Merge tag 'drm-for-v4.17' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "Cannonlake and Vega12 support are probably the two major things. This
  pull lacks nouveau, Ben had some unforseen leave and a few other
  blockers so we'll see how things look or maybe leave it for this merge
  window.

  core:
   - Device links to handle sound/gpu pm dependency
   - Color encoding/range properties
   - Plane clipping into plane check helper
   - Backlight helpers
   - DP TP4 + HBR3 helper support

  amdgpu:
   - Vega12 support
   - Enable DC by default on all supported GPUs
   - Powerplay restructuring and cleanup
   - DC bandwidth calc updates
   - DC backlight on pre-DCE11
   - TTM backing store dropping support
   - SR-IOV fixes
   - Adding "wattman" like functionality
   - DC crc support
   - Improved DC dual-link handling

  amdkfd:
   - GPUVM support for dGPU
   - KFD events for dGPU
   - Enable PCIe atomics for dGPUs
   - HSA process eviction support
   - Live-lock fixes for process eviction
   - VM page table allocation fix for large-bar systems

  panel:
   - Raydium RM68200
   - AUO G104SN02 V2
   - KEO TX31D200VM0BAA
   - ARM Versatile panels

  i915:
   - Cannonlake support enabled
   - AUX-F port support added
   - Icelake base enabling until internal milestone of forcewake support
   - Query uAPI interface (used for GPU topology information currently)
   - Compressed framebuffer support for sprites
   - kmem cache shrinking when GPU is idle
   - Avoid boosting GPU when waited item is being processed already
   - Avoid retraining LSPCON link unnecessarily
   - Decrease request signaling latency
   - Deprecation of I915_SET_COLORKEY_NONE
   - Kerneldoc and compiler warning cleanup for upcoming CI enforcements
   - Full range ycbcr toggling
   - HDCP support

  i915/gvt:
   - Big refactor for shadow ppgtt
   - KBL context save/restore via LRI cmd (Weinan)
   - Properly unmap dma for guest page (Changbin)

  vmwgfx:
   - Lots of various improvements

  etnaviv:
   - Use the drm gpu scheduler
   - prep work for GC7000L support

  vc4:
   - fix alpha blending
   - Expose perf counters to userspace

  pl111:
   - Bandwidth checking/limiting
   - Versatile panel support

  sun4i:
   - A83T HDMI support
   - A80 support
   - YUV plane support
   - H3/H5 HDMI support

  omapdrm:
   - HPD support for DVI connector
   - remove lots of static variables

  msm:
   - DSI updates from 10nm / SDM845
   - fix for race condition with a3xx/a4xx fence completion irq
   - some refactoring/prep work for eventual a6xx support (ie. when we
     have a userspace)
   - a5xx debugfs enhancements
   - some mdp5 fixes/cleanups to prepare for eventually merging
     writeback
   - support (ie. when we have a userspace)

  tegra:
   - mmap() fixes for fbdev devices
   - Overlay plane for hw cursor fix
   - dma-buf cache maintenance support

  mali-dp:
   - YUV->RGB conversion support

  rockchip:
   - rk3399/chromebook fixes and improvements

  rcar-du:
   - LVDS support move to drm bridge
   - DT bindings for R8A77995
   - Driver/DT support for R8A77970

  tilcdc:
   - DRM panel support"

* tag 'drm-for-v4.17' of git://people.freedesktop.org/~airlied/linux: (1646 commits)
  drm/i915: Fix hibernation with ACPI S0 target state
  drm/i915/execlists: Use a locked clear_bit() for synchronisation with interrupt
  drm/i915: Specify which engines to reset following semaphore/event lockups
  drm/i915/dp: Write to SET_POWER dpcd to enable MST hub.
  drm/amdkfd: Use ordered workqueue to restore processes
  drm/amdgpu: Fix acquiring VM on large-BAR systems
  drm/amd/pp: clean header file hwmgr.h
  drm/amd/pp: use mlck_table.count for array loop index limit
  drm: Fix uabi regression by allowing garbage mode->type from userspace
  drm/amdgpu: Add an ATPX quirk for hybrid laptop
  drm/amdgpu: fix spelling mistake: "asssert" -> "assert"
  drm/amd/pp: Add new asic support in pp_psm.c
  drm/amd/pp: Clean up powerplay code on Vega12
  drm/amd/pp: Add smu irq handlers for legacy asics
  drm/amd/pp: Fix set wrong temperature range on smu7
  drm/amdgpu: Don't change preferred domian when fallback GTT v5
  drm/vmwgfx: Bump version patchlevel and date
  drm/vmwgfx: use monotonic event timestamps
  drm/vmwgfx: Unpin the screen object backup buffer when not used
  drm/vmwgfx: Stricter count of legacy surface device resources
  ...

