_LCPLL:
 		refclk = LC_FREQ;
 		break;
 	default:
@@ -622,11 +576,10 @@ static void intel_ddi_clock_get(struct intel_encoder *encoder,
 				struct intel_crtc_config *pipe_config)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
-	enum port port = intel_ddi_get_encoder_port(encoder);
 	int link_clock = 0;
 	u32 val, pll;
 
-	val = I915_READ(PORT_CLK_SEL(port));
+	val = pipe_config->ddi_pll_sel;
 	switch (val & PORT_CLK_SEL_MASK) {
 	case PORT_CLK_SEL_LCPLL_810:
 		link_clock = 81000;
@@ -750,173 +703,37 @@ bool intel_ddi_pll_select(struct intel_crtc *intel_crtc)
 {
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_encoder *intel_encoder = intel_ddi_get_crtc_encoder(crtc);
-	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
-	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
 	int type = intel_encoder->type;
-	enum pipe pipe = intel_crtc->pipe;
 	int clock = intel_crtc->config.port_clock;
 
-	intel_ddi_put_crtc_pll(crtc);
-
-	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
-		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-
-		switch (intel_dp->link_bw) {
-		case DP_LINK_BW_1_62:
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_LCPLL_810;
-			break;
-		case DP_LINK_BW_2_7:
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_LCPLL_1350;
-			break;
-		case DP_LINK_BW_5_4:
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_LCPLL_2700;
-			break;
-		default:
-			DRM_ERROR("Link bandwidth %d unsupported\n",
-				  intel_dp->link_bw);
-			return false;
-		}
+	intel_put_shared_dpll(intel_crtc);
 
-	} else if (type == INTEL_OUTPUT_HDMI) {
-		uint32_t reg, val;
+	if (type == INTEL_OUTPUT_HDMI) {
+		struct intel_shared_dpll *pll;
+		uint32_t val;
 		unsigned p, n2, r2;
 
 		intel_ddi_calculate_wrpll(clock * 1000, &r2, &n2, &p);
 
-		val = WRPLL_PLL_ENABLE | WRPLL_PLL_SELECT_LCPLL_2700 |
+		val = WRPLL_PLL_ENABLE | WRPLL_PLL_LCPLL |
 		      WRPLL_DIVIDER_REFERENCE(r2) | WRPLL_DIVIDER_FEEDBACK(n2) |
 		      WRPLL_DIVIDER_POST(p);
 
-		if (val == I915_READ(WRPLL_CTL1)) {
-			DRM_DEBUG_KMS("Reusing WRPLL 1 on pipe %c\n",
-				      pipe_name(pipe));
-			reg = WRPLL_CTL1;
-		} else if (val == I915_READ(WRPLL_CTL2)) {
-			DRM_DEBUG_KMS("Reusing WRPLL 2 on pipe %c\n",
-				      pipe_name(pipe));
-			reg = WRPLL_CTL2;
-		} else if (plls->wrpll1_refcount == 0) {
-			DRM_DEBUG_KMS("Using WRPLL 1 on pipe %c\n",
-				      pipe_name(pipe));
-			reg = WRPLL_CTL1;
-		} else if (plls->wrpll2_refcount == 0) {
-			DRM_DEBUG_KMS("Using WRPLL 2 on pipe %c\n",
-				      pipe_name(pipe));
-			reg = WRPLL_CTL2;
-		} else {
-			DRM_ERROR("No WRPLLs available!\n");
-			return false;
-		}
-
-		DRM_DEBUG_KMS("WRPLL: %dKHz refresh rate with p=%d, n2=%d r2=%d\n",
-			      clock, p, n2, r2);
-
-		if (reg == WRPLL_CTL1) {
-			plls->wrpll1_refcount++;
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_WRPLL1;
-		} else {
-			plls->wrpll2_refcount++;
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_WRPLL2;
-		}
+		intel_crtc->config.dpll_hw_state.wrpll = val;
 
-	} else if (type == INTEL_OUTPUT_ANALOG) {
-		if (plls->spll_refcount == 0) {
-			DRM_DEBUG_KMS("Using SPLL on pipe %c\n",
-				      pipe_name(pipe));
-			plls->spll_refcount++;
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_SPLL;
-		} else {
-			DRM_ERROR("SPLL already in use\n");
+		pll = intel_get_shared_dpll(intel_crtc);
+		if (pll == NULL) {
+			DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
+					 pipe_name(intel_crtc->pipe));
 			return false;
 		}
 
-	} else {
-		WARN(1, "Invalid DDI encoder type %d\n", type);
-		return false;
+		intel_crtc->config.ddi_pll_sel = PORT_CLK_SEL_WRPLL(pll->id);
 	}
 
 	return true;
 }
 
-/*
- * To be called after intel_ddi_pll_select(). That one selects the PLL to be
- * used, this one actually enables the PLL.
- */
-void intel_ddi_pll_enable(struct intel_crtc *crtc)
-{
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
-	int clock = crtc->config.port_clock;
-	uint32_t reg, cur_val, new_val;
-	int refcount;
-	const char *pll_name;
-	uint32_t enable_bit = (1 << 31);
-	unsigned int p, n2, r2;
-
-	BUILD_BUG_ON(enable_bit != SPLL_PLL_ENABLE);
-	BUILD_BUG_ON(enable_bit != WRPLL_PLL_ENABLE);
-
-	switch (crtc->ddi_pll_sel) {
-	case PORT_CLK_SEL_LCPLL_2700:
-	case PORT_CLK_SEL_LCPLL_1350:
-	case PORT_CLK_SEL_LCPLL_810:
-		/*
-		 * LCPLL should always be enabled at this point of the mode set
-		 * sequence, so nothing to do.
-		 */
-		return;
-
-	case PORT_CLK_SEL_SPLL:
-		pll_name = "SPLL";
-		reg = SPLL_CTL;
-		refcount = plls->spll_refcount;
-		new_val = SPLL_PLL_ENABLE | SPLL_PLL_FREQ_1350MHz |
-			  SPLL_PLL_SSC;
-		break;
-
-	case PORT_CLK_SEL_WRPLL1:
-	case PORT_CLK_SEL_WRPLL2:
-		if (crtc->ddi_pll_sel == PORT_CLK_SEL_WRPLL1) {
-			pll_name = "WRPLL1";
-			reg = WRPLL_CTL1;
-			refcount = plls->wrpll1_refcount;
-		} else {
-			pll_name = "WRPLL2";
-			reg = WRPLL_CTL2;
-			refcount = plls->wrpll2_refcount;
-		}
-
-		intel_ddi_calculate_wrpll(clock * 1000, &r2, &n2, &p);
-
-		new_val = WRPLL_PLL_ENABLE | WRPLL_PLL_SELECT_LCPLL_2700 |
-			  WRPLL_DIVIDER_REFERENCE(r2) |
-			  WRPLL_DIVIDER_FEEDBACK(n2) | WRPLL_DIVIDER_POST(p);
-
-		break;
-
-	case PORT_CLK_SEL_NONE:
-		WARN(1, "Bad selected pll: PORT_CLK_SEL_NONE\n");
-		return;
-	default:
-		WARN(1, "Bad selected pll: 0x%08x\n", crtc->ddi_pll_sel);
-		return;
-	}
-
-	cur_val = I915_READ(reg);
-
-	WARN(refcount < 1, "Bad %s refcount: %d\n", pll_name, refcount);
-	if (refcount == 1) {
-		WARN(cur_val & enable_bit, "%s already enabled\n", pll_name);
-		I915_WRITE(reg, new_val);
-		POSTING_READ(reg);
-		udelay(20);
-	} else {
-		WARN((cur_val & enable_bit) == 0, "%s disabled\n", pll_name);
-	}
-}
-
 void intel_ddi_set_pipe_settings(struct drm_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
@@ -995,7 +812,9 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 			 * eDP when not using the panel fitter, and when not
 			 * using motion blur mitigation (which we don't
 			 * support). */
-			if (IS_HASWELL(dev) && intel_crtc->config.pch_pfit.enabled)
+			if (IS_HASWELL(dev) &&
+			    (intel_crtc->config.pch_pfit.enabled ||
+			     intel_crtc->config.pch_pfit.force_thru))
 				temp |= TRANS_DDI_EDP_INPUT_A_ONOFF;
 			else
 				temp |= TRANS_DDI_EDP_INPUT_A_ON;
@@ -1146,76 +965,6 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
 	return false;
 }
 
-static uint32_t intel_ddi_get_crtc_pll(struct drm_i915_private *dev_priv,
-				       enum pipe pipe)
-{
-	uint32_t temp, ret;
-	enum port port = I915_MAX_PORTS;
-	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
-								      pipe);
-	int i;
-
-	if (cpu_transcoder == TRANSCODER_EDP) {
-		port = PORT_A;
-	} else {
-		temp = I915_READ(TRANS_DDI_FUNC_CTL(cpu_transcoder));
-		temp &= TRANS_DDI_PORT_MASK;
-
-		for (i = PORT_B; i <= PORT_E; i++)
-			if (temp == TRANS_DDI_SELECT_PORT(i))
-				port = i;
-	}
-
-	if (port == I915_MAX_PORTS) {
-		WARN(1, "Pipe %c enabled on an unknown port\n",
-		     pipe_name(pipe));
-		ret = PORT_CLK_SEL_NONE;
-	} else {
-		ret = I915_READ(PORT_CLK_SEL(port));
-		DRM_DEBUG_KMS("Pipe %c connected to port %c using clock "
-			      "0x%08x\n", pipe_name(pipe), port_name(port),
-			      ret);
-	}
-
-	return ret;
-}
-
-void intel_ddi_setup_hw_pll_state(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum pipe pipe;
-	struct intel_crtc *intel_crtc;
-
-	dev_priv->ddi_plls.spll_refcount = 0;
-	dev_priv->ddi_plls.wrpll1_refcount = 0;
-	dev_priv->ddi_plls.wrpll2_refcount = 0;
-
-	for_each_pipe(pipe) {
-		intel_crtc =
-			to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
-
-		if (!intel_crtc->active) {
-			intel_crtc->ddi_pll_sel = PORT_CLK_SEL_NONE;
-			continue;
-		}
-
-		intel_crtc->ddi_pll_sel = intel_ddi_get_crtc_pll(dev_priv,
-								 pipe);
-
-		switch (intel_crtc->ddi_pll_sel) {
-		case PORT_CLK_SEL_SPLL:
-			dev_priv->ddi_plls.spll_refcount++;
-			break;
-		case PORT_CLK_SEL_WRPLL1:
-			dev_priv->ddi_plls.wrpll1_refcount++;
-			break;
-		case PORT_CLK_SEL_WRPLL2:
-			dev_priv->ddi_plls.wrpll2_refcount++;
-			break;
-		}
-	}
-}
-
 void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc)
 {
 	struct drm_crtc *crtc = &intel_crtc->base;
@@ -1261,8 +1010,8 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 		intel_edp_panel_on(intel_dp);
 	}
 
-	WARN_ON(crtc->ddi_pll_sel == PORT_CLK_SEL_NONE);
-	I915_WRITE(PORT_CLK_SEL(port), crtc->ddi_pll_sel);
+	WARN_ON(crtc->config.ddi_pll_sel == PORT_CLK_SEL_NONE);
+	I915_WRITE(PORT_CLK_SEL(port), crtc->config.ddi_pll_sel);
 
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
@@ -1418,10 +1167,60 @@ int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv)
 	}
 }
 
+static void hsw_ddi_pll_enable(struct drm_i915_private *dev_priv,
+			       struct intel_shared_dpll *pll)
+{
+	I915_WRITE(WRPLL_CTL(pll->id), pll->hw_state.wrpll);
+	POSTING_READ(WRPLL_CTL(pll->id));
+	udelay(20);
+}
+
+static void hsw_ddi_pll_disable(struct drm_i915_private *dev_priv,
+				struct intel_shared_dpll *pll)
+{
+	uint32_t val;
+
+	val = I915_READ(WRPLL_CTL(pll->id));
+	I915_WRITE(WRPLL_CTL(pll->id), val & ~WRPLL_PLL_ENABLE);
+	POSTING_READ(WRPLL_CTL(pll->id));
+}
+
+static bool hsw_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
+				     struct intel_shared_dpll *pll,
+				     struct intel_dpll_hw_state *hw_state)
+{
+	uint32_t val;
+
+	if (!intel_display_power_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
+	val = I915_READ(WRPLL_CTL(pll->id));
+	hw_state->wrpll = val;
+
+	return val & WRPLL_PLL_ENABLE;
+}
+
+static char *hsw_ddi_pll_names[] = {
+	"WRPLL 1",
+	"WRPLL 2",
+};
+
 void intel_ddi_pll_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t val = I915_READ(LCPLL_CTL);
+	int i;
+
+	dev_priv->num_shared_dpll = 2;
+
+	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
+		dev_priv->shared_dplls[i].id = i;
+		dev_priv->shared_dplls[i].name = hsw_ddi_pll_names[i];
+		dev_priv->shared_dplls[i].disable = hsw_ddi_pll_disable;
+		dev_priv->shared_dplls[i].enable = hsw_ddi_pll_enable;
+		dev_priv->shared_dplls[i].get_hw_state =
+			hsw_ddi_pll_get_hw_state;
+	}
 
 	/* The LCPLL register should be turned on by the BIOS. For now let's
 	 * just check its state and print errors in case something is wrong.
@@ -1705,6 +1504,9 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 	intel_encoder->cloneable = 0;
 	intel_encoder->hot_plug = intel_ddi_hot_plug;
 
+	intel_dig_port->hpd_pulse = intel_dp_hpd_pulse;
+	dev_priv->hpd_irq_port[port] = intel_dig_port;
+
 	if (init_dp)
 		dp_connector = intel_ddi_init_dp_connector(intel_dig_port);
 

drivers/gpu/drm/i915/intel_display.c

@@ -1094,11 +1094,6 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv,
 	bool cur_state;
 	struct intel_dpll_hw_state hw_state;
 
-	if (HAS_PCH_LPT(dev_priv->dev)) {
-		DRM_DEBUG_DRIVER("LPT detected: skipping PCH PLL test\n");
-		return;
-	}
-
 	if (WARN (!pll,
 		  "asserting DPLL %s with no DPLL\n", state_string(state)))
 		return;
@@ -1514,9 +1509,6 @@ static void intel_reset_dpio(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (!IS_VALLEYVIEW(dev))
-		return;
-
 	if (IS_CHERRYVIEW(dev)) {
 		enum dpio_phy phy;
 		u32 val;
@@ -1538,26 +1530,6 @@ static void intel_reset_dpio(struct drm_device *dev)
 			I915_WRITE(DISPLAY_PHY_CONTROL,
 				PHY_COM_LANE_RESET_DEASSERT(phy, val));
 		}
-
-	} else {
-		/*
-		 * If DPIO has already been reset, e.g. by BIOS, just skip all
-		 * this.
-		 */
-		if (I915_READ(DPIO_CTL) & DPIO_CMNRST)
-			return;
-
-		/*
-		 * From VLV2A0_DP_eDP_HDMI_DPIO_driver_vbios_notes_11.docx:
-		 * Need to assert and de-assert PHY SB reset by gating the
-		 * common lane power, then un-gating it.
-		 * Simply ungating isn't enough to reset the PHY enough to get
-		 * ports and lanes running.
-		 */
-		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
-				     false);
-		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
-				     true);
 	}
 }
 
@@ -1837,12 +1809,14 @@ static void intel_enable_shared_dpll(struct intel_crtc *crtc)
 	}
 	WARN_ON(pll->on);
 
