Merge tag 'drm-next-du-20170803' of git://linuxtv.org/pinchartl/media into drm-next

rcar-du updates, contains vsp1 updates as well.

* tag 'drm-next-du-20170803' of git://linuxtv.org/pinchartl/media: (24 commits)
  drm: rcar-du: Use new iterator macros
  drm: rcar-du: Repair vblank for DRM page flips using the VSP
  drm: rcar-du: Fix race condition when disabling planes at CRTC stop
  drm: rcar-du: Wait for flip completion instead of vblank in commit tail
  drm: rcar-du: Use the VBK interrupt for vblank events
  drm: rcar-du: Add HDMI outputs to R8A7796 device description
  drm: rcar-du: Remove an unneeded NULL check
  drm: rcar-du: Setup planes before enabling CRTC to avoid flicker
  drm: rcar-du: Configure DPAD0 routing through last group on Gen3
  drm: rcar-du: Restrict DPLL duty cycle workaround to H3 ES1.x
  drm: rcar-du: Support multiple sources from the same VSP
  drm: rcar-du: Fix comments to comply with the kernel coding style
  drm: rcar-du: Use of_graph_get_remote_endpoint()
  v4l: vsp1: Add support for header display lists in continuous mode
  v4l: vsp1: Add support for multiple DRM pipelines
  v4l: vsp1: Add support for multiple LIF instances
  v4l: vsp1: Add support for new VSP2-BS, VSP2-DL and VSP2-D instances
  v4l: vsp1: Add support for the BRS entity
  v4l: vsp1: Add pipe index argument to the VSP-DU API
  v4l: vsp1: Don't create links for DRM pipeline
  ...

drivers/gpu/drm/rcar-du/rcar_du_crtc.c

@@ -13,6 +13,7 @@
 
 #include <linux/clk.h>
 #include <linux/mutex.h>
+#include <linux/sys_soc.h>
 
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
@@ -129,10 +130,8 @@ static void rcar_du_dpll_divider(struct rcar_du_crtc *rcrtc,
 			for (fdpll = 1; fdpll < 32; fdpll++) {
 				unsigned long output;
 
-				/* 1/2 (FRQSEL=1) for duty rate 50% */
 				output = input * (n + 1) / (m + 1)
-				       / (fdpll + 1) / 2;
-
+				       / (fdpll + 1);
 				if (output >= 400000000)
 					continue;
 
@@ -158,6 +157,11 @@ done:
 		 best_diff);
 }
 
+static const struct soc_device_attribute rcar_du_r8a7795_es1[] = {
+	{ .soc_id = "r8a7795", .revision = "ES1.*" },
+	{ /* sentinel */ }
+};
+
 static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 {
 	const struct drm_display_mode *mode = &rcrtc->crtc.state->adjusted_mode;
@@ -168,7 +172,8 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 	u32 escr;
 	u32 div;
 
-	/* Compute the clock divisor and select the internal or external dot
+	/*
+	 * Compute the clock divisor and select the internal or external dot
 	 * clock based on the requested frequency.
 	 */
 	clk = clk_get_rate(rcrtc->clock);
@@ -185,7 +190,20 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 
 		extclk = clk_get_rate(rcrtc->extclock);
 		if (rcdu->info->dpll_ch & (1 << rcrtc->index)) {
-			rcar_du_dpll_divider(rcrtc, &dpll, extclk, mode_clock);
+			unsigned long target = mode_clock;
+
+			/*
+			 * The H3 ES1.x exhibits dot clock duty cycle stability
+			 * issues. We can work around them by configuring the
+			 * DPLL to twice the desired frequency, coupled with a
+			 * /2 post-divider. This isn't needed on other SoCs and
+			 * breaks HDMI output on M3-W for a currently unknown
+			 * reason, so restrict the workaround to H3 ES1.x.
+			 */
+			if (soc_device_match(rcar_du_r8a7795_es1))
+				target *= 2;
+
+			rcar_du_dpll_divider(rcrtc, &dpll, extclk, target);
 			extclk = dpll.output;
 		}
 
@@ -197,8 +215,6 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 
 		if (abs((long)extrate - (long)mode_clock) <
 		    abs((long)rate - (long)mode_clock)) {
-			dev_dbg(rcrtc->group->dev->dev,
-				"crtc%u: using external clock\n", rcrtc->index);
 
 			if (rcdu->info->dpll_ch & (1 << rcrtc->index)) {
 				u32 dpllcr = DPLLCR_CODE | DPLLCR_CLKE
@@ -215,12 +231,14 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 
 				rcar_du_group_write(rcrtc->group, DPLLCR,
 						    dpllcr);
-
-				escr = ESCR_DCLKSEL_DCLKIN | 1;
-			} else {
-				escr = ESCR_DCLKSEL_DCLKIN | extdiv;
 			}
+
+			escr = ESCR_DCLKSEL_DCLKIN | extdiv;
 		}
+
+		dev_dbg(rcrtc->group->dev->dev,
+			"mode clock %lu extrate %lu rate %lu ESCR 0x%08x\n",
+			mode_clock, extrate, rate, escr);
 	}
 
 	rcar_du_group_write(rcrtc->group, rcrtc->index % 2 ? ESCR2 : ESCR,
@@ -261,12 +279,14 @@ void rcar_du_crtc_route_output(struct drm_crtc *crtc,
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 	struct rcar_du_device *rcdu = rcrtc->group->dev;
 
-	/* Store the route from the CRTC output to the DU output. The DU will be
+	/*
+	 * Store the route from the CRTC output to the DU output. The DU will be
 	 * configured when starting the CRTC.
 	 */
 	rcrtc->outputs |= BIT(output);
 
-	/* Store RGB routing to DPAD0, the hardware will be configured when
+	/*
+	 * Store RGB routing to DPAD0, the hardware will be configured when
 	 * starting the CRTC.
 	 */
 	if (output == RCAR_DU_OUTPUT_DPAD0)
@@ -342,7 +362,8 @@ static void rcar_du_crtc_update_planes(struct rcar_du_crtc *rcrtc)
 		}
 	}
 
-	/* Update the planes to display timing and dot clock generator
+	/*
+	 * Update the planes to display timing and dot clock generator
 	 * associations.
 	 *
 	 * Updating the DPTSR register requires restarting the CRTC group,
@@ -431,14 +452,8 @@ static void rcar_du_crtc_wait_page_flip(struct rcar_du_crtc *rcrtc)
  * Start/Stop and Suspend/Resume
  */
 
-static void rcar_du_crtc_start(struct rcar_du_crtc *rcrtc)
+static void rcar_du_crtc_setup(struct rcar_du_crtc *rcrtc)
 {
-	struct drm_crtc *crtc = &rcrtc->crtc;
-	bool interlaced;
-
-	if (rcrtc->started)
-		return;
-
 	/* Set display off and background to black */
 	rcar_du_crtc_write(rcrtc, DOOR, DOOR_RGB(0, 0, 0));
 	rcar_du_crtc_write(rcrtc, BPOR, BPOR_RGB(0, 0, 0));
@@ -450,7 +465,20 @@ static void rcar_du_crtc_start(struct rcar_du_crtc *rcrtc)
 	/* Start with all planes disabled. */
 	rcar_du_group_write(rcrtc->group, rcrtc->index % 2 ? DS2PR : DS1PR, 0);
 
-	/* Select master sync mode. This enables display operation in master
+	/* Enable the VSP compositor. */
+	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
+		rcar_du_vsp_enable(rcrtc);
+
+	/* Turn vertical blanking interrupt reporting on. */
+	drm_crtc_vblank_on(&rcrtc->crtc);
+}
+
+static void rcar_du_crtc_start(struct rcar_du_crtc *rcrtc)
+{
+	bool interlaced;
+
+	/*
+	 * Select master sync mode. This enables display operation in master
 	 * sync mode (with the HSYNC and VSYNC signals configured as outputs and
 	 * actively driven).
 	 */
@@ -460,38 +488,56 @@ static void rcar_du_crtc_start(struct rcar_du_crtc *rcrtc)
 			     DSYSR_TVM_MASTER);
 
 	rcar_du_group_start_stop(rcrtc->group, true);
+}
 
-	/* Enable the VSP compositor. */
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
-		rcar_du_vsp_enable(rcrtc);
+static void rcar_du_crtc_disable_planes(struct rcar_du_crtc *rcrtc)
+{
+	struct rcar_du_device *rcdu = rcrtc->group->dev;
+	struct drm_crtc *crtc = &rcrtc->crtc;
+	u32 status;
 
-	/* Turn vertical blanking interrupt reporting back on. */
-	drm_crtc_vblank_on(crtc);
+	/* Make sure vblank interrupts are enabled. */
+	drm_crtc_vblank_get(crtc);
 
-	rcrtc->started = true;
+	/*
+	 * Disable planes and calculate how many vertical blanking interrupts we
+	 * have to wait for. If a vertical blanking interrupt has been triggered
+	 * but not processed yet, we don't know whether it occurred before or
+	 * after the planes got disabled. We thus have to wait for two vblank
+	 * interrupts in that case.
+	 */
+	spin_lock_irq(&rcrtc->vblank_lock);
+	rcar_du_group_write(rcrtc->group, rcrtc->index % 2 ? DS2PR : DS1PR, 0);
+	status = rcar_du_crtc_read(rcrtc, DSSR);
+	rcrtc->vblank_count = status & DSSR_VBK ? 2 : 1;
+	spin_unlock_irq(&rcrtc->vblank_lock);
+
+	if (!wait_event_timeout(rcrtc->vblank_wait, rcrtc->vblank_count == 0,
+				msecs_to_jiffies(100)))
+		dev_warn(rcdu->dev, "vertical blanking timeout\n");
+
+	drm_crtc_vblank_put(crtc);
 }
 
 static void rcar_du_crtc_stop(struct rcar_du_crtc *rcrtc)
 {
 	struct drm_crtc *crtc = &rcrtc->crtc;
 
-	if (!rcrtc->started)
-		return;
-
-	/* Disable all planes and wait for the change to take effect. This is
-	 * required as the DSnPR registers are updated on vblank, and no vblank
-	 * will occur once the CRTC is stopped. Disabling planes when starting
-	 * the CRTC thus wouldn't be enough as it would start scanning out
-	 * immediately from old frame buffers until the next vblank.
+	/*
+	 * Disable all planes and wait for the change to take effect. This is
+	 * required as the plane enable registers are updated on vblank, and no
+	 * vblank will occur once the CRTC is stopped. Disabling planes when
+	 * starting the CRTC thus wouldn't be enough as it would start scanning
+	 * out immediately from old frame buffers until the next vblank.
 	 *
 	 * This increases the CRTC stop delay, especially when multiple CRTCs
 	 * are stopped in one operation as we now wait for one vblank per CRTC.
 	 * Whether this can be improved needs to be researched.
 	 */
-	rcar_du_group_write(rcrtc->group, rcrtc->index % 2 ? DS2PR : DS1PR, 0);
-	drm_crtc_wait_one_vblank(crtc);
+	rcar_du_crtc_disable_planes(rcrtc);
 
-	/* Disable vertical blanking interrupt reporting. We first need to wait
+	/*
+	 * Disable vertical blanking interrupt reporting. We first need to wait
 	 * for page flip completion before stopping the CRTC as userspace
 	 * expects page flips to eventually complete.
 	 */
@@ -502,14 +548,13 @@ static void rcar_du_crtc_stop(struct rcar_du_crtc *rcrtc)
 	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_disable(rcrtc);
 
-	/* Select switch sync mode. This stops display operation and configures
+	/*
+	 * Select switch sync mode. This stops display operation and configures
 	 * the HSYNC and VSYNC signals as inputs.
 	 */
 	rcar_du_crtc_clr_set(rcrtc, DSYSR, DSYSR_TVM_MASK, DSYSR_TVM_SWITCH);
 
 	rcar_du_group_start_stop(rcrtc->group, false);
-
-	rcrtc->started = false;
 }
 
 void rcar_du_crtc_suspend(struct rcar_du_crtc *rcrtc)
@@ -529,12 +574,10 @@ void rcar_du_crtc_resume(struct rcar_du_crtc *rcrtc)
 		return;
 
 	rcar_du_crtc_get(rcrtc);
-	rcar_du_crtc_start(rcrtc);
+	rcar_du_crtc_setup(rcrtc);
 
 	/* Commit the planes state. */
-	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE)) {
-		rcar_du_vsp_enable(rcrtc);
-	} else {
+	if (!rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE)) {
 		for (i = 0; i < rcrtc->group->num_planes; ++i) {
 			struct rcar_du_plane *plane = &rcrtc->group->planes[i];
 
@@ -546,6 +589,7 @@ void rcar_du_crtc_resume(struct rcar_du_crtc *rcrtc)
 	}
 
 	rcar_du_crtc_update_planes(rcrtc);
+	rcar_du_crtc_start(rcrtc);
 }
 
 /* -----------------------------------------------------------------------------
@@ -557,7 +601,16 @@ static void rcar_du_crtc_atomic_enable(struct drm_crtc *crtc,
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 
-	rcar_du_crtc_get(rcrtc);
+	/*
+	 * If the CRTC has already been setup by the .atomic_begin() handler we
+	 * can skip the setup stage.
+	 */
+	if (!rcrtc->initialized) {
+		rcar_du_crtc_get(rcrtc);
+		rcar_du_crtc_setup(rcrtc);
+		rcrtc->initialized = true;
+	}
+
 	rcar_du_crtc_start(rcrtc);
 }
 
@@ -576,6 +629,7 @@ static void rcar_du_crtc_atomic_disable(struct drm_crtc *crtc,
 	}
 	spin_unlock_irq(&crtc->dev->event_lock);
 
+	rcrtc->initialized = false;
 	rcrtc->outputs = 0;
 }
 
@@ -584,6 +638,19 @@ static void rcar_du_crtc_atomic_begin(struct drm_crtc *crtc,
 {
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 
+	WARN_ON(!crtc->state->enable);
+
+	/*
+	 * If a mode set is in progress we can be called with the CRTC disabled.
+	 * We then need to first setup the CRTC in order to configure planes.
+	 * The .atomic_enable() handler will notice and skip the CRTC setup.
+	 */
+	if (!rcrtc->initialized) {
+		rcar_du_crtc_get(rcrtc);
+		rcar_du_crtc_setup(rcrtc);
+		rcrtc->initialized = true;
+	}
+
 	if (rcar_du_has(rcrtc->group->dev, RCAR_DU_FEATURE_VSP1_SOURCE))
 		rcar_du_vsp_atomic_begin(rcrtc);
 }
@@ -623,6 +690,7 @@ static int rcar_du_crtc_enable_vblank(struct drm_crtc *crtc)
 
 	rcar_du_crtc_write(rcrtc, DSRCR, DSRCR_VBCL);
 	rcar_du_crtc_set(rcrtc, DIER, DIER_VBE);
+	rcrtc->vblank_enable = true;
 
 	return 0;
 }
@@ -632,6 +700,7 @@ static void rcar_du_crtc_disable_vblank(struct drm_crtc *crtc)
 	struct rcar_du_crtc *rcrtc = to_rcar_crtc(crtc);
 
 	rcar_du_crtc_clr(rcrtc, DIER, DIER_VBE);
+	rcrtc->vblank_enable = false;
 }
 
 static const struct drm_crtc_funcs crtc_funcs = {
@@ -656,14 +725,30 @@ static irqreturn_t rcar_du_crtc_irq(int irq, void *arg)
 	irqreturn_t ret = IRQ_NONE;
 	u32 status;
 
+	spin_lock(&rcrtc->vblank_lock);
+
 	status = rcar_du_crtc_read(rcrtc, DSSR);
 	rcar_du_crtc_write(rcrtc, DSRCR, status & DSRCR_MASK);
 
-	if (status & DSSR_FRM) {
-		drm_crtc_handle_vblank(&rcrtc->crtc);
+	if (status & DSSR_VBK) {
+		/*
+		 * Wake up the vblank wait if the counter reaches 0. This must
+		 * be protected by the vblank_lock to avoid races in
+		 * rcar_du_crtc_disable_planes().
+		 */
+		if (rcrtc->vblank_count) {
+			if (--rcrtc->vblank_count == 0)
+				wake_up(&rcrtc->vblank_wait);
+		}
+	}
 
-		if (rcdu->info->gen < 3)
+	spin_unlock(&rcrtc->vblank_lock);
+
+	if (status & DSSR_VBK) {
+		if (rcdu->info->gen < 3) {
+			drm_crtc_handle_vblank(&rcrtc->crtc);
 			rcar_du_crtc_finish_page_flip(rcrtc);
+		}
 
 		ret = IRQ_HANDLED;
 	}
@@ -717,13 +802,15 @@ int rcar_du_crtc_create(struct rcar_du_group *rgrp, unsigned int index)
 	}
 
 	init_waitqueue_head(&rcrtc->flip_wait);
+	init_waitqueue_head(&rcrtc->vblank_wait);
+	spin_lock_init(&rcrtc->vblank_lock);
 
 	rcrtc->group = rgrp;
 	rcrtc->mmio_offset = mmio_offsets[index];
 	rcrtc->index = index;
 
 	if (rcar_du_has(rcdu, RCAR_DU_FEATURE_VSP1_SOURCE))
-		primary = &rcrtc->vsp->planes[0].plane;
+		primary = &rcrtc->vsp->planes[rcrtc->vsp_pipe].plane;
 	else
 		primary = &rgrp->planes[index % 2].plane;
 

drivers/gpu/drm/rcar-du/rcar_du_crtc.h

@@ -15,6 +15,7 @@
 #define __RCAR_DU_CRTC_H__
 
 #include <linux/mutex.h>
+#include <linux/spinlock.h>
 #include <linux/wait.h>
 
 #include <drm/drmP.h>
@@ -30,11 +31,17 @@ struct rcar_du_vsp;
  * @extclock: external pixel dot clock (optional)
  * @mmio_offset: offset of the CRTC registers in the DU MMIO block
  * @index: CRTC software and hardware index
- * @started: whether the CRTC has been started and is running
+ * @initialized: whether the CRTC has been initialized and clocks enabled
+ * @vblank_enable: whether vblank events are enabled on this CRTC
  * @event: event to post when the pending page flip completes
  * @flip_wait: wait queue used to signal page flip completion
+ * @vblank_lock: protects vblank_wait and vblank_count
+ * @vblank_wait: wait queue used to signal vertical blanking
+ * @vblank_count: number of vertical blanking interrupts to wait for
  * @outputs: bitmask of the outputs (enum rcar_du_output) driven by this CRTC
  * @group: CRTC group this CRTC belongs to
+ * @vsp: VSP feeding video to this CRTC
+ * @vsp_pipe: index of the VSP pipeline feeding video to this CRTC
  */
 struct rcar_du_crtc {
 	struct drm_crtc crtc;
@@ -43,15 +50,21 @@ struct rcar_du_crtc {
 	struct clk *extclock;
 	unsigned int mmio_offset;
 	unsigned int index;
-	bool started;
+	bool initialized;
 