4 files changed, 306 insertions, 83 deletions

diff --git a/drivers/gpu/drm/pl111/pl111_display.c b/drivers/gpu/drm/pl111/pl111_display.c
index 06c4bf756b69..310646427907 100644
--- a/drivers/gpu/drm/pl111/pl111_display.c
+++ b/drivers/gpu/drm/pl111/pl111_display.c
@@ -50,6 +50,41 @@ irqreturn_t pl111_irq(int irq, void *data)
 	return status;
 }
 
+static enum drm_mode_status
+pl111_mode_valid(struct drm_crtc *crtc,
+		 const struct drm_display_mode *mode)
+{
+	struct drm_device *drm = crtc->dev;
+	struct pl111_drm_dev_private *priv = drm->dev_private;
+	u32 cpp = priv->variant->fb_bpp / 8;
+	u64 bw;
+
+	/*
+	 * We use the pixelclock to also account for interlaced modes, the
+	 * resulting bandwidth is in bytes per second.
+	 */
+	bw = mode->clock * 1000; /* In Hz */
+	bw = bw * mode->hdisplay * mode->vdisplay * cpp;
+	bw = div_u64(bw, mode->htotal * mode->vtotal);
+
+	/*
+	 * If no bandwidth constraints, anything goes, else
+	 * check if we are too fast.
+	 */
+	if (priv->memory_bw && (bw > priv->memory_bw)) {
+		DRM_DEBUG_KMS("%d x %d @ %d Hz, %d cpp, bw %llu too fast\n",
+			      mode->hdisplay, mode->vdisplay,
+			      mode->clock * 1000, cpp, bw);
+
+		return MODE_BAD;
+	}
+	DRM_DEBUG_KMS("%d x %d @ %d Hz, %d cpp, bw %llu bytes/s OK\n",
+		      mode->hdisplay, mode->vdisplay,
+		      mode->clock * 1000, cpp, bw);
+
+	return MODE_OK;
+}
+
 static int pl111_display_check(struct drm_simple_display_pipe *pipe,
 			       struct drm_plane_state *pstate,
 			       struct drm_crtc_state *cstate)
@@ -94,6 +129,7 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 	const struct drm_display_mode *mode = &cstate->mode;
 	struct drm_framebuffer *fb = plane->state->fb;
 	struct drm_connector *connector = priv->connector;
+	struct drm_bridge *bridge = priv->bridge;
 	u32 cntl;
 	u32 ppl, hsw, hfp, hbp;
 	u32 lpp, vsw, vfp, vbp;
@@ -137,17 +173,46 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 	tim2 = readl(priv->regs + CLCD_TIM2);
 	tim2 &= (TIM2_BCD | TIM2_PCD_LO_MASK | TIM2_PCD_HI_MASK);
 
+	if (priv->variant->broken_clockdivider)
+		tim2 |= TIM2_BCD;
+
 	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
 		tim2 |= TIM2_IHS;
 
 	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
 		tim2 |= TIM2_IVS;
 
-	if (connector->display_info.bus_flags & DRM_BUS_FLAG_DE_LOW)
-		tim2 |= TIM2_IOE;
+	if (connector) {
+		if (connector->display_info.bus_flags & DRM_BUS_FLAG_DE_LOW)
+			tim2 |= TIM2_IOE;
+
+		if (connector->display_info.bus_flags &
+		    DRM_BUS_FLAG_PIXDATA_NEGEDGE)
+			tim2 |= TIM2_IPC;
+	}
 