+	intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
+
 	DRM_DEBUG_KMS("enabling %s\n", pll->name);
 	pll->enable(dev_priv, pll);
 	pll->on = true;
 }
 
-static void intel_disable_shared_dpll(struct intel_crtc *crtc)
+void intel_disable_shared_dpll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1873,6 +1847,8 @@ static void intel_disable_shared_dpll(struct intel_crtc *crtc)
 	DRM_DEBUG_KMS("disabling %s\n", pll->name);
 	pll->disable(dev_priv, pll);
 	pll->on = false;
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
 }
 
 static void ironlake_enable_pch_transcoder(struct drm_i915_private *dev_priv,
@@ -2219,6 +2195,8 @@ intel_pin_and_fence_fb_obj(struct drm_device *dev,
 	u32 alignment;
 	int ret;
 
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
 	switch (obj->tiling_mode) {
 	case I915_TILING_NONE:
 		if (IS_BROADWATER(dev) || IS_CRESTLINE(dev))
@@ -2275,6 +2253,8 @@ err_interruptible:
 
 void intel_unpin_fb_obj(struct drm_i915_gem_object *obj)
 {
+	WARN_ON(!mutex_is_locked(&obj->base.dev->struct_mutex));
+
 	i915_gem_object_unpin_fence(obj);
 	i915_gem_object_unpin_from_display_plane(obj);
 }
@@ -2379,7 +2359,7 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct drm_crtc *c;
 	struct intel_crtc *i;
-	struct intel_framebuffer *fb;
+	struct drm_i915_gem_object *obj;
 
 	if (!intel_crtc->base.primary->fb)
 		return;
@@ -2400,14 +2380,17 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 		if (c == &intel_crtc->base)
 			continue;
 
-		if (!i->active || !c->primary->fb)
+		if (!i->active)
 			continue;
 
-		fb = to_intel_framebuffer(c->primary->fb);
-		if (i915_gem_obj_ggtt_offset(fb->obj) == plane_config->base) {
+		obj = intel_fb_obj(c->primary->fb);
+		if (obj == NULL)
+			continue;
+
+		if (i915_gem_obj_ggtt_offset(obj) == plane_config->base) {
 			drm_framebuffer_reference(c->primary->fb);
 			intel_crtc->base.primary->fb = c->primary->fb;
-			fb->obj->frontbuffer_bits |= INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe);
+			obj->frontbuffer_bits |= INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe);
 			break;
 		}
 	}
@@ -2420,16 +2403,12 @@ static void i9xx_update_primary_plane(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_framebuffer *intel_fb;
-	struct drm_i915_gem_object *obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int plane = intel_crtc->plane;
 	unsigned long linear_offset;
 	u32 dspcntr;
 	u32 reg;
 
-	intel_fb = to_intel_framebuffer(fb);
-	obj = intel_fb->obj;
-
 	reg = DSPCNTR(plane);
 	dspcntr = I915_READ(reg);
 	/* Mask out pixel format bits in case we change it */
@@ -2510,16 +2489,12 @@ static void ironlake_update_primary_plane(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_framebuffer *intel_fb;
-	struct drm_i915_gem_object *obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int plane = intel_crtc->plane;
 	unsigned long linear_offset;
 	u32 dspcntr;
 	u32 reg;
 
-	intel_fb = to_intel_framebuffer(fb);
-	obj = intel_fb->obj;
-
 	reg = DSPCNTR(plane);
 	dspcntr = I915_READ(reg);
 	/* Mask out pixel format bits in case we change it */
@@ -2650,7 +2625,7 @@ void intel_display_handle_reset(struct drm_device *dev)
 static int
 intel_finish_fb(struct drm_framebuffer *old_fb)
 {
-	struct drm_i915_gem_object *obj = to_intel_framebuffer(old_fb)->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(old_fb);
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	bool was_interruptible = dev_priv->mm.interruptible;
 	int ret;
@@ -2697,8 +2672,9 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
-	struct drm_framebuffer *old_fb;
-	struct drm_i915_gem_object *obj = to_intel_framebuffer(fb)->obj;
+	struct drm_framebuffer *old_fb = crtc->primary->fb;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct drm_i915_gem_object *old_obj = intel_fb_obj(old_fb);
 	int ret;
 
 	if (intel_crtc_has_pending_flip(crtc)) {
@@ -2719,12 +2695,10 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		return -EINVAL;
 	}
 
-	old_fb = crtc->primary->fb;
-
 	mutex_lock(&dev->struct_mutex);
 	ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
 	if (ret == 0)
-		i915_gem_track_fb(to_intel_framebuffer(old_fb)->obj, obj,
+		i915_gem_track_fb(old_obj, obj,
 				  INTEL_FRONTBUFFER_PRIMARY(pipe));
 	mutex_unlock(&dev->struct_mutex);
 	if (ret != 0) {
@@ -2776,7 +2750,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		if (intel_crtc->active && old_fb != fb)
 			intel_wait_for_vblank(dev, intel_crtc->pipe);
 		mutex_lock(&dev->struct_mutex);
-		intel_unpin_fb_obj(to_intel_framebuffer(old_fb)->obj);
+		intel_unpin_fb_obj(old_obj);
 		mutex_unlock(&dev->struct_mutex);
 	}
 
@@ -3642,7 +3616,7 @@ static void lpt_pch_enable(struct drm_crtc *crtc)
 	lpt_enable_pch_transcoder(dev_priv, cpu_transcoder);
 }
 
-static void intel_put_shared_dpll(struct intel_crtc *crtc)
+void intel_put_shared_dpll(struct intel_crtc *crtc)
 {
 	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
 
@@ -3662,7 +3636,7 @@ static void intel_put_shared_dpll(struct intel_crtc *crtc)
 	crtc->config.shared_dpll = DPLL_ID_PRIVATE;
 }
 
-static struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
+struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
 	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
@@ -3915,30 +3889,6 @@ static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
 	 */
 }
 
-/**
- * i9xx_fixup_plane - ugly workaround for G45 to fire up the hardware
- * cursor plane briefly if not already running after enabling the display
- * plane.
- * This workaround avoids occasional blank screens when self refresh is
- * enabled.
- */
-static void
-g4x_fixup_plane(struct drm_i915_private *dev_priv, enum pipe pipe)
-{
-	u32 cntl = I915_READ(CURCNTR(pipe));
-
-	if ((cntl & CURSOR_MODE) == 0) {
-		u32 fw_bcl_self = I915_READ(FW_BLC_SELF);
-
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self & ~FW_BLC_SELF_EN);
-		I915_WRITE(CURCNTR(pipe), CURSOR_MODE_64_ARGB_AX);
-		intel_wait_for_vblank(dev_priv->dev, pipe);
-		I915_WRITE(CURCNTR(pipe), cntl);
-		I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe)));
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self);
-	}
-}
-
 static void intel_crtc_enable_planes(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -3951,9 +3901,6 @@ static void intel_crtc_enable_planes(struct drm_crtc *crtc)
 
 	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
 	intel_enable_planes(crtc);
-	/* The fixup needs to happen before cursor is enabled */
-	if (IS_G4X(dev))
-		g4x_fixup_plane(dev_priv, pipe);
 	intel_crtc_update_cursor(crtc, true);
 	intel_crtc_dpms_overlay(intel_crtc, true);
 
@@ -4128,6 +4075,9 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	if (intel_crtc->active)
 		return;
 
+	if (intel_crtc_to_shared_dpll(intel_crtc))
+		intel_enable_shared_dpll(intel_crtc);
+
 	if (intel_crtc->config.has_dp_encoder)
 		intel_dp_set_m_n(intel_crtc);
 
@@ -4152,16 +4102,15 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
-	if (intel_crtc->config.has_pch_encoder)
-		intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, true);
-
-	if (intel_crtc->config.has_pch_encoder)
-		dev_priv->display.fdi_link_train(crtc);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
 
+	if (intel_crtc->config.has_pch_encoder) {
+		intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, true);
+		dev_priv->display.fdi_link_train(crtc);
+	}
+
 	intel_ddi_enable_pipe_clock(intel_crtc);
 
 	ironlake_pfit_enable(intel_crtc);
@@ -4299,22 +4248,25 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
 	intel_ddi_disable_pipe_clock(intel_crtc);
 
-	for_each_encoder_on_crtc(dev, crtc, encoder)
-		if (encoder->post_disable)
-			encoder->post_disable(encoder);
-
 	if (intel_crtc->config.has_pch_encoder) {
 		lpt_disable_pch_transcoder(dev_priv);
 		intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, true);
 		intel_ddi_fdi_disable(crtc);
 	}
 
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->post_disable)
+			encoder->post_disable(encoder);
+
 	intel_crtc->active = false;
 	intel_update_watermarks(crtc);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
 	mutex_unlock(&dev->struct_mutex);
+
+	if (intel_crtc_to_shared_dpll(intel_crtc))
+		intel_disable_shared_dpll(intel_crtc);
 }
 
 static void ironlake_crtc_off(struct drm_crtc *crtc)
@@ -4323,10 +4275,6 @@ static void ironlake_crtc_off(struct drm_crtc *crtc)
 	intel_put_shared_dpll(intel_crtc);
 }
 
-static void haswell_crtc_off(struct drm_crtc *crtc)
-{
-	intel_ddi_put_crtc_pll(crtc);
-}
 
 static void i9xx_pfit_enable(struct intel_crtc *crtc)
 {
@@ -4398,7 +4346,6 @@ static unsigned long get_crtc_power_domains(struct drm_crtc *crtc)
 	struct intel_encoder *intel_encoder;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
-	bool pfit_enabled = intel_crtc->config.pch_pfit.enabled;
 	unsigned long mask;
 	enum transcoder transcoder;
 
@@ -4406,7 +4353,8 @@ static unsigned long get_crtc_power_domains(struct drm_crtc *crtc)
 
 	mask = BIT(POWER_DOMAIN_PIPE(pipe));
 	mask |= BIT(POWER_DOMAIN_TRANSCODER(transcoder));
-	if (pfit_enabled)
+	if (intel_crtc->config.pch_pfit.enabled ||
+	    intel_crtc->config.pch_pfit.force_thru)
 		mask |= BIT(POWER_DOMAIN_PIPE_PANEL_FITTER(pipe));
 
 	for_each_encoder_on_crtc(dev, crtc, intel_encoder)
@@ -4463,7 +4411,8 @@ static void modeset_update_crtc_power_domains(struct drm_device *dev)
 	intel_display_set_init_power(dev_priv, false);
 }
 
-int valleyview_get_vco(struct drm_i915_private *dev_priv)
+/* returns HPLL frequency in kHz */
+static int valleyview_get_vco(struct drm_i915_private *dev_priv)
 {
 	int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 };
 
@@ -4473,7 +4422,23 @@ int valleyview_get_vco(struct drm_i915_private *dev_priv)
 		CCK_FUSE_HPLL_FREQ_MASK;
 	mutex_unlock(&dev_priv->dpio_lock);
 
-	return vco_freq[hpll_freq];
+	return vco_freq[hpll_freq] * 1000;
+}
+
+static void vlv_update_cdclk(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->vlv_cdclk_freq = dev_priv->display.get_display_clock_speed(dev);
+	DRM_DEBUG_DRIVER("Current CD clock rate: %d kHz",
+			 dev_priv->vlv_cdclk_freq);
+
+	/*
+	 * Program the gmbus_freq based on the cdclk frequency.
+	 * BSpec erroneously claims we should aim for 4MHz, but
+	 * in fact 1MHz is the correct frequency.
+	 */
+	I915_WRITE(GMBUSFREQ_VLV, dev_priv->vlv_cdclk_freq);
 }
 
 /* Adjust CDclk dividers to allow high res or save power if possible */
@@ -4482,12 +4447,11 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, cmd;
 
-	WARN_ON(valleyview_cur_cdclk(dev_priv) != dev_priv->vlv_cdclk_freq);
-	dev_priv->vlv_cdclk_freq = cdclk;
+	WARN_ON(dev_priv->display.get_display_clock_speed(dev) != dev_priv->vlv_cdclk_freq);
 
-	if (cdclk >= 320) /* jump to highest voltage for 400MHz too */
+	if (cdclk >= 320000) /* jump to highest voltage for 400MHz too */
 		cmd = 2;
-	else if (cdclk == 266)
+	else if (cdclk == 266667)
 		cmd = 1;
 	else
 		cmd = 0;
@@ -4504,18 +4468,23 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	}
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
-	if (cdclk == 400) {
+	if (cdclk == 400000) {
 		u32 divider, vco;
 
 		vco = valleyview_get_vco(dev_priv);
-		divider = ((vco << 1) / cdclk) - 1;
+		divider = DIV_ROUND_CLOSEST(vco << 1, cdclk) - 1;
 
 		mutex_lock(&dev_priv->dpio_lock);
 		/* adjust cdclk divider */
 		val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
-		val &= ~0xf;
+		val &= ~DISPLAY_FREQUENCY_VALUES;
 		val |= divider;
 		vlv_cck_write(dev_priv, CCK_DISPLAY_CLOCK_CONTROL, val);
+
+		if (wait_for((vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL) &
+			      DISPLAY_FREQUENCY_STATUS) == (divider << DISPLAY_FREQUENCY_STATUS_SHIFT),
+			     50))
+			DRM_ERROR("timed out waiting for CDclk change\n");
 		mutex_unlock(&dev_priv->dpio_lock);
 	}
 
@@ -4528,54 +4497,43 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	 * For high bandwidth configs, we set a higher latency in the bunit
 	 * so that the core display fetch happens in time to avoid underruns.
 	 */
-	if (cdclk == 400)
+	if (cdclk == 400000)
 		val |= 4500 / 250; /* 4.5 usec */
 	else
 		val |= 3000 / 250; /* 3.0 usec */
 	vlv_bunit_write(dev_priv, BUNIT_REG_BISOC, val);
 	mutex_unlock(&dev_priv->dpio_lock);
 
-	/* Since we changed the CDclk, we need to update the GMBUSFREQ too */
-	intel_i2c_reset(dev);
-}
-
-int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
-{
-	int cur_cdclk, vco;
-	int divider;
-
-	vco = valleyview_get_vco(dev_priv);
-
-	mutex_lock(&dev_priv->dpio_lock);
-	divider = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
-	mutex_unlock(&dev_priv->dpio_lock);
-
-	divider &= 0xf;
-
-	cur_cdclk = (vco << 1) / (divider + 1);
-
-	return cur_cdclk;
+	vlv_update_cdclk(dev);
 }
 
 static int valleyview_calc_cdclk(struct drm_i915_private *dev_priv,
 				 int max_pixclk)
 {
+	int vco = valleyview_get_vco(dev_priv);
+	int freq_320 = (vco <<  1) % 320000 != 0 ? 333333 : 320000;
+
 	/*
 	 * Really only a few cases to deal with, as only 4 CDclks are supported:
 	 *   200MHz
 	 *   267MHz
-	 *   320MHz
+	 *   320/333MHz (depends on HPLL freq)
 	 *   400MHz
 	 * So we check to see whether we're above 90% of the lower bin and
 	 * adjust if needed.
+	 *
+	 * We seem to get an unstable or solid color picture at 200MHz.
+	 * Not sure what's wrong. For now use 200MHz only when all pipes
+	 * are off.
 	 */
-	if (max_pixclk > 288000) {
-		return 400;
-	} else if (max_pixclk > 240000) {
-		return 320;
-	} else
-		return 266;
-	/* Looks like the 200MHz CDclk freq doesn't work on some configs */
+	if (max_pixclk > freq_320*9/10)
+		return 400000;
+	else if (max_pixclk > 266667*9/10)
+		return freq_320;
+	else if (max_pixclk > 0)
+		return 266667;
+	else
+		return 200000;
 }
 