+	bool vblank_enable;
 	struct drm_pending_vblank_event *event;
 	wait_queue_head_t flip_wait;
 
+	spinlock_t vblank_lock;
+	wait_queue_head_t vblank_wait;
+	unsigned int vblank_count;
+
 	unsigned int outputs;
 
 	struct rcar_du_group *group;
 	struct rcar_du_vsp *vsp;
+	unsigned int vsp_pipe;
 };
 
 #define to_rcar_crtc(c)	container_of(c, struct rcar_du_crtc, crtc)

drivers/gpu/drm/rcar-du/rcar_du_drv.c

@@ -39,7 +39,8 @@ static const struct rcar_du_device_info rcar_du_r8a7779_info = {
 	.features = 0,
 	.num_crtcs = 2,
 	.routes = {
-		/* R8A7779 has two RGB outputs and one (currently unsupported)
+		/*
+		 * R8A7779 has two RGB outputs and one (currently unsupported)
 		 * TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -61,7 +62,8 @@ static const struct rcar_du_device_info rcar_du_r8a7790_info = {
 	.quirks = RCAR_DU_QUIRK_ALIGN_128B | RCAR_DU_QUIRK_LVDS_LANES,
 	.num_crtcs = 3,
 	.routes = {
-		/* R8A7790 has one RGB output, two LVDS outputs and one
+		/*
+		 * R8A7790 has one RGB output, two LVDS outputs and one
 		 * (currently unsupported) TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -87,7 +89,8 @@ static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 		  | RCAR_DU_FEATURE_EXT_CTRL_REGS,
 	.num_crtcs = 2,
 	.routes = {
-		/* R8A779[13] has one RGB output, one LVDS output and one
+		/*
+		 * R8A779[13] has one RGB output, one LVDS output and one
 		 * (currently unsupported) TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -127,7 +130,8 @@ static const struct rcar_du_device_info rcar_du_r8a7794_info = {
 		  | RCAR_DU_FEATURE_EXT_CTRL_REGS,
 	.num_crtcs = 2,
 	.routes = {
-		/* R8A7794 has two RGB outputs and one (currently unsupported)
+		/*
+		 * R8A7794 has two RGB outputs and one (currently unsupported)
 		 * TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -149,7 +153,8 @@ static const struct rcar_du_device_info rcar_du_r8a7795_info = {
 		  | RCAR_DU_FEATURE_VSP1_SOURCE,
 	.num_crtcs = 4,
 	.routes = {
-		/* R8A7795 has one RGB output, two HDMI outputs and one
+		/*
+		 * R8A7795 has one RGB output, two HDMI outputs and one
 		 * LVDS output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -180,19 +185,25 @@ static const struct rcar_du_device_info rcar_du_r8a7796_info = {
 		  | RCAR_DU_FEATURE_VSP1_SOURCE,
 	.num_crtcs = 3,
 	.routes = {
-		/* R8A7796 has one RGB output, one LVDS output and one
-		 * (currently unsupported) HDMI output.
+		/*
+		 * R8A7796 has one RGB output, one LVDS output and one HDMI
+		 * output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
 			.possible_crtcs = BIT(2),
 			.port = 0,
 		},
+		[RCAR_DU_OUTPUT_HDMI0] = {
+			.possible_crtcs = BIT(1),
+			.port = 1,
+		},
 		[RCAR_DU_OUTPUT_LVDS0] = {
 			.possible_crtcs = BIT(0),
 			.port = 2,
 		},
 	},
 	.num_lvds = 1,
+	.dpll_ch =  BIT(1),
 };
 
 static const struct of_device_id rcar_du_of_table[] = {
@@ -341,7 +352,8 @@ static int rcar_du_probe(struct platform_device *pdev)
 
 	ddev->irq_enabled = 1;
 
-	/* Register the DRM device with the core and the connectors with
+	/*
+	 * Register the DRM device with the core and the connectors with
 	 * sysfs.
 	 */
 	ret = drm_dev_register(ddev, 0);

drivers/gpu/drm/rcar-du/rcar_du_group.c

@@ -64,7 +64,8 @@ static void rcar_du_group_setup_defr8(struct rcar_du_group *rgrp)
 	if (rcdu->info->gen < 3) {
 		defr8 |= DEFR8_DEFE8;
 
-		/* On Gen2 the DEFR8 register for the first group also controls
+		/*
+		 * On Gen2 the DEFR8 register for the first group also controls
 		 * RGB output routing to DPAD0 and VSPD1 routing to DU0/1/2 for
 		 * DU instances that support it.
 		 */
@@ -75,7 +76,8 @@ static void rcar_du_group_setup_defr8(struct rcar_du_group *rgrp)
 				defr8 |= DEFR8_VSCS;
 		}
 	} else {
-		/* On Gen3 VSPD routing can't be configured, but DPAD routing
+		/*
+		 * On Gen3 VSPD routing can't be configured, but DPAD routing
 		 * needs to be set despite having a single option available.
 		 */
 		u32 crtc = ffs(possible_crtcs) - 1;
@@ -124,7 +126,8 @@ static void rcar_du_group_setup(struct rcar_du_group *rgrp)
 	if (rcdu->info->gen >= 3)
 		rcar_du_group_write(rgrp, DEFR10, DEFR10_CODE | DEFR10_DEFE10);
 
-	/* Use DS1PR and DS2PR to configure planes priorities and connects the
+	/*
+	 * Use DS1PR and DS2PR to configure planes priorities and connects the
 	 * superposition 0 to DU0 pins. DU1 pins will be configured dynamically.
 	 */
 	rcar_du_group_write(rgrp, DORCR, DORCR_PG1D_DS1 | DORCR_DPRS);
@@ -177,7 +180,8 @@ static void __rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start)
 
 void rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start)
 {
-	/* Many of the configuration bits are only updated when the display
+	/*
+	 * Many of the configuration bits are only updated when the display
 	 * reset (DRES) bit in DSYSR is set to 1, disabling *both* CRTCs. Some
 	 * of those bits could be pre-configured, but others (especially the
 	 * bits related to plane assignment to display timing controllers) need
@@ -208,23 +212,32 @@ void rcar_du_group_restart(struct rcar_du_group *rgrp)
 
 int rcar_du_set_dpad0_vsp1_routing(struct rcar_du_device *rcdu)
 {
+	struct rcar_du_group *rgrp;
+	struct rcar_du_crtc *crtc;
+	unsigned int index;
 	int ret;
 
 	if (!rcar_du_has(rcdu, RCAR_DU_FEATURE_EXT_CTRL_REGS))
 		return 0;
 
-	/* RGB output routing to DPAD0 and VSP1D routing to DU0/1/2 are
-	 * configured in the DEFR8 register of the first group. As this function
-	 * can be called with the DU0 and DU1 CRTCs disabled, we need to enable
-	 * the first group clock before accessing the register.
+	/*
+	 * RGB output routing to DPAD0 and VSP1D routing to DU0/1/2 are
+	 * configured in the DEFR8 register of the first group on Gen2 and the
+	 * last group on Gen3. As this function can be called with the DU
+	 * channels of the corresponding CRTCs disabled, we need to enable the
+	 * group clock before accessing the register.
 	 */
-	ret = clk_prepare_enable(rcdu->crtcs[0].clock);
+	index = rcdu->info->gen < 3 ? 0 : DIV_ROUND_UP(rcdu->num_crtcs, 2) - 1;
+	rgrp = &rcdu->groups[index];
+	crtc = &rcdu->crtcs[index * 2];
+
+	ret = clk_prepare_enable(crtc->clock);
 	if (ret < 0)
 		return ret;
 
-	rcar_du_group_setup_defr8(&rcdu->groups[0]);
+	rcar_du_group_setup_defr8(rgrp);
 
-	clk_disable_unprepare(rcdu->crtcs[0].clock);
+	clk_disable_unprepare(crtc->clock);
 
 	return 0;
 }
@@ -236,7 +249,8 @@ int rcar_du_group_set_routing(struct rcar_du_group *rgrp)
 
 	dorcr &= ~(DORCR_PG2T | DORCR_DK2S | DORCR_PG2D_MASK);
 
-	/* Set the DPAD1 pins sources. Select CRTC 0 if explicitly requested and
+	/*
+	 * Set the DPAD1 pins sources. Select CRTC 0 if explicitly requested and
 	 * CRTC 1 in all other cases to avoid cloning CRTC 0 to DPAD0 and DPAD1
 	 * by default.
 	 */

drivers/gpu/drm/rcar-du/rcar_du_kms.c

@@ -96,7 +96,8 @@ static const struct rcar_du_format_info rcar_du_format_infos[] = {
 		.pnmr = PnMR_SPIM_TP_OFF | PnMR_DDDF_YC,
 		.edf = PnDDCR4_EDF_NONE,
 	},
-	/* The following formats are not supported on Gen2 and thus have no
+	/*
+	 * The following formats are not supported on Gen2 and thus have no
 	 * associated .pnmr or .edf settings.
 	 */
 	{
@@ -153,7 +154,8 @@ int rcar_du_dumb_create(struct drm_file *file, struct drm_device *dev,
 	unsigned int min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8);
 	unsigned int align;
 
-	/* The R8A7779 DU requires a 16 pixels pitch alignment as documented,
+	/*
+	 * The R8A7779 DU requires a 16 pixels pitch alignment as documented,
 	 * but the R8A7790 DU seems to require a 128 bytes pitch alignment.
 	 */
 	if (rcar_du_needs(rcdu, RCAR_DU_QUIRK_ALIGN_128B))
@@ -255,12 +257,12 @@ static void rcar_du_atomic_commit_tail(struct drm_atomic_state *old_state)
 
 	/* Apply the atomic update. */
 	drm_atomic_helper_commit_modeset_disables(dev, old_state);
-	drm_atomic_helper_commit_modeset_enables(dev, old_state);
 	drm_atomic_helper_commit_planes(dev, old_state,
 					DRM_PLANE_COMMIT_ACTIVE_ONLY);
+	drm_atomic_helper_commit_modeset_enables(dev, old_state);
 
 	drm_atomic_helper_commit_hw_done(old_state);
-	drm_atomic_helper_wait_for_vblanks(dev, old_state);
+	drm_atomic_helper_wait_for_flip_done(dev, old_state);
 
 	drm_atomic_helper_cleanup_planes(dev, old_state);
 }
@@ -309,7 +311,7 @@ static int rcar_du_encoders_init_one(struct rcar_du_device *rcdu,
 		return -ENODEV;
 	}
 
-	entity_ep_node = of_parse_phandle(ep->local_node, "remote-endpoint", 0);
+	entity_ep_node = of_graph_get_remote_endpoint(ep->local_node);
 
 	for_each_endpoint_of_node(entity, ep_node) {
 		if (ep_node == entity_ep_node)
@@ -419,7 +421,8 @@ static int rcar_du_properties_init(struct rcar_du_device *rcdu)
 	if (rcdu->props.alpha == NULL)
 		return -ENOMEM;
 
-	/* The color key is expressed as an RGB888 triplet stored in a 32-bit
+	/*
+	 * The color key is expressed as an RGB888 triplet stored in a 32-bit
 	 * integer in XRGB8888 format. Bit 24 is used as a flag to disable (0)
 	 * or enable source color keying (1).
 	 */
@@ -432,6 +435,81 @@ static int rcar_du_properties_init(struct rcar_du_device *rcdu)
 	return 0;
 }
 
+static int rcar_du_vsps_init(struct rcar_du_device *rcdu)
+{
+	const struct device_node *np = rcdu->dev->of_node;
+	struct of_phandle_args args;
+	struct {
+		struct device_node *np;
+		unsigned int crtcs_mask;
+	} vsps[RCAR_DU_MAX_VSPS] = { { 0, }, };
+	unsigned int vsps_count = 0;
+	unsigned int cells;
+	unsigned int i;
+	int ret;
+
+	/*
+	 * First parse the DT vsps property to populate the list of VSPs. Each
+	 * entry contains a pointer to the VSP DT node and a bitmask of the
+	 * connected DU CRTCs.
+	 */
+	cells = of_property_count_u32_elems(np, "vsps") / rcdu->num_crtcs - 1;
+	if (cells > 1)
+		return -EINVAL;
+
+	for (i = 0; i < rcdu->num_crtcs; ++i) {
+		unsigned int j;
+
+		ret = of_parse_phandle_with_fixed_args(np, "vsps", cells, i,
+						       &args);
+		if (ret < 0)
+			goto error;
+
+		/*
+		 * Add the VSP to the list or update the corresponding existing
+		 * entry if the VSP has already been added.
+		 */
+		for (j = 0; j < vsps_count; ++j) {
+			if (vsps[j].np == args.np)
+				break;
+		}
+
+		if (j < vsps_count)
+			of_node_put(args.np);
+		else
+			vsps[vsps_count++].np = args.np;
+
+		vsps[j].crtcs_mask |= BIT(i);
+
+		/* Store the VSP pointer and pipe index in the CRTC. */
+		rcdu->crtcs[i].vsp = &rcdu->vsps[j];
+		rcdu->crtcs[i].vsp_pipe = cells >= 1 ? args.args[0] : 0;
+	}
+
+	/*
+	 * Then initialize all the VSPs from the node pointers and CRTCs bitmask
+	 * computed previously.
+	 */
+	for (i = 0; i < vsps_count; ++i) {
+		struct rcar_du_vsp *vsp = &rcdu->vsps[i];
+
+		vsp->index = i;
+		vsp->dev = rcdu;
+
+		ret = rcar_du_vsp_init(vsp, vsps[i].np, vsps[i].crtcs_mask);
+		if (ret < 0)
+			goto error;
+	}
+
+	return 0;
+
+error:
+	for (i = 0; i < ARRAY_SIZE(vsps); ++i)
+		of_node_put(vsps[i].np);
+
+	return ret;
+}
+
 int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 {
 	static const unsigned int mmio_offsets[] = {
@@ -461,7 +539,8 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 	if (ret < 0)
 		return ret;
 
-	/* Initialize vertical blanking interrupts handling. Start with vblank
+	/*
+	 * Initialize vertical blanking interrupts handling. Start with vblank
 	 * disabled for all CRTCs.
 	 */
 	ret = drm_vblank_init(dev, (1 << rcdu->info->num_crtcs) - 1);
@@ -481,7 +560,8 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 		rgrp->index = i;
 		rgrp->num_crtcs = min(rcdu->num_crtcs - 2 * i, 2U);
 
-		/* If we have more than one CRTCs in this group pre-associate
+		/*
+		 * If we have more than one CRTCs in this group pre-associate
 		 * the low-order planes with CRTC 0 and the high-order planes
 		 * with CRTC 1 to minimize flicker occurring when the
 		 * association is changed.
@@ -499,17 +579,9 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 
 	/* Initialize the compositors. */
 	if (rcar_du_has(rcdu, RCAR_DU_FEATURE_VSP1_SOURCE)) {
-		for (i = 0; i < rcdu->num_crtcs; ++i) {
-			struct rcar_du_vsp *vsp = &rcdu->vsps[i];
-
-			vsp->index = i;
-			vsp->dev = rcdu;
-			rcdu->crtcs[i].vsp = vsp;
-
-			ret = rcar_du_vsp_init(vsp);
-			if (ret < 0)
-				return ret;
-		}
+		ret = rcar_du_vsps_init(rcdu);
+		if (ret < 0)
+			return ret;
 	}
 
 	/* Create the CRTCs. */
@@ -537,7 +609,8 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu)
 
 	num_encoders = ret;
 
-	/* Set the possible CRTCs and possible clones. There's always at least
+	/*
+	 * Set the possible CRTCs and possible clones. There's always at least
 	 * one way for all encoders to clone each other, set all bits in the
 	 * possible clones field.
 	 */

drivers/gpu/drm/rcar-du/rcar_du_lvdsenc.c

@@ -59,7 +59,8 @@ static void rcar_du_lvdsenc_start_gen2(struct rcar_du_lvdsenc *lvds,
 
 	rcar_lvds_write(lvds, LVDPLLCR, pllcr);
 
-	/* Select the input, hardcode mode 0, enable LVDS operation and turn
+	/*
+	 * Select the input, hardcode mode 0, enable LVDS operation and turn
 	 * bias circuitry on.
 	 */
 	lvdcr0 = (lvds->mode << LVDCR0_LVMD_SHIFT) | LVDCR0_BEN | LVDCR0_LVEN;
@@ -73,7 +74,8 @@ static void rcar_du_lvdsenc_start_gen2(struct rcar_du_lvdsenc *lvds,
 			LVDCR1_CHSTBY_GEN2(1) | LVDCR1_CHSTBY_GEN2(0) |
 			LVDCR1_CLKSTBY_GEN2);
 
