diff --git a/Documentation/devicetree/bindings/display/msm/hdmi.txt b/Documentation/devicetree/bindings/display/msm/hdmi.txt
index b63f614e0c04..2ad578984fcf 100644
--- a/Documentation/devicetree/bindings/display/msm/hdmi.txt
+++ b/Documentation/devicetree/bindings/display/msm/hdmi.txt
@@ -14,17 +14,16 @@ Required properties:
 - power-domains: Should be <&mmcc MDSS_GDSC>.
 - clocks: device clocks
   See ../clocks/clock-bindings.txt for details.
-- qcom,hdmi-tx-ddc-clk-gpio: ddc clk pin
-- qcom,hdmi-tx-ddc-data-gpio: ddc data pin
-- qcom,hdmi-tx-hpd-gpio: hpd pin
 - core-vdda-supply: phandle to supply regulator
 - hdmi-mux-supply: phandle to mux regulator
 - phys: the phandle for the HDMI PHY device
 - phy-names: the name of the corresponding PHY device
 
 Optional properties:
-- qcom,hdmi-tx-mux-en-gpio: hdmi mux enable pin
-- qcom,hdmi-tx-mux-sel-gpio: hdmi mux select pin
+- hpd-gpios: hpd pin
+- qcom,hdmi-tx-mux-en-gpios: hdmi mux enable pin
+- qcom,hdmi-tx-mux-sel-gpios: hdmi mux select pin
+- qcom,hdmi-tx-mux-lpm-gpios: hdmi mux lpm pin
 - power-domains: reference to the power domain(s), if available.
 - pinctrl-names: the pin control state names; should contain "default"
 - pinctrl-0: the default pinctrl state (active)
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 7c7a0314a756..d96b2b6898a3 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -11,6 +11,7 @@ config DRM_MSM
 	select TMPFS
 	select QCOM_SCM
 	select SND_SOC_HDMI_CODEC if SND_SOC
+	select SYNC_FILE
 	default y
 	help
 	  DRM/KMS driver for MSM/snapdragon.
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 973720792236..a968cad509c2 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -422,11 +422,28 @@ static const struct {
 
 static int msm_hdmi_get_gpio(struct device_node *of_node, const char *name)
 {
-	int gpio = of_get_named_gpio(of_node, name, 0);
+	int gpio;
+
+	/* try with the gpio names as in the table (downstream bindings) */
+	gpio = of_get_named_gpio(of_node, name, 0);
 	if (gpio < 0) {
 		char name2[32];
-		snprintf(name2, sizeof(name2), "%s-gpio", name);
+
+		/* try with the gpio names as in the upstream bindings */
+		snprintf(name2, sizeof(name2), "%s-gpios", name);
 		gpio = of_get_named_gpio(of_node, name2, 0);
+		if (gpio < 0) {
+			char name3[32];
+
+			/*
+			 * try again after stripping out the "qcom,hdmi-tx"
+			 * prefix. This is mainly to match "hpd-gpios" used
+			 * in the upstream bindings
+			 */
+			if (sscanf(name2, "qcom,hdmi-tx-%s", name3))
+				gpio = of_get_named_gpio(of_node, name3, 0);
+		}
+
 		if (gpio < 0) {
 			DBG("failed to get gpio: %s (%d)", name, gpio);
 			gpio = -1;
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
index de9007e72f4e..73e20219d431 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
@@ -243,7 +243,6 @@ void msm_hdmi_i2c_destroy(struct i2c_adapter *i2c)
 
 struct i2c_adapter *msm_hdmi_i2c_init(struct hdmi *hdmi)
 {
-	struct drm_device *dev = hdmi->dev;
 	struct hdmi_i2c_adapter *hdmi_i2c;
 	struct i2c_adapter *i2c = NULL;
 	int ret;
@@ -267,10 +266,8 @@ struct i2c_adapter *msm_hdmi_i2c_init(struct hdmi *hdmi)
 	i2c->algo = &msm_hdmi_i2c_algorithm;
 
 	ret = i2c_add_adapter(i2c);
-	if (ret) {
-		dev_err(dev->dev, "failed to register hdmi i2c: %d\n", ret);
+	if (ret)
 		goto fail;
-	}
 
 	return i2c;
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
index 7b39e89fbc2b..571a91ee9607 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
@@ -228,18 +228,21 @@ static struct device_node *mdp4_detect_lcdc_panel(struct drm_device *dev)
 	struct device_node *endpoint, *panel_node;
 	struct device_node *np = dev->dev->of_node;
 