 /* compute the max pixel clock for new configuration */
@@ -4829,6 +4787,16 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	if (IS_GEN2(dev))
 		intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 
+	/*
+	 * Vblank time updates from the shadow to live plane control register
+	 * are blocked if the memory self-refresh mode is active at that
+	 * moment. So to make sure the plane gets truly disabled, disable
+	 * first the self-refresh mode. The self-refresh enable bit in turn
+	 * will be checked/applied by the HW only at the next frame start
+	 * event which is after the vblank start event, so we need to have a
+	 * wait-for-vblank between disabling the plane and the pipe.
+	 */
+	intel_set_memory_cxsr(dev_priv, false);
 	intel_crtc_disable_planes(crtc);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
@@ -4837,9 +4805,10 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	/*
 	 * On gen2 planes are double buffered but the pipe isn't, so we must
 	 * wait for planes to fully turn off before disabling the pipe.
+	 * We also need to wait on all gmch platforms because of the
+	 * self-refresh mode constraint explained above.
 	 */
-	if (IS_GEN2(dev))
-		intel_wait_for_vblank(dev, pipe);
+	intel_wait_for_vblank(dev, pipe);
 
 	intel_disable_pipe(dev_priv, pipe);
 
@@ -4956,7 +4925,7 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_connector *connector;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *old_obj;
+	struct drm_i915_gem_object *old_obj = intel_fb_obj(crtc->primary->fb);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 
 	/* crtc should still be enabled when we disable it. */
@@ -4971,7 +4940,6 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	assert_pipe_disabled(dev->dev_private, pipe);
 
 	if (crtc->primary->fb) {
-		old_obj = to_intel_framebuffer(crtc->primary->fb)->obj;
 		mutex_lock(&dev->struct_mutex);
 		intel_unpin_fb_obj(old_obj);
 		i915_gem_track_fb(old_obj, NULL,
@@ -5253,9 +5221,11 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 	if (HAS_IPS(dev))
 		hsw_compute_ips_config(crtc, pipe_config);
 
-	/* XXX: PCH clock sharing is done in ->mode_set, so make sure the old
-	 * clock survives for now. */
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
+	/*
+	 * XXX: PCH/WRPLL clock sharing is done in ->mode_set, so make sure the
+	 * old clock survives for now.
+	 */
+	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev) || HAS_DDI(dev))
 		pipe_config->shared_dpll = crtc->config.shared_dpll;
 
 	if (pipe_config->has_pch_encoder)
@@ -5266,7 +5236,22 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 
 static int valleyview_get_display_clock_speed(struct drm_device *dev)
 {
-	return 400000; /* FIXME */
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int vco = valleyview_get_vco(dev_priv);
+	u32 val;
+	int divider;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	divider = val & DISPLAY_FREQUENCY_VALUES;
+
+	WARN((val & DISPLAY_FREQUENCY_STATUS) !=
+	     (divider << DISPLAY_FREQUENCY_STATUS_SHIFT),
+	     "cdclk change in progress\n");
+
+	return DIV_ROUND_CLOSEST(vco << 1, divider + 1);
 }
 
 static int i945_get_display_clock_speed(struct drm_device *dev)
@@ -6217,8 +6202,8 @@ static void i9xx_get_plane_config(struct intel_crtc *crtc,
 	aligned_height = intel_align_height(dev, crtc->base.primary->fb->height,
 					    plane_config->tiled);
 
-	plane_config->size = ALIGN(crtc->base.primary->fb->pitches[0] *
-				   aligned_height, PAGE_SIZE);
+	plane_config->size = PAGE_ALIGN(crtc->base.primary->fb->pitches[0] *
+					aligned_height);
 
 	DRM_DEBUG_KMS("pipe/plane %d/%d with fb: size=%dx%d@%d, offset=%x, pitch %d, size 0x%x\n",
 		      pipe, plane, crtc->base.primary->fb->width,
@@ -7237,8 +7222,8 @@ static void ironlake_get_plane_config(struct intel_crtc *crtc,
 	aligned_height = intel_align_height(dev, crtc->base.primary->fb->height,
 					    plane_config->tiled);
 
-	plane_config->size = ALIGN(crtc->base.primary->fb->pitches[0] *
-				   aligned_height, PAGE_SIZE);
+	plane_config->size = PAGE_ALIGN(crtc->base.primary->fb->pitches[0] *
+					aligned_height);
 
 	DRM_DEBUG_KMS("pipe/plane %d/%d with fb: size=%dx%d@%d, offset=%x, pitch %d, size 0x%x\n",
 		      pipe, plane, crtc->base.primary->fb->width,
@@ -7255,6 +7240,10 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 
+	if (!intel_display_power_enabled(dev_priv,
+					 POWER_DOMAIN_PIPE(crtc->pipe)))
+		return false;
+
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
@@ -7329,7 +7318,6 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
-	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
 	struct intel_crtc *crtc;
 
 	for_each_intel_crtc(dev, crtc)
@@ -7337,9 +7325,9 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 		     pipe_name(crtc->pipe));
 
 	WARN(I915_READ(HSW_PWR_WELL_DRIVER), "Power well on\n");
-	WARN(plls->spll_refcount, "SPLL enabled\n");
-	WARN(plls->wrpll1_refcount, "WRPLL1 enabled\n");
-	WARN(plls->wrpll2_refcount, "WRPLL2 enabled\n");
+	WARN(I915_READ(SPLL_CTL) & SPLL_PLL_ENABLE, "SPLL enabled\n");
+	WARN(I915_READ(WRPLL_CTL1) & WRPLL_PLL_ENABLE, "WRPLL1 enabled\n");
+	WARN(I915_READ(WRPLL_CTL2) & WRPLL_PLL_ENABLE, "WRPLL2 enabled\n");
 	WARN(I915_READ(PCH_PP_STATUS) & PP_ON, "Panel power on\n");
 	WARN(I915_READ(BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE,
 	     "CPU PWM1 enabled\n");
@@ -7360,6 +7348,16 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	WARN(!dev_priv->pm.irqs_disabled, "IRQs enabled\n");
 }
 
+static uint32_t hsw_read_dcomp(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	if (IS_HASWELL(dev))
+		return I915_READ(D_COMP_HSW);
+	else
+		return I915_READ(D_COMP_BDW);
+}
+
 static void hsw_write_dcomp(struct drm_i915_private *dev_priv, uint32_t val)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -7368,12 +7366,12 @@ static void hsw_write_dcomp(struct drm_i915_private *dev_priv, uint32_t val)
 		mutex_lock(&dev_priv->rps.hw_lock);
 		if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP,
 					    val))
-			DRM_ERROR("Failed to disable D_COMP\n");
+			DRM_ERROR("Failed to write to D_COMP\n");
 		mutex_unlock(&dev_priv->rps.hw_lock);
 	} else {
-		I915_WRITE(D_COMP, val);
+		I915_WRITE(D_COMP_BDW, val);
+		POSTING_READ(D_COMP_BDW);
 	}
-	POSTING_READ(D_COMP);
 }
 
 /*
@@ -7411,12 +7409,13 @@ static void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
 	if (wait_for((I915_READ(LCPLL_CTL) & LCPLL_PLL_LOCK) == 0, 1))
 		DRM_ERROR("LCPLL still locked\n");
 
-	val = I915_READ(D_COMP);
+	val = hsw_read_dcomp(dev_priv);
 	val |= D_COMP_COMP_DISABLE;
 	hsw_write_dcomp(dev_priv, val);
 	ndelay(100);
 
-	if (wait_for((I915_READ(D_COMP) & D_COMP_RCOMP_IN_PROGRESS) == 0, 1))
+	if (wait_for((hsw_read_dcomp(dev_priv) & D_COMP_RCOMP_IN_PROGRESS) == 0,
+		     1))
 		DRM_ERROR("D_COMP RCOMP still in progress\n");
 
 	if (allow_power_down) {
@@ -7465,7 +7464,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 		POSTING_READ(LCPLL_CTL);
 	}
 
-	val = I915_READ(D_COMP);
+	val = hsw_read_dcomp(dev_priv);
 	val |= D_COMP_COMP_FORCE;
 	val &= ~D_COMP_COMP_DISABLE;
 	hsw_write_dcomp(dev_priv, val);
@@ -7571,13 +7570,59 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 
 	if (!intel_ddi_pll_select(intel_crtc))
 		return -EINVAL;
-	intel_ddi_pll_enable(intel_crtc);
 
 	intel_crtc->lowfreq_avail = false;
 
 	return 0;
 }
 
+static void haswell_get_ddi_port_state(struct intel_crtc *crtc,
+				       struct intel_crtc_config *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll;
+	enum port port;
+	uint32_t tmp;
+
+	tmp = I915_READ(TRANS_DDI_FUNC_CTL(pipe_config->cpu_transcoder));
+
+	port = (tmp & TRANS_DDI_PORT_MASK) >> TRANS_DDI_PORT_SHIFT;
+
+	pipe_config->ddi_pll_sel = I915_READ(PORT_CLK_SEL(port));
+
+	switch (pipe_config->ddi_pll_sel) {
+	case PORT_CLK_SEL_WRPLL1:
+		pipe_config->shared_dpll = DPLL_ID_WRPLL1;
+		break;
+	case PORT_CLK_SEL_WRPLL2:
+		pipe_config->shared_dpll = DPLL_ID_WRPLL2;
+		break;
+	}
+
+	if (pipe_config->shared_dpll >= 0) {
+		pll = &dev_priv->shared_dplls[pipe_config->shared_dpll];
+
+		WARN_ON(!pll->get_hw_state(dev_priv, pll,
+					   &pipe_config->dpll_hw_state));
+	}
+
+	/*
+	 * Haswell has only FDI/PCH transcoder A. It is which is connected to
+	 * DDI E. So just check whether this pipe is wired to DDI E and whether
+	 * the PCH transcoder is on.
+	 */
+	if ((port == PORT_E) && I915_READ(LPT_TRANSCONF) & TRANS_ENABLE) {
+		pipe_config->has_pch_encoder = true;
+
+		tmp = I915_READ(FDI_RX_CTL(PIPE_A));
+		pipe_config->fdi_lanes = ((FDI_DP_PORT_WIDTH_MASK & tmp) >>
+					  FDI_DP_PORT_WIDTH_SHIFT) + 1;
+
+		ironlake_get_fdi_m_n_config(crtc, pipe_config);
+	}
+}
+
 static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 				    struct intel_crtc_config *pipe_config)
 {
@@ -7623,22 +7668,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	if (!(tmp & PIPECONF_ENABLE))
 		return false;
 
-	/*
-	 * Haswell has only FDI/PCH transcoder A. It is which is connected to
-	 * DDI E. So just check whether this pipe is wired to DDI E and whether
-	 * the PCH transcoder is on.
-	 */
-	tmp = I915_READ(TRANS_DDI_FUNC_CTL(pipe_config->cpu_transcoder));
-	if ((tmp & TRANS_DDI_PORT_MASK) == TRANS_DDI_SELECT_PORT(PORT_E) &&
-	    I915_READ(LPT_TRANSCONF) & TRANS_ENABLE) {
-		pipe_config->has_pch_encoder = true;
-
-		tmp = I915_READ(FDI_RX_CTL(PIPE_A));
-		pipe_config->fdi_lanes = ((FDI_DP_PORT_WIDTH_MASK & tmp) >>
-					  FDI_DP_PORT_WIDTH_SHIFT) + 1;
-
-		ironlake_get_fdi_m_n_config(crtc, pipe_config);
-	}
+	haswell_get_ddi_port_state(crtc, pipe_config);
 
 	intel_get_pipe_timings(crtc, pipe_config);
 
@@ -8326,7 +8356,7 @@ static u32
 intel_framebuffer_size_for_mode(struct drm_display_mode *mode, int bpp)
 {
 	u32 pitch = intel_framebuffer_pitch_for_width(mode->hdisplay, bpp);
-	return ALIGN(pitch * mode->vdisplay, PAGE_SIZE);
+	return PAGE_ALIGN(pitch * mode->vdisplay);
 }
 
 static struct drm_framebuffer *
@@ -9447,6 +9477,9 @@ static bool use_mmio_flip(struct intel_engine_cs *ring,
 	 * So using MMIO flips there would disrupt this mechanism.
 	 */
 
+	if (ring == NULL)
+		return true;
+
 	if (INTEL_INFO(ring->dev)->gen < 5)
 		return false;
 
@@ -9595,7 +9628,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *old_fb = crtc->primary->fb;
-	struct drm_i915_gem_object *obj = to_intel_framebuffer(fb)->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
 	struct intel_unpin_work *work;
@@ -9603,6 +9636,14 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	unsigned long flags;
 	int ret;
 
+	/*
+	 * drm_mode_page_flip_ioctl() should already catch this, but double
+	 * check to be safe.  In the future we may enable pageflipping from
+	 * a disabled primary plane.
+	 */
+	if (WARN_ON(intel_fb_obj(old_fb) == NULL))
+		return -EBUSY;
+
 	/* Can't change pixel format via MI display flips. */
 	if (fb->pixel_format != crtc->primary->fb->pixel_format)
 		return -EINVAL;
@@ -9625,7 +9666,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	work->event = event;
 	work->crtc = crtc;
-	work->old_fb_obj = to_intel_framebuffer(old_fb)->obj;
+	work->old_fb_obj = intel_fb_obj(old_fb);
 	INIT_WORK(&work->work, intel_unpin_work_fn);
 
 	ret = drm_crtc_vblank_get(crtc);
@@ -9670,6 +9711,11 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	if (IS_VALLEYVIEW(dev)) {
 		ring = &dev_priv->ring[BCS];
+		if (obj->tiling_mode != work->old_fb_obj->tiling_mode)
+			/* vlv: DISPLAY_FLIP fails to change tiling */
+			ring = NULL;
+	} else if (IS_IVYBRIDGE(dev)) {
+		ring = &dev_priv->ring[BCS];
 	} else if (INTEL_INFO(dev)->gen >= 7) {
 		ring = obj->ring;
 		if (ring == NULL || ring->id != RCS)
@@ -10401,11 +10447,14 @@ intel_pipe_config_compare(struct drm_device *dev,
 
 	PIPE_CONF_CHECK_I(double_wide);
 
+	PIPE_CONF_CHECK_X(ddi_pll_sel);
+
 	PIPE_CONF_CHECK_I(shared_dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll_md);
 	PIPE_CONF_CHECK_X(dpll_hw_state.fp0);
 	PIPE_CONF_CHECK_X(dpll_hw_state.fp1);
+	PIPE_CONF_CHECK_X(dpll_hw_state.wrpll);
 