-	/* Turn the PLL on, wait for the startup delay, and turn the output
+	/*
+	 * Turn the PLL on, wait for the startup delay, and turn the output
 	 * on.
 	 */
 	lvdcr0 |= LVDCR0_PLLON;
@@ -140,7 +142,8 @@ static int rcar_du_lvdsenc_start(struct rcar_du_lvdsenc *lvds,
 	if (ret < 0)
 		return ret;
 
-	/* Hardcode the channels and control signals routing for now.
+	/*
+	 * Hardcode the channels and control signals routing for now.
 	 *
 	 * HSYNC -> CTRL0
 	 * VSYNC -> CTRL1
@@ -202,7 +205,8 @@ void rcar_du_lvdsenc_atomic_check(struct rcar_du_lvdsenc *lvds,
 {
 	struct rcar_du_device *rcdu = lvds->dev;
 
-	/* The internal LVDS encoder has a restricted clock frequency operating
+	/*
+	 * The internal LVDS encoder has a restricted clock frequency operating
 	 * range (30MHz to 150MHz on Gen2, 25.175MHz to 148.5MHz on Gen3). Clamp
 	 * the clock accordingly.
 	 */

drivers/gpu/drm/rcar-du/rcar_du_plane.c

@@ -50,23 +50,21 @@
  * automatically when the core swaps the old and new states.
  */
 
-static bool rcar_du_plane_needs_realloc(struct rcar_du_plane *plane,
-					struct rcar_du_plane_state *new_state)
+static bool rcar_du_plane_needs_realloc(
+				const struct rcar_du_plane_state *old_state,
+				const struct rcar_du_plane_state *new_state)
 {
-	struct rcar_du_plane_state *cur_state;
-
-	cur_state = to_rcar_plane_state(plane->plane.state);
-
-	/* Lowering the number of planes doesn't strictly require reallocation
+	/*
+	 * Lowering the number of planes doesn't strictly require reallocation
 	 * as the extra hardware plane will be freed when committing, but doing
 	 * so could lead to more fragmentation.
 	 */
-	if (!cur_state->format ||
-	    cur_state->format->planes != new_state->format->planes)
+	if (!old_state->format ||
+	    old_state->format->planes != new_state->format->planes)
 		return true;
 
 	/* Reallocate hardware planes if the source has changed. */
-	if (cur_state->source != new_state->source)
+	if (old_state->source != new_state->source)
 		return true;
 
 	return false;
@@ -141,37 +139,43 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 	unsigned int groups = 0;
 	unsigned int i;
 	struct drm_plane *drm_plane;
-	struct drm_plane_state *drm_plane_state;
+	struct drm_plane_state *old_drm_plane_state;
+	struct drm_plane_state *new_drm_plane_state;
 
 	/* Check if hardware planes need to be reallocated. */
-	for_each_plane_in_state(state, drm_plane, drm_plane_state, i) {
-		struct rcar_du_plane_state *plane_state;
+	for_each_oldnew_plane_in_state(state, drm_plane, old_drm_plane_state,
+				       new_drm_plane_state, i) {
+		struct rcar_du_plane_state *old_plane_state;
+		struct rcar_du_plane_state *new_plane_state;
 		struct rcar_du_plane *plane;
 		unsigned int index;
 
 		plane = to_rcar_plane(drm_plane);
-		plane_state = to_rcar_plane_state(drm_plane_state);
+		old_plane_state = to_rcar_plane_state(old_drm_plane_state);
+		new_plane_state = to_rcar_plane_state(new_drm_plane_state);
 
 		dev_dbg(rcdu->dev, "%s: checking plane (%u,%tu)\n", __func__,
 			plane->group->index, plane - plane->group->planes);
 
-		/* If the plane is being disabled we don't need to go through
+		/*
+		 * If the plane is being disabled we don't need to go through
 		 * the full reallocation procedure. Just mark the hardware
 		 * plane(s) as freed.
 		 */
-		if (!plane_state->format) {
+		if (!new_plane_state->format) {
 			dev_dbg(rcdu->dev, "%s: plane is being disabled\n",
 				__func__);
 			index = plane - plane->group->planes;
 			group_freed_planes[plane->group->index] |= 1 << index;
-			plane_state->hwindex = -1;
+			new_plane_state->hwindex = -1;
 			continue;
 		}
 
-		/* If the plane needs to be reallocated mark it as such, and
+		/*
+		 * If the plane needs to be reallocated mark it as such, and
 		 * mark the hardware plane(s) as free.
 		 */
-		if (rcar_du_plane_needs_realloc(plane, plane_state)) {
+		if (rcar_du_plane_needs_realloc(old_plane_state, new_plane_state)) {
 			dev_dbg(rcdu->dev, "%s: plane needs reallocation\n",
 				__func__);
 			groups |= 1 << plane->group->index;
@@ -179,14 +183,15 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 
 			index = plane - plane->group->planes;
 			group_freed_planes[plane->group->index] |= 1 << index;
-			plane_state->hwindex = -1;
+			new_plane_state->hwindex = -1;
 		}
 	}
 
 	if (!needs_realloc)
 		return 0;
 
-	/* Grab all plane states for the groups that need reallocation to ensure
+	/*
+	 * Grab all plane states for the groups that need reallocation to ensure
 	 * locking and avoid racy updates. This serializes the update operation,
 	 * but there's not much we can do about it as that's the hardware
 	 * design.
@@ -204,14 +209,15 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 
 		for (i = 0; i < group->num_planes; ++i) {
 			struct rcar_du_plane *plane = &group->planes[i];
-			struct rcar_du_plane_state *plane_state;
+			struct rcar_du_plane_state *new_plane_state;
 			struct drm_plane_state *s;
 
 			s = drm_atomic_get_plane_state(state, &plane->plane);
 			if (IS_ERR(s))
 				return PTR_ERR(s);
 
-			/* If the plane has been freed in the above loop its
+			/*
+			 * If the plane has been freed in the above loop its
 			 * hardware planes must not be added to the used planes
 			 * bitmask. However, the current state doesn't reflect
 			 * the free state yet, as we've modified the new state
@@ -226,16 +232,16 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 				continue;
 			}
 
-			plane_state = to_rcar_plane_state(plane->plane.state);
-			used_planes |= rcar_du_plane_hwmask(plane_state);
+			new_plane_state = to_rcar_plane_state(s);
+			used_planes |= rcar_du_plane_hwmask(new_plane_state);
 
 			dev_dbg(rcdu->dev,
 				"%s: plane (%u,%tu) uses %u hwplanes (index %d)\n",
 				__func__, plane->group->index,
 				plane - plane->group->planes,
-				plane_state->format ?
-				plane_state->format->planes : 0,
-				plane_state->hwindex);
+				new_plane_state->format ?
+				new_plane_state->format->planes : 0,
+				new_plane_state->hwindex);
 		}
 
 		group_free_planes[index] = 0xff & ~used_planes;
@@ -246,40 +252,45 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 	}
 
 	/* Reallocate hardware planes for each plane that needs it. */
-	for_each_plane_in_state(state, drm_plane, drm_plane_state, i) {
-		struct rcar_du_plane_state *plane_state;
+	for_each_oldnew_plane_in_state(state, drm_plane, old_drm_plane_state,
+				       new_drm_plane_state, i) {
+		struct rcar_du_plane_state *old_plane_state;
+		struct rcar_du_plane_state *new_plane_state;
 		struct rcar_du_plane *plane;
 		unsigned int crtc_planes;
 		unsigned int free;
 		int idx;
 
 		plane = to_rcar_plane(drm_plane);
-		plane_state = to_rcar_plane_state(drm_plane_state);
+		old_plane_state = to_rcar_plane_state(old_drm_plane_state);
+		new_plane_state = to_rcar_plane_state(new_drm_plane_state);
 
 		dev_dbg(rcdu->dev, "%s: allocating plane (%u,%tu)\n", __func__,
 			plane->group->index, plane - plane->group->planes);
 
-		/* Skip planes that are being disabled or don't need to be
+		/*
+		 * Skip planes that are being disabled or don't need to be
 		 * reallocated.
 		 */
-		if (!plane_state->format ||
-		    !rcar_du_plane_needs_realloc(plane, plane_state))
+		if (!new_plane_state->format ||
+		    !rcar_du_plane_needs_realloc(old_plane_state, new_plane_state))
 			continue;
 
-		/* Try to allocate the plane from the free planes currently
+		/*
+		 * Try to allocate the plane from the free planes currently
 		 * associated with the target CRTC to avoid restarting the CRTC
 		 * group and thus minimize flicker. If it fails fall back to
 		 * allocating from all free planes.
 		 */
-		crtc_planes = to_rcar_crtc(plane_state->state.crtc)->index % 2
+		crtc_planes = to_rcar_crtc(new_plane_state->state.crtc)->index % 2
 			    ? plane->group->dptsr_planes
 			    : ~plane->group->dptsr_planes;
 		free = group_free_planes[plane->group->index];
 
-		idx = rcar_du_plane_hwalloc(plane, plane_state,
+		idx = rcar_du_plane_hwalloc(plane, new_plane_state,
 					    free & crtc_planes);
 		if (idx < 0)
-			idx = rcar_du_plane_hwalloc(plane, plane_state,
+			idx = rcar_du_plane_hwalloc(plane, new_plane_state,
 						    free);
 		if (idx < 0) {
 			dev_dbg(rcdu->dev, "%s: no available hardware plane\n",
@@ -288,12 +299,12 @@ int rcar_du_atomic_check_planes(struct drm_device *dev,
 		}
 
 		dev_dbg(rcdu->dev, "%s: allocated %u hwplanes (index %u)\n",
-			__func__, plane_state->format->planes, idx);
+			__func__, new_plane_state->format->planes, idx);
 
-		plane_state->hwindex = idx;
+		new_plane_state->hwindex = idx;
 
 		group_free_planes[plane->group->index] &=
-			~rcar_du_plane_hwmask(plane_state);
+			~rcar_du_plane_hwmask(new_plane_state);
 
 		dev_dbg(rcdu->dev, "%s: group %u free planes mask 0x%02x\n",
 			__func__, plane->group->index,
@@ -351,14 +362,16 @@ static void rcar_du_plane_setup_scanout(struct rcar_du_group *rgrp,
 		dma[1] = 0;
 	}
 
-	/* Memory pitch (expressed in pixels). Must be doubled for interlaced
+	/*
+	 * Memory pitch (expressed in pixels). Must be doubled for interlaced
 	 * operation with 32bpp formats.
 	 */
 	rcar_du_plane_write(rgrp, index, PnMWR,
 			    (interlaced && state->format->bpp == 32) ?
 			    pitch * 2 : pitch);
 
-	/* The Y position is expressed in raster line units and must be doubled
+	/*
+	 * The Y position is expressed in raster line units and must be doubled
 	 * for 32bpp formats, according to the R8A7790 datasheet. No mention of
 	 * doubling the Y position is found in the R8A7779 datasheet, but the
 	 * rule seems to apply there as well.
@@ -396,7 +409,8 @@ static void rcar_du_plane_setup_mode(struct rcar_du_group *rgrp,
 	u32 colorkey;
 	u32 pnmr;
 
-	/* The PnALPHAR register controls alpha-blending in 16bpp formats
+	/*
+	 * The PnALPHAR register controls alpha-blending in 16bpp formats
 	 * (ARGB1555 and XRGB1555).
 	 *
 	 * For ARGB, set the alpha value to 0, and enable alpha-blending when
@@ -413,7 +427,8 @@ static void rcar_du_plane_setup_mode(struct rcar_du_group *rgrp,
 
 	pnmr = PnMR_BM_MD | state->format->pnmr;
 
-	/* Disable color keying when requested. YUV formats have the
+	/*
+	 * Disable color keying when requested. YUV formats have the
 	 * PnMR_SPIM_TP_OFF bit set in their pnmr field, disabling color keying
 	 * automatically.
 	 */
@@ -457,7 +472,8 @@ static void rcar_du_plane_setup_format_gen2(struct rcar_du_group *rgrp,
 	u32 ddcr2 = PnDDCR2_CODE;
 	u32 ddcr4;
 
-	/* Data format
+	/*
+	 * Data format
 	 *
 	 * The data format is selected by the DDDF field in PnMR and the EDF
 	 * field in DDCR4.
@@ -589,7 +605,8 @@ static void rcar_du_plane_atomic_update(struct drm_plane *plane,
 
 	rcar_du_plane_setup(rplane);
 
-	/* Check whether the source has changed from memory to live source or
+	/*
+	 * Check whether the source has changed from memory to live source or
 	 * from live source to memory. The source has been configured by the
 	 * VSPS bit in the PnDDCR4 register. Although the datasheet states that
 	 * the bit is updated during vertical blanking, it seems that updates
@@ -726,7 +743,8 @@ int rcar_du_planes_init(struct rcar_du_group *rgrp)
 	unsigned int i;
 	int ret;
 
-	 /* Create one primary plane per CRTC in this group and seven overlay
+	 /*
+	  * Create one primary plane per CRTC in this group and seven overlay
 	  * planes.
 	  */
 	rgrp->num_planes = rgrp->num_crtcs + 7;

drivers/gpu/drm/rcar-du/rcar_du_plane.h

@@ -20,7 +20,8 @@
 struct rcar_du_format_info;
 struct rcar_du_group;
 
-/* The RCAR DU has 8 hardware planes, shared between primary and overlay planes.
+/*
+ * The RCAR DU has 8 hardware planes, shared between primary and overlay planes.
  * As using overlay planes requires at least one of the CRTCs being enabled, no
  * more than 7 overlay planes can be available. We thus create 1 primary plane
  * per CRTC and 7 overlay planes, for a total of up to 9 KMS planes.

drivers/gpu/drm/rcar-du/rcar_du_vsp.c

@@ -19,6 +19,7 @@
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_plane_helper.h>
 
+#include <linux/bitops.h>
 #include <linux/dma-mapping.h>
 #include <linux/of_platform.h>
 #include <linux/scatterlist.h>
@@ -30,11 +31,15 @@
 #include "rcar_du_kms.h"
 #include "rcar_du_vsp.h"
 
-static void rcar_du_vsp_complete(void *private)
+static void rcar_du_vsp_complete(void *private, bool completed)
 {
 	struct rcar_du_crtc *crtc = private;
 
-	rcar_du_crtc_finish_page_flip(crtc);
+	if (crtc->vblank_enable)
+		drm_crtc_handle_vblank(&crtc->crtc);
+
+	if (completed)
+		rcar_du_crtc_finish_page_flip(crtc);
 }
 
 void rcar_du_vsp_enable(struct rcar_du_crtc *crtc)
@@ -73,7 +78,8 @@ void rcar_du_vsp_enable(struct rcar_du_crtc *crtc)
 
 	__rcar_du_plane_setup(crtc->group, &state);
 
-	/* Ensure that the plane source configuration takes effect by requesting
+	/*
+	 * Ensure that the plane source configuration takes effect by requesting
 	 * a restart of the group. See rcar_du_plane_atomic_update() for a more
 	 * detailed explanation.
 	 *
@@ -81,22 +87,22 @@ void rcar_du_vsp_enable(struct rcar_du_crtc *crtc)
 	 */
 	crtc->group->need_restart = true;
 
-	vsp1_du_setup_lif(crtc->vsp->vsp, &cfg);
+	vsp1_du_setup_lif(crtc->vsp->vsp, crtc->vsp_pipe, &cfg);
 }
 
 void rcar_du_vsp_disable(struct rcar_du_crtc *crtc)
 {
-	vsp1_du_setup_lif(crtc->vsp->vsp, NULL);
+	vsp1_du_setup_lif(crtc->vsp->vsp, crtc->vsp_pipe, NULL);
 }
 
 void rcar_du_vsp_atomic_begin(struct rcar_du_crtc *crtc)
 {
-	vsp1_du_atomic_begin(crtc->vsp->vsp);
+	vsp1_du_atomic_begin(crtc->vsp->vsp, crtc->vsp_pipe);
 }
 
 void rcar_du_vsp_atomic_flush(struct rcar_du_crtc *crtc)
 {
-	vsp1_du_atomic_flush(crtc->vsp->vsp);
+	vsp1_du_atomic_flush(crtc->vsp->vsp, crtc->vsp_pipe);
 }
 
 /* Keep the two tables in sync. */
@@ -162,6 +168,7 @@ static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane)
 {
 	struct rcar_du_vsp_plane_state *state =
 		to_rcar_vsp_plane_state(plane->plane.state);
+	struct rcar_du_crtc *crtc = to_rcar_crtc(state->state.crtc);
 	struct drm_framebuffer *fb = plane->plane.state->fb;
 	struct vsp1_du_atomic_config cfg = {
 		.pixelformat = 0,
@@ -192,7 +199,8 @@ static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane)
 		}
 	}
 
-	vsp1_du_atomic_update(plane->vsp->vsp, plane->index, &cfg);
+	vsp1_du_atomic_update(plane->vsp->vsp, crtc->vsp_pipe,
+			      plane->index, &cfg);
 }
 
 static int rcar_du_vsp_plane_prepare_fb(struct drm_plane *plane,
@@ -288,11 +296,13 @@ static void rcar_du_vsp_plane_atomic_update(struct drm_plane *plane,
 					struct drm_plane_state *old_state)
 {
 	struct rcar_du_vsp_plane *rplane = to_rcar_vsp_plane(plane);
+	struct rcar_du_crtc *crtc = to_rcar_crtc(old_state->crtc);
 
 	if (plane->state->crtc)
 		rcar_du_vsp_plane_setup(rplane);
 	else
-		vsp1_du_atomic_update(rplane->vsp->vsp, rplane->index, NULL);
+		vsp1_du_atomic_update(rplane->vsp->vsp, crtc->vsp_pipe,
+				      rplane->index, NULL);
 }
 
 static const struct drm_plane_helper_funcs rcar_du_vsp_plane_helper_funcs = {
@@ -391,23 +401,17 @@ static const struct drm_plane_funcs rcar_du_vsp_plane_funcs = {
 	.atomic_get_property = rcar_du_vsp_plane_atomic_get_property,
 };
 