-	if (connector->display_info.bus_flags & DRM_BUS_FLAG_PIXDATA_NEGEDGE)
-		tim2 |= TIM2_IPC;
+	if (bridge) {
+		const struct drm_bridge_timings *btimings = bridge->timings;
+
+		/*
+		 * Here is when things get really fun. Sometimes the bridge
+		 * timings are such that the signal out from PL11x is not
+		 * stable before the receiving bridge (such as a dumb VGA DAC
+		 * or similar) samples it. If that happens, we compensate by
+		 * the only method we have: output the data on the opposite
+		 * edge of the clock so it is for sure stable when it gets
+		 * sampled.
+		 *
+		 * The PL111 manual does not contain proper timining diagrams
+		 * or data for these details, but we know from experiments
+		 * that the setup time is more than 3000 picoseconds (3 ns).
+		 * If we have a bridge that requires the signal to be stable
+		 * earlier than 3000 ps before the clock pulse, we have to
+		 * output the data on the opposite edge to avoid flicker.
+		 */
+		if (btimings && btimings->setup_time_ps >= 3000)
+			tim2 ^= TIM2_IPC;
+	}
 
 	tim2 |= cpl << 16;
 	writel(tim2, priv->regs + CLCD_TIM2);
@@ -172,10 +237,17 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 		cntl |= CNTL_LCDBPP24 | CNTL_BGR;
 		break;
 	case DRM_FORMAT_BGR565:
-		cntl |= CNTL_LCDBPP16_565;
+		if (priv->variant->is_pl110)
+			cntl |= CNTL_LCDBPP16;
+		else
+			cntl |= CNTL_LCDBPP16_565;
 		break;
 	case DRM_FORMAT_RGB565:
-		cntl |= CNTL_LCDBPP16_565 | CNTL_BGR;
+		if (priv->variant->is_pl110)
+			cntl |= CNTL_LCDBPP16;
+		else
+			cntl |= CNTL_LCDBPP16_565;
+		cntl |= CNTL_BGR;
 		break;
 	case DRM_FORMAT_ABGR1555:
 	case DRM_FORMAT_XBGR1555:
@@ -199,6 +271,10 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 		break;
 	}
 
+	/* The PL110 in Integrator/Versatile does the BGR routing externally */
+	if (priv->variant->external_bgr)
+		cntl &= ~CNTL_BGR;
+
 	/* Power sequence: first enable and chill */
 	writel(cntl, priv->regs + priv->ctrl);
 
@@ -215,7 +291,8 @@ static void pl111_display_enable(struct drm_simple_display_pipe *pipe,
 	cntl |= CNTL_LCDPWR;
 	writel(cntl, priv->regs + priv->ctrl);
 
-	drm_crtc_vblank_on(crtc);
+	if (!priv->variant->broken_vblank)
+		drm_crtc_vblank_on(crtc);
 }
 
 void pl111_display_disable(struct drm_simple_display_pipe *pipe)
@@ -225,7 +302,8 @@ void pl111_display_disable(struct drm_simple_display_pipe *pipe)
 	struct pl111_drm_dev_private *priv = drm->dev_private;
 	u32 cntl;
 
-	drm_crtc_vblank_off(crtc);
+	if (!priv->variant->broken_vblank)
+		drm_crtc_vblank_off(crtc);
 
 	/* Power Down */
 	cntl = readl(priv->regs + priv->ctrl);
@@ -278,8 +356,10 @@ static void pl111_display_update(struct drm_simple_display_pipe *pipe,
 	}
 }
 
-int pl111_enable_vblank(struct drm_device *drm, unsigned int crtc)
+static int pl111_display_enable_vblank(struct drm_simple_display_pipe *pipe)
 {
+	struct drm_crtc *crtc = &pipe->crtc;
+	struct drm_device *drm = crtc->dev;
 	struct pl111_drm_dev_private *priv = drm->dev_private;
 
 	writel(CLCD_IRQ_NEXTBASE_UPDATE, priv->regs + priv->ienb);
@@ -287,8 +367,10 @@ int pl111_enable_vblank(struct drm_device *drm, unsigned int crtc)
 	return 0;
 }
 