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5)
 		PIPE_CONF_CHECK_I(pipe_bpp);
@@ -10762,10 +10811,9 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	 * on the DPLL.
 	 */
 	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		struct drm_framebuffer *old_fb;
-		struct drm_i915_gem_object *old_obj = NULL;
-		struct drm_i915_gem_object *obj =
-			to_intel_framebuffer(fb)->obj;
+		struct drm_framebuffer *old_fb = crtc->primary->fb;
+		struct drm_i915_gem_object *old_obj = intel_fb_obj(old_fb);
+		struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 
 		mutex_lock(&dev->struct_mutex);
 		ret = intel_pin_and_fence_fb_obj(dev,
@@ -10776,11 +10824,8 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 			mutex_unlock(&dev->struct_mutex);
 			goto done;
 		}
-		old_fb = crtc->primary->fb;
-		if (old_fb) {
-			old_obj = to_intel_framebuffer(old_fb)->obj;
+		if (old_fb)
 			intel_unpin_fb_obj(old_obj);
-		}
 		i915_gem_track_fb(old_obj, obj,
 				  INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe));
 		mutex_unlock(&dev->struct_mutex);
@@ -11266,18 +11311,15 @@ static const struct drm_crtc_funcs intel_crtc_funcs = {
 	.page_flip = intel_crtc_page_flip,
 };
 
-static void intel_cpu_pll_init(struct drm_device *dev)
-{
-	if (HAS_DDI(dev))
-		intel_ddi_pll_init(dev);
-}
-
 static bool ibx_pch_dpll_get_hw_state(struct drm_i915_private *dev_priv,
 				      struct intel_shared_dpll *pll,
 				      struct intel_dpll_hw_state *hw_state)
 {
 	uint32_t val;
 
+	if (!intel_display_power_enabled(dev_priv, POWER_DOMAIN_PLLS))
+		return false;
+
 	val = I915_READ(PCH_DPLL(pll->id));
 	hw_state->dpll = val;
 	hw_state->fp0 = I915_READ(PCH_FP0(pll->id));
@@ -11359,7 +11401,9 @@ static void intel_shared_dpll_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
+	if (HAS_DDI(dev))
+		intel_ddi_pll_init(dev);
+	else if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
 		ibx_pch_dpll_init(dev);
 	else
 		dev_priv->num_shared_dpll = 0;
@@ -11398,9 +11442,11 @@ intel_primary_plane_disable(struct drm_plane *plane)
 	intel_disable_primary_hw_plane(dev_priv, intel_plane->plane,
 				       intel_plane->pipe);
 disable_unpin:
-	i915_gem_track_fb(to_intel_framebuffer(plane->fb)->obj, NULL,
+	mutex_lock(&dev->struct_mutex);
+	i915_gem_track_fb(intel_fb_obj(plane->fb), NULL,
 			  INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe));
-	intel_unpin_fb_obj(to_intel_framebuffer(plane->fb)->obj);
+	intel_unpin_fb_obj(intel_fb_obj(plane->fb));
+	mutex_unlock(&dev->struct_mutex);
 	plane->fb = NULL;
 
 	return 0;
@@ -11417,7 +11463,8 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct drm_i915_gem_object *obj, *old_obj = NULL;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct drm_i915_gem_object *old_obj = intel_fb_obj(plane->fb);
 	struct drm_rect dest = {
 		/* integer pixels */
 		.x1 = crtc_x,
@@ -11449,10 +11496,6 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (ret)
 		return ret;
 
-	if (plane->fb)
-		old_obj = to_intel_framebuffer(plane->fb)->obj;
-	obj = to_intel_framebuffer(fb)->obj;
-
 	/*
 	 * If the CRTC isn't enabled, we're just pinning the framebuffer,
 	 * updating the fb pointer, and returning without touching the
@@ -11460,6 +11503,8 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * turn on the display with all planes setup as desired.
 	 */
 	if (!crtc->enabled) {
+		mutex_lock(&dev->struct_mutex);
+
 		/*
 		 * If we already called setplane while the crtc was disabled,
 		 * we may have an fb pinned; unpin it.
@@ -11471,7 +11516,10 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 				  INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe));
 
 		/* Pin and return without programming hardware */
-		return intel_pin_and_fence_fb_obj(dev, obj, NULL);
+		ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
+		mutex_unlock(&dev->struct_mutex);
+
+		return ret;
 	}
 
 	intel_crtc_wait_for_pending_flips(crtc);
@@ -11483,14 +11531,18 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * because plane->fb still gets set and pinned.
 	 */
 	if (!visible) {
+		mutex_lock(&dev->struct_mutex);
+
 		/*
 		 * Try to pin the new fb first so that we can bail out if we
 		 * fail.
 		 */
 		if (plane->fb != fb) {
 			ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
-			if (ret)
+			if (ret) {
+				mutex_unlock(&dev->struct_mutex);
 				return ret;
+			}
 		}
 
 		i915_gem_track_fb(old_obj, obj,
@@ -11506,6 +11558,8 @@ intel_primary_plane_setplane(struct drm_plane *plane, struct drm_crtc *crtc,
 			if (plane->fb)
 				intel_unpin_fb_obj(old_obj);
 
+		mutex_unlock(&dev->struct_mutex);
+
 		return 0;
 	}
 
@@ -12159,7 +12213,7 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.crtc_mode_set = haswell_crtc_mode_set;
 		dev_priv->display.crtc_enable = haswell_crtc_enable;
 		dev_priv->display.crtc_disable = haswell_crtc_disable;
-		dev_priv->display.off = haswell_crtc_off;
+		dev_priv->display.off = ironlake_crtc_off;
 		dev_priv->display.update_primary_plane =
 			ironlake_update_primary_plane;
 	} else if (HAS_PCH_SPLIT(dev)) {
@@ -12426,6 +12480,9 @@ void intel_modeset_init_hw(struct drm_device *dev)
 {
 	intel_prepare_ddi(dev);
 
+	if (IS_VALLEYVIEW(dev))
+		vlv_update_cdclk(dev);
+
 	intel_init_clock_gating(dev);
 
 	intel_reset_dpio(dev);
@@ -12502,7 +12559,6 @@ void intel_modeset_init(struct drm_device *dev)
 	intel_init_dpio(dev);
 	intel_reset_dpio(dev);
 
-	intel_cpu_pll_init(dev);
 	intel_shared_dpll_init(dev);
 
 	/* Just disable it once at startup */
@@ -12811,10 +12867,6 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 			      crtc->active ? "enabled" : "disabled");
 	}
 
-	/* FIXME: Smash this into the new shared dpll infrastructure. */
-	if (HAS_DDI(dev))
-		intel_ddi_setup_hw_pll_state(dev);
-
 	for (i = 0; i < dev_priv->num_shared_dpll; i++) {
 		struct intel_shared_dpll *pll = &dev_priv->shared_dplls[i];
 
@@ -12828,6 +12880,9 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 
 		DRM_DEBUG_KMS("%s hw state readout: refcount %i, on %i\n",
 			      pll->name, pll->refcount, pll->on);
+
+		if (pll->refcount)
+			intel_display_power_get(dev_priv, POWER_DOMAIN_PLLS);
 	}
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
@@ -12945,7 +13000,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 void intel_modeset_gem_init(struct drm_device *dev)
 {
 	struct drm_crtc *c;
-	struct intel_framebuffer *fb;
+	struct drm_i915_gem_object *obj;
 
 	mutex_lock(&dev->struct_mutex);
 	intel_init_gt_powersave(dev);
@@ -12962,11 +13017,11 @@ void intel_modeset_gem_init(struct drm_device *dev)
 	 */
 	mutex_lock(&dev->struct_mutex);
 	for_each_crtc(dev, c) {
-		if (!c->primary->fb)
+		obj = intel_fb_obj(c->primary->fb);
+		if (obj == NULL)
 			continue;
 
-		fb = to_intel_framebuffer(c->primary->fb);
-		if (intel_pin_and_fence_fb_obj(dev, fb->obj, NULL)) {
+		if (intel_pin_and_fence_fb_obj(dev, obj, NULL)) {
 			DRM_ERROR("failed to pin boot fb on pipe %d\n",
 				  to_intel_crtc(c)->pipe);
 			drm_framebuffer_unreference(c->primary->fb);

drivers/gpu/drm/i915/intel_dp.c

@@ -745,6 +745,22 @@ intel_dp_connector_unregister(struct intel_connector *intel_connector)
 	intel_connector_unregister(intel_connector);
 }
 
+static void
+hsw_dp_set_ddi_pll_sel(struct intel_crtc_config *pipe_config, int link_bw)
+{
+	switch (link_bw) {
+	case DP_LINK_BW_1_62:
+		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_810;
+		break;
+	case DP_LINK_BW_2_7:
+		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_1350;
+		break;
+	case DP_LINK_BW_5_4:
+		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_2700;
+		break;
+	}
+}
+
 static void
 intel_dp_set_clock(struct intel_encoder *encoder,
 		   struct intel_crtc_config *pipe_config, int link_bw)
@@ -756,8 +772,6 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 	if (IS_G4X(dev)) {
 		divisor = gen4_dpll;
 		count = ARRAY_SIZE(gen4_dpll);
-	} else if (IS_HASWELL(dev)) {
-		/* Haswell has special-purpose DP DDI clocks. */
 	} else if (HAS_PCH_SPLIT(dev)) {
 		divisor = pch_dpll;
 		count = ARRAY_SIZE(pch_dpll);
@@ -928,7 +942,10 @@ found:
 				&pipe_config->dp_m2_n2);
 	}
 
-	intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
+	if (HAS_DDI(dev))
+		hsw_dp_set_ddi_pll_sel(pipe_config, intel_dp->link_bw);
+	else
+		intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
 
 	return true;
 }
@@ -1316,8 +1333,6 @@ void intel_edp_panel_off(struct intel_dp *intel_dp)
 
 	DRM_DEBUG_KMS("Turn eDP power off\n");
 
-	edp_wait_backlight_off(intel_dp);
-
 	WARN(!intel_dp->want_panel_vdd, "Need VDD to turn off panel\n");
 
 	pp = ironlake_get_pp_control(intel_dp);
@@ -1353,6 +1368,9 @@ void intel_edp_backlight_on(struct intel_dp *intel_dp)
 		return;
 
 	DRM_DEBUG_KMS("\n");
+
+	intel_panel_enable_backlight(intel_dp->attached_connector);
+
 	/*
 	 * If we enable the backlight right away following a panel power
 	 * on, we may see slight flicker as the panel syncs with the eDP
@@ -1367,8 +1385,6 @@ void intel_edp_backlight_on(struct intel_dp *intel_dp)
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
-
-	intel_panel_enable_backlight(intel_dp->attached_connector);
 }
 
 void intel_edp_backlight_off(struct intel_dp *intel_dp)
@@ -1381,8 +1397,6 @@ void intel_edp_backlight_off(struct intel_dp *intel_dp)
 	if (!is_edp(intel_dp))
 		return;
 
-	intel_panel_disable_backlight(intel_dp->attached_connector);
-
 	DRM_DEBUG_KMS("\n");
 	pp = ironlake_get_pp_control(intel_dp);
 	pp &= ~EDP_BLC_ENABLE;
@@ -1392,6 +1406,10 @@ void intel_edp_backlight_off(struct intel_dp *intel_dp)
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
 	intel_dp->last_backlight_off = jiffies;
+
+	edp_wait_backlight_off(intel_dp);
+
+	intel_panel_disable_backlight(intel_dp->attached_connector);
 }
 
 static void ironlake_edp_pll_on(struct intel_dp *intel_dp)
@@ -1751,7 +1769,7 @@ static bool intel_edp_psr_match_conditions(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = dig_port->base.base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_i915_gem_object *obj = to_intel_framebuffer(crtc->primary->fb)->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(crtc->primary->fb);
 	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
 
 	dev_priv->psr.source_ok = false;
@@ -1784,7 +1802,6 @@ static bool intel_edp_psr_match_conditions(struct intel_dp *intel_dp)
 		return false;
 	}
 
-	obj = to_intel_framebuffer(crtc->primary->fb)->obj;
 	if (obj->tiling_mode != I915_TILING_X ||
 	    obj->fence_reg == I915_FENCE_REG_NONE) {
 		DRM_DEBUG_KMS("PSR condition failed: fb not tiled or fenced\n");
@@ -3815,6 +3832,22 @@ intel_dp_hot_plug(struct intel_encoder *intel_encoder)
 	intel_dp_check_link_status(intel_dp);
 }
 
+bool
+intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
+{
+	struct intel_dp *intel_dp = &intel_dig_port->dp;
+
+	if (long_hpd)
+		return true;
+
+	/*
+	 * we'll check the link status via the normal hot plug path later -
+	 * but for short hpds we should check it now
+	 */
+	intel_dp_check_link_status(intel_dp);
+	return false;
+}
+
 /* Return which DP Port should be selected for Transcoder DP control */
 int
 intel_trans_dp_port_sel(struct drm_crtc *crtc)
@@ -4387,6 +4420,7 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 void
 intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_digital_port *intel_dig_port;
 	struct intel_encoder *intel_encoder;
 	struct drm_encoder *encoder;
@@ -4443,6 +4477,9 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 	intel_encoder->cloneable = 0;
 	intel_encoder->hot_plug = intel_dp_hot_plug;
 
+	intel_dig_port->hpd_pulse = intel_dp_hpd_pulse;
+	dev_priv->hpd_irq_port[port] = intel_dig_port;
+
 	if (!intel_dp_init_connector(intel_dig_port, intel_connector)) {
 		drm_encoder_cleanup(encoder);
 		kfree(intel_dig_port);

drivers/gpu/drm/i915/intel_drv.h

@@ -307,6 +307,9 @@ struct intel_crtc_config {
 	/* Selected dpll when shared or DPLL_ID_PRIVATE. */
 	enum intel_dpll_id shared_dpll;
 
+	/* PORT_CLK_SEL for DDI ports. */
+	uint32_t ddi_pll_sel;
+
 	/* Actual register state of the dpll, for shared dpll cross-checking. */
 	struct intel_dpll_hw_state dpll_hw_state;
 
@@ -338,6 +341,7 @@ struct intel_crtc_config {
 		u32 pos;
 		u32 size;
 		bool enabled;
+		bool force_thru;
 	} pch_pfit;
 
 	/* FDI configuration, only valid if has_pch_encoder is set. */
@@ -398,8 +402,6 @@ struct intel_crtc {
 	struct intel_crtc_config *new_config;
 	bool new_enabled;
 
-	uint32_t ddi_pll_sel;
-
 	/* reset counter value when the last flip was submitted */
 	unsigned int reset_counter;
 