-int rcar_du_vsp_init(struct rcar_du_vsp *vsp)
+int rcar_du_vsp_init(struct rcar_du_vsp *vsp, struct device_node *np,
+		     unsigned int crtcs)
 {
 	struct rcar_du_device *rcdu = vsp->dev;
 	struct platform_device *pdev;
-	struct device_node *np;
+	unsigned int num_crtcs = hweight32(crtcs);
 	unsigned int i;
 	int ret;
 
 	/* Find the VSP device and initialize it. */
-	np = of_parse_phandle(rcdu->dev->of_node, "vsps", vsp->index);
-	if (!np) {
-		dev_err(rcdu->dev, "vsps node not found\n");
-		return -ENXIO;
-	}
-
 	pdev = of_find_device_by_node(np);
-	of_node_put(np);
 	if (!pdev)
 		return -ENXIO;
 
@@ -417,7 +421,8 @@ int rcar_du_vsp_init(struct rcar_du_vsp *vsp)
 	if (ret < 0)
 		return ret;
 
-	 /* The VSP2D (Gen3) has 5 RPFs, but the VSP1D (Gen2) is limited to
+	 /*
+	  * The VSP2D (Gen3) has 5 RPFs, but the VSP1D (Gen2) is limited to
 	  * 4 RPFs.
 	  */
 	vsp->num_planes = rcdu->info->gen >= 3 ? 5 : 4;
@@ -428,15 +433,15 @@ int rcar_du_vsp_init(struct rcar_du_vsp *vsp)
 		return -ENOMEM;
 
 	for (i = 0; i < vsp->num_planes; ++i) {
-		enum drm_plane_type type = i ? DRM_PLANE_TYPE_OVERLAY
-					 : DRM_PLANE_TYPE_PRIMARY;
+		enum drm_plane_type type = i < num_crtcs
+					 ? DRM_PLANE_TYPE_PRIMARY
+					 : DRM_PLANE_TYPE_OVERLAY;
 		struct rcar_du_vsp_plane *plane = &vsp->planes[i];
 
 		plane->vsp = vsp;
 		plane->index = i;
 
-		ret = drm_universal_plane_init(rcdu->ddev, &plane->plane,
-					       1 << vsp->index,
+		ret = drm_universal_plane_init(rcdu->ddev, &plane->plane, crtcs,
 					       &rcar_du_vsp_plane_funcs,
 					       formats_kms,
 					       ARRAY_SIZE(formats_kms), type,

drivers/gpu/drm/rcar-du/rcar_du_vsp.h

@@ -64,13 +64,19 @@ to_rcar_vsp_plane_state(struct drm_plane_state *state)
 }
 
 #ifdef CONFIG_DRM_RCAR_VSP
-int rcar_du_vsp_init(struct rcar_du_vsp *vsp);
+int rcar_du_vsp_init(struct rcar_du_vsp *vsp, struct device_node *np,
+		     unsigned int crtcs);
 void rcar_du_vsp_enable(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_disable(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_atomic_begin(struct rcar_du_crtc *crtc);
 void rcar_du_vsp_atomic_flush(struct rcar_du_crtc *crtc);
 #else
-static inline int rcar_du_vsp_init(struct rcar_du_vsp *vsp) { return -ENXIO; };
+static inline int rcar_du_vsp_init(struct rcar_du_vsp *vsp,
+				   struct device_node *np,
+				   unsigned int crtcs)
+{
+	return -ENXIO;
+}
 static inline void rcar_du_vsp_enable(struct rcar_du_crtc *crtc) { };
 static inline void rcar_du_vsp_disable(struct rcar_du_crtc *crtc) { };
 static inline void rcar_du_vsp_atomic_begin(struct rcar_du_crtc *crtc) { };

drivers/gpu/drm/rcar-du/rcar_dw_hdmi.c

@@ -45,7 +45,7 @@ static int rcar_hdmi_phy_configure(struct dw_hdmi *hdmi,
 {
 	const struct rcar_hdmi_phy_params *params = rcar_hdmi_phy_params;
 
-	for (; params && params->mpixelclock != ~0UL; ++params) {
+	for (; params->mpixelclock != ~0UL; ++params) {
 		if (mpixelclock <= params->mpixelclock)
 			break;
 	}

drivers/media/platform/vsp1/vsp1.h

@@ -41,11 +41,11 @@ struct vsp1_rwpf;
 struct vsp1_sru;
 struct vsp1_uds;
 
+#define VSP1_MAX_LIF		2
 #define VSP1_MAX_RPF		5
 #define VSP1_MAX_UDS		3
 #define VSP1_MAX_WPF		4
 
-#define VSP1_HAS_LIF		(1 << 0)
 #define VSP1_HAS_LUT		(1 << 1)
 #define VSP1_HAS_SRU		(1 << 2)
 #define VSP1_HAS_BRU		(1 << 3)
@@ -54,12 +54,14 @@ struct vsp1_uds;
 #define VSP1_HAS_WPF_HFLIP	(1 << 6)
 #define VSP1_HAS_HGO		(1 << 7)
 #define VSP1_HAS_HGT		(1 << 8)
+#define VSP1_HAS_BRS		(1 << 9)
 
 struct vsp1_device_info {
 	u32 version;
 	const char *model;
 	unsigned int gen;
 	unsigned int features;
+	unsigned int lif_count;
 	unsigned int rpf_count;
 	unsigned int uds_count;
 	unsigned int wpf_count;
@@ -76,13 +78,14 @@ struct vsp1_device {
 	struct rcar_fcp_device *fcp;
 	struct device *bus_master;
 
+	struct vsp1_bru *brs;
 	struct vsp1_bru *bru;
 	struct vsp1_clu *clu;
 	struct vsp1_hgo *hgo;
 	struct vsp1_hgt *hgt;
 	struct vsp1_hsit *hsi;
 	struct vsp1_hsit *hst;
-	struct vsp1_lif *lif;
+	struct vsp1_lif *lif[VSP1_MAX_LIF];
 	struct vsp1_lut *lut;
 	struct vsp1_rwpf *rpf[VSP1_MAX_RPF];
 	struct vsp1_sru *sru;

drivers/media/platform/vsp1/vsp1_bru.c

@@ -33,7 +33,7 @@
 static inline void vsp1_bru_write(struct vsp1_bru *bru, struct vsp1_dl_list *dl,
 				  u32 reg, u32 data)
 {
-	vsp1_dl_list_write(dl, reg, data);
+	vsp1_dl_list_write(dl, bru->base + reg, data);
 }
 
 /* -----------------------------------------------------------------------------
@@ -332,11 +332,14 @@ static void bru_configure(struct vsp1_entity *entity,
 	/*
 	 * Route BRU input 1 as SRC input to the ROP unit and configure the ROP
 	 * unit with a NOP operation to make BRU input 1 available as the
-	 * Blend/ROP unit B SRC input.
+	 * Blend/ROP unit B SRC input. Only needed for BRU, the BRS has no ROP
+	 * unit.
 	 */
-	vsp1_bru_write(bru, dl, VI6_BRU_ROP, VI6_BRU_ROP_DSTSEL_BRUIN(1) |
-		       VI6_BRU_ROP_CROP(VI6_ROP_NOP) |
-		       VI6_BRU_ROP_AROP(VI6_ROP_NOP));
+	if (entity->type == VSP1_ENTITY_BRU)
+		vsp1_bru_write(bru, dl, VI6_BRU_ROP,
+			       VI6_BRU_ROP_DSTSEL_BRUIN(1) |
+			       VI6_BRU_ROP_CROP(VI6_ROP_NOP) |
+			       VI6_BRU_ROP_AROP(VI6_ROP_NOP));
 
 	for (i = 0; i < bru->entity.source_pad; ++i) {
 		bool premultiplied = false;
@@ -366,12 +369,13 @@ static void bru_configure(struct vsp1_entity *entity,
 			ctrl |= VI6_BRU_CTRL_DSTSEL_VRPF;
 
 		/*
-		 * Route BRU inputs 0 to 3 as SRC inputs to Blend/ROP units A to
-		 * D in that order. The Blend/ROP unit B SRC is hardwired to the
-		 * ROP unit output, the corresponding register bits must be set
-		 * to 0.
+		 * Route inputs 0 to 3 as SRC inputs to Blend/ROP units A to D
+		 * in that order. In the BRU the Blend/ROP unit B SRC is
+		 * hardwired to the ROP unit output, the corresponding register
+		 * bits must be set to 0. The BRS has no ROP unit and doesn't
+		 * need any special processing.
 		 */
-		if (i != 1)
+		if (!(entity->type == VSP1_ENTITY_BRU && i == 1))
 			ctrl |= VI6_BRU_CTRL_SRCSEL_BRUIN(i);
 
 		vsp1_bru_write(bru, dl, VI6_BRU_CTRL(i), ctrl);
@@ -407,20 +411,31 @@ static const struct vsp1_entity_operations bru_entity_ops = {
  * Initialization and Cleanup
  */
 
-struct vsp1_bru *vsp1_bru_create(struct vsp1_device *vsp1)
+struct vsp1_bru *vsp1_bru_create(struct vsp1_device *vsp1,
+				 enum vsp1_entity_type type)
 {
 	struct vsp1_bru *bru;
+	unsigned int num_pads;
+	const char *name;
 	int ret;
 
 	bru = devm_kzalloc(vsp1->dev, sizeof(*bru), GFP_KERNEL);
 	if (bru == NULL)
 		return ERR_PTR(-ENOMEM);
 
+	bru->base = type == VSP1_ENTITY_BRU ? VI6_BRU_BASE : VI6_BRS_BASE;
 	bru->entity.ops = &bru_entity_ops;
-	bru->entity.type = VSP1_ENTITY_BRU;
+	bru->entity.type = type;
+
+	if (type == VSP1_ENTITY_BRU) {
+		num_pads = vsp1->info->num_bru_inputs + 1;
+		name = "bru";
+	} else {
+		num_pads = 3;
+		name = "brs";
+	}
 
-	ret = vsp1_entity_init(vsp1, &bru->entity, "bru",
-			       vsp1->info->num_bru_inputs + 1, &bru_ops,
+	ret = vsp1_entity_init(vsp1, &bru->entity, name, num_pads, &bru_ops,
 			       MEDIA_ENT_F_PROC_VIDEO_COMPOSER);
 	if (ret < 0)
 		return ERR_PTR(ret);
@@ -435,7 +450,7 @@ struct vsp1_bru *vsp1_bru_create(struct vsp1_device *vsp1)
 	bru->entity.subdev.ctrl_handler = &bru->ctrls;
 
 	if (bru->ctrls.error) {
-		dev_err(vsp1->dev, "bru: failed to initialize controls\n");
+		dev_err(vsp1->dev, "%s: failed to initialize controls\n", name);
 		ret = bru->ctrls.error;
 		vsp1_entity_destroy(&bru->entity);
 		return ERR_PTR(ret);

drivers/media/platform/vsp1/vsp1_bru.h

@@ -26,6 +26,7 @@ struct vsp1_rwpf;
 
 struct vsp1_bru {
 	struct vsp1_entity entity;
+	unsigned int base;
 
 	struct v4l2_ctrl_handler ctrls;
 
@@ -41,6 +42,7 @@ static inline struct vsp1_bru *to_bru(struct v4l2_subdev *subdev)
 	return container_of(subdev, struct vsp1_bru, entity.subdev);
 }
 
-struct vsp1_bru *vsp1_bru_create(struct vsp1_device *vsp1);
+struct vsp1_bru *vsp1_bru_create(struct vsp1_device *vsp1,
+				 enum vsp1_entity_type type);
 
 #endif /* __VSP1_BRU_H__ */

drivers/media/platform/vsp1/vsp1_dl.c

@@ -95,6 +95,7 @@ enum vsp1_dl_mode {
  * struct vsp1_dl_manager - Display List manager
  * @index: index of the related WPF
  * @mode: display list operation mode (header or headerless)
+ * @singleshot: execute the display list in single-shot mode
  * @vsp1: the VSP1 device
  * @lock: protects the free, active, queued, pending and gc_fragments lists
  * @free: array of all free display lists
@@ -107,6 +108,7 @@ enum vsp1_dl_mode {
 struct vsp1_dl_manager {
 	unsigned int index;
 	enum vsp1_dl_mode mode;
+	bool singleshot;
 	struct vsp1_device *vsp1;
 
 	spinlock_t lock;
@@ -437,6 +439,7 @@ int vsp1_dl_list_add_chain(struct vsp1_dl_list *head,
 
 static void vsp1_dl_list_fill_header(struct vsp1_dl_list *dl, bool is_last)
 {
+	struct vsp1_dl_manager *dlm = dl->dlm;
 	struct vsp1_dl_header_list *hdr = dl->header->lists;
 	struct vsp1_dl_body *dlb;
 	unsigned int num_lists = 0;
@@ -461,106 +464,152 @@ static void vsp1_dl_list_fill_header(struct vsp1_dl_list *dl, bool is_last)
 
 	dl->header->num_lists = num_lists;
 
-	/*
-	 * If this display list's chain is not empty, we are on a list, where
-	 * the next item in the list is the display list entity which should be
-	 * automatically queued by the hardware.
-	 */
 	if (!list_empty(&dl->chain) && !is_last) {
+		/*
+		 * If this display list's chain is not empty, we are on a list,
+		 * and the next item is the display list that we must queue for
+		 * automatic processing by the hardware.
+		 */
 		struct vsp1_dl_list *next = list_next_entry(dl, chain);
 
 		dl->header->next_header = next->dma;
 		dl->header->flags = VSP1_DLH_AUTO_START;
+	} else if (!dlm->singleshot) {
+		/*
+		 * if the display list manager works in continuous mode, the VSP
+		 * should loop over the display list continuously until
+		 * instructed to do otherwise.
+		 */
+		dl->header->next_header = dl->dma;
+		dl->header->flags = VSP1_DLH_INT_ENABLE | VSP1_DLH_AUTO_START;
 	} else {
+		/*
+		 * Otherwise, in mem-to-mem mode, we work in single-shot mode
+		 * and the next display list must not be started automatically.
+		 */
 		dl->header->flags = VSP1_DLH_INT_ENABLE;
 	}
 }
 
-void vsp1_dl_list_commit(struct vsp1_dl_list *dl)
+static bool vsp1_dl_list_hw_update_pending(struct vsp1_dl_manager *dlm)
 {
-	struct vsp1_dl_manager *dlm = dl->dlm;
 	struct vsp1_device *vsp1 = dlm->vsp1;
-	unsigned long flags;
-	bool update;
 
-	spin_lock_irqsave(&dlm->lock, flags);
+	if (!dlm->queued)
+		return false;
 
-	if (dl->dlm->mode == VSP1_DL_MODE_HEADER) {
-		struct vsp1_dl_list *dl_child;
+	/*
+	 * Check whether the VSP1 has taken the update. In headerless mode the
+	 * hardware indicates this by clearing the UPD bit in the DL_BODY_SIZE
+	 * register, and in header mode by clearing the UPDHDR bit in the CMD
+	 * register.
+	 */
+	if (dlm->mode == VSP1_DL_MODE_HEADERLESS)
+		return !!(vsp1_read(vsp1, VI6_DL_BODY_SIZE)
+			  & VI6_DL_BODY_SIZE_UPD);
+	else
+		return !!(vsp1_read(vsp1, VI6_CMD(dlm->index) & VI6_CMD_UPDHDR));
+}
 
+static void vsp1_dl_list_hw_enqueue(struct vsp1_dl_list *dl)
+{
+	struct vsp1_dl_manager *dlm = dl->dlm;
+	struct vsp1_device *vsp1 = dlm->vsp1;
+
+	if (dlm->mode == VSP1_DL_MODE_HEADERLESS) {
 		/*
-		 * In header mode the caller guarantees that the hardware is
-		 * idle at this point.
+		 * In headerless mode, program the hardware directly with the
+		 * display list body address and size and set the UPD bit. The
+		 * bit will be cleared by the hardware when the display list
+		 * processing starts.
 		 */
-
-		/* Fill the header for the head and chained display lists. */
-		vsp1_dl_list_fill_header(dl, list_empty(&dl->chain));
-
-		list_for_each_entry(dl_child, &dl->chain, chain) {
-			bool last = list_is_last(&dl_child->chain, &dl->chain);
-
-			vsp1_dl_list_fill_header(dl_child, last);
-		}
-
+		vsp1_write(vsp1, VI6_DL_HDR_ADDR(0), dl->body0.dma);
+		vsp1_write(vsp1, VI6_DL_BODY_SIZE, VI6_DL_BODY_SIZE_UPD |
+			   (dl->body0.num_entries * sizeof(*dl->header->lists)));
+	} else {
 		/*
-		 * Commit the head display list to hardware. Chained headers
-		 * will auto-start.
+		 * In header mode, program the display list header address. If
+		 * the hardware is idle (single-shot mode or first frame in
+		 * continuous mode) it will then be started independently. If
+		 * the hardware is operating, the VI6_DL_HDR_REF_ADDR register
+		 * will be updated with the display list address.
 		 */
 		vsp1_write(vsp1, VI6_DL_HDR_ADDR(dlm->index), dl->dma);
-
-		dlm->active = dl;
-		goto done;
 	}
+}
+
+static void vsp1_dl_list_commit_continuous(struct vsp1_dl_list *dl)
+{
+	struct vsp1_dl_manager *dlm = dl->dlm;
 