-	endpoint = of_graph_get_next_endpoint(np, NULL);
+	/*
+	 * LVDS/LCDC is the first port described in the list of ports in the
+	 * MDP4 DT node.
+	 */
+	endpoint = of_graph_get_endpoint_by_regs(np, 0, -1);
 	if (!endpoint) {
-		DBG("no endpoint in MDP4 to fetch LVDS panel\n");
+		DBG("no LVDS remote endpoint\n");
 		return NULL;
 	}
 
-	/* don't proceed if we have an endpoint but no panel_node tied to it */
 	panel_node = of_graph_get_remote_port_parent(endpoint);
 	if (!panel_node) {
-		dev_err(dev->dev, "no valid panel node\n");
+		DBG("no valid panel node in LVDS endpoint\n");
 		of_node_put(endpoint);
-		return ERR_PTR(-ENODEV);
+		return NULL;
 	}
 
 	of_node_put(endpoint);
@@ -262,14 +265,12 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms,
 	switch (intf_type) {
 	case DRM_MODE_ENCODER_LVDS:
 		/*
-		 * bail out early if:
-		 * - there is no panel node (no need to initialize lcdc
-		 *   encoder and lvds connector), or
-		 * - panel node is a bad pointer
+		 * bail out early if there is no panel node (no need to
+		 * initialize LCDC encoder and LVDS connector)
 		 */
 		panel_node = mdp4_detect_lcdc_panel(dev);
-		if (IS_ERR_OR_NULL(panel_node))
-			return PTR_ERR(panel_node);
+		if (!panel_node)
+			return 0;
 
 		encoder = mdp4_lcdc_encoder_init(dev, panel_node);
 		if (IS_ERR(encoder)) {
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
index bc3d8e719c6c..a06b064f86c1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
@@ -93,7 +93,7 @@ static const struct drm_encoder_funcs mdp4_lcdc_encoder_funcs = {
 };
 
 /* this should probably be a helper: */
-struct drm_connector *get_connector(struct drm_encoder *encoder)
+static struct drm_connector *get_connector(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_connector *connector;
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
index 7c9626d92019..3903dbcda763 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
@@ -81,7 +81,7 @@ static void mdp4_plane_install_properties(struct drm_plane *plane,
 	// XXX
 }
 
-int mdp4_plane_set_property(struct drm_plane *plane,
+static int mdp4_plane_set_property(struct drm_plane *plane,
 		struct drm_property *property, uint64_t val)
 {
 	// XXX
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index ba8f43278a44..951c002b05df 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -78,12 +78,12 @@ static void mdp5_plane_install_rotation_property(struct drm_device *dev,
 	if (!dev->mode_config.rotation_property)
 		dev->mode_config.rotation_property =
 			drm_mode_create_rotation_property(dev,
-			DRM_REFLECT_X | DRM_REFLECT_Y);
+				DRM_ROTATE_0 | DRM_REFLECT_X | DRM_REFLECT_Y);
 
 	if (dev->mode_config.rotation_property)
 		drm_object_attach_property(&plane->base,
 			dev->mode_config.rotation_property,
-			0);
+			DRM_ROTATE_0);
 }
 
 /* helper to install properties which are common to planes and crtcs */
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 042bde48200d..fb5c0b0a7594 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -26,9 +26,10 @@
  * MSM driver version:
  * - 1.0.0 - initial interface
  * - 1.1.0 - adds madvise, and support for submits with > 4 cmd buffers
+ * - 1.2.0 - adds explicit fence support for submit ioctl
  */
 #define MSM_VERSION_MAJOR	1
-#define MSM_VERSION_MINOR	1
+#define MSM_VERSION_MINOR	2
 #define MSM_VERSION_PATCHLEVEL	0
 
 static void msm_fb_output_poll_changed(struct drm_device *dev)
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 6cd4af443139..0a9b5580b2e9 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -584,18 +584,16 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	bool write = !!(op & MSM_PREP_WRITE);
+	unsigned long remain =
+		op & MSM_PREP_NOSYNC ? 0 : timeout_to_jiffies(timeout);
+	long ret;
 