@@ -485,6 +487,7 @@ struct cxsr_latency {
 #define to_intel_encoder(x) container_of(x, struct intel_encoder, base)
 #define to_intel_framebuffer(x) container_of(x, struct intel_framebuffer, base)
 #define to_intel_plane(x) container_of(x, struct intel_plane, base)
+#define intel_fb_obj(x) (x ? to_intel_framebuffer(x)->obj : NULL)
 
 struct intel_hdmi {
 	u32 hdmi_reg;
@@ -567,6 +570,7 @@ struct intel_digital_port {
 	u32 saved_port_bits;
 	struct intel_dp dp;
 	struct intel_hdmi hdmi;
+	bool (*hpd_pulse)(struct intel_digital_port *, bool);
 };
 
 static inline int
@@ -706,10 +710,7 @@ void intel_ddi_disable_transcoder_func(struct drm_i915_private *dev_priv,
 				       enum transcoder cpu_transcoder);
 void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc);
 void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc);
-void intel_ddi_setup_hw_pll_state(struct drm_device *dev);
 bool intel_ddi_pll_select(struct intel_crtc *crtc);
-void intel_ddi_pll_enable(struct intel_crtc *crtc);
-void intel_ddi_put_crtc_pll(struct drm_crtc *crtc);
 void intel_ddi_set_pipe_settings(struct drm_crtc *crtc);
 void intel_ddi_prepare_link_retrain(struct drm_encoder *encoder);
 bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
@@ -722,7 +723,6 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 const char *intel_output_name(int output);
 bool intel_has_pending_fb_unpin(struct drm_device *dev);
 int intel_pch_rawclk(struct drm_device *dev);
-int valleyview_cur_cdclk(struct drm_i915_private *dev_priv);
 void intel_mark_busy(struct drm_device *dev);
 void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
 			     struct intel_engine_cs *ring);
@@ -793,12 +793,18 @@ __intel_framebuffer_create(struct drm_device *dev,
 void intel_prepare_page_flip(struct drm_device *dev, int plane);
 void intel_finish_page_flip(struct drm_device *dev, int pipe);
 void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
+
+/* shared dpll functions */
 struct intel_shared_dpll *intel_crtc_to_shared_dpll(struct intel_crtc *crtc);
 void assert_shared_dpll(struct drm_i915_private *dev_priv,
 			struct intel_shared_dpll *pll,
 			bool state);
 #define assert_shared_dpll_enabled(d, p) assert_shared_dpll(d, p, true)
 #define assert_shared_dpll_disabled(d, p) assert_shared_dpll(d, p, false)
+struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc);
+void intel_put_shared_dpll(struct intel_crtc *crtc);
+
+/* modesetting asserts */
 void assert_pll(struct drm_i915_private *dev_priv,
 		enum pipe pipe, bool state);
 #define assert_pll_enabled(d, p) assert_pll(d, p, true)
@@ -831,7 +837,6 @@ void hsw_disable_ips(struct intel_crtc *crtc);
 void intel_display_set_init_power(struct drm_i915_private *dev, bool enable);
 enum intel_display_power_domain
 intel_display_port_power_domain(struct intel_encoder *intel_encoder);
-int valleyview_get_vco(struct drm_i915_private *dev_priv);
 void intel_mode_from_pipe_config(struct drm_display_mode *mode,
 				 struct intel_crtc_config *pipe_config);
 int intel_format_to_fourcc(int format);
@@ -852,6 +857,8 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc);
 bool intel_dp_compute_config(struct intel_encoder *encoder,
 			     struct intel_crtc_config *pipe_config);
 bool intel_dp_is_edp(struct drm_device *dev, enum port port);
+bool intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port,
+			bool long_hpd);
 void intel_edp_backlight_on(struct intel_dp *intel_dp);
 void intel_edp_backlight_off(struct intel_dp *intel_dp);
 void intel_edp_panel_vdd_on(struct intel_dp *intel_dp);
@@ -863,7 +870,6 @@ void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate);
 void intel_edp_psr_exit(struct drm_device *dev);
 void intel_edp_psr_init(struct drm_device *dev);
 
-
 /* intel_dsi.c */
 void intel_dsi_init(struct drm_device *dev);
 
@@ -1005,8 +1011,7 @@ void intel_runtime_pm_put(struct drm_i915_private *dev_priv);
 void intel_init_runtime_pm(struct drm_i915_private *dev_priv);
 void intel_fini_runtime_pm(struct drm_i915_private *dev_priv);
 void ilk_wm_get_hw_state(struct drm_device *dev);
-void __vlv_set_power_well(struct drm_i915_private *dev_priv,
-			  enum punit_power_well power_well_id, bool enable);
+
 
 /* intel_sdvo.c */
 bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);

drivers/gpu/drm/i915/intel_fbdev.c

@@ -107,7 +107,7 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 							  sizes->surface_depth);
 
 	size = mode_cmd.pitches[0] * mode_cmd.height;
-	size = ALIGN(size, PAGE_SIZE);
+	size = PAGE_ALIGN(size);
 	obj = i915_gem_object_create_stolen(dev, size);
 	if (obj == NULL)
 		obj = i915_gem_alloc_object(dev, size);

drivers/gpu/drm/i915/intel_i2c.c

@@ -34,11 +34,6 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
-enum disp_clk {
-	CDCLK,
-	CZCLK
-};
-
 struct gmbus_port {
 	const char *name;
 	int reg;
@@ -63,60 +58,11 @@ to_intel_gmbus(struct i2c_adapter *i2c)
 	return container_of(i2c, struct intel_gmbus, adapter);
 }
 
-static int get_disp_clk_div(struct drm_i915_private *dev_priv,
-			    enum disp_clk clk)
-{
-	u32 reg_val;
-	int clk_ratio;
-
-	reg_val = I915_READ(CZCLK_CDCLK_FREQ_RATIO);
-
-	if (clk == CDCLK)
-		clk_ratio =
-			((reg_val & CDCLK_FREQ_MASK) >> CDCLK_FREQ_SHIFT) + 1;
-	else
-		clk_ratio = (reg_val & CZCLK_FREQ_MASK) + 1;
-
-	return clk_ratio;
-}
-
-static void gmbus_set_freq(struct drm_i915_private *dev_priv)
-{
-	int vco, gmbus_freq = 0, cdclk_div;
-
-	BUG_ON(!IS_VALLEYVIEW(dev_priv->dev));
-
-	vco = valleyview_get_vco(dev_priv);
-
-	/* Get the CDCLK divide ratio */
-	cdclk_div = get_disp_clk_div(dev_priv, CDCLK);
-
-	/*
-	 * Program the gmbus_freq based on the cdclk frequency.
-	 * BSpec erroneously claims we should aim for 4MHz, but
-	 * in fact 1MHz is the correct frequency.
-	 */
-	if (cdclk_div)
-		gmbus_freq = (vco << 1) / cdclk_div;
-
-	if (WARN_ON(gmbus_freq == 0))
-		return;
-
-	I915_WRITE(GMBUSFREQ_VLV, gmbus_freq);
-}
-
 void
 intel_i2c_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	/*
-	 * In BIOS-less system, program the correct gmbus frequency
-	 * before reading edid.
-	 */
-	if (IS_VALLEYVIEW(dev))
-		gmbus_set_freq(dev_priv);
-
 	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS0, 0);
 	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS4, 0);
 }

drivers/gpu/drm/i915/intel_lvds.c

@@ -51,6 +51,7 @@ struct intel_lvds_encoder {
 
 	bool is_dual_link;
 	u32 reg;
+	u32 a3_power;
 
 	struct intel_lvds_connector *attached_connector;
 };
@@ -71,8 +72,13 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
+	enum intel_display_power_domain power_domain;
 	u32 tmp;
 
+	power_domain = intel_display_port_power_domain(encoder);
+	if (!intel_display_power_enabled(dev_priv, power_domain))
+		return false;
+
 	tmp = I915_READ(lvds_encoder->reg);
 
 	if (!(tmp & LVDS_PORT_EN))
@@ -165,8 +171,11 @@ static void intel_pre_enable_lvds(struct intel_encoder *encoder)
 
 	/* It would be nice to set 24 vs 18-bit mode (LVDS_A3_POWER_UP)
 	 * appropriately here, but we need to look more thoroughly into how
-	 * panels behave in the two modes.
+	 * panels behave in the two modes. For now, let's just maintain the
+	 * value we got from the BIOS.
 	 */
+	 temp &= ~LVDS_A3_POWER_MASK;
+	 temp |= lvds_encoder->a3_power;
 
 	/* Set the dithering flag on LVDS as needed, note that there is no
 	 * special lvds dither control bit on pch-split platforms, dithering is
@@ -264,7 +273,6 @@ static bool intel_lvds_compute_config(struct intel_encoder *intel_encoder,
 				      struct intel_crtc_config *pipe_config)
 {
 	struct drm_device *dev = intel_encoder->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_lvds_encoder *lvds_encoder =
 		to_lvds_encoder(&intel_encoder->base);
 	struct intel_connector *intel_connector =
@@ -279,8 +287,7 @@ static bool intel_lvds_compute_config(struct intel_encoder *intel_encoder,
 		return false;
 	}
 
-	if ((I915_READ(lvds_encoder->reg) & LVDS_A3_POWER_MASK) ==
-	    LVDS_A3_POWER_UP)
+	if (lvds_encoder->a3_power == LVDS_A3_POWER_UP)
 		lvds_bpp = 8*3;
 	else
 		lvds_bpp = 6*3;
@@ -1081,6 +1088,9 @@ out:
 	DRM_DEBUG_KMS("detected %s-link lvds configuration\n",
 		      lvds_encoder->is_dual_link ? "dual" : "single");
 
+	lvds_encoder->a3_power = I915_READ(lvds_encoder->reg) &
+				 LVDS_A3_POWER_MASK;
+
 	/*
 	 * Unlock registers and just
 	 * leave them unlocked

drivers/gpu/drm/i915/intel_pm.c

@@ -93,8 +93,7 @@ static void i8xx_enable_fbc(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *fb = crtc->primary->fb;
-	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
-	struct drm_i915_gem_object *obj = intel_fb->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int cfb_pitch;
 	int i;
@@ -150,8 +149,7 @@ static void g4x_enable_fbc(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *fb = crtc->primary->fb;
-	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
-	struct drm_i915_gem_object *obj = intel_fb->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 dpfc_ctl;
 
@@ -222,16 +220,26 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *fb = crtc->primary->fb;
-	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
-	struct drm_i915_gem_object *obj = intel_fb->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 dpfc_ctl;
 
 	dpfc_ctl = DPFC_CTL_PLANE(intel_crtc->plane);
 	if (drm_format_plane_cpp(fb->pixel_format, 0) == 2)
+		dev_priv->fbc.threshold++;
+
+	switch (dev_priv->fbc.threshold) {
+	case 4:
+	case 3:
+		dpfc_ctl |= DPFC_CTL_LIMIT_4X;
+		break;
+	case 2:
 		dpfc_ctl |= DPFC_CTL_LIMIT_2X;
-	else
+		break;
+	case 1:
 		dpfc_ctl |= DPFC_CTL_LIMIT_1X;
+		break;
+	}
 	dpfc_ctl |= DPFC_CTL_FENCE_EN;
 	if (IS_GEN5(dev))
 		dpfc_ctl |= obj->fence_reg;
@@ -278,16 +286,27 @@ static void gen7_enable_fbc(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_framebuffer *fb = crtc->primary->fb;
-	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
-	struct drm_i915_gem_object *obj = intel_fb->obj;
+	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 dpfc_ctl;
 
 	dpfc_ctl = IVB_DPFC_CTL_PLANE(intel_crtc->plane);
 	if (drm_format_plane_cpp(fb->pixel_format, 0) == 2)
+		dev_priv->fbc.threshold++;
+
+	switch (dev_priv->fbc.threshold) {
+	case 4:
+	case 3:
+		dpfc_ctl |= DPFC_CTL_LIMIT_4X;
+		break;
+	case 2:
 		dpfc_ctl |= DPFC_CTL_LIMIT_2X;
-	else
+		break;
+	case 1:
 		dpfc_ctl |= DPFC_CTL_LIMIT_1X;
+		break;
+	}
+
 	dpfc_ctl |= IVB_DPFC_CTL_FENCE_EN;
 
 	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
@@ -462,7 +481,6 @@ void intel_update_fbc(struct drm_device *dev)
 	struct drm_crtc *crtc = NULL, *tmp_crtc;
 	struct intel_crtc *intel_crtc;
 	struct drm_framebuffer *fb;
-	struct intel_framebuffer *intel_fb;
 	struct drm_i915_gem_object *obj;
 	const struct drm_display_mode *adjusted_mode;
 	unsigned int max_width, max_height;
@@ -507,8 +525,7 @@ void intel_update_fbc(struct drm_device *dev)
 
 	intel_crtc = to_intel_crtc(crtc);
 	fb = crtc->primary->fb;
-	intel_fb = to_intel_framebuffer(fb);
-	obj = intel_fb->obj;
+	obj = intel_fb_obj(fb);
 	adjusted_mode = &intel_crtc->config.adjusted_mode;
 
 	if (i915.enable_fbc < 0) {
@@ -566,7 +583,8 @@ void intel_update_fbc(struct drm_device *dev)
 	if (in_dbg_master())
 		goto out_disable;
 
-	if (i915_gem_stolen_setup_compression(dev, intel_fb->obj->base.size)) {
+	if (i915_gem_stolen_setup_compression(dev, obj->base.size,
+					      drm_format_plane_cpp(fb->pixel_format, 0))) {
 		if (set_no_fbc_reason(dev_priv, FBC_STOLEN_TOO_SMALL))
 			DRM_DEBUG_KMS("framebuffer too large, disabling compression\n");
 		goto out_disable;
@@ -792,12 +810,33 @@ static const struct cxsr_latency *intel_get_cxsr_latency(int is_desktop,
 	return NULL;
 }
 
-static void pineview_disable_cxsr(struct drm_device *dev)
+void intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enable)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_device *dev = dev_priv->dev;
+	u32 val;
 
-	/* deactivate cxsr */
-	I915_WRITE(DSPFW3, I915_READ(DSPFW3) & ~PINEVIEW_SELF_REFRESH_EN);
+	if (IS_VALLEYVIEW(dev)) {
+		I915_WRITE(FW_BLC_SELF_VLV, enable ? FW_CSPWRDWNEN : 0);
+	} else if (IS_G4X(dev) || IS_CRESTLINE(dev)) {
+		I915_WRITE(FW_BLC_SELF, enable ? FW_BLC_SELF_EN : 0);
+	} else if (IS_PINEVIEW(dev)) {
+		val = I915_READ(DSPFW3) & ~PINEVIEW_SELF_REFRESH_EN;
+		val |= enable ? PINEVIEW_SELF_REFRESH_EN : 0;
+		I915_WRITE(DSPFW3, val);
+	} else if (IS_I945G(dev) || IS_I945GM(dev)) {
+		val = enable ? _MASKED_BIT_ENABLE(FW_BLC_SELF_EN) :
+			       _MASKED_BIT_DISABLE(FW_BLC_SELF_EN);
+		I915_WRITE(FW_BLC_SELF, val);
+	} else if (IS_I915GM(dev)) {
+		val = enable ? _MASKED_BIT_ENABLE(INSTPM_SELF_EN) :
+			       _MASKED_BIT_DISABLE(INSTPM_SELF_EN);
+		I915_WRITE(INSTPM, val);
+	} else {
+		return;
+	}
+
+	DRM_DEBUG_KMS("memory self-refresh is %s\n",
+		      enable ? "enabled" : "disabled");
 }
 
 /*
@@ -1036,7 +1075,7 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 					 dev_priv->fsb_freq, dev_priv->mem_freq);
 	if (!latency) {
 		DRM_DEBUG_KMS("Unknown FSB/MEM found, disable CxSR\n");
-		pineview_disable_cxsr(dev);
+		intel_set_memory_cxsr(dev_priv, false);
 		return;
 	}
 
@@ -1087,13 +1126,9 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 		I915_WRITE(DSPFW3, reg);
 		DRM_DEBUG_KMS("DSPFW3 register is %x\n", reg);
 
-		/* activate cxsr */
-		I915_WRITE(DSPFW3,
-			   I915_READ(DSPFW3) | PINEVIEW_SELF_REFRESH_EN);
-		DRM_DEBUG_KMS("Self-refresh is enabled\n");
+		intel_set_memory_cxsr(dev_priv, true);
 	} else {
-		pineview_disable_cxsr(dev);
-		DRM_DEBUG_KMS("Self-refresh is disabled\n");
+		intel_set_memory_cxsr(dev_priv, false);
 	}
 }
 
@@ -1319,6 +1354,7 @@ static void valleyview_update_wm(struct drm_crtc *crtc)
 	int plane_sr, cursor_sr;
 	int ignore_plane_sr, ignore_cursor_sr;
 	unsigned int enabled = 0;
+	bool cxsr_enabled;
 
 	vlv_update_drain_latency(dev);
 