 	/*
-	 * Once the UPD bit has been set the hardware can start processing the
-	 * display list at any time and we can't touch the address and size
-	 * registers. In that case mark the update as pending, it will be
-	 * queued up to the hardware by the frame end interrupt handler.
+	 * If a previous display list has been queued to the hardware but not
+	 * processed yet, the VSP can start processing it at any time. In that
+	 * case we can't replace the queued list by the new one, as we could
+	 * race with the hardware. We thus mark the update as pending, it will
+	 * be queued up to the hardware by the frame end interrupt handler.
 	 */
-	update = !!(vsp1_read(vsp1, VI6_DL_BODY_SIZE) & VI6_DL_BODY_SIZE_UPD);
-	if (update) {
+	if (vsp1_dl_list_hw_update_pending(dlm)) {
 		__vsp1_dl_list_put(dlm->pending);
 		dlm->pending = dl;
-		goto done;
+		return;
 	}
 
 	/*
-	 * Program the hardware with the display list body address and size.
-	 * The UPD bit will be cleared by the device when the display list is
-	 * processed.
+	 * Pass the new display list to the hardware and mark it as queued. It
+	 * will become active when the hardware starts processing it.
 	 */
-	vsp1_write(vsp1, VI6_DL_HDR_ADDR(0), dl->body0.dma);
-	vsp1_write(vsp1, VI6_DL_BODY_SIZE, VI6_DL_BODY_SIZE_UPD |
-		   (dl->body0.num_entries * sizeof(*dl->header->lists)));
+	vsp1_dl_list_hw_enqueue(dl);
 
 	__vsp1_dl_list_put(dlm->queued);
 	dlm->queued = dl;
-
-done:
-	spin_unlock_irqrestore(&dlm->lock, flags);
 }
 
-/* -----------------------------------------------------------------------------
- * Display List Manager
- */
-
-/* Interrupt Handling */
-void vsp1_dlm_irq_display_start(struct vsp1_dl_manager *dlm)
+static void vsp1_dl_list_commit_singleshot(struct vsp1_dl_list *dl)
 {
-	spin_lock(&dlm->lock);
+	struct vsp1_dl_manager *dlm = dl->dlm;
 
 	/*
-	 * The display start interrupt signals the end of the display list
-	 * processing by the device. The active display list, if any, won't be
-	 * accessed anymore and can be reused.
+	 * When working in single-shot mode, the caller guarantees that the
+	 * hardware is idle at this point. Just commit the head display list
+	 * to hardware. Chained lists will be started automatically.
 	 */
-	__vsp1_dl_list_put(dlm->active);
-	dlm->active = NULL;
+	vsp1_dl_list_hw_enqueue(dl);
 
-	spin_unlock(&dlm->lock);
+	dlm->active = dl;
+}
+
+void vsp1_dl_list_commit(struct vsp1_dl_list *dl)
+{
+	struct vsp1_dl_manager *dlm = dl->dlm;
+	struct vsp1_dl_list *dl_child;
+	unsigned long flags;
+
+	if (dlm->mode == VSP1_DL_MODE_HEADER) {
+		/* Fill the header for the head and chained display lists. */
+		vsp1_dl_list_fill_header(dl, list_empty(&dl->chain));
+
+		list_for_each_entry(dl_child, &dl->chain, chain) {
+			bool last = list_is_last(&dl_child->chain, &dl->chain);
+
+			vsp1_dl_list_fill_header(dl_child, last);
+		}
+	}
+
+	spin_lock_irqsave(&dlm->lock, flags);
+
+	if (dlm->singleshot)
+		vsp1_dl_list_commit_singleshot(dl);
+	else
+		vsp1_dl_list_commit_continuous(dl);
+
+	spin_unlock_irqrestore(&dlm->lock, flags);
 }
 
+/* -----------------------------------------------------------------------------
+ * Display List Manager
+ */
+
 /**
  * vsp1_dlm_irq_frame_end - Display list handler for the frame end interrupt
  * @dlm: the display list manager
@@ -572,31 +621,28 @@ void vsp1_dlm_irq_display_start(struct vsp1_dl_manager *dlm)
  */
 bool vsp1_dlm_irq_frame_end(struct vsp1_dl_manager *dlm)
 {
-	struct vsp1_device *vsp1 = dlm->vsp1;
 	bool completed = false;
 
 	spin_lock(&dlm->lock);
 
-	__vsp1_dl_list_put(dlm->active);
-	dlm->active = NULL;
-
 	/*
-	 * Header mode is used for mem-to-mem pipelines only. We don't need to
-	 * perform any operation as there can't be any new display list queued
-	 * in that case.
+	 * The mem-to-mem pipelines work in single-shot mode. No new display
+	 * list can be queued, we don't have to do anything.
 	 */
-	if (dlm->mode == VSP1_DL_MODE_HEADER) {
+	if (dlm->singleshot) {
+		__vsp1_dl_list_put(dlm->active);
+		dlm->active = NULL;
 		completed = true;
 		goto done;
 	}
 
 	/*
-	 * The UPD bit set indicates that the commit operation raced with the
-	 * interrupt and occurred after the frame end event and UPD clear but
-	 * before interrupt processing. The hardware hasn't taken the update
-	 * into account yet, we'll thus skip one frame and retry.
+	 * If the commit operation raced with the interrupt and occurred after
+	 * the frame end event but before interrupt processing, the hardware
+	 * hasn't taken the update into account yet. We have to skip one frame
+	 * and retry.
 	 */
-	if (vsp1_read(vsp1, VI6_DL_BODY_SIZE) & VI6_DL_BODY_SIZE_UPD)
+	if (vsp1_dl_list_hw_update_pending(dlm))
 		goto done;
 
 	/*
@@ -604,24 +650,20 @@ bool vsp1_dlm_irq_frame_end(struct vsp1_dl_manager *dlm)
 	 * frame end interrupt. The display list thus becomes active.
 	 */
 	if (dlm->queued) {
+		__vsp1_dl_list_put(dlm->active);
 		dlm->active = dlm->queued;
 		dlm->queued = NULL;
 		completed = true;
 	}
 
 	/*
-	 * Now that the UPD bit has been cleared we can queue the next display
-	 * list to the hardware if one has been prepared.
+	 * Now that the VSP has started processing the queued display list, we
+	 * can queue the pending display list to the hardware if one has been
+	 * prepared.
 	 */
 	if (dlm->pending) {
-		struct vsp1_dl_list *dl = dlm->pending;
-
-		vsp1_write(vsp1, VI6_DL_HDR_ADDR(0), dl->body0.dma);
-		vsp1_write(vsp1, VI6_DL_BODY_SIZE, VI6_DL_BODY_SIZE_UPD |
-			   (dl->body0.num_entries *
-			    sizeof(*dl->header->lists)));
-
-		dlm->queued = dl;
+		vsp1_dl_list_hw_enqueue(dlm->pending);
+		dlm->queued = dlm->pending;
 		dlm->pending = NULL;
 	}
 
@@ -714,6 +756,7 @@ struct vsp1_dl_manager *vsp1_dlm_create(struct vsp1_device *vsp1,
 	dlm->index = index;
 	dlm->mode = index == 0 && !vsp1->info->uapi
 		  ? VSP1_DL_MODE_HEADERLESS : VSP1_DL_MODE_HEADER;
+	dlm->singleshot = vsp1->info->uapi;
 	dlm->vsp1 = vsp1;
 
 	spin_lock_init(&dlm->lock);

drivers/media/platform/vsp1/vsp1_dl.h

@@ -27,7 +27,6 @@ struct vsp1_dl_manager *vsp1_dlm_create(struct vsp1_device *vsp1,
 					unsigned int prealloc);
 void vsp1_dlm_destroy(struct vsp1_dl_manager *dlm);
 void vsp1_dlm_reset(struct vsp1_dl_manager *dlm);
-void vsp1_dlm_irq_display_start(struct vsp1_dl_manager *dlm);
 bool vsp1_dlm_irq_frame_end(struct vsp1_dl_manager *dlm);
 
 struct vsp1_dl_list *vsp1_dl_list_get(struct vsp1_dl_manager *dlm);

drivers/media/platform/vsp1/vsp1_drm.c

@@ -32,17 +32,13 @@
  * Interrupt Handling
  */
 
-void vsp1_drm_display_start(struct vsp1_device *vsp1)
+static void vsp1_du_pipeline_frame_end(struct vsp1_pipeline *pipe,
+				       bool completed)
 {
-	vsp1_dlm_irq_display_start(vsp1->drm->pipe.output->dlm);
-}
-
-static void vsp1_du_pipeline_frame_end(struct vsp1_pipeline *pipe)
-{
-	struct vsp1_drm *drm = to_vsp1_drm(pipe);
+	struct vsp1_drm_pipeline *drm_pipe = to_vsp1_drm_pipeline(pipe);
 
-	if (drm->du_complete)
-		drm->du_complete(drm->du_private);
+	if (drm_pipe->du_complete)
+		drm_pipe->du_complete(drm_pipe->du_private, completed);
 }
 
 /* -----------------------------------------------------------------------------
@@ -63,29 +59,44 @@ EXPORT_SYMBOL_GPL(vsp1_du_init);
 /**
  * vsp1_du_setup_lif - Setup the output part of the VSP pipeline
  * @dev: the VSP device
+ * @pipe_index: the DRM pipeline index
  * @cfg: the LIF configuration
  *
  * Configure the output part of VSP DRM pipeline for the given frame @cfg.width
- * and @cfg.height. This sets up formats on the BRU source pad, the WPF0 sink
- * and source pads, and the LIF sink pad.
+ * and @cfg.height. This sets up formats on the blend unit (BRU or BRS) source
+ * pad, the WPF sink and source pads, and the LIF sink pad.
+ *
+ * The @pipe_index argument selects which DRM pipeline to setup. The number of
+ * available pipelines depend on the VSP instance.
  *
- * As the media bus code on the BRU source pad is conditioned by the
- * configuration of the BRU sink 0 pad, we also set up the formats on all BRU
+ * As the media bus code on the blend unit source pad is conditioned by the
+ * configuration of its sink 0 pad, we also set up the formats on all blend unit
  * sinks, even if the configuration will be overwritten later by
- * vsp1_du_setup_rpf(). This ensures that the BRU configuration is set to a well
- * defined state.
+ * vsp1_du_setup_rpf(). This ensures that the blend unit configuration is set to
+ * a well defined state.
  *
  * Return 0 on success or a negative error code on failure.
  */
-int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
+int vsp1_du_setup_lif(struct device *dev, unsigned int pipe_index,
+		      const struct vsp1_du_lif_config *cfg)
 {
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
-	struct vsp1_pipeline *pipe = &vsp1->drm->pipe;
-	struct vsp1_bru *bru = vsp1->bru;
+	struct vsp1_drm_pipeline *drm_pipe;
+	struct vsp1_pipeline *pipe;
+	struct vsp1_bru *bru;
 	struct v4l2_subdev_format format;
+	const char *bru_name;
 	unsigned int i;
 	int ret;
 
+	if (pipe_index >= vsp1->info->lif_count)
+		return -EINVAL;
+
+	drm_pipe = &vsp1->drm->pipe[pipe_index];
+	pipe = &drm_pipe->pipe;
+	bru = to_bru(&pipe->bru->subdev);
+	bru_name = pipe->bru->type == VSP1_ENTITY_BRU ? "BRU" : "BRS";
+
 	if (!cfg) {
 		/*
 		 * NULL configuration means the CRTC is being disabled, stop
@@ -97,14 +108,25 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 
 		media_pipeline_stop(&pipe->output->entity.subdev.entity);
 
-		for (i = 0; i < bru->entity.source_pad; ++i) {
-			vsp1->drm->inputs[i].enabled = false;
-			bru->inputs[i].rpf = NULL;
+		for (i = 0; i < ARRAY_SIZE(pipe->inputs); ++i) {
+			struct vsp1_rwpf *rpf = pipe->inputs[i];
+
+			if (!rpf)
+				continue;
+
+			/*
+			 * Remove the RPF from the pipe and the list of BRU
+			 * inputs.
+			 */
+			WARN_ON(list_empty(&rpf->entity.list_pipe));
+			list_del_init(&rpf->entity.list_pipe);
 			pipe->inputs[i] = NULL;
+
+			bru->inputs[rpf->bru_input].rpf = NULL;
 		}
 
+		drm_pipe->du_complete = NULL;
 		pipe->num_inputs = 0;
-		vsp1->drm->du_complete = NULL;
 
 		vsp1_dlm_reset(pipe->output->dlm);
 		vsp1_device_put(vsp1);
@@ -114,8 +136,8 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 		return 0;
 	}
 
-	dev_dbg(vsp1->dev, "%s: configuring LIF with format %ux%u\n",
-		__func__, cfg->width, cfg->height);
+	dev_dbg(vsp1->dev, "%s: configuring LIF%u with format %ux%u\n",
+		__func__, pipe_index, cfg->width, cfg->height);
 
 	/*
 	 * Configure the format at the BRU sinks and propagate it through the
@@ -124,7 +146,7 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 	memset(&format, 0, sizeof(format));
 	format.which = V4L2_SUBDEV_FORMAT_ACTIVE;
 
-	for (i = 0; i < bru->entity.source_pad; ++i) {
+	for (i = 0; i < pipe->bru->source_pad; ++i) {
 		format.pad = i;
 
 		format.format.width = cfg->width;
@@ -132,60 +154,60 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 		format.format.code = MEDIA_BUS_FMT_ARGB8888_1X32;
 		format.format.field = V4L2_FIELD_NONE;
 
-		ret = v4l2_subdev_call(&bru->entity.subdev, pad,
+		ret = v4l2_subdev_call(&pipe->bru->subdev, pad,
 				       set_fmt, NULL, &format);
 		if (ret < 0)
 			return ret;
 
-		dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on BRU pad %u\n",
+		dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on %s pad %u\n",
 			__func__, format.format.width, format.format.height,
-			format.format.code, i);
+			format.format.code, bru_name, i);
 	}
 
-	format.pad = bru->entity.source_pad;
+	format.pad = pipe->bru->source_pad;
 	format.format.width = cfg->width;
 	format.format.height = cfg->height;
 	format.format.code = MEDIA_BUS_FMT_ARGB8888_1X32;
 	format.format.field = V4L2_FIELD_NONE;
 
-	ret = v4l2_subdev_call(&bru->entity.subdev, pad, set_fmt, NULL,
+	ret = v4l2_subdev_call(&pipe->bru->subdev, pad, set_fmt, NULL,
 			       &format);
 	if (ret < 0)
 		return ret;
 
-	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on BRU pad %u\n",
+	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on %s pad %u\n",
 		__func__, format.format.width, format.format.height,
-		format.format.code, i);
+		format.format.code, bru_name, i);
 
 	format.pad = RWPF_PAD_SINK;
-	ret = v4l2_subdev_call(&vsp1->wpf[0]->entity.subdev, pad, set_fmt, NULL,
+	ret = v4l2_subdev_call(&pipe->output->entity.subdev, pad, set_fmt, NULL,
 			       &format);
 	if (ret < 0)
 		return ret;
 
-	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on WPF0 sink\n",
+	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on WPF%u sink\n",
 		__func__, format.format.width, format.format.height,
-		format.format.code);
+		format.format.code, pipe->output->entity.index);
 
 	format.pad = RWPF_PAD_SOURCE;
-	ret = v4l2_subdev_call(&vsp1->wpf[0]->entity.subdev, pad, get_fmt, NULL,
+	ret = v4l2_subdev_call(&pipe->output->entity.subdev, pad, get_fmt, NULL,
 			       &format);
 	if (ret < 0)
 		return ret;
 
-	dev_dbg(vsp1->dev, "%s: got format %ux%u (%x) on WPF0 source\n",
+	dev_dbg(vsp1->dev, "%s: got format %ux%u (%x) on WPF%u source\n",
 		__func__, format.format.width, format.format.height,
-		format.format.code);
+		format.format.code, pipe->output->entity.index);
 
 	format.pad = LIF_PAD_SINK;
-	ret = v4l2_subdev_call(&vsp1->lif->entity.subdev, pad, set_fmt, NULL,
+	ret = v4l2_subdev_call(&pipe->lif->subdev, pad, set_fmt, NULL,
 			       &format);
 	if (ret < 0)
 		return ret;
 
-	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on LIF sink\n",
+	dev_dbg(vsp1->dev, "%s: set format %ux%u (%x) on LIF%u sink\n",
 		__func__, format.format.width, format.format.height,
-		format.format.code);
+		format.format.code, pipe_index);
 
 	/*
 	 * Verify that the format at the output of the pipeline matches the
@@ -213,8 +235,8 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 	 * Register a callback to allow us to notify the DRM driver of frame
 	 * completion events.
 	 */
-	vsp1->drm->du_complete = cfg->callback;
-	vsp1->drm->du_private = cfg->callback_data;
+	drm_pipe->du_complete = cfg->callback;
+	drm_pipe->du_private = cfg->callback_data;
 
 	ret = media_pipeline_start(&pipe->output->entity.subdev.entity,
 					  &pipe->pipe);
@@ -224,6 +246,10 @@ int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg)
 		return ret;
 	}
 
+	/* Disable the display interrupts. */
+	vsp1_write(vsp1, VI6_DISP_IRQ_STA, 0);
+	vsp1_write(vsp1, VI6_DISP_IRQ_ENB, 0);
+
 	dev_dbg(vsp1->dev, "%s: pipeline enabled\n", __func__);
 
 	return 0;
@@ -233,19 +259,21 @@ EXPORT_SYMBOL_GPL(vsp1_du_setup_lif);
 /**
  * vsp1_du_atomic_begin - Prepare for an atomic update
  * @dev: the VSP device
+ * @pipe_index: the DRM pipeline index
  */
-void vsp1_du_atomic_begin(struct device *dev)
+void vsp1_du_atomic_begin(struct device *dev, unsigned int pipe_index)
 {
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
-	struct vsp1_pipeline *pipe = &vsp1->drm->pipe;
+	struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[pipe_index];
 