-void pl111_disable_vblank(struct drm_device *drm, unsigned int crtc)
+static void pl111_display_disable_vblank(struct drm_simple_display_pipe *pipe)
 {
+	struct drm_crtc *crtc = &pipe->crtc;
+	struct drm_device *drm = crtc->dev;
 	struct pl111_drm_dev_private *priv = drm->dev_private;
 
 	writel(0, priv->regs + priv->ienb);
@@ -300,7 +382,8 @@ static int pl111_display_prepare_fb(struct drm_simple_display_pipe *pipe,
 	return drm_gem_fb_prepare_fb(&pipe->plane, plane_state);
 }
 
-static const struct drm_simple_display_pipe_funcs pl111_display_funcs = {
+static struct drm_simple_display_pipe_funcs pl111_display_funcs = {
+	.mode_valid = pl111_mode_valid,
 	.check = pl111_display_check,
 	.enable = pl111_display_enable,
 	.disable = pl111_display_disable,
@@ -417,6 +500,11 @@ pl111_init_clock_divider(struct drm_device *drm)
 		dev_err(drm->dev, "CLCD: unable to get clcdclk.\n");
 		return PTR_ERR(parent);
 	}
+	/* If the clock divider is broken, use the parent directly */
+	if (priv->variant->broken_clockdivider) {
+		priv->clk = parent;
+		return 0;
+	}
 	parent_name = __clk_get_name(parent);
 
 	spin_lock_init(&priv->tim2_lock);
@@ -454,6 +542,11 @@ int pl111_display_init(struct drm_device *drm)
 	if (ret)
 		return ret;
 
+	if (!priv->variant->broken_vblank) {
+		pl111_display_funcs.enable_vblank = pl111_display_enable_vblank;
+		pl111_display_funcs.disable_vblank = pl111_display_disable_vblank;
+	}
+
 	ret = drm_simple_display_pipe_init(drm, &priv->pipe,
 					   &pl111_display_funcs,
 					   priv->variant->formats,
diff --git a/drivers/gpu/drm/pl111/pl111_drm.h b/drivers/gpu/drm/pl111/pl111_drm.h
index 07fa2cdb364a..8639b2d4ddf7 100644
--- a/drivers/gpu/drm/pl111/pl111_drm.h
+++ b/drivers/gpu/drm/pl111/pl111_drm.h
@@ -36,14 +36,24 @@ struct drm_minor;
  * struct pl111_variant_data - encodes IP differences
  * @name: the name of this variant
  * @is_pl110: this is the early PL110 variant
+ * @external_bgr: this is the Versatile Pl110 variant with external
+ *	BGR/RGB routing
+ * @broken_clockdivider: the clock divider is broken and we need to
+ *	use the supplied clock directly
+ * @broken_vblank: the vblank IRQ is broken on this variant
  * @formats: array of supported pixel formats on this variant
  * @nformats: the length of the array of supported pixel formats
+ * @fb_bpp: desired bits per pixel on the default framebuffer
  */
 struct pl111_variant_data {
 	const char *name;
 	bool is_pl110;
+	bool external_bgr;
+	bool broken_clockdivider;
+	bool broken_vblank;
 	const u32 *formats;
 	unsigned int nformats;
+	unsigned int fb_bpp;
 };
 
 struct pl111_drm_dev_private {
@@ -55,6 +65,7 @@ struct pl111_drm_dev_private {
 	struct drm_simple_display_pipe pipe;
 
 	void *regs;
+	u32 memory_bw;
 	u32 ienb;
 	u32 ctrl;
 	/* The pixel clock (a reference to our clock divider off of CLCDCLK). */
@@ -71,8 +82,6 @@ struct pl111_drm_dev_private {
 };
 
 int pl111_display_init(struct drm_device *dev);
-int pl111_enable_vblank(struct drm_device *drm, unsigned int crtc);
-void pl111_disable_vblank(struct drm_device *drm, unsigned int crtc);
 irqreturn_t pl111_irq(int irq, void *data);
 int pl111_debugfs_init(struct drm_minor *minor);
 