-	if (op & MSM_PREP_NOSYNC) {
-		if (!reservation_object_test_signaled_rcu(msm_obj->resv, write))
-			return -EBUSY;
-	} else {
-		int ret;
-
-		ret = reservation_object_wait_timeout_rcu(msm_obj->resv, write,
-				true, timeout_to_jiffies(timeout));
-		if (ret <= 0)
-			return ret == 0 ? -ETIMEDOUT : ret;
-	}
+	ret = reservation_object_wait_timeout_rcu(msm_obj->resv, write,
+						  true,  remain);
+	if (ret == 0)
+		return remain == 0 ? -EBUSY : -ETIMEDOUT;
+	else if (ret < 0)
+		return ret;
 
 	/* TODO cache maintenance */
 
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 9766f9ae4b7d..3ac14cd1e5b9 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -15,6 +15,8 @@
  * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
+#include <linux/sync_file.h>
+
 #include "msm_drv.h"
 #include "msm_gpu.h"
 #include "msm_gem.h"
@@ -361,6 +363,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct msm_file_private *ctx = file->driver_priv;
 	struct msm_gem_submit *submit;
 	struct msm_gpu *gpu = priv->gpu;
+	struct fence *in_fence = NULL;
+	struct sync_file *sync_file = NULL;
+	int out_fence_fd = -1;
 	unsigned i;
 	int ret;
 
@@ -370,13 +375,24 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	/* for now, we just have 3d pipe.. eventually this would need to
 	 * be more clever to dispatch to appropriate gpu module:
 	 */
-	if (args->pipe != MSM_PIPE_3D0)
+	if (MSM_PIPE_ID(args->flags) != MSM_PIPE_3D0)
+		return -EINVAL;
+
+	if (MSM_PIPE_FLAGS(args->flags) & ~MSM_SUBMIT_FLAGS)
 		return -EINVAL;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
 		return ret;
 
+	if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
+		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
+		if (out_fence_fd < 0) {
+			ret = out_fence_fd;
+			goto out_unlock;
+		}
+	}
+
 	submit = submit_create(dev, gpu, args->nr_bos, args->nr_cmds);
 	if (!submit) {
 		ret = -ENOMEM;
@@ -391,9 +407,32 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (ret)
 		goto out;
 
-	ret = submit_fence_sync(submit);
-	if (ret)
-		goto out;
+	if (args->flags & MSM_SUBMIT_FENCE_FD_IN) {
+		in_fence = sync_file_get_fence(args->fence_fd);
+
+		if (!in_fence) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		/* TODO if we get an array-fence due to userspace merging multiple
+		 * fences, we need a way to determine if all the backing fences
+		 * are from our own context..
+		 */
+
+		if (in_fence->context != gpu->fctx->context) {
+			ret = fence_wait(in_fence, true);
+			if (ret)
+				goto out;
+		}
+
+	}
+
+	if (!(args->fence & MSM_SUBMIT_NO_IMPLICIT)) {
+		ret = submit_fence_sync(submit);
+		if (ret)
+			goto out;
+	}
 
 	ret = submit_pin_objects(submit);
 	if (ret)
@@ -459,15 +498,39 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 
 	submit->nr_cmds = i;
 
-	ret = msm_gpu_submit(gpu, submit, ctx);
+	submit->fence = msm_fence_alloc(gpu->fctx);
+	if (IS_ERR(submit->fence)) {
+		ret = PTR_ERR(submit->fence);
+		submit->fence = NULL;
+		goto out;
+	}
+
+	if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
+		sync_file = sync_file_create(submit->fence);
+		if (!sync_file) {
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+
+	msm_gpu_submit(gpu, submit, ctx);
 
 	args->fence = submit->fence->seqno;
 