-	vsp1->drm->num_inputs = pipe->num_inputs;
+	drm_pipe->enabled = drm_pipe->pipe.num_inputs != 0;
 }
 EXPORT_SYMBOL_GPL(vsp1_du_atomic_begin);
 
 /**
  * vsp1_du_atomic_update - Setup one RPF input of the VSP pipeline
  * @dev: the VSP device
+ * @pipe_index: the DRM pipeline index
  * @rpf_index: index of the RPF to setup (0-based)
  * @cfg: the RPF configuration
  *
@@ -272,10 +300,12 @@ EXPORT_SYMBOL_GPL(vsp1_du_atomic_begin);
  *
  * Return 0 on success or a negative error code on failure.
  */
-int vsp1_du_atomic_update(struct device *dev, unsigned int rpf_index,
+int vsp1_du_atomic_update(struct device *dev, unsigned int pipe_index,
+			  unsigned int rpf_index,
 			  const struct vsp1_du_atomic_config *cfg)
 {
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
+	struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[pipe_index];
 	const struct vsp1_format_info *fmtinfo;
 	struct vsp1_rwpf *rpf;
 
@@ -288,7 +318,12 @@ int vsp1_du_atomic_update(struct device *dev, unsigned int rpf_index,
 		dev_dbg(vsp1->dev, "%s: RPF%u: disable requested\n", __func__,
 			rpf_index);
 
-		vsp1->drm->inputs[rpf_index].enabled = false;
+		/*
+		 * Remove the RPF from the pipe's inputs. The atomic flush
+		 * handler will disable the input and remove the entity from the
+		 * pipe's entities list.
+		 */
+		drm_pipe->pipe.inputs[rpf_index] = NULL;
 		return 0;
 	}
 
@@ -324,13 +359,15 @@ int vsp1_du_atomic_update(struct device *dev, unsigned int rpf_index,
 	vsp1->drm->inputs[rpf_index].crop = cfg->src;
 	vsp1->drm->inputs[rpf_index].compose = cfg->dst;
 	vsp1->drm->inputs[rpf_index].zpos = cfg->zpos;
-	vsp1->drm->inputs[rpf_index].enabled = true;
+
+	drm_pipe->pipe.inputs[rpf_index] = rpf;
 
 	return 0;
 }
 EXPORT_SYMBOL_GPL(vsp1_du_atomic_update);
 
 static int vsp1_du_setup_rpf_pipe(struct vsp1_device *vsp1,
+				  struct vsp1_pipeline *pipe,
 				  struct vsp1_rwpf *rpf, unsigned int bru_input)
 {
 	struct v4l2_subdev_selection sel;
@@ -404,7 +441,7 @@ static int vsp1_du_setup_rpf_pipe(struct vsp1_device *vsp1,
 	/* BRU sink, propagate the format from the RPF source. */
 	format.pad = bru_input;
 
-	ret = v4l2_subdev_call(&vsp1->bru->entity.subdev, pad, set_fmt, NULL,
+	ret = v4l2_subdev_call(&pipe->bru->subdev, pad, set_fmt, NULL,
 			       &format);
 	if (ret < 0)
 		return ret;
@@ -417,8 +454,8 @@ static int vsp1_du_setup_rpf_pipe(struct vsp1_device *vsp1,
 	sel.target = V4L2_SEL_TGT_COMPOSE;
 	sel.r = vsp1->drm->inputs[rpf->entity.index].compose;
 
-	ret = v4l2_subdev_call(&vsp1->bru->entity.subdev, pad, set_selection,
-			       NULL, &sel);
+	ret = v4l2_subdev_call(&pipe->bru->subdev, pad, set_selection, NULL,
+			       &sel);
 	if (ret < 0)
 		return ret;
 
@@ -438,18 +475,25 @@ static unsigned int rpf_zpos(struct vsp1_device *vsp1, struct vsp1_rwpf *rpf)
 /**
  * vsp1_du_atomic_flush - Commit an atomic update
  * @dev: the VSP device
+ * @pipe_index: the DRM pipeline index
  */
-void vsp1_du_atomic_flush(struct device *dev)
+void vsp1_du_atomic_flush(struct device *dev, unsigned int pipe_index)
 {
 	struct vsp1_device *vsp1 = dev_get_drvdata(dev);
-	struct vsp1_pipeline *pipe = &vsp1->drm->pipe;
+	struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[pipe_index];
+	struct vsp1_pipeline *pipe = &drm_pipe->pipe;
 	struct vsp1_rwpf *inputs[VSP1_MAX_RPF] = { NULL, };
+	struct vsp1_bru *bru = to_bru(&pipe->bru->subdev);
 	struct vsp1_entity *entity;
+	struct vsp1_entity *next;
 	struct vsp1_dl_list *dl;
+	const char *bru_name;
 	unsigned long flags;
 	unsigned int i;
 	int ret;
 
+	bru_name = pipe->bru->type == VSP1_ENTITY_BRU ? "BRU" : "BRS";
+
 	/* Prepare the display list. */
 	dl = vsp1_dl_list_get(pipe->output->dlm);
 
@@ -460,12 +504,8 @@ void vsp1_du_atomic_flush(struct device *dev)
 		struct vsp1_rwpf *rpf = vsp1->rpf[i];
 		unsigned int j;
 
-		if (!vsp1->drm->inputs[i].enabled) {
-			pipe->inputs[i] = NULL;
+		if (!pipe->inputs[i])
 			continue;
-		}
-
-		pipe->inputs[i] = rpf;
 
 		/* Insert the RPF in the sorted RPFs array. */
 		for (j = pipe->num_inputs++; j > 0; --j) {
@@ -478,22 +518,26 @@ void vsp1_du_atomic_flush(struct device *dev)
 	}
 
 	/* Setup the RPF input pipeline for every enabled input. */
-	for (i = 0; i < vsp1->info->num_bru_inputs; ++i) {
+	for (i = 0; i < pipe->bru->source_pad; ++i) {
 		struct vsp1_rwpf *rpf = inputs[i];
 
 		if (!rpf) {
-			vsp1->bru->inputs[i].rpf = NULL;
+			bru->inputs[i].rpf = NULL;
 			continue;
 		}
 
-		vsp1->bru->inputs[i].rpf = rpf;
+		if (list_empty(&rpf->entity.list_pipe))
+			list_add_tail(&rpf->entity.list_pipe, &pipe->entities);
+
+		bru->inputs[i].rpf = rpf;
 		rpf->bru_input = i;
+		rpf->entity.sink = pipe->bru;
 		rpf->entity.sink_pad = i;
 
-		dev_dbg(vsp1->dev, "%s: connecting RPF.%u to BRU:%u\n",
-			__func__, rpf->entity.index, i);
+		dev_dbg(vsp1->dev, "%s: connecting RPF.%u to %s:%u\n",
+			__func__, rpf->entity.index, bru_name, i);
 
-		ret = vsp1_du_setup_rpf_pipe(vsp1, rpf, i);
+		ret = vsp1_du_setup_rpf_pipe(vsp1, pipe, rpf, i);
 		if (ret < 0)
 			dev_err(vsp1->dev,
 				"%s: failed to setup RPF.%u\n",
@@ -501,16 +545,16 @@ void vsp1_du_atomic_flush(struct device *dev)
 	}
 
 	/* Configure all entities in the pipeline. */
-	list_for_each_entry(entity, &pipe->entities, list_pipe) {
+	list_for_each_entry_safe(entity, next, &pipe->entities, list_pipe) {
 		/* Disconnect unused RPFs from the pipeline. */
-		if (entity->type == VSP1_ENTITY_RPF) {
-			struct vsp1_rwpf *rpf = to_rwpf(&entity->subdev);
+		if (entity->type == VSP1_ENTITY_RPF &&
+		    !pipe->inputs[entity->index]) {
+			vsp1_dl_list_write(dl, entity->route->reg,
+					   VI6_DPR_NODE_UNUSED);
 
-			if (!pipe->inputs[rpf->entity.index]) {
-				vsp1_dl_list_write(dl, entity->route->reg,
-						   VI6_DPR_NODE_UNUSED);
-				continue;
-			}
+			list_del_init(&entity->list_pipe);
+
+			continue;
 		}
 
 		vsp1_entity_route_setup(entity, pipe, dl);
@@ -528,14 +572,11 @@ void vsp1_du_atomic_flush(struct device *dev)
 	vsp1_dl_list_commit(dl);
 
 	/* Start or stop the pipeline if needed. */
-	if (!vsp1->drm->num_inputs && pipe->num_inputs) {
-		vsp1_write(vsp1, VI6_DISP_IRQ_STA, 0);
-		vsp1_write(vsp1, VI6_DISP_IRQ_ENB, VI6_DISP_IRQ_ENB_DSTE);
+	if (!drm_pipe->enabled && pipe->num_inputs) {
 		spin_lock_irqsave(&pipe->irqlock, flags);
 		vsp1_pipeline_run(pipe);
 		spin_unlock_irqrestore(&pipe->irqlock, flags);
-	} else if (vsp1->drm->num_inputs && !pipe->num_inputs) {
-		vsp1_write(vsp1, VI6_DISP_IRQ_ENB, 0);
+	} else if (drm_pipe->enabled && !pipe->num_inputs) {
 		vsp1_pipeline_stop(pipe);
 	}
 }
@@ -568,83 +609,48 @@ EXPORT_SYMBOL_GPL(vsp1_du_unmap_sg);
  * Initialization
  */
 
-int vsp1_drm_create_links(struct vsp1_device *vsp1)
-{
-	const u32 flags = MEDIA_LNK_FL_ENABLED | MEDIA_LNK_FL_IMMUTABLE;
-	unsigned int i;
-	int ret;
-
-	/*
-	 * VSPD instances require a BRU to perform composition and a LIF to
-	 * output to the DU.
-	 */
-	if (!vsp1->bru || !vsp1->lif)
-		return -ENXIO;
-
-	for (i = 0; i < vsp1->info->rpf_count; ++i) {
-		struct vsp1_rwpf *rpf = vsp1->rpf[i];
-
-		ret = media_create_pad_link(&rpf->entity.subdev.entity,
-					    RWPF_PAD_SOURCE,
-					    &vsp1->bru->entity.subdev.entity,
-					    i, flags);
-		if (ret < 0)
-			return ret;
-
-		rpf->entity.sink = &vsp1->bru->entity.subdev.entity;
-		rpf->entity.sink_pad = i;
-	}
-
-	ret = media_create_pad_link(&vsp1->bru->entity.subdev.entity,
-				    vsp1->bru->entity.source_pad,
-				    &vsp1->wpf[0]->entity.subdev.entity,
-				    RWPF_PAD_SINK, flags);
-	if (ret < 0)
-		return ret;
-
-	vsp1->bru->entity.sink = &vsp1->wpf[0]->entity.subdev.entity;
-	vsp1->bru->entity.sink_pad = RWPF_PAD_SINK;
-
-	ret = media_create_pad_link(&vsp1->wpf[0]->entity.subdev.entity,
-				    RWPF_PAD_SOURCE,
-				    &vsp1->lif->entity.subdev.entity,
-				    LIF_PAD_SINK, flags);
-	if (ret < 0)
-		return ret;
-
-	return 0;
-}
-
 int vsp1_drm_init(struct vsp1_device *vsp1)
 {
-	struct vsp1_pipeline *pipe;
 	unsigned int i;
 
 	vsp1->drm = devm_kzalloc(vsp1->dev, sizeof(*vsp1->drm), GFP_KERNEL);
 	if (!vsp1->drm)
 		return -ENOMEM;
 
-	pipe = &vsp1->drm->pipe;
+	/* Create one DRM pipeline per LIF. */
+	for (i = 0; i < vsp1->info->lif_count; ++i) {
+		struct vsp1_drm_pipeline *drm_pipe = &vsp1->drm->pipe[i];
+		struct vsp1_pipeline *pipe = &drm_pipe->pipe;
 
-	vsp1_pipeline_init(pipe);
+		vsp1_pipeline_init(pipe);
 
-	/* The DRM pipeline is static, add entities manually. */
+		/*
+		 * The DRM pipeline is static, add entities manually. The first
+		 * pipeline uses the BRU and the second pipeline the BRS.
+		 */
+		pipe->bru = i == 0 ? &vsp1->bru->entity : &vsp1->brs->entity;
+		pipe->lif = &vsp1->lif[i]->entity;
+		pipe->output = vsp1->wpf[i];
+		pipe->output->pipe = pipe;
+		pipe->frame_end = vsp1_du_pipeline_frame_end;
+
+		pipe->bru->sink = &pipe->output->entity;
+		pipe->bru->sink_pad = 0;
+		pipe->output->entity.sink = pipe->lif;
+		pipe->output->entity.sink_pad = 0;
+
+		list_add_tail(&pipe->bru->list_pipe, &pipe->entities);
+		list_add_tail(&pipe->lif->list_pipe, &pipe->entities);
+		list_add_tail(&pipe->output->entity.list_pipe, &pipe->entities);
+	}
+
+	/* Disable all RPFs initially. */
 	for (i = 0; i < vsp1->info->rpf_count; ++i) {
 		struct vsp1_rwpf *input = vsp1->rpf[i];
 
-		list_add_tail(&input->entity.list_pipe, &pipe->entities);
+		INIT_LIST_HEAD(&input->entity.list_pipe);
 	}
 
-	list_add_tail(&vsp1->bru->entity.list_pipe, &pipe->entities);
-	list_add_tail(&vsp1->wpf[0]->entity.list_pipe, &pipe->entities);
-	list_add_tail(&vsp1->lif->entity.list_pipe, &pipe->entities);
-
-	pipe->bru = &vsp1->bru->entity;
-	pipe->lif = &vsp1->lif->entity;
-	pipe->output = vsp1->wpf[0];
-	pipe->output->pipe = pipe;
-	pipe->frame_end = vsp1_du_pipeline_frame_end;
-
 	return 0;
 }
 

drivers/media/platform/vsp1/vsp1_drm.h

@@ -18,38 +18,44 @@
 #include "vsp1_pipe.h"
 
 /**
- * vsp1_drm - State for the API exposed to the DRM driver
+ * vsp1_drm_pipeline - State for the API exposed to the DRM driver
  * @pipe: the VSP1 pipeline used for display
- * @num_inputs: number of active pipeline inputs at the beginning of an update
- * @inputs: source crop rectangle, destination compose rectangle and z-order
- *	position for every input
+ * @enabled: pipeline state at the beginning of an update
  * @du_complete: frame completion callback for the DU driver (optional)
  * @du_private: data to be passed to the du_complete callback
  */
-struct vsp1_drm {
+struct vsp1_drm_pipeline {
 	struct vsp1_pipeline pipe;
-	unsigned int num_inputs;
+	bool enabled;
+
+	/* Frame synchronisation */
+	void (*du_complete)(void *, bool);
+	void *du_private;
+};
+
+/**
+ * vsp1_drm - State for the API exposed to the DRM driver
+ * @pipe: the VSP1 DRM pipeline used for display
+ * @inputs: source crop rectangle, destination compose rectangle and z-order
+ *	position for every input (indexed by RPF index)
+ */
+struct vsp1_drm {
+	struct vsp1_drm_pipeline pipe[VSP1_MAX_LIF];
+
 	struct {
-		bool enabled;
 		struct v4l2_rect crop;
 		struct v4l2_rect compose;
 		unsigned int zpos;
 	} inputs[VSP1_MAX_RPF];
-
-	/* Frame synchronisation */
-	void (*du_complete)(void *);
-	void *du_private;
 };
 
-static inline struct vsp1_drm *to_vsp1_drm(struct vsp1_pipeline *pipe)
+static inline struct vsp1_drm_pipeline *
+to_vsp1_drm_pipeline(struct vsp1_pipeline *pipe)
 {
-	return container_of(pipe, struct vsp1_drm, pipe);
+	return container_of(pipe, struct vsp1_drm_pipeline, pipe);
 }
 
 int vsp1_drm_init(struct vsp1_device *vsp1);
 void vsp1_drm_cleanup(struct vsp1_device *vsp1);
-int vsp1_drm_create_links(struct vsp1_device *vsp1);
-
-void vsp1_drm_display_start(struct vsp1_device *vsp1);
 
 #endif /* __VSP1_DRM_H__ */

drivers/media/platform/vsp1/vsp1_drv.c

@@ -68,14 +68,6 @@ static irqreturn_t vsp1_irq_handler(int irq, void *data)
 		}
 	}
 
-	status = vsp1_read(vsp1, VI6_DISP_IRQ_STA);
-	vsp1_write(vsp1, VI6_DISP_IRQ_STA, ~status & VI6_DISP_IRQ_STA_DST);
-
-	if (status & VI6_DISP_IRQ_STA_DST) {
-		vsp1_drm_display_start(vsp1);
-		ret = IRQ_HANDLED;
-	}
-
 	return ret;
 }
 
@@ -92,6 +84,10 @@ static irqreturn_t vsp1_irq_handler(int irq, void *data)
  *
  * - from a UDS to a UDS (UDS entities can't be chained)
  * - from an entity to itself (no loops are allowed)
+ *
+ * Furthermore, the BRS can't be connected to histogram generators, but no
+ * special check is currently needed as all VSP instances that include a BRS
+ * have no histogram generator.
  */
 static int vsp1_create_sink_links(struct vsp1_device *vsp1,
 				  struct vsp1_entity *sink)
@@ -129,7 +125,7 @@ static int vsp1_create_sink_links(struct vsp1_device *vsp1,
 				return ret;
 
 			if (flags & MEDIA_LNK_FL_ENABLED)
-				source->sink = entity;
+				source->sink = sink;
 		}
 	}
 
@@ -172,10 +168,13 @@ static int vsp1_uapi_create_links(struct vsp1_device *vsp1)
 			return ret;
 	}
 