diff --git a/drivers/gpu/drm/pl111/pl111_drv.c b/drivers/gpu/drm/pl111/pl111_drv.c
index acb738c69873..4621259d5387 100644
--- a/drivers/gpu/drm/pl111/pl111_drv.c
+++ b/drivers/gpu/drm/pl111/pl111_drv.c
@@ -58,6 +58,8 @@
 #include <linux/dma-buf.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_graph.h>
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic_helper.h>
@@ -85,9 +87,13 @@ static int pl111_modeset_init(struct drm_device *dev)
 {
 	struct drm_mode_config *mode_config;
 	struct pl111_drm_dev_private *priv = dev->dev_private;
-	struct drm_panel *panel;
-	struct drm_bridge *bridge;
+	struct device_node *np = dev->dev->of_node;
+	struct device_node *remote;
+	struct drm_panel *panel = NULL;
+	struct drm_bridge *bridge = NULL;
+	bool defer = false;
 	int ret = 0;
+	int i;
 
 	drm_mode_config_init(dev);
 	mode_config = &dev->mode_config;
@@ -97,10 +103,54 @@ static int pl111_modeset_init(struct drm_device *dev)
 	mode_config->min_height = 1;
 	mode_config->max_height = 768;
 
-	ret = drm_of_find_panel_or_bridge(dev->dev->of_node,
-					  0, 0, &panel, &bridge);
-	if (ret && ret != -ENODEV)
-		return ret;
+	i = 0;
+	for_each_endpoint_of_node(np, remote) {
+		struct drm_panel *tmp_panel;
+		struct drm_bridge *tmp_bridge;
+
+		dev_dbg(dev->dev, "checking endpoint %d\n", i);
+
+		ret = drm_of_find_panel_or_bridge(dev->dev->of_node,
+						  0, i,
+						  &tmp_panel,
+						  &tmp_bridge);
+		if (ret) {
+			if (ret == -EPROBE_DEFER) {
+				/*
+				 * Something deferred, but that is often just
+				 * another way of saying -ENODEV, but let's
+				 * cast a vote for later deferral.
+				 */
+				defer = true;
+			} else if (ret != -ENODEV) {
+				/* Continue, maybe something else is working */
+				dev_err(dev->dev,
+					"endpoint %d returns %d\n", i, ret);
+			}
+		}
+
+		if (tmp_panel) {
+			dev_info(dev->dev,
+				 "found panel on endpoint %d\n", i);
+			panel = tmp_panel;
+		}
+		if (tmp_bridge) {
+			dev_info(dev->dev,
+				 "found bridge on endpoint %d\n", i);
+			bridge = tmp_bridge;
+		}
+
+		i++;
+	}
+
+	/*
+	 * If we can't find neither panel nor bridge on any of the
+	 * endpoints, and any of them retured -EPROBE_DEFER, then
+	 * let's defer this driver too.
+	 */
+	if ((!panel && !bridge) && defer)
+		return -EPROBE_DEFER;
+
 	if (panel) {
 		bridge = drm_panel_bridge_add(panel,
 					      DRM_MODE_CONNECTOR_Unknown);
@@ -108,11 +158,17 @@ static int pl111_modeset_init(struct drm_device *dev)
 			ret = PTR_ERR(bridge);
 			goto out_config;
 		}
-		/*
-		 * TODO: when we are using a different bridge than a panel
-		 * (such as a dumb VGA connector) we need to devise a different
-		 * method to get the connector out of the bridge.
-		 */
+	} else if (bridge) {
+		dev_info(dev->dev, "Using non-panel bridge\n");
+	} else {
+		dev_err(dev->dev, "No bridge, exiting\n");
+		return -ENODEV;
+	}
+
+	priv->bridge = bridge;
+	if (panel) {
+		priv->panel = panel;
+		priv->connector = panel->connector;
 	}
 
 	ret = pl111_display_init(dev);
@@ -126,19 +182,17 @@ static int pl111_modeset_init(struct drm_device *dev)
 	if (ret)
 		return ret;
 
-	priv->bridge = bridge;
-	priv->panel = panel;
-	priv->connector = panel->connector;
-
-	ret = drm_vblank_init(dev, 1);
-	if (ret != 0) {
-		dev_err(dev->dev, "Failed to init vblank\n");
-		goto out_bridge;
+	if (!priv->variant->broken_vblank) {
+		ret = drm_vblank_init(dev, 1);
+		if (ret != 0) {
+			dev_err(dev->dev, "Failed to init vblank\n");
+			goto out_bridge;
+		}
 	}
 
 	drm_mode_config_reset(dev);
 