@@ -1345,10 +1381,10 @@ static void valleyview_update_wm(struct drm_crtc *crtc)
 			     &valleyview_wm_info,
 			     &valleyview_cursor_wm_info,
 			     &ignore_plane_sr, &cursor_sr)) {
-		I915_WRITE(FW_BLC_SELF_VLV, FW_CSPWRDWNEN);
+		cxsr_enabled = true;
 	} else {
-		I915_WRITE(FW_BLC_SELF_VLV,
-			   I915_READ(FW_BLC_SELF_VLV) & ~FW_CSPWRDWNEN);
+		cxsr_enabled = false;
+		intel_set_memory_cxsr(dev_priv, false);
 		plane_sr = cursor_sr = 0;
 	}
 
@@ -1368,6 +1404,9 @@ static void valleyview_update_wm(struct drm_crtc *crtc)
 	I915_WRITE(DSPFW3,
 		   (I915_READ(DSPFW3) & ~DSPFW_CURSOR_SR_MASK) |
 		   (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
+
+	if (cxsr_enabled)
+		intel_set_memory_cxsr(dev_priv, true);
 }
 
 static void g4x_update_wm(struct drm_crtc *crtc)
@@ -1378,6 +1417,7 @@ static void g4x_update_wm(struct drm_crtc *crtc)
 	int planea_wm, planeb_wm, cursora_wm, cursorb_wm;
 	int plane_sr, cursor_sr;
 	unsigned int enabled = 0;
+	bool cxsr_enabled;
 
 	if (g4x_compute_wm0(dev, PIPE_A,
 			    &g4x_wm_info, latency_ns,
@@ -1397,10 +1437,10 @@ static void g4x_update_wm(struct drm_crtc *crtc)
 			     &g4x_wm_info,
 			     &g4x_cursor_wm_info,
 			     &plane_sr, &cursor_sr)) {
-		I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN);
+		cxsr_enabled = true;
 	} else {
-		I915_WRITE(FW_BLC_SELF,
-			   I915_READ(FW_BLC_SELF) & ~FW_BLC_SELF_EN);
+		cxsr_enabled = false;
+		intel_set_memory_cxsr(dev_priv, false);
 		plane_sr = cursor_sr = 0;
 	}
 
@@ -1421,6 +1461,9 @@ static void g4x_update_wm(struct drm_crtc *crtc)
 	I915_WRITE(DSPFW3,
 		   (I915_READ(DSPFW3) & ~(DSPFW_HPLL_SR_EN | DSPFW_CURSOR_SR_MASK)) |
 		   (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
+
+	if (cxsr_enabled)
+		intel_set_memory_cxsr(dev_priv, true);
 }
 
 static void i965_update_wm(struct drm_crtc *unused_crtc)
@@ -1430,6 +1473,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 	struct drm_crtc *crtc;
 	int srwm = 1;
 	int cursor_sr = 16;
+	bool cxsr_enabled;
 
 	/* Calc sr entries for one plane configs */
 	crtc = single_enabled_crtc(dev);
@@ -1471,13 +1515,11 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 		DRM_DEBUG_KMS("self-refresh watermark: display plane %d "
 			      "cursor %d\n", srwm, cursor_sr);
 
-		if (IS_CRESTLINE(dev))
-			I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN);
+		cxsr_enabled = true;
 	} else {
+		cxsr_enabled = false;
 		/* Turn off self refresh if both pipes are enabled */
-		if (IS_CRESTLINE(dev))
-			I915_WRITE(FW_BLC_SELF, I915_READ(FW_BLC_SELF)
-				   & ~FW_BLC_SELF_EN);
+		intel_set_memory_cxsr(dev_priv, false);
 	}
 
 	DRM_DEBUG_KMS("Setting FIFO watermarks - A: 8, B: 8, C: 8, SR %d\n",
@@ -1489,6 +1531,9 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 	I915_WRITE(DSPFW2, (8 << 8) | (8 << 0));
 	/* update cursor SR watermark */
 	I915_WRITE(DSPFW3, (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
+
+	if (cxsr_enabled)
+		intel_set_memory_cxsr(dev_priv, true);
 }
 
 static void i9xx_update_wm(struct drm_crtc *unused_crtc)
@@ -1548,12 +1593,12 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	DRM_DEBUG_KMS("FIFO watermarks - A: %d, B: %d\n", planea_wm, planeb_wm);
 
 	if (IS_I915GM(dev) && enabled) {
-		struct intel_framebuffer *fb;
+		struct drm_i915_gem_object *obj;
 
-		fb = to_intel_framebuffer(enabled->primary->fb);
+		obj = intel_fb_obj(enabled->primary->fb);
 
 		/* self-refresh seems busted with untiled */
-		if (fb->obj->tiling_mode == I915_TILING_NONE)
+		if (obj->tiling_mode == I915_TILING_NONE)
 			enabled = NULL;
 	}
 
@@ -1563,10 +1608,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	cwm = 2;
 
 	/* Play safe and disable self-refresh before adjusting watermarks. */
-	if (IS_I945G(dev) || IS_I945GM(dev))
-		I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN_MASK | 0);
-	else if (IS_I915GM(dev))
-		I915_WRITE(INSTPM, _MASKED_BIT_DISABLE(INSTPM_SELF_EN));
+	intel_set_memory_cxsr(dev_priv, false);
 
 	/* Calc sr entries for one plane configs */
 	if (HAS_FW_BLC(dev) && enabled) {
@@ -1612,17 +1654,8 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	I915_WRITE(FW_BLC, fwater_lo);
 	I915_WRITE(FW_BLC2, fwater_hi);
 
-	if (HAS_FW_BLC(dev)) {
-		if (enabled) {
-			if (IS_I945G(dev) || IS_I945GM(dev))
-				I915_WRITE(FW_BLC_SELF,
-					   FW_BLC_SELF_EN_MASK | FW_BLC_SELF_EN);
-			else if (IS_I915GM(dev))
-				I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_SELF_EN));
-			DRM_DEBUG_KMS("memory self refresh enabled\n");
-		} else
-			DRM_DEBUG_KMS("memory self refresh disabled\n");
-	}
+	if (enabled)
+		intel_set_memory_cxsr(dev_priv, true);
 }
 
 static void i845_update_wm(struct drm_crtc *unused_crtc)
@@ -3150,6 +3183,9 @@ static u32 gen6_rps_pm_mask(struct drm_i915_private *dev_priv, u8 val)
 	if (val < dev_priv->rps.max_freq_softlimit)
 		mask |= GEN6_PM_RP_UP_THRESHOLD;
 
+	mask |= dev_priv->pm_rps_events & (GEN6_PM_RP_DOWN_EI_EXPIRED | GEN6_PM_RP_UP_EI_EXPIRED);
+	mask &= dev_priv->pm_rps_events;
+
 	/* IVB and SNB hard hangs on looping batchbuffer
 	 * if GEN6_PM_UP_EI_EXPIRED is masked.
 	 */
@@ -3493,15 +3529,23 @@ static void gen8_enable_rps(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, unused)
 		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
 	I915_WRITE(GEN6_RC_SLEEP, 0);
-	I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
+	if (IS_BROADWELL(dev))
+		I915_WRITE(GEN6_RC6_THRESHOLD, 625); /* 800us/1.28 for TO */
+	else
+		I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
 
 	/* 3: Enable RC6 */
 	if (intel_enable_rc6(dev) & INTEL_RC6_ENABLE)
 		rc6_mask = GEN6_RC_CTL_RC6_ENABLE;
 	intel_print_rc6_info(dev, rc6_mask);
-	I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
-				    GEN6_RC_CTL_EI_MODE(1) |
-				    rc6_mask);
+	if (IS_BROADWELL(dev))
+		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
+				GEN7_RC_CTL_TO_MODE |
+				rc6_mask);
+	else
+		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
+				GEN6_RC_CTL_EI_MODE(1) |
+				rc6_mask);
 
 	/* 4 Program defaults and thresholds for RPS*/
 	I915_WRITE(GEN6_RPNSWREQ,
@@ -4078,6 +4122,7 @@ static void valleyview_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RP_DOWN_EI, 350000);
 
 	I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
+	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 0xf4240);
 
 	I915_WRITE(GEN6_RP_CONTROL,
 		   GEN6_RP_MEDIA_TURBO |
@@ -4098,9 +4143,11 @@ static void valleyview_enable_rps(struct drm_device *dev)
 
 	/* allows RC6 residency counter to work */
 	I915_WRITE(VLV_COUNTER_CONTROL,
-		   _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
+		   _MASKED_BIT_ENABLE(VLV_MEDIA_RC0_COUNT_EN |
+				      VLV_RENDER_RC0_COUNT_EN |
 				      VLV_MEDIA_RC6_COUNT_EN |
 				      VLV_RENDER_RC6_COUNT_EN));
+
 	if (intel_enable_rc6(dev) & INTEL_RC6_ENABLE)
 		rc6_mode = GEN7_RC_CTL_TO_MODE | VLV_RC_CTL_CTX_RST_PARALLEL;
 
@@ -5328,7 +5375,7 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GAMTARBMODE, _MASKED_BIT_ENABLE(ARB_MODE_BWGTLB_DISABLE));
 
 	I915_WRITE(_3D_CHICKEN3,
-		   _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(2));
+		   _MASKED_BIT_ENABLE(_3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(2)));
 
 	I915_WRITE(COMMON_SLICE_CHICKEN2,
 		   _MASKED_BIT_ENABLE(GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE));
@@ -5563,10 +5610,6 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	}
 	DRM_DEBUG_DRIVER("DDR speed: %d MHz", dev_priv->mem_freq);
 
-	dev_priv->vlv_cdclk_freq = valleyview_cur_cdclk(dev_priv);
-	DRM_DEBUG_DRIVER("Current CD clock rate: %d MHz",
-			 dev_priv->vlv_cdclk_freq);
-
 	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
 
 	/* WaDisableEarlyCull:vlv */
@@ -5982,34 +6025,13 @@ static bool i9xx_always_on_power_well_enabled(struct drm_i915_private *dev_priv,
 	return true;
 }
 
-void __vlv_set_power_well(struct drm_i915_private *dev_priv,
-			  enum punit_power_well power_well_id, bool enable)
+static void vlv_set_power_well(struct drm_i915_private *dev_priv,
+			       struct i915_power_well *power_well, bool enable)
 {
-	struct drm_device *dev = dev_priv->dev;
+	enum punit_power_well power_well_id = power_well->data;
 	u32 mask;
 	u32 state;
 	u32 ctrl;
-	enum pipe pipe;
-
-	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC) {
-		if (enable) {
-			/*
-			 * Enable the CRI clock source so we can get at the
-			 * display and the reference clock for VGA
-			 * hotplug / manual detection.
-			 */
-			I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
-				   DPLL_REFA_CLK_ENABLE_VLV |
-				   DPLL_INTEGRATED_CRI_CLK_VLV);
-			udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
-		} else {
-			for_each_pipe(pipe)
-				assert_pll_disabled(dev_priv, pipe);
-			/* Assert common reset */
-			I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) &
-				   ~DPIO_CMNRST);
-		}
-	}
 
 	mask = PUNIT_PWRGT_MASK(power_well_id);
 	state = enable ? PUNIT_PWRGT_PWR_ON(power_well_id) :
@@ -6037,28 +6059,6 @@ void __vlv_set_power_well(struct drm_i915_private *dev_priv,
 
 out:
 	mutex_unlock(&dev_priv->rps.hw_lock);
-
-	/*
-	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
-	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
-	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
-	 *   b.	The other bits such as sfr settings / modesel may all
-	 *	be set to 0.
-	 *
-	 * This should only be done on init and resume from S3 with
-	 * both PLLs disabled, or we risk losing DPIO and PLL
-	 * synchronization.
-	 */
-	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC && enable)
-		I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
-}
-
-static void vlv_set_power_well(struct drm_i915_private *dev_priv,
-			       struct i915_power_well *power_well, bool enable)
-{
-	enum punit_power_well power_well_id = power_well->data;
-
-	__vlv_set_power_well(dev_priv, power_well_id, enable);
 }
 
 static void vlv_power_well_sync_hw(struct drm_i915_private *dev_priv,
@@ -6150,6 +6150,53 @@ static void vlv_display_power_well_disable(struct drm_i915_private *dev_priv,
 	vlv_set_power_well(dev_priv, power_well, false);
 }
 
+static void vlv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
+					   struct i915_power_well *power_well)
+{
+	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DPIO_CMN_BC);
+
+	/*
+	 * Enable the CRI clock source so we can get at the
+	 * display and the reference clock for VGA
+	 * hotplug / manual detection.
+	 */
+	I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
+		   DPLL_REFA_CLK_ENABLE_VLV | DPLL_INTEGRATED_CRI_CLK_VLV);
+	udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
+
+	vlv_set_power_well(dev_priv, power_well, true);
+
+	/*
+	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
+	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
+	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
+	 *   b.	The other bits such as sfr settings / modesel may all
+	 *	be set to 0.
+	 *
+	 * This should only be done on init and resume from S3 with
+	 * both PLLs disabled, or we risk losing DPIO and PLL
+	 * synchronization.
+	 */
+	I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+}
+
+static void vlv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
+					    struct i915_power_well *power_well)
+{
+	struct drm_device *dev = dev_priv->dev;
+	enum pipe pipe;
+
+	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DPIO_CMN_BC);
+
+	for_each_pipe(pipe)
+		assert_pll_disabled(dev_priv, pipe);
+
+	/* Assert common reset */
+	I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) & ~DPIO_CMNRST);
+
+	vlv_set_power_well(dev_priv, power_well, false);
+}
+
 static void check_power_well_state(struct drm_i915_private *dev_priv,
 				   struct i915_power_well *power_well)
 {
@@ -6299,6 +6346,7 @@ EXPORT_SYMBOL_GPL(i915_get_cdclk_freq);
 	BIT(POWER_DOMAIN_PORT_DDI_D_2_LANES) |		\
 	BIT(POWER_DOMAIN_PORT_DDI_D_4_LANES) |		\
 	BIT(POWER_DOMAIN_PORT_CRT) |			\
+	BIT(POWER_DOMAIN_PLLS) |			\
 	BIT(POWER_DOMAIN_INIT))
 #define HSW_DISPLAY_POWER_DOMAINS (				\
 	(POWER_DOMAIN_MASK & ~HSW_ALWAYS_ON_POWER_DOMAINS) |	\
@@ -6398,6 +6446,13 @@ static const struct i915_power_well_ops vlv_display_power_well_ops = {
 	.is_enabled = vlv_power_well_enabled,
 };
 
+static const struct i915_power_well_ops vlv_dpio_cmn_power_well_ops = {
+	.sync_hw = vlv_power_well_sync_hw,
+	.enable = vlv_dpio_cmn_power_well_enable,
+	.disable = vlv_dpio_cmn_power_well_disable,
+	.is_enabled = vlv_power_well_enabled,
+};
+
 static const struct i915_power_well_ops vlv_dpio_power_well_ops = {
 	.sync_hw = vlv_power_well_sync_hw,
 	.enable = vlv_power_well_enable,
@@ -6458,10 +6513,25 @@ static struct i915_power_well vlv_power_wells[] = {
 		.name = "dpio-common",
 		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
 		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
-		.ops = &vlv_dpio_power_well_ops,
+		.ops = &vlv_dpio_cmn_power_well_ops,
 	},
 };
 