-	if (vsp1->lif) {
-		ret = media_create_pad_link(&vsp1->wpf[0]->entity.subdev.entity,
+	for (i = 0; i < vsp1->info->lif_count; ++i) {
+		if (!vsp1->lif[i])
+			continue;
+
+		ret = media_create_pad_link(&vsp1->wpf[i]->entity.subdev.entity,
 					    RWPF_PAD_SOURCE,
-					    &vsp1->lif->entity.subdev.entity,
+					    &vsp1->lif[i]->entity.subdev.entity,
 					    LIF_PAD_SINK, 0);
 		if (ret < 0)
 			return ret;
@@ -269,8 +268,18 @@ static int vsp1_create_entities(struct vsp1_device *vsp1)
 	}
 
 	/* Instantiate all the entities. */
+	if (vsp1->info->features & VSP1_HAS_BRS) {
+		vsp1->brs = vsp1_bru_create(vsp1, VSP1_ENTITY_BRS);
+		if (IS_ERR(vsp1->brs)) {
+			ret = PTR_ERR(vsp1->brs);
+			goto done;
+		}
+
+		list_add_tail(&vsp1->brs->entity.list_dev, &vsp1->entities);
+	}
+
 	if (vsp1->info->features & VSP1_HAS_BRU) {
-		vsp1->bru = vsp1_bru_create(vsp1);
+		vsp1->bru = vsp1_bru_create(vsp1, VSP1_ENTITY_BRU);
 		if (IS_ERR(vsp1->bru)) {
 			ret = PTR_ERR(vsp1->bru);
 			goto done;
@@ -328,18 +337,23 @@ static int vsp1_create_entities(struct vsp1_device *vsp1)
 	}
 
 	/*
-	 * The LIF is only supported when used in conjunction with the DU, in
+	 * The LIFs are only supported when used in conjunction with the DU, in
 	 * which case the userspace API is disabled. If the userspace API is
-	 * enabled skip the LIF, even when present.
+	 * enabled skip the LIFs, even when present.
 	 */
-	if (vsp1->info->features & VSP1_HAS_LIF && !vsp1->info->uapi) {
-		vsp1->lif = vsp1_lif_create(vsp1);
-		if (IS_ERR(vsp1->lif)) {
-			ret = PTR_ERR(vsp1->lif);
-			goto done;
-		}
+	if (!vsp1->info->uapi) {
+		for (i = 0; i < vsp1->info->lif_count; ++i) {
+			struct vsp1_lif *lif;
+
+			lif = vsp1_lif_create(vsp1, i);
+			if (IS_ERR(lif)) {
+				ret = PTR_ERR(lif);
+				goto done;
+			}
 
-		list_add_tail(&vsp1->lif->entity.list_dev, &vsp1->entities);
+			vsp1->lif[i] = lif;
+			list_add_tail(&lif->entity.list_dev, &vsp1->entities);
+		}
 	}
 
 	if (vsp1->info->features & VSP1_HAS_LUT) {
@@ -420,7 +434,6 @@ static int vsp1_create_entities(struct vsp1_device *vsp1)
 			}
 
 			list_add_tail(&video->list, &vsp1->videos);
-			wpf->entity.sink = &video->video.entity;
 		}
 	}
 
@@ -432,19 +445,15 @@ static int vsp1_create_entities(struct vsp1_device *vsp1)
 			goto done;
 	}
 
-	/* Create links. */
-	if (vsp1->info->uapi)
-		ret = vsp1_uapi_create_links(vsp1);
-	else
-		ret = vsp1_drm_create_links(vsp1);
-	if (ret < 0)
-		goto done;
-
 	/*
-	 * Register subdev nodes if the userspace API is enabled or initialize
-	 * the DRM pipeline otherwise.
+	 * Create links and register subdev nodes if the userspace API is
+	 * enabled or initialize the DRM pipeline otherwise.
 	 */
 	if (vsp1->info->uapi) {
+		ret = vsp1_uapi_create_links(vsp1);
+		if (ret < 0)
+			goto done;
+
 		ret = v4l2_device_register_subdev_nodes(&vsp1->v4l2_dev);
 		if (ret < 0)
 			goto done;
@@ -515,6 +524,9 @@ static int vsp1_device_init(struct vsp1_device *vsp1)
 	vsp1_write(vsp1, VI6_DPR_HSI_ROUTE, VI6_DPR_NODE_UNUSED);
 	vsp1_write(vsp1, VI6_DPR_BRU_ROUTE, VI6_DPR_NODE_UNUSED);
 
+	if (vsp1->info->features & VSP1_HAS_BRS)
+		vsp1_write(vsp1, VI6_DPR_ILV_BRS_ROUTE, VI6_DPR_NODE_UNUSED);
+
 	vsp1_write(vsp1, VI6_DPR_HGO_SMPPT, (7 << VI6_DPR_SMPPT_TGW_SHIFT) |
 		   (VI6_DPR_NODE_UNUSED << VI6_DPR_SMPPT_PT_SHIFT));
 	vsp1_write(vsp1, VI6_DPR_HGT_SMPPT, (7 << VI6_DPR_SMPPT_TGW_SHIFT) |
@@ -634,8 +646,8 @@ static const struct vsp1_device_info vsp1_device_infos[] = {
 		.version = VI6_IP_VERSION_MODEL_VSPD_GEN2,
 		.model = "VSP1-D",
 		.gen = 2,
-		.features = VSP1_HAS_BRU | VSP1_HAS_HGO | VSP1_HAS_LIF
-			  | VSP1_HAS_LUT,
+		.features = VSP1_HAS_BRU | VSP1_HAS_HGO | VSP1_HAS_LUT,
+		.lif_count = 1,
 		.rpf_count = 4,
 		.uds_count = 1,
 		.wpf_count = 1,
@@ -668,8 +680,8 @@ static const struct vsp1_device_info vsp1_device_infos[] = {
 		.version = VI6_IP_VERSION_MODEL_VSPD_V2H,
 		.model = "VSP1V-D",
 		.gen = 2,
-		.features = VSP1_HAS_BRU | VSP1_HAS_CLU | VSP1_HAS_LUT
-			  | VSP1_HAS_LIF,
+		.features = VSP1_HAS_BRU | VSP1_HAS_CLU | VSP1_HAS_LUT,
+		.lif_count = 1,
 		.rpf_count = 4,
 		.uds_count = 1,
 		.wpf_count = 1,
@@ -705,11 +717,38 @@ static const struct vsp1_device_info vsp1_device_infos[] = {
 		.wpf_count = 1,
 		.num_bru_inputs = 5,
 		.uapi = true,
+	}, {
+		.version = VI6_IP_VERSION_MODEL_VSPBS_GEN3,
+		.model = "VSP2-BS",
+		.gen = 3,
+		.features = VSP1_HAS_BRS | VSP1_HAS_WPF_VFLIP,
+		.rpf_count = 2,
+		.wpf_count = 1,
+		.uapi = true,
 	}, {
 		.version = VI6_IP_VERSION_MODEL_VSPD_GEN3,
 		.model = "VSP2-D",
 		.gen = 3,
-		.features = VSP1_HAS_BRU | VSP1_HAS_LIF | VSP1_HAS_WPF_VFLIP,
+		.features = VSP1_HAS_BRU | VSP1_HAS_WPF_VFLIP,
+		.lif_count = 1,
+		.rpf_count = 5,
+		.wpf_count = 2,
+		.num_bru_inputs = 5,
+	}, {
+		.version = VI6_IP_VERSION_MODEL_VSPD_V3,
+		.model = "VSP2-D",
+		.gen = 3,
+		.features = VSP1_HAS_BRS | VSP1_HAS_BRU,
+		.lif_count = 1,
+		.rpf_count = 5,
+		.wpf_count = 1,
+		.num_bru_inputs = 5,
+	}, {
+		.version = VI6_IP_VERSION_MODEL_VSPDL_GEN3,
+		.model = "VSP2-DL",
+		.gen = 3,
+		.features = VSP1_HAS_BRS | VSP1_HAS_BRU,
+		.lif_count = 2,
 		.rpf_count = 5,
 		.wpf_count = 2,
 		.num_bru_inputs = 5,

drivers/media/platform/vsp1/vsp1_entity.c

@@ -24,18 +24,12 @@
 #include "vsp1_pipe.h"
 #include "vsp1_rwpf.h"
 
-static inline struct vsp1_entity *
-media_entity_to_vsp1_entity(struct media_entity *entity)
-{
-	return container_of(entity, struct vsp1_entity, subdev.entity);
-}
-
 void vsp1_entity_route_setup(struct vsp1_entity *entity,
 			     struct vsp1_pipeline *pipe,
 			     struct vsp1_dl_list *dl)
 {
 	struct vsp1_entity *source;
-	struct vsp1_entity *sink;
+	u32 route;
 
 	if (entity->type == VSP1_ENTITY_HGO) {
 		u32 smppt;
@@ -44,7 +38,7 @@ void vsp1_entity_route_setup(struct vsp1_entity *entity,
 		 * The HGO is a special case, its routing is configured on the
 		 * sink pad.
 		 */
-		source = media_entity_to_vsp1_entity(entity->sources[0]);
+		source = entity->sources[0];
 		smppt = (pipe->output->entity.index << VI6_DPR_SMPPT_TGW_SHIFT)
 		      | (source->route->output << VI6_DPR_SMPPT_PT_SHIFT);
 
@@ -57,7 +51,7 @@ void vsp1_entity_route_setup(struct vsp1_entity *entity,
 		 * The HGT is a special case, its routing is configured on the
 		 * sink pad.
 		 */
-		source = media_entity_to_vsp1_entity(entity->sources[0]);
+		source = entity->sources[0];
 		smppt = (pipe->output->entity.index << VI6_DPR_SMPPT_TGW_SHIFT)
 		      | (source->route->output << VI6_DPR_SMPPT_PT_SHIFT);
 
@@ -69,9 +63,14 @@ void vsp1_entity_route_setup(struct vsp1_entity *entity,
 	if (source->route->reg == 0)
 		return;
 
-	sink = media_entity_to_vsp1_entity(source->sink);
-	vsp1_dl_list_write(dl, source->route->reg,
-			   sink->route->inputs[source->sink_pad]);
+	route = source->sink->route->inputs[source->sink_pad];
+	/*
+	 * The ILV and BRS share the same data path route. The extra BRSSEL bit
+	 * selects between the ILV and BRS.
+	 */
+	if (source->type == VSP1_ENTITY_BRS)
+		route |= VI6_DPR_ROUTE_BRSSEL;
+	vsp1_dl_list_write(dl, source->route->reg, route);
 }
 
 /* -----------------------------------------------------------------------------
@@ -316,6 +315,12 @@ done:
  * Media Operations
  */
 
+static inline struct vsp1_entity *
+media_entity_to_vsp1_entity(struct media_entity *entity)
+{
+	return container_of(entity, struct vsp1_entity, subdev.entity);
+}
+
 static int vsp1_entity_link_setup_source(const struct media_pad *source_pad,
 					 const struct media_pad *sink_pad,
 					 u32 flags)
@@ -339,7 +344,7 @@ static int vsp1_entity_link_setup_source(const struct media_pad *source_pad,
 		    sink->type != VSP1_ENTITY_HGT) {
 			if (source->sink)
 				return -EBUSY;
-			source->sink = sink_pad->entity;
+			source->sink = sink;
 			source->sink_pad = sink_pad->index;
 		}
 	} else {
@@ -355,15 +360,17 @@ static int vsp1_entity_link_setup_sink(const struct media_pad *source_pad,
 				       u32 flags)
 {
 	struct vsp1_entity *sink;
+	struct vsp1_entity *source;
 
 	sink = media_entity_to_vsp1_entity(sink_pad->entity);
+	source = media_entity_to_vsp1_entity(source_pad->entity);
 
 	if (flags & MEDIA_LNK_FL_ENABLED) {
 		/* Fan-in is limited to one. */
 		if (sink->sources[sink_pad->index])
 			return -EBUSY;
 
-		sink->sources[sink_pad->index] = source_pad->entity;
+		sink->sources[sink_pad->index] = source;
 	} else {
 		sink->sources[sink_pad->index] = NULL;
 	}
@@ -450,6 +457,8 @@ struct media_pad *vsp1_entity_remote_pad(struct media_pad *pad)
 	  { VI6_DPR_NODE_WPF(idx) }, VI6_DPR_NODE_WPF(idx) }
 
 static const struct vsp1_route vsp1_routes[] = {
+	{ VSP1_ENTITY_BRS, 0, VI6_DPR_ILV_BRS_ROUTE,
+	  { VI6_DPR_NODE_BRS_IN(0), VI6_DPR_NODE_BRS_IN(1) }, 0 },
 	{ VSP1_ENTITY_BRU, 0, VI6_DPR_BRU_ROUTE,
 	  { VI6_DPR_NODE_BRU_IN(0), VI6_DPR_NODE_BRU_IN(1),
 	    VI6_DPR_NODE_BRU_IN(2), VI6_DPR_NODE_BRU_IN(3),
@@ -459,7 +468,8 @@ static const struct vsp1_route vsp1_routes[] = {
 	{ VSP1_ENTITY_HGT, 0, 0, { 0, }, 0 },
 	VSP1_ENTITY_ROUTE(HSI),
 	VSP1_ENTITY_ROUTE(HST),
-	{ VSP1_ENTITY_LIF, 0, 0, { VI6_DPR_NODE_LIF, }, VI6_DPR_NODE_LIF },
+	{ VSP1_ENTITY_LIF, 0, 0, { 0, }, 0 },
+	{ VSP1_ENTITY_LIF, 1, 0, { 0, }, 0 },
 	VSP1_ENTITY_ROUTE(LUT),
 	VSP1_ENTITY_ROUTE_RPF(0),
 	VSP1_ENTITY_ROUTE_RPF(1),

drivers/media/platform/vsp1/vsp1_entity.h

@@ -23,6 +23,7 @@ struct vsp1_dl_list;
 struct vsp1_pipeline;
 
 enum vsp1_entity_type {
+	VSP1_ENTITY_BRS,
 	VSP1_ENTITY_BRU,
 	VSP1_ENTITY_CLU,
 	VSP1_ENTITY_HGO,
@@ -104,8 +105,8 @@ struct vsp1_entity {
 	struct media_pad *pads;
 	unsigned int source_pad;
 
-	struct media_entity **sources;
-	struct media_entity *sink;
+	struct vsp1_entity **sources;
+	struct vsp1_entity *sink;
 	unsigned int sink_pad;
 
 	struct v4l2_subdev subdev;

drivers/media/platform/vsp1/vsp1_lif.c

@@ -30,7 +30,7 @@
 static inline void vsp1_lif_write(struct vsp1_lif *lif, struct vsp1_dl_list *dl,
 				  u32 reg, u32 data)
 {
-	vsp1_dl_list_write(dl, reg, data);
+	vsp1_dl_list_write(dl, reg + lif->entity.index * VI6_LIF_OFFSET, data);
 }
 
 /* -----------------------------------------------------------------------------
@@ -165,7 +165,7 @@ static const struct vsp1_entity_operations lif_entity_ops = {
  * Initialization and Cleanup
  */
 
-struct vsp1_lif *vsp1_lif_create(struct vsp1_device *vsp1)
+struct vsp1_lif *vsp1_lif_create(struct vsp1_device *vsp1, unsigned int index)
 {
 	struct vsp1_lif *lif;
 	int ret;
@@ -176,6 +176,7 @@ struct vsp1_lif *vsp1_lif_create(struct vsp1_device *vsp1)
 
 	lif->entity.ops = &lif_entity_ops;
 	lif->entity.type = VSP1_ENTITY_LIF;
+	lif->entity.index = index;
 
 	/*
 	 * The LIF is never exposed to userspace, but media entity registration

drivers/media/platform/vsp1/vsp1_lif.h

@@ -32,6 +32,6 @@ static inline struct vsp1_lif *to_lif(struct v4l2_subdev *subdev)
 	return container_of(subdev, struct vsp1_lif, entity.subdev);
 }
 
-struct vsp1_lif *vsp1_lif_create(struct vsp1_device *vsp1);
+struct vsp1_lif *vsp1_lif_create(struct vsp1_device *vsp1, unsigned int index);
 
 #endif /* __VSP1_LIF_H__ */

drivers/media/platform/vsp1/vsp1_pipe.c

@@ -335,16 +335,12 @@ void vsp1_pipeline_frame_end(struct vsp1_pipeline *pipe)
 	if (pipe == NULL)
 		return;
 
+	/*
+	 * If the DL commit raced with the frame end interrupt, the commit ends
+	 * up being postponed by one frame. @completed represents whether the
+	 * active frame was finished or postponed.
+	 */
 	completed = vsp1_dlm_irq_frame_end(pipe->output->dlm);
-	if (!completed) {
-		/*
-		 * If the DL commit raced with the frame end interrupt, the
-		 * commit ends up being postponed by one frame. Return
-		 * immediately without calling the pipeline's frame end handler
-		 * or incrementing the sequence number.
-		 */
-		return;
-	}
 
 	if (pipe->hgo)
 		vsp1_hgo_frame_end(pipe->hgo);
@@ -352,8 +348,12 @@ void vsp1_pipeline_frame_end(struct vsp1_pipeline *pipe)
 	if (pipe->hgt)
 		vsp1_hgt_frame_end(pipe->hgt);
 
+	/*
+	 * Regardless of frame completion we still need to notify the pipe
+	 * frame_end to account for vblank events.
+	 */
 	if (pipe->frame_end)
-		pipe->frame_end(pipe);
+		pipe->frame_end(pipe, completed);
 
 	pipe->sequence++;
 }
@@ -373,10 +373,11 @@ void vsp1_pipeline_propagate_alpha(struct vsp1_pipeline *pipe,
 		return;
 