-	drm_fb_cma_fbdev_init(dev, 32, 0);
+	drm_fb_cma_fbdev_init(dev, priv->variant->fb_bpp, 0);
 
 	drm_kms_helper_poll_init(dev);
 
@@ -170,10 +224,6 @@ static struct drm_driver pl111_drm_driver = {
 	.dumb_create = drm_gem_cma_dumb_create,
 	.gem_free_object_unlocked = drm_gem_cma_free_object,
 	.gem_vm_ops = &drm_gem_cma_vm_ops,
-
-	.enable_vblank = pl111_enable_vblank,
-	.disable_vblank = pl111_disable_vblank,
-
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_import = drm_gem_prime_import,
@@ -191,7 +241,7 @@ static int pl111_amba_probe(struct amba_device *amba_dev,
 {
 	struct device *dev = &amba_dev->dev;
 	struct pl111_drm_dev_private *priv;
-	struct pl111_variant_data *variant = id->data;
+	const struct pl111_variant_data *variant = id->data;
 	struct drm_device *drm;
 	int ret;
 
@@ -207,27 +257,16 @@ static int pl111_amba_probe(struct amba_device *amba_dev,
 	drm->dev_private = priv;
 	priv->variant = variant;
 
-	/*
-	 * The PL110 and PL111 variants have two registers
-	 * swapped: interrupt enable and control. For this reason
-	 * we use offsets that we can change per variant.
-	 */
+	if (of_property_read_u32(dev->of_node, "max-memory-bandwidth",
+				 &priv->memory_bw)) {
+		dev_info(dev, "no max memory bandwidth specified, assume unlimited\n");
+		priv->memory_bw = 0;
+	}
+
+	/* The two variants swap this register */
 	if (variant->is_pl110) {
-		/*
-		 * The ARM Versatile boards are even more special:
-		 * their PrimeCell ID say they are PL110 but the
-		 * control and interrupt enable registers are anyway
-		 * swapped to the PL111 order so they are not following
-		 * the PL110 datasheet.
-		 */
-		if (of_machine_is_compatible("arm,versatile-ab") ||
-		    of_machine_is_compatible("arm,versatile-pb")) {
-			priv->ienb = CLCD_PL111_IENB;
-			priv->ctrl = CLCD_PL111_CNTL;
-		} else {
-			priv->ienb = CLCD_PL110_IENB;
-			priv->ctrl = CLCD_PL110_CNTL;
-		}
+		priv->ienb = CLCD_PL110_IENB;
+		priv->ctrl = CLCD_PL110_CNTL;
 	} else {
 		priv->ienb = CLCD_PL111_IENB;
 		priv->ctrl = CLCD_PL111_CNTL;
@@ -239,6 +278,11 @@ static int pl111_amba_probe(struct amba_device *amba_dev,
 		return PTR_ERR(priv->regs);
 	}
 
+	/* This may override some variant settings */
+	ret = pl111_versatile_init(dev, priv);
+	if (ret)
+		goto dev_unref;
+
 	/* turn off interrupts before requesting the irq */
 	writel(0, priv->regs + priv->ienb);
 
@@ -249,10 +293,6 @@ static int pl111_amba_probe(struct amba_device *amba_dev,
 		return ret;
 	}
 
-	ret = pl111_versatile_init(dev, priv);
-	if (ret)
-		goto dev_unref;
-
 	ret = pl111_modeset_init(drm);
 	if (ret != 0)
 		goto dev_unref;
@@ -284,8 +324,7 @@ static int pl111_amba_remove(struct amba_device *amba_dev)
 }
 
 /*
- * This variant exist in early versions like the ARM Integrator
- * and this version lacks the 565 and 444 pixel formats.
+ * This early variant lacks the 565 and 444 pixel formats.
  */
 static const u32 pl110_pixel_formats[] = {
 	DRM_FORMAT_ABGR8888,
@@ -303,6 +342,7 @@ static const struct pl111_variant_data pl110_variant = {
 	.is_pl110 = true,
 	.formats = pl110_pixel_formats,
 	.nformats = ARRAY_SIZE(pl110_pixel_formats),
+	.fb_bpp = 16,
 };
 