+	if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) {
+		fd_install(out_fence_fd, sync_file->file);
+		args->fence_fd = out_fence_fd;
+	}
+
 out:
+	if (in_fence)
+		fence_put(in_fence);
 	submit_cleanup(submit);
 	if (ret)
 		msm_gem_submit_free(submit);
 out_unlock:
+	if (ret && (out_fence_fd >= 0))
+		put_unused_fd(out_fence_fd);
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
 }
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 36ed53e661fe..5bb09838b5ae 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -509,22 +509,15 @@ void msm_gpu_retire(struct msm_gpu *gpu)
 }
 
 /* add bo's to gpu's ring, and kick gpu: */
-int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
+void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		struct msm_file_private *ctx)
 {
 	struct drm_device *dev = gpu->dev;
 	struct msm_drm_private *priv = dev->dev_private;
-	int i, ret;
+	int i;
 
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
 
-	submit->fence = msm_fence_alloc(gpu->fctx);
-	if (IS_ERR(submit->fence)) {
-		ret = PTR_ERR(submit->fence);
-		submit->fence = NULL;
-		return ret;
-	}
-
 	inactive_cancel(gpu);
 
 	list_add_tail(&submit->node, &gpu->submit_list);
@@ -557,8 +550,6 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 	priv->lastctx = ctx;
 
 	hangcheck_timer_reset(gpu);
-
-	return 0;
 }
 
 /*
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index c9022837a1a4..d61d98a6e047 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -163,7 +163,7 @@ int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
 		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs);
 
 void msm_gpu_retire(struct msm_gpu *gpu);
-int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
+void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		struct msm_file_private *ctx);
 
 int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
index 49f778de8e06..8c51e8a0df89 100644
--- a/include/uapi/drm/msm_drm.h
+++ b/include/uapi/drm/msm_drm.h
@@ -42,6 +42,15 @@ extern "C" {
 #define MSM_PIPE_2D1         0x02
 #define MSM_PIPE_3D0         0x10
 
+/* The pipe-id just uses the lower bits, so can be OR'd with flags in
+ * the upper 16 bits (which could be extended further, if needed, maybe
+ * we extend/overload the pipe-id some day to deal with multiple rings,
+ * but even then I don't think we need the full lower 16 bits).
+ */
+#define MSM_PIPE_ID_MASK     0xffff
+#define MSM_PIPE_ID(x)       ((x) & MSM_PIPE_ID_MASK)
+#define MSM_PIPE_FLAGS(x)    ((x) & ~MSM_PIPE_ID_MASK)
+
 /* timeouts are specified in clock-monotonic absolute times (to simplify
  * restarting interrupted ioctls).  The following struct is logically the
  * same as 'struct timespec' but 32/64b ABI safe.
@@ -175,17 +184,28 @@ struct drm_msm_gem_submit_bo {
 	__u64 presumed;       /* in/out, presumed buffer address */
 };
 
+/* Valid submit ioctl flags: */
+#define MSM_SUBMIT_NO_IMPLICIT   0x80000000 /* disable implicit sync */
+#define MSM_SUBMIT_FENCE_FD_IN   0x40000000 /* enable input fence_fd */
+#define MSM_SUBMIT_FENCE_FD_OUT  0x20000000 /* enable output fence_fd */
+#define MSM_SUBMIT_FLAGS                ( \
+		MSM_SUBMIT_NO_IMPLICIT   | \
+		MSM_SUBMIT_FENCE_FD_IN   | \
+		MSM_SUBMIT_FENCE_FD_OUT  | \
+		0)
+
 /* Each cmdstream submit consists of a table of buffers involved, and
  * one or more cmdstream buffers.  This allows for conditional execution
  * (context-restore), and IB buffers needed for per tile/bin draw cmds.
  */
 struct drm_msm_gem_submit {
-	__u32 pipe;           /* in, MSM_PIPE_x */
+	__u32 flags;          /* MSM_PIPE_x | MSM_SUBMIT_x */
 	__u32 fence;          /* out */
 	__u32 nr_bos;         /* in, number of submit_bo's */
 	__u32 nr_cmds;        /* in, number of submit_cmd's */
 	__u64 __user bos;     /* in, ptr to array of submit_bo's */
 	__u64 __user cmds;    /* in, ptr to array of submit_cmd's */
+	__s32 fence_fd;       /* in/out fence fd (see MSM_SUBMIT_FENCE_FD_IN/OUT) */
 };
 
 /* The normal way to synchronize with the GPU is just to CPU_PREP on