+static struct i915_power_well *lookup_power_well(struct drm_i915_private *dev_priv,
+						 enum punit_power_well power_well_id)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *power_well;
+	int i;
+
+	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains) {
+		if (power_well->data == power_well_id)
+			return power_well;
+	}
+
+	return NULL;
+}
+
 #define set_power_wells(power_domains, __power_wells) ({		\
 	(power_domains)->power_wells = (__power_wells);			\
 	(power_domains)->power_well_count = ARRAY_SIZE(__power_wells);	\
@@ -6512,11 +6582,50 @@ static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
 	mutex_unlock(&power_domains->lock);
 }
 
+static void vlv_cmnlane_wa(struct drm_i915_private *dev_priv)
+{
+	struct i915_power_well *cmn =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC);
+	struct i915_power_well *disp2d =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DISP2D);
+
+	/* nothing to do if common lane is already off */
+	if (!cmn->ops->is_enabled(dev_priv, cmn))
+		return;
+
+	/* If the display might be already active skip this */
+	if (disp2d->ops->is_enabled(dev_priv, disp2d) &&
+	    I915_READ(DPIO_CTL) & DPIO_CMNRST)
+		return;
+
+	DRM_DEBUG_KMS("toggling display PHY side reset\n");
+
+	/* cmnlane needs DPLL registers */
+	disp2d->ops->enable(dev_priv, disp2d);
+
+	/*
+	 * From VLV2A0_DP_eDP_HDMI_DPIO_driver_vbios_notes_11.docx:
+	 * Need to assert and de-assert PHY SB reset by gating the
+	 * common lane power, then un-gating it.
+	 * Simply ungating isn't enough to reset the PHY enough to get
+	 * ports and lanes running.
+	 */
+	cmn->ops->disable(dev_priv, cmn);
+}
+
 void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 {
+	struct drm_device *dev = dev_priv->dev;
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 
 	power_domains->initializing = true;
+
+	if (IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev)) {
+		mutex_lock(&power_domains->lock);
+		vlv_cmnlane_wa(dev_priv);
+		mutex_unlock(&power_domains->lock);
+	}
+
 	/* For now, we need the power well to be always enabled. */
 	intel_display_set_init_power(dev_priv, true);
 	intel_power_domains_resume(dev_priv);
@@ -6689,7 +6798,7 @@ void intel_init_pm(struct drm_device *dev)
 				 (dev_priv->is_ddr3 == 1) ? "3" : "2",
 				 dev_priv->fsb_freq, dev_priv->mem_freq);
 			/* Disable CxSR and never update its watermark again */
-			pineview_disable_cxsr(dev);
+			intel_set_memory_cxsr(dev_priv, false);
 			dev_priv->display.update_wm = NULL;
 		} else
 			dev_priv->display.update_wm = pineview_update_wm;

drivers/gpu/drm/i915/intel_ringbuffer.c

@@ -48,9 +48,8 @@ static inline int __ring_space(int head, int tail, int size)
 	return space;
 }
 
-static inline int ring_space(struct intel_engine_cs *ring)
+static inline int ring_space(struct intel_ringbuffer *ringbuf)
 {
-	struct intel_ringbuffer *ringbuf = ring->buffer;
 	return __ring_space(ringbuf->head & HEAD_ADDR, ringbuf->tail, ringbuf->size);
 }
 
@@ -545,7 +544,7 @@ static int init_ring_common(struct intel_engine_cs *ring)
 	else {
 		ringbuf->head = I915_READ_HEAD(ring);
 		ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
-		ringbuf->space = ring_space(ring);
+		ringbuf->space = ring_space(ringbuf);
 		ringbuf->last_retired_head = -1;
 	}
 
@@ -660,6 +659,13 @@ static int init_render_ring(struct intel_engine_cs *ring)
 static void render_ring_cleanup(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->semaphore_obj) {
+		i915_gem_object_ggtt_unpin(dev_priv->semaphore_obj);
+		drm_gem_object_unreference(&dev_priv->semaphore_obj->base);
+		dev_priv->semaphore_obj = NULL;
+	}
 
 	if (ring->scratch.obj == NULL)
 		return;
@@ -673,29 +679,96 @@ static void render_ring_cleanup(struct intel_engine_cs *ring)
 	ring->scratch.obj = NULL;
 }
 
+static int gen8_rcs_signal(struct intel_engine_cs *signaller,
+			   unsigned int num_dwords)
+{
+#define MBOX_UPDATE_DWORDS 8
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *waiter;
+	int i, ret, num_rings;
+
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
+#undef MBOX_UPDATE_DWORDS
+
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+
+	for_each_ring(waiter, dev_priv, i) {
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
+			continue;
+
+		intel_ring_emit(signaller, GFX_OP_PIPE_CONTROL(6));
+		intel_ring_emit(signaller, PIPE_CONTROL_GLOBAL_GTT_IVB |
+					   PIPE_CONTROL_QW_WRITE |
+					   PIPE_CONTROL_FLUSH_ENABLE);
+		intel_ring_emit(signaller, lower_32_bits(gtt_offset));
+		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
+		intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+		intel_ring_emit(signaller, 0);
+		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
+					   MI_SEMAPHORE_TARGET(waiter->id));
+		intel_ring_emit(signaller, 0);
+	}
+
+	return 0;
+}
+
+static int gen8_xcs_signal(struct intel_engine_cs *signaller,
+			   unsigned int num_dwords)
+{
+#define MBOX_UPDATE_DWORDS 6
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *waiter;
+	int i, ret, num_rings;
+
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += (num_rings-1) * MBOX_UPDATE_DWORDS;
+#undef MBOX_UPDATE_DWORDS
+
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+
+	for_each_ring(waiter, dev_priv, i) {
+		u64 gtt_offset = signaller->semaphore.signal_ggtt[i];
+		if (gtt_offset == MI_SEMAPHORE_SYNC_INVALID)
+			continue;
+
+		intel_ring_emit(signaller, (MI_FLUSH_DW + 1) |
+					   MI_FLUSH_DW_OP_STOREDW);
+		intel_ring_emit(signaller, lower_32_bits(gtt_offset) |
+					   MI_FLUSH_DW_USE_GTT);
+		intel_ring_emit(signaller, upper_32_bits(gtt_offset));
+		intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+		intel_ring_emit(signaller, MI_SEMAPHORE_SIGNAL |
+					   MI_SEMAPHORE_TARGET(waiter->id));
+		intel_ring_emit(signaller, 0);
+	}
+
+	return 0;
+}
+
 static int gen6_signal(struct intel_engine_cs *signaller,
 		       unsigned int num_dwords)
 {
 	struct drm_device *dev = signaller->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *useless;
-	int i, ret;
+	int i, ret, num_rings;
 
-	/* NB: In order to be able to do semaphore MBOX updates for varying
-	 * number of rings, it's easiest if we round up each individual update
-	 * to a multiple of 2 (since ring updates must always be a multiple of
-	 * 2) even though the actual update only requires 3 dwords.
-	 */
-#define MBOX_UPDATE_DWORDS 4
-	if (i915_semaphore_is_enabled(dev))
-		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
-	else
-		return intel_ring_begin(signaller, num_dwords);
+#define MBOX_UPDATE_DWORDS 3
+	num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+	num_dwords += round_up((num_rings-1) * MBOX_UPDATE_DWORDS, 2);
+#undef MBOX_UPDATE_DWORDS
 
 	ret = intel_ring_begin(signaller, num_dwords);
 	if (ret)
 		return ret;
-#undef MBOX_UPDATE_DWORDS
 
 	for_each_ring(useless, dev_priv, i) {
 		u32 mbox_reg = signaller->semaphore.mbox.signal[i];
@@ -703,15 +776,13 @@ static int gen6_signal(struct intel_engine_cs *signaller,
 			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
 			intel_ring_emit(signaller, mbox_reg);
 			intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
-			intel_ring_emit(signaller, MI_NOOP);
-		} else {
-			intel_ring_emit(signaller, MI_NOOP);
-			intel_ring_emit(signaller, MI_NOOP);
-			intel_ring_emit(signaller, MI_NOOP);
-			intel_ring_emit(signaller, MI_NOOP);
 		}
 	}
 
+	/* If num_dwords was rounded, make sure the tail pointer is correct */
+	if (num_rings % 2 == 0)
+		intel_ring_emit(signaller, MI_NOOP);
+
 	return 0;
 }
 
@@ -729,7 +800,11 @@ gen6_add_request(struct intel_engine_cs *ring)
 {
 	int ret;
 
-	ret = ring->semaphore.signal(ring, 4);
+	if (ring->semaphore.signal)
+		ret = ring->semaphore.signal(ring, 4);
+	else
+		ret = intel_ring_begin(ring, 4);
+
 	if (ret)
 		return ret;
 
@@ -756,6 +831,32 @@ static inline bool i915_gem_has_seqno_wrapped(struct drm_device *dev,
  * @signaller - ring which has, or will signal
  * @seqno - seqno which the waiter will block on
  */
+
+static int
+gen8_ring_sync(struct intel_engine_cs *waiter,
+	       struct intel_engine_cs *signaller,
+	       u32 seqno)
+{
+	struct drm_i915_private *dev_priv = waiter->dev->dev_private;
+	int ret;
+
+	ret = intel_ring_begin(waiter, 4);
+	if (ret)
+		return ret;
+
+	intel_ring_emit(waiter, MI_SEMAPHORE_WAIT |
+				MI_SEMAPHORE_GLOBAL_GTT |
+				MI_SEMAPHORE_POLL |
+				MI_SEMAPHORE_SAD_GTE_SDD);
+	intel_ring_emit(waiter, seqno);
+	intel_ring_emit(waiter,
+			lower_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
+	intel_ring_emit(waiter,
+			upper_32_bits(GEN8_WAIT_OFFSET(waiter, signaller->id)));
+	intel_ring_advance(waiter);
+	return 0;
+}
+
 static int
 gen6_ring_sync(struct intel_engine_cs *waiter,
 	       struct intel_engine_cs *signaller,
@@ -1331,6 +1432,7 @@ static int init_status_page(struct intel_engine_cs *ring)
 	struct drm_i915_gem_object *obj;
 
 	if ((obj = ring->status_page.obj) == NULL) {
+		unsigned flags;
 		int ret;
 
 		obj = i915_gem_alloc_object(ring->dev, 4096);
@@ -1343,7 +1445,20 @@ static int init_status_page(struct intel_engine_cs *ring)
 		if (ret)
 			goto err_unref;
 
-		ret = i915_gem_obj_ggtt_pin(obj, 4096, 0);
+		flags = 0;
+		if (!HAS_LLC(ring->dev))
+			/* On g33, we cannot place HWS above 256MiB, so
+			 * restrict its pinning to the low mappable arena.
+			 * Though this restriction is not documented for
+			 * gen4, gen5, or byt, they also behave similarly
+			 * and hang if the HWS is placed at the top of the
+			 * GTT. To generalise, it appears that all !llc
+			 * platforms have issues with us placing the HWS
+			 * above the mappable region (even though we never
+			 * actualy map it).
+			 */
+			flags |= PIN_MAPPABLE;
+		ret = i915_gem_obj_ggtt_pin(obj, 4096, flags);
 		if (ret) {
 err_unref:
 			drm_gem_object_unreference(&obj->base);
@@ -1380,15 +1495,25 @@ static int init_phys_status_page(struct intel_engine_cs *ring)
 	return 0;
 }
 
-static int allocate_ring_buffer(struct intel_engine_cs *ring)
+static void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
+{
+	if (!ringbuf->obj)
+		return;
+
+	iounmap(ringbuf->virtual_start);
+	i915_gem_object_ggtt_unpin(ringbuf->obj);
+	drm_gem_object_unreference(&ringbuf->obj->base);
+	ringbuf->obj = NULL;
+}
+
+static int intel_alloc_ringbuffer_obj(struct drm_device *dev,
+				      struct intel_ringbuffer *ringbuf)
 {
-	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	if (intel_ring_initialized(ring))
+	if (ringbuf->obj)
 		return 0;
 
 	obj = NULL;
@@ -1460,7 +1585,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 			goto error;
 	}
 
-	ret = allocate_ring_buffer(ring);
+	ret = intel_alloc_ringbuffer_obj(dev, ringbuf);
 	if (ret) {
 		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n", ring->name, ret);
 		goto error;
@@ -1501,11 +1626,7 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 	intel_stop_ring_buffer(ring);
 	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
 
-	iounmap(ringbuf->virtual_start);
-
-	i915_gem_object_ggtt_unpin(ringbuf->obj);
-	drm_gem_object_unreference(&ringbuf->obj->base);
-	ringbuf->obj = NULL;
+	intel_destroy_ringbuffer_obj(ringbuf);
 	ring->preallocated_lazy_request = NULL;
 	ring->outstanding_lazy_seqno = 0;
 
@@ -1531,7 +1652,7 @@ static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
 		ringbuf->head = ringbuf->last_retired_head;
 		ringbuf->last_retired_head = -1;
 
-		ringbuf->space = ring_space(ring);
+		ringbuf->space = ring_space(ringbuf);
 		if (ringbuf->space >= n)
 			return 0;
 	}
@@ -1554,7 +1675,7 @@ static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
 	ringbuf->head = ringbuf->last_retired_head;
 	ringbuf->last_retired_head = -1;
 
-	ringbuf->space = ring_space(ring);
+	ringbuf->space = ring_space(ringbuf);
 	return 0;
 }
 
@@ -1583,7 +1704,7 @@ static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
 	trace_i915_ring_wait_begin(ring);
 	do {
 		ringbuf->head = I915_READ_HEAD(ring);
-		ringbuf->space = ring_space(ring);
+		ringbuf->space = ring_space(ringbuf);
 		if (ringbuf->space >= n) {
 			ret = 0;
 			break;
@@ -1635,7 +1756,7 @@ static int intel_wrap_ring_buffer(struct intel_engine_cs *ring)
 		iowrite32(MI_NOOP, virt++);
 
 	ringbuf->tail = 0;
-	ringbuf->space = ring_space(ring);
+	ringbuf->space = ring_space(ringbuf);
 
 	return 0;
 }
@@ -1947,45 +2068,74 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct drm_i915_gem_object *obj;
+	int ret;
 
 	ring->name = "render ring";
 	ring->id = RCS;
 	ring->mmio_base = RENDER_RING_BASE;
 