 /* RealView, Versatile Express etc use this modern variant */
@@ -327,6 +367,7 @@ static const struct pl111_variant_data pl111_variant = {
 	.name = "PL111",
 	.formats = pl111_pixel_formats,
 	.nformats = ARRAY_SIZE(pl111_pixel_formats),
+	.fb_bpp = 32,
 };
 
 static const struct amba_id pl111_id_table[] = {
diff --git a/drivers/gpu/drm/pl111/pl111_versatile.c b/drivers/gpu/drm/pl111/pl111_versatile.c
index 97d4af6925a3..9302f516045e 100644
--- a/drivers/gpu/drm/pl111/pl111_versatile.c
+++ b/drivers/gpu/drm/pl111/pl111_versatile.c
@@ -1,3 +1,4 @@
+#include <linux/amba/clcd-regs.h>
 #include <linux/device.h>
 #include <linux/of.h>
 #include <linux/regmap.h>
@@ -64,10 +65,8 @@ static const struct of_device_id versatile_clcd_of_match[] = {
 #define INTEGRATOR_CLCD_LCDBIASEN	BIT(8)
 #define INTEGRATOR_CLCD_LCDBIASUP	BIT(9)
 #define INTEGRATOR_CLCD_LCDBIASDN	BIT(10)
-/* Bits 11,12,13 controls the LCD type */
-#define INTEGRATOR_CLCD_LCDMUX_MASK	(BIT(11)|BIT(12)|BIT(13))
+/* Bits 11,12,13 controls the LCD or VGA bridge type */
 #define INTEGRATOR_CLCD_LCDMUX_LCD24	BIT(11)
-#define INTEGRATOR_CLCD_LCDMUX_VGA565	BIT(12)
 #define INTEGRATOR_CLCD_LCDMUX_SHARP	(BIT(11)|BIT(12))
 #define INTEGRATOR_CLCD_LCDMUX_VGA555	BIT(13)
 #define INTEGRATOR_CLCD_LCDMUX_VGA24	(BIT(11)|BIT(12)|BIT(13))
@@ -82,16 +81,7 @@ static const struct of_device_id versatile_clcd_of_match[] = {
 /* 0 = 24bit VGA, 1 = 18bit VGA */
 #define INTEGRATOR_CLCD_LCD_N24BITEN	BIT(19)
 
-#define INTEGRATOR_CLCD_MASK		(INTEGRATOR_CLCD_LCDBIASEN | \
-					 INTEGRATOR_CLCD_LCDBIASUP | \
-					 INTEGRATOR_CLCD_LCDBIASDN | \
-					 INTEGRATOR_CLCD_LCDMUX_MASK | \
-					 INTEGRATOR_CLCD_LCD0_EN | \
-					 INTEGRATOR_CLCD_LCD1_EN | \
-					 INTEGRATOR_CLCD_LCD_STATIC1 | \
-					 INTEGRATOR_CLCD_LCD_STATIC2 | \
-					 INTEGRATOR_CLCD_LCD_STATIC | \
-					 INTEGRATOR_CLCD_LCD_N24BITEN)
+#define INTEGRATOR_CLCD_MASK		GENMASK(19, 8)
 
 static void pl111_integrator_enable(struct drm_device *drm, u32 format)
 {
@@ -106,11 +96,8 @@ static void pl111_integrator_enable(struct drm_device *drm, u32 format)
 	switch (format) {
 	case DRM_FORMAT_XBGR8888:
 	case DRM_FORMAT_XRGB8888:
-		break;
-	case DRM_FORMAT_BGR565:
-	case DRM_FORMAT_RGB565:
-		/* truecolor RGB565 */
-		val |= INTEGRATOR_CLCD_LCDMUX_VGA565;
+		/* 24bit formats */
+		val |= INTEGRATOR_CLCD_LCDMUX_VGA24;
 		break;
 	case DRM_FORMAT_XBGR1555:
 	case DRM_FORMAT_XRGB1555:
@@ -217,6 +204,88 @@ static void pl111_realview_clcd_enable(struct drm_device *drm, u32 format)
 			   SYS_CLCD_NLCDIOON | SYS_CLCD_PWR3V5SWITCH);
 }
 