 	/*
-	 * The BRU background color has a fixed alpha value set to 255, the
-	 * output alpha value is thus always equal to 255.
+	 * The BRU and BRS background color has a fixed alpha value set to 255,
+	 * the output alpha value is thus always equal to 255.
 	 */
-	if (pipe->uds_input->type == VSP1_ENTITY_BRU)
+	if (pipe->uds_input->type == VSP1_ENTITY_BRU ||
+	    pipe->uds_input->type == VSP1_ENTITY_BRS)
 		alpha = 255;
 
 	vsp1_uds_set_alpha(pipe->uds, dl, alpha);

drivers/media/platform/vsp1/vsp1_pipe.h

@@ -91,7 +91,7 @@ struct vsp1_pipeline {
 	enum vsp1_pipeline_state state;
 	wait_queue_head_t wq;
 
-	void (*frame_end)(struct vsp1_pipeline *pipe);
+	void (*frame_end)(struct vsp1_pipeline *pipe, bool completed);
 
 	struct mutex lock;
 	struct kref kref;

drivers/media/platform/vsp1/vsp1_regs.h

@@ -18,6 +18,7 @@
  */
 
 #define VI6_CMD(n)			(0x0000 + (n) * 4)
+#define VI6_CMD_UPDHDR			(1 << 4)
 #define VI6_CMD_STRCMD			(1 << 0)
 
 #define VI6_CLK_DCSWT			0x0018
@@ -238,6 +239,10 @@
 #define VI6_WPF_SRCRPF_VIRACT_SUB	(1 << 28)
 #define VI6_WPF_SRCRPF_VIRACT_MST	(2 << 28)
 #define VI6_WPF_SRCRPF_VIRACT_MASK	(3 << 28)
+#define VI6_WPF_SRCRPF_VIRACT2_DIS	(0 << 24)
+#define VI6_WPF_SRCRPF_VIRACT2_SUB	(1 << 24)
+#define VI6_WPF_SRCRPF_VIRACT2_MST	(2 << 24)
+#define VI6_WPF_SRCRPF_VIRACT2_MASK	(3 << 24)
 #define VI6_WPF_SRCRPF_RPF_ACT_DIS(n)	(0 << ((n) * 2))
 #define VI6_WPF_SRCRPF_RPF_ACT_SUB(n)	(1 << ((n) * 2))
 #define VI6_WPF_SRCRPF_RPF_ACT_MST(n)	(2 << ((n) * 2))
@@ -321,6 +326,8 @@
 #define VI6_DPR_HST_ROUTE		0x2044
 #define VI6_DPR_HSI_ROUTE		0x2048
 #define VI6_DPR_BRU_ROUTE		0x204c
+#define VI6_DPR_ILV_BRS_ROUTE		0x2050
+#define VI6_DPR_ROUTE_BRSSEL		(1 << 28)
 #define VI6_DPR_ROUTE_FXA_MASK		(0xff << 16)
 #define VI6_DPR_ROUTE_FXA_SHIFT		16
 #define VI6_DPR_ROUTE_FP_MASK		(0x3f << 8)
@@ -344,7 +351,8 @@
 #define VI6_DPR_NODE_CLU		29
 #define VI6_DPR_NODE_HST		30
 #define VI6_DPR_NODE_HSI		31
-#define VI6_DPR_NODE_LIF		55
+#define VI6_DPR_NODE_BRS_IN(n)		(38 + (n))
+#define VI6_DPR_NODE_LIF		55		/* Gen2 only */
 #define VI6_DPR_NODE_WPF(n)		(56 + (n))
 #define VI6_DPR_NODE_UNUSED		63
 
@@ -476,7 +484,7 @@
 #define VI6_HSI_CTRL_EN			(1 << 0)
 
 /* -----------------------------------------------------------------------------
- * BRU Control Registers
+ * BRS and BRU Control Registers
  */
 
 #define VI6_ROP_NOP			0
@@ -496,7 +504,10 @@
 #define VI6_ROP_NAND			14
 #define VI6_ROP_SET			15
 
-#define VI6_BRU_INCTRL			0x2c00
+#define VI6_BRU_BASE			0x2c00
+#define VI6_BRS_BASE			0x3900
+
+#define VI6_BRU_INCTRL			0x0000
 #define VI6_BRU_INCTRL_NRM		(1 << 28)
 #define VI6_BRU_INCTRL_DnON		(1 << (16 + (n)))
 #define VI6_BRU_INCTRL_DITHn_OFF	(0 << ((n) * 4))
@@ -508,19 +519,19 @@
 #define VI6_BRU_INCTRL_DITHn_MASK	(7 << ((n) * 4))
 #define VI6_BRU_INCTRL_DITHn_SHIFT	((n) * 4)
 
-#define VI6_BRU_VIRRPF_SIZE		0x2c04
+#define VI6_BRU_VIRRPF_SIZE		0x0004
 #define VI6_BRU_VIRRPF_SIZE_HSIZE_MASK	(0x1fff << 16)
 #define VI6_BRU_VIRRPF_SIZE_HSIZE_SHIFT	16
 #define VI6_BRU_VIRRPF_SIZE_VSIZE_MASK	(0x1fff << 0)
 #define VI6_BRU_VIRRPF_SIZE_VSIZE_SHIFT	0
 
-#define VI6_BRU_VIRRPF_LOC		0x2c08
+#define VI6_BRU_VIRRPF_LOC		0x0008
 #define VI6_BRU_VIRRPF_LOC_HCOORD_MASK	(0x1fff << 16)
 #define VI6_BRU_VIRRPF_LOC_HCOORD_SHIFT	16
 #define VI6_BRU_VIRRPF_LOC_VCOORD_MASK	(0x1fff << 0)
 #define VI6_BRU_VIRRPF_LOC_VCOORD_SHIFT	0
 
-#define VI6_BRU_VIRRPF_COL		0x2c0c
+#define VI6_BRU_VIRRPF_COL		0x000c
 #define VI6_BRU_VIRRPF_COL_A_MASK	(0xff << 24)
 #define VI6_BRU_VIRRPF_COL_A_SHIFT	24
 #define VI6_BRU_VIRRPF_COL_RCR_MASK	(0xff << 16)
@@ -530,7 +541,7 @@
 #define VI6_BRU_VIRRPF_COL_BCB_MASK	(0xff << 0)
 #define VI6_BRU_VIRRPF_COL_BCB_SHIFT	0
 
-#define VI6_BRU_CTRL(n)			(0x2c10 + (n) * 8 + ((n) <= 3 ? 0 : 4))
+#define VI6_BRU_CTRL(n)			(0x0010 + (n) * 8 + ((n) <= 3 ? 0 : 4))
 #define VI6_BRU_CTRL_RBC		(1 << 31)
 #define VI6_BRU_CTRL_DSTSEL_BRUIN(n)	(((n) <= 3 ? (n) : (n)+1) << 20)
 #define VI6_BRU_CTRL_DSTSEL_VRPF	(4 << 20)
@@ -543,7 +554,7 @@
 #define VI6_BRU_CTRL_AROP(rop)		((rop) << 0)
 #define VI6_BRU_CTRL_AROP_MASK		(0xf << 0)
 
-#define VI6_BRU_BLD(n)			(0x2c14 + (n) * 8 + ((n) <= 3 ? 0 : 4))
+#define VI6_BRU_BLD(n)			(0x0014 + (n) * 8 + ((n) <= 3 ? 0 : 4))
 #define VI6_BRU_BLD_CBES		(1 << 31)
 #define VI6_BRU_BLD_CCMDX_DST_A		(0 << 28)
 #define VI6_BRU_BLD_CCMDX_255_DST_A	(1 << 28)
@@ -576,7 +587,7 @@
 #define VI6_BRU_BLD_COEFY_MASK		(0xff << 0)
 #define VI6_BRU_BLD_COEFY_SHIFT		0
 
-#define VI6_BRU_ROP			0x2c30
+#define VI6_BRU_ROP			0x0030	/* Only available on BRU */
 #define VI6_BRU_ROP_DSTSEL_BRUIN(n)	(((n) <= 3 ? (n) : (n)+1) << 20)
 #define VI6_BRU_ROP_DSTSEL_VRPF		(4 << 20)
 #define VI6_BRU_ROP_DSTSEL_MASK		(7 << 20)
@@ -653,6 +664,8 @@
  * LIF Control Registers
  */
 
+#define VI6_LIF_OFFSET			(-0x100)
+
 #define VI6_LIF_CTRL			0x3b00
 #define VI6_LIF_CTRL_OBTH_MASK		(0x7ff << 16)
 #define VI6_LIF_CTRL_OBTH_SHIFT		16
@@ -689,9 +702,20 @@
 #define VI6_IP_VERSION_MODEL_VSPBD_GEN3	(0x15 << 8)
 #define VI6_IP_VERSION_MODEL_VSPBC_GEN3	(0x16 << 8)
 #define VI6_IP_VERSION_MODEL_VSPD_GEN3	(0x17 << 8)
+#define VI6_IP_VERSION_MODEL_VSPD_V3	(0x18 << 8)
+#define VI6_IP_VERSION_MODEL_VSPDL_GEN3	(0x19 << 8)
+#define VI6_IP_VERSION_MODEL_VSPBS_GEN3	(0x1a << 8)
 #define VI6_IP_VERSION_SOC_MASK		(0xff << 0)
-#define VI6_IP_VERSION_SOC_H		(0x01 << 0)
-#define VI6_IP_VERSION_SOC_M		(0x02 << 0)
+#define VI6_IP_VERSION_SOC_H2		(0x01 << 0)
+#define VI6_IP_VERSION_SOC_V2H		(0x01 << 0)
+#define VI6_IP_VERSION_SOC_V3M		(0x01 << 0)
+#define VI6_IP_VERSION_SOC_M2		(0x02 << 0)
+#define VI6_IP_VERSION_SOC_M3W		(0x02 << 0)
+#define VI6_IP_VERSION_SOC_V3H		(0x02 << 0)
+#define VI6_IP_VERSION_SOC_H3		(0x03 << 0)
+#define VI6_IP_VERSION_SOC_D3		(0x04 << 0)
+#define VI6_IP_VERSION_SOC_M3N		(0x04 << 0)
+#define VI6_IP_VERSION_SOC_E3		(0x04 << 0)
 
 /* -----------------------------------------------------------------------------
  * RPF CLUT Registers

drivers/media/platform/vsp1/vsp1_video.c

@@ -440,13 +440,17 @@ static void vsp1_video_pipeline_run(struct vsp1_pipeline *pipe)
 	vsp1_pipeline_run(pipe);
 }
 
-static void vsp1_video_pipeline_frame_end(struct vsp1_pipeline *pipe)
+static void vsp1_video_pipeline_frame_end(struct vsp1_pipeline *pipe,
+					  bool completed)
 {
 	struct vsp1_device *vsp1 = pipe->output->entity.vsp1;
 	enum vsp1_pipeline_state state;
 	unsigned long flags;
 	unsigned int i;
 
+	/* M2M Pipelines should never call here with an incomplete frame. */
+	WARN_ON_ONCE(!completed);
+
 	spin_lock_irqsave(&pipe->irqlock, flags);
 
 	/* Complete buffers on all video nodes. */
@@ -481,7 +485,7 @@ static int vsp1_video_pipeline_build_branch(struct vsp1_pipeline *pipe,
 	struct media_entity_enum ent_enum;
 	struct vsp1_entity *entity;
 	struct media_pad *pad;
-	bool bru_found = false;
+	struct vsp1_bru *bru = NULL;
 	int ret;
 
 	ret = media_entity_enum_init(&ent_enum, &input->entity.vsp1->media_dev);
@@ -511,16 +515,20 @@ static int vsp1_video_pipeline_build_branch(struct vsp1_pipeline *pipe,
 			media_entity_to_v4l2_subdev(pad->entity));
 
 		/*
-		 * A BRU is present in the pipeline, store the BRU input pad
+		 * A BRU or BRS is present in the pipeline, store its input pad
 		 * number in the input RPF for use when configuring the RPF.
 		 */
-		if (entity->type == VSP1_ENTITY_BRU) {
-			struct vsp1_bru *bru = to_bru(&entity->subdev);
+		if (entity->type == VSP1_ENTITY_BRU ||
+		    entity->type == VSP1_ENTITY_BRS) {
+			/* BRU and BRS can't be chained. */
+			if (bru) {
+				ret = -EPIPE;
+				goto out;
+			}
 
+			bru = to_bru(&entity->subdev);
 			bru->inputs[pad->index].rpf = input;
 			input->bru_input = pad->index;
-
-			bru_found = true;
 		}
 
 		/* We've reached the WPF, we're done. */
@@ -542,8 +550,7 @@ static int vsp1_video_pipeline_build_branch(struct vsp1_pipeline *pipe,
 			}
 
 			pipe->uds = entity;
-			pipe->uds_input = bru_found ? pipe->bru
-					: &input->entity;
+			pipe->uds_input = bru ? &bru->entity : &input->entity;
 		}
 
 		/* Follow the source link, ignoring any HGO or HGT. */
@@ -589,30 +596,42 @@ static int vsp1_video_pipeline_build(struct vsp1_pipeline *pipe,
 		e = to_vsp1_entity(subdev);
 		list_add_tail(&e->list_pipe, &pipe->entities);
 
-		if (e->type == VSP1_ENTITY_RPF) {
+		switch (e->type) {
+		case VSP1_ENTITY_RPF:
 			rwpf = to_rwpf(subdev);
 			pipe->inputs[rwpf->entity.index] = rwpf;
 			rwpf->video->pipe_index = ++pipe->num_inputs;
 			rwpf->pipe = pipe;
-		} else if (e->type == VSP1_ENTITY_WPF) {
+			break;
+
+		case VSP1_ENTITY_WPF:
 			rwpf = to_rwpf(subdev);
 			pipe->output = rwpf;
 			rwpf->video->pipe_index = 0;
 			rwpf->pipe = pipe;
-		} else if (e->type == VSP1_ENTITY_LIF) {
+			break;
+
+		case VSP1_ENTITY_LIF:
 			pipe->lif = e;
-		} else if (e->type == VSP1_ENTITY_BRU) {
+			break;
+
+		case VSP1_ENTITY_BRU:
+		case VSP1_ENTITY_BRS:
 			pipe->bru = e;
-		} else if (e->type == VSP1_ENTITY_HGO) {
-			struct vsp1_hgo *hgo = to_hgo(subdev);
+			break;
 
+		case VSP1_ENTITY_HGO:
 			pipe->hgo = e;
-			hgo->histo.pipe = pipe;
-		} else if (e->type == VSP1_ENTITY_HGT) {
-			struct vsp1_hgt *hgt = to_hgt(subdev);
+			to_hgo(subdev)->histo.pipe = pipe;
+			break;
 
+		case VSP1_ENTITY_HGT:
 			pipe->hgt = e;
-			hgt->histo.pipe = pipe;
+			to_hgt(subdev)->histo.pipe = pipe;
+			break;
+
+		default:
+			break;
 		}
 	}
 
@@ -796,12 +815,14 @@ static int vsp1_video_setup_pipeline(struct vsp1_pipeline *pipe)
 		struct vsp1_uds *uds = to_uds(&pipe->uds->subdev);
 
 		/*
-		 * If a BRU is present in the pipeline before the UDS, the alpha
-		 * component doesn't need to be scaled as the BRU output alpha
-		 * value is fixed to 255. Otherwise we need to scale the alpha
-		 * component only when available at the input RPF.
+		 * If a BRU or BRS is present in the pipeline before the UDS,
+		 * the alpha component doesn't need to be scaled as the BRU and
+		 * BRS output alpha value is fixed to 255. Otherwise we need to
+		 * scale the alpha component only when available at the input
+		 * RPF.
 		 */
-		if (pipe->uds_input->type == VSP1_ENTITY_BRU) {
+		if (pipe->uds_input->type == VSP1_ENTITY_BRU ||
+		    pipe->uds_input->type == VSP1_ENTITY_BRS) {
 			uds->scale_alpha = false;
 		} else {
 			struct vsp1_rwpf *rpf =

drivers/media/platform/vsp1/vsp1_wpf.c

@@ -453,7 +453,9 @@ static void wpf_configure(struct vsp1_entity *entity,
 	}
 
 	if (pipe->bru || pipe->num_inputs > 1)
-		srcrpf |= VI6_WPF_SRCRPF_VIRACT_MST;
+		srcrpf |= pipe->bru->type == VSP1_ENTITY_BRU
+			? VI6_WPF_SRCRPF_VIRACT_MST
+			: VI6_WPF_SRCRPF_VIRACT2_MST;
 
 	vsp1_wpf_write(wpf, dl, VI6_WPF_SRCRPF, srcrpf);
 

include/media/vsp1.h

@@ -34,11 +34,12 @@ struct vsp1_du_lif_config {
 	unsigned int width;
 	unsigned int height;
 
-	void (*callback)(void *);
+	void (*callback)(void *, bool);
 	void *callback_data;
 };
 
-int vsp1_du_setup_lif(struct device *dev, const struct vsp1_du_lif_config *cfg);
+int vsp1_du_setup_lif(struct device *dev, unsigned int pipe_index,
+		      const struct vsp1_du_lif_config *cfg);
 
 struct vsp1_du_atomic_config {
 	u32 pixelformat;
@@ -50,10 +51,11 @@ struct vsp1_du_atomic_config {
 	unsigned int zpos;
 };
 
-void vsp1_du_atomic_begin(struct device *dev);
-int vsp1_du_atomic_update(struct device *dev, unsigned int rpf,
+void vsp1_du_atomic_begin(struct device *dev, unsigned int pipe_index);
+int vsp1_du_atomic_update(struct device *dev, unsigned int pipe_index,
+			  unsigned int rpf,
 			  const struct vsp1_du_atomic_config *cfg);
-void vsp1_du_atomic_flush(struct device *dev);
+void vsp1_du_atomic_flush(struct device *dev, unsigned int pipe_index);
 int vsp1_du_map_sg(struct device *dev, struct sg_table *sgt);
 void vsp1_du_unmap_sg(struct device *dev, struct sg_table *sgt);