-	if (INTEL_INFO(dev)->gen >= 6) {
+	if (INTEL_INFO(dev)->gen >= 8) {
+		if (i915_semaphore_is_enabled(dev)) {
+			obj = i915_gem_alloc_object(dev, 4096);
+			if (obj == NULL) {
+				DRM_ERROR("Failed to allocate semaphore bo. Disabling semaphores\n");
+				i915.semaphores = 0;
+			} else {
+				i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+				ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_NONBLOCK);
+				if (ret != 0) {
+					drm_gem_object_unreference(&obj->base);
+					DRM_ERROR("Failed to pin semaphore bo. Disabling semaphores\n");
+					i915.semaphores = 0;
+				} else
+					dev_priv->semaphore_obj = obj;
+			}
+		}
+		ring->add_request = gen6_add_request;
+		ring->flush = gen8_render_ring_flush;
+		ring->irq_get = gen8_ring_get_irq;
+		ring->irq_put = gen8_ring_put_irq;
+		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
+		ring->get_seqno = gen6_ring_get_seqno;
+		ring->set_seqno = ring_set_seqno;
+		if (i915_semaphore_is_enabled(dev)) {
+			WARN_ON(!dev_priv->semaphore_obj);
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_rcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
+	} else if (INTEL_INFO(dev)->gen >= 6) {
 		ring->add_request = gen6_add_request;
 		ring->flush = gen7_render_ring_flush;
 		if (INTEL_INFO(dev)->gen == 6)
 			ring->flush = gen6_render_ring_flush;
-		if (INTEL_INFO(dev)->gen >= 8) {
-			ring->flush = gen8_render_ring_flush;
-			ring->irq_get = gen8_ring_get_irq;
-			ring->irq_put = gen8_ring_put_irq;
-		} else {
-			ring->irq_get = gen6_ring_get_irq;
-			ring->irq_put = gen6_ring_put_irq;
-		}
+		ring->irq_get = gen6_ring_get_irq;
+		ring->irq_put = gen6_ring_put_irq;
 		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
 		ring->get_seqno = gen6_ring_get_seqno;
 		ring->set_seqno = ring_set_seqno;
-		ring->semaphore.sync_to = gen6_ring_sync;
-		ring->semaphore.signal = gen6_signal;
-		/*
-		 * The current semaphore is only applied on pre-gen8 platform.
-		 * And there is no VCS2 ring on the pre-gen8 platform. So the
-		 * semaphore between RCS and VCS2 is initialized as INVALID.
-		 * Gen8 will initialize the sema between VCS2 and RCS later.
-		 */
-		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
-		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
-		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
-		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
-		ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
-		ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
-		ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
-		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen6_ring_sync;
+			ring->semaphore.signal = gen6_signal;
+			/*
+			 * The current semaphore is only applied on pre-gen8
+			 * platform.  And there is no VCS2 ring on the pre-gen8
+			 * platform. So the semaphore between RCS and VCS2 is
+			 * initialized as INVALID.  Gen8 will initialize the
+			 * sema between VCS2 and RCS later.
+			 */
+			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
+			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
+			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
+			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+			ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
+			ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
+			ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
+			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		}
 	} else if (IS_GEN5(dev)) {
 		ring->add_request = pc_render_add_request;
 		ring->flush = gen4_render_ring_flush;
@@ -2013,6 +2163,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->irq_enable_mask = I915_USER_INTERRUPT;
 	}
 	ring->write_tail = ring_write_tail;
+
 	if (IS_HASWELL(dev))
 		ring->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
 	else if (IS_GEN8(dev))
@@ -2030,9 +2181,6 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 
 	/* Workaround batchbuffer to combat CS tlb bug. */
 	if (HAS_BROKEN_CS_TLB(dev)) {
-		struct drm_i915_gem_object *obj;
-		int ret;
-
 		obj = i915_gem_alloc_object(dev, I830_BATCH_LIMIT);
 		if (obj == NULL) {
 			DRM_ERROR("Failed to allocate batch bo\n");
@@ -2163,31 +2311,32 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 			ring->irq_put = gen8_ring_put_irq;
 			ring->dispatch_execbuffer =
 				gen8_ring_dispatch_execbuffer;
+			if (i915_semaphore_is_enabled(dev)) {
+				ring->semaphore.sync_to = gen8_ring_sync;
+				ring->semaphore.signal = gen8_xcs_signal;
+				GEN8_RING_SEMAPHORE_INIT;
+			}
 		} else {
 			ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
 			ring->irq_get = gen6_ring_get_irq;
 			ring->irq_put = gen6_ring_put_irq;
 			ring->dispatch_execbuffer =
 				gen6_ring_dispatch_execbuffer;
+			if (i915_semaphore_is_enabled(dev)) {
+				ring->semaphore.sync_to = gen6_ring_sync;
+				ring->semaphore.signal = gen6_signal;
+				ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
+				ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+				ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
+				ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
+				ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+				ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
+				ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+				ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
+				ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
+				ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+			}
 		}
-		ring->semaphore.sync_to = gen6_ring_sync;
-		ring->semaphore.signal = gen6_signal;
-		/*
-		 * The current semaphore is only applied on pre-gen8 platform.
-		 * And there is no VCS2 ring on the pre-gen8 platform. So the
-		 * semaphore between VCS and VCS2 is initialized as INVALID.
-		 * Gen8 will initialize the sema between VCS2 and VCS later.
-		 */
-		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
-		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
-		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
-		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
-		ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
-		ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
-		ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
-		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	} else {
 		ring->mmio_base = BSD_RING_BASE;
 		ring->flush = bsd_ring_flush;
@@ -2224,7 +2373,7 @@ int intel_init_bsd2_ring_buffer(struct drm_device *dev)
 		return -EINVAL;
 	}
 
-	ring->name = "bds2_ring";
+	ring->name = "bsd2 ring";
 	ring->id = VCS2;
 
 	ring->write_tail = ring_write_tail;
@@ -2239,25 +2388,11 @@ int intel_init_bsd2_ring_buffer(struct drm_device *dev)
 	ring->irq_put = gen8_ring_put_irq;
 	ring->dispatch_execbuffer =
 			gen8_ring_dispatch_execbuffer;
-	ring->semaphore.sync_to = gen6_ring_sync;
-	ring->semaphore.signal = gen6_signal;
-	/*
-	 * The current semaphore is only applied on the pre-gen8. And there
-	 * is no bsd2 ring on the pre-gen8. So now the semaphore_register
-	 * between VCS2 and other ring is initialized as invalid.
-	 * Gen8 will initialize the sema between VCS2 and other ring later.
-	 */
-	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
-
+	if (i915_semaphore_is_enabled(dev)) {
+		ring->semaphore.sync_to = gen8_ring_sync;
+		ring->semaphore.signal = gen8_xcs_signal;
+		GEN8_RING_SEMAPHORE_INIT;
+	}
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
@@ -2283,30 +2418,38 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 		ring->irq_get = gen8_ring_get_irq;
 		ring->irq_put = gen8_ring_put_irq;
 		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
 	} else {
 		ring->irq_enable_mask = GT_BLT_USER_INTERRUPT;
 		ring->irq_get = gen6_ring_get_irq;
 		ring->irq_put = gen6_ring_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.signal = gen6_signal;
+			ring->semaphore.sync_to = gen6_ring_sync;
+			/*
+			 * The current semaphore is only applied on pre-gen8
+			 * platform.  And there is no VCS2 ring on the pre-gen8
+			 * platform. So the semaphore between BCS and VCS2 is
+			 * initialized as INVALID.  Gen8 will initialize the
+			 * sema between BCS and VCS2 later.
+			 */
+			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
+			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
+			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
+			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
+			ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
+			ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+			ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
+			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		}
 	}
-	ring->semaphore.sync_to = gen6_ring_sync;
-	ring->semaphore.signal = gen6_signal;
-	/*
-	 * The current semaphore is only applied on pre-gen8 platform. And
-	 * there is no VCS2 ring on the pre-gen8 platform. So the semaphore
-	 * between BCS and VCS2 is initialized as INVALID.
-	 * Gen8 will initialize the sema between BCS and VCS2 later.
-	 */
-	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
-	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
-	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
-	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
-	ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
-	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
-	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
@@ -2333,24 +2476,31 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev)
 		ring->irq_get = gen8_ring_get_irq;
 		ring->irq_put = gen8_ring_put_irq;
 		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen8_ring_sync;
+			ring->semaphore.signal = gen8_xcs_signal;
+			GEN8_RING_SEMAPHORE_INIT;
+		}
 	} else {
 		ring->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
 		ring->irq_get = hsw_vebox_get_irq;
 		ring->irq_put = hsw_vebox_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		if (i915_semaphore_is_enabled(dev)) {
+			ring->semaphore.sync_to = gen6_ring_sync;
+			ring->semaphore.signal = gen6_signal;
+			ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
+			ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
+			ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
+			ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+			ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
+			ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
+			ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
+			ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+			ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+		}
 	}
-	ring->semaphore.sync_to = gen6_ring_sync;
-	ring->semaphore.signal = gen6_signal;
-	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
-	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
-	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
-	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
-	ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
-	ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
-	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
-	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);

drivers/gpu/drm/i915/intel_ringbuffer.h

@@ -40,6 +40,32 @@ struct  intel_hw_status_page {
 #define I915_READ_MODE(ring) I915_READ(RING_MI_MODE((ring)->mmio_base))
 #define I915_WRITE_MODE(ring, val) I915_WRITE(RING_MI_MODE((ring)->mmio_base), val)
 
+/* seqno size is actually only a uint32, but since we plan to use MI_FLUSH_DW to
+ * do the writes, and that must have qw aligned offsets, simply pretend it's 8b.
+ */
+#define i915_semaphore_seqno_size sizeof(uint64_t)
+#define GEN8_SIGNAL_OFFSET(__ring, to)			     \
+	(i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj) + \
+	((__ring)->id * I915_NUM_RINGS * i915_semaphore_seqno_size) +	\
+	(i915_semaphore_seqno_size * (to)))
+
+#define GEN8_WAIT_OFFSET(__ring, from)			     \
+	(i915_gem_obj_ggtt_offset(dev_priv->semaphore_obj) + \
+	((from) * I915_NUM_RINGS * i915_semaphore_seqno_size) + \
+	(i915_semaphore_seqno_size * (__ring)->id))
+
+#define GEN8_RING_SEMAPHORE_INIT do { \
+	if (!dev_priv->semaphore_obj) { \
+		break; \
+	} \
+	ring->semaphore.signal_ggtt[RCS] = GEN8_SIGNAL_OFFSET(ring, RCS); \
+	ring->semaphore.signal_ggtt[VCS] = GEN8_SIGNAL_OFFSET(ring, VCS); \
+	ring->semaphore.signal_ggtt[BCS] = GEN8_SIGNAL_OFFSET(ring, BCS); \
+	ring->semaphore.signal_ggtt[VECS] = GEN8_SIGNAL_OFFSET(ring, VECS); \
+	ring->semaphore.signal_ggtt[VCS2] = GEN8_SIGNAL_OFFSET(ring, VCS2); \
+	ring->semaphore.signal_ggtt[ring->id] = MI_SEMAPHORE_SYNC_INVALID; \
+	} while(0)
+
 enum intel_ring_hangcheck_action {
 	HANGCHECK_IDLE = 0,
 	HANGCHECK_WAIT,
@@ -127,15 +153,55 @@ struct  intel_engine_cs {
 #define I915_DISPATCH_PINNED 0x2
 	void		(*cleanup)(struct intel_engine_cs *ring);
 
+	/* GEN8 signal/wait table - never trust comments!
+	 *	  signal to	signal to    signal to   signal to      signal to
+	 *	    RCS		   VCS          BCS        VECS		 VCS2
+	 *      --------------------------------------------------------------------
+	 *  RCS | NOP (0x00) | VCS (0x08) | BCS (0x10) | VECS (0x18) | VCS2 (0x20) |
+	 *	|-------------------------------------------------------------------
+	 *  VCS | RCS (0x28) | NOP (0x30) | BCS (0x38) | VECS (0x40) | VCS2 (0x48) |
+	 *	|-------------------------------------------------------------------
+	 *  BCS | RCS (0x50) | VCS (0x58) | NOP (0x60) | VECS (0x68) | VCS2 (0x70) |
+	 *	|-------------------------------------------------------------------
+	 * VECS | RCS (0x78) | VCS (0x80) | BCS (0x88) |  NOP (0x90) | VCS2 (0x98) |
+	 *	|-------------------------------------------------------------------
+	 * VCS2 | RCS (0xa0) | VCS (0xa8) | BCS (0xb0) | VECS (0xb8) | NOP  (0xc0) |
+	 *	|-------------------------------------------------------------------
+	 *
+	 * Generalization:
+	 *  f(x, y) := (x->id * NUM_RINGS * seqno_size) + (seqno_size * y->id)
+	 *  ie. transpose of g(x, y)
+	 *
+	 *	 sync from	sync from    sync from    sync from	sync from
+	 *	    RCS		   VCS          BCS        VECS		 VCS2
+	 *      --------------------------------------------------------------------
+	 *  RCS | NOP (0x00) | VCS (0x28) | BCS (0x50) | VECS (0x78) | VCS2 (0xa0) |
+	 *	|-------------------------------------------------------------------
+	 *  VCS | RCS (0x08) | NOP (0x30) | BCS (0x58) | VECS (0x80) | VCS2 (0xa8) |
+	 *	|-------------------------------------------------------------------
+	 *  BCS | RCS (0x10) | VCS (0x38) | NOP (0x60) | VECS (0x88) | VCS2 (0xb0) |
+	 *	|-------------------------------------------------------------------
+	 * VECS | RCS (0x18) | VCS (0x40) | BCS (0x68) |  NOP (0x90) | VCS2 (0xb8) |
+	 *	|-------------------------------------------------------------------
+	 * VCS2 | RCS (0x20) | VCS (0x48) | BCS (0x70) | VECS (0x98) |  NOP (0xc0) |
+	 *	|-------------------------------------------------------------------
+	 *
+	 * Generalization:
+	 *  g(x, y) := (y->id * NUM_RINGS * seqno_size) + (seqno_size * x->id)
+	 *  ie. transpose of f(x, y)
+	 */
 	struct {
 		u32	sync_seqno[I915_NUM_RINGS-1];
 
-		struct {
-			/* our mbox written by others */
-			u32		wait[I915_NUM_RINGS];
-			/* mboxes this ring signals to */
-			u32		signal[I915_NUM_RINGS];
-		} mbox;
+		union {
+			struct {
+				/* our mbox written by others */
+				u32		wait[I915_NUM_RINGS];
+				/* mboxes this ring signals to */
+				u32		signal[I915_NUM_RINGS];
+			} mbox;
+			u64		signal_ggtt[I915_NUM_RINGS];
+		};
 
 		/* AKA wait() */
 		int	(*sync_to)(struct intel_engine_cs *ring,
@@ -238,9 +304,11 @@ intel_ring_sync_index(struct intel_engine_cs *ring,
 	int idx;
 
 	/*
-	 * cs -> 0 = vcs, 1 = bcs
-	 * vcs -> 0 = bcs, 1 = cs,
-	 * bcs -> 0 = cs, 1 = vcs.
+	 * rcs -> 0 = vcs, 1 = bcs, 2 = vecs, 3 = vcs2;
+	 * vcs -> 0 = bcs, 1 = vecs, 2 = vcs2, 3 = rcs;
+	 * bcs -> 0 = vecs, 1 = vcs2. 2 = rcs, 3 = vcs;
+	 * vecs -> 0 = vcs2, 1 = rcs, 2 = vcs, 3 = bcs;
+	 * vcs2 -> 0 = rcs, 1 = vcs, 2 = bcs, 3 = vecs;
 	 */
 
 	idx = (other - ring) - 1;
@@ -318,9 +386,9 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev);
 u64 intel_ring_get_active_head(struct intel_engine_cs *ring);
 void intel_ring_setup_status_page(struct intel_engine_cs *ring);
 
-static inline u32 intel_ring_get_tail(struct intel_engine_cs *ring)
+static inline u32 intel_ring_get_tail(struct intel_ringbuffer *ringbuf)
 {
-	return ring->buffer->tail;
+	return ringbuf->tail;
 }
 
 static inline u32 intel_ring_get_seqno(struct intel_engine_cs *ring)

drivers/gpu/drm/i915/intel_uncore.c

@@ -1010,7 +1010,7 @@ int i915_get_reset_stats_ioctl(struct drm_device *dev,
 	if (args->flags || args->pad)
 		return -EINVAL;
 
-	if (args->ctx_id == DEFAULT_CONTEXT_ID && !capable(CAP_SYS_ADMIN))
+	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);