+/* PL110 pixel formats for Integrator, vanilla PL110 */
+static const u32 pl110_integrator_pixel_formats[] = {
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_XRGB1555,
+};
+
+/* Extended PL110 pixel formats for Integrator and Versatile */
+static const u32 pl110_versatile_pixel_formats[] = {
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_BGR565, /* Uses external PLD */
+	DRM_FORMAT_RGB565, /* Uses external PLD */
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_XRGB1555,
+};
+
+static const u32 pl111_realview_pixel_formats[] = {
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_BGR565,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ABGR4444,
+	DRM_FORMAT_XBGR4444,
+	DRM_FORMAT_ARGB4444,
+	DRM_FORMAT_XRGB4444,
+};
+
+/*
+ * The Integrator variant is a PL110 with a bunch of broken, or not
+ * yet implemented features
+ */
+static const struct pl111_variant_data pl110_integrator = {
+	.name = "PL110 Integrator",
+	.is_pl110 = true,
+	.broken_clockdivider = true,
+	.broken_vblank = true,
+	.formats = pl110_integrator_pixel_formats,
+	.nformats = ARRAY_SIZE(pl110_integrator_pixel_formats),
+	.fb_bpp = 16,
+};
+
+/*
+ * This is the in-between PL110 variant found in the ARM Versatile,
+ * supporting RGB565/BGR565
+ */
+static const struct pl111_variant_data pl110_versatile = {
+	.name = "PL110 Versatile",
+	.is_pl110 = true,
+	.external_bgr = true,
+	.formats = pl110_versatile_pixel_formats,
+	.nformats = ARRAY_SIZE(pl110_versatile_pixel_formats),
+	.fb_bpp = 16,
+};
+
+/*
+ * RealView PL111 variant, the only real difference from the vanilla
+ * PL111 is that we select 16bpp framebuffer by default to be able
+ * to get 1024x768 without saturating the memory bus.
+ */
+static const struct pl111_variant_data pl111_realview = {
+	.name = "PL111 RealView",
+	.formats = pl111_realview_pixel_formats,
+	.nformats = ARRAY_SIZE(pl111_realview_pixel_formats),
+	.fb_bpp = 16,
+};
+
 int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 {
 	const struct of_device_id *clcd_id;
@@ -241,14 +310,24 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 	switch (versatile_clcd_type) {
 	case INTEGRATOR_CLCD_CM:
 		versatile_syscon_map = map;
+		priv->variant = &pl110_integrator;
 		priv->variant_display_enable = pl111_integrator_enable;
 		dev_info(dev, "set up callbacks for Integrator PL110\n");
 		break;
 	case VERSATILE_CLCD:
 		versatile_syscon_map = map;
+		/* This can do RGB565 with external PLD */
+		priv->variant = &pl110_versatile;
 		priv->variant_display_enable = pl111_versatile_enable;
 		priv->variant_display_disable = pl111_versatile_disable;
-		dev_info(dev, "set up callbacks for Versatile PL110+\n");
+		/*
+		 * The Versatile has a variant halfway between PL110
+		 * and PL111 where these two registers have already been
+		 * swapped.
+		 */
+		priv->ienb = CLCD_PL111_IENB;
+		priv->ctrl = CLCD_PL111_CNTL;
+		dev_info(dev, "set up callbacks for Versatile PL110\n");
 		break;
 	case REALVIEW_CLCD_EB:
 	case REALVIEW_CLCD_PB1176:
@@ -256,6 +335,7 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 	case REALVIEW_CLCD_PBA8:
 	case REALVIEW_CLCD_PBX:
 		versatile_syscon_map = map;
+		priv->variant = &pl111_realview;
 		priv->variant_display_enable = pl111_realview_clcd_enable;
 		priv->variant_display_disable = pl111_realview_clcd_disable;
 		dev_info(dev, "set up callbacks for RealView PL111\n");