Merge tag 'drm-next-2019-05-09' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie: "This has two exciting community drivers for ARM Mali accelerators. Since ARM has never been open source friendly on the GPU side of the house, the community has had to create open source drivers for the Mali GPUs. Lima covers the older t4xx and panfrost the newer 6xx/7xx series. Well done to all involved and hopefully this will help ARM head in the right direction. There is also now the ability if you don't have any of the legacy drivers enabled (pre-KMS) to remove all the pre-KMS support code from the core drm, this saves 10% or so in codesize on my machine. i915 also enable Icelake/Elkhart Lake Gen11 GPUs by default, vboxvideo moves out of staging. There are also some rcar-du patches which crossover with media tree but all should be acked by Mauro. Summary: uapi changes: - Colorspace connector property - fourcc - new YUV formts - timeline sync objects initially merged - expose FB_DAMAGE_CLIPS to atomic userspace new drivers: - vboxvideo: moved out of staging - aspeed: ASPEED SoC BMC chip display support - lima: ARM Mali4xx GPU acceleration driver support - panfrost: ARM Mali6xx/7xx Midgard/Bitfrost acceleration driver support core: - component helper docs - unplugging fixes - devm device init - MIPI/DSI rate control - shmem backed gem objects - connector, display_info, edid_quirks cleanups - dma_buf fence chain support - 64-bit dma-fence seqno comparison fixes - move initial fb config code to core - gem fence array helpers for Lima - ability to remove legacy support code if no drivers requires it (removes 10% of drm.ko size) - lease fixes ttm: - unified DRM_FILE_PAGE_OFFSET handling - Account for kernel allocations in kernel zone only panel: - OSD070T1718-19TS panel support - panel-tpo-td028ttec1 backlight support - Ronbo RB070D30 MIPI/DSI - Feiyang FY07024DI26A30-D MIPI-DSI panel - Rocktech jh057n00900 MIPI-DSI panel i915: - Comet Lake (Gen9) PCI IDs - Updated Icelake PCI IDs - Elkhartlake (Gen11) support - DP MST property addtions - plane and watermark fixes - Icelake port sync and VEBOX disable fixes - struct_mutex usage reduction - Icelake gamma fix - GuC reset fixes - make mmap more asynchronous - sound display power well race fixes - DDI/MIPI-DSI clocks for Icelake - Icelake RPS frequency changing support - Icelake workarounds amdgpu: - Use HMM for userptr - vega20 experimental smu11 support - RAS support for vega20 - BACO support for vega12 + fixes for vega20 - reworked IH interrupt handling - amdkfd RAS support - Freesync improvements - initial timeline sync object support - DC Z ordering fixes - NV12 planes support - colorspace properties for planes= - eDP opts if eDP already initialized nouveau: - misc fixes etnaviv: - misc fixes msm: - GPU zap shader support expansion - robustness ABI addition exynos: - Logging cleanups tegra: - Shared reset fix - CPU cache maintenance fix cirrus: - driver rewritten using simple helpers meson: - G12A support vmwgfx: - Resource dirtying management improvements - Userspace logging improvements virtio: - PRIME fixes rockchip: - rk3066 hdmi support sun4i: - DSI burst mode support vc4: - load tracker to detect underflow v3d: - v3d v4.2 support malidp: - initial Mali D71 support in komeda driver tfp410: - omap related improvement omapdrm: - drm bridge/panel support - drop some omap specific panels rcar-du: - Display writeback support" * tag 'drm-next-2019-05-09' of git://anongit.freedesktop.org/drm/drm: (1507 commits) drm/msm/a6xx: No zap shader is not an error drm/cma-helper: Fix drm_gem_cma_free_object() drm: Fix timestamp docs for variable refresh properties. drm/komeda: Mark the local functions as static drm/komeda: Fixed warning: Function parameter or member not described drm/komeda: Expose bus_width to Komeda-CORE drm/komeda: Add sysfs attribute: core_id and config_id drm: add non-desktop quirk for Valve HMDs drm/panfrost: Show stored feature registers drm/panfrost: Don't scream about deferred probe drm/panfrost: Disable PM on probe failure drm/panfrost: Set DMA masks earlier drm/panfrost: Add sanity checks to submit IOCTL drm/etnaviv: initialize idle mask before querying the HW db drm: introduce a capability flag for syncobj timeline support drm: report consistent errors when checking syncobj capibility drm/nouveau/nouveau: forward error generated while resuming objects tree drm/nouveau/fb/ramgk104: fix spelling mistake "sucessfully" -> "successfully" drm/nouveau/i2c: Disable i2c bus access after ->fini() drm/nouveau: Remove duplicate ACPI_VIDEO_NOTIFY_PROBE definition ...
author: Linus Torvalds <torvalds@linux-foundation.org> 2019-05-08 21:35:19 -0700
committer: Linus Torvalds <torvalds@linux-foundation.org> 2019-05-08 21:35:19 -0700
commit: a2d635decbfa9c1e4ae15cb05b68b2559f7f827c (patch)
tree: 1c3766c35215450ff9e4228efed578d5e6ba65d1 /drivers/gpu/drm/vc4
parent: 89c3b37af87ec183b666d83428cb28cc421671a6 (diff)
parent: eb85d03e01c3e9f3b0ba7282b2e3515a635decb2 (diff)
download: linux-a2d635decbfa9c1e4ae15cb05b68b2559f7f827c.tar.gz
19 files changed, 934 insertions, 709 deletions
diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
index 8dcce7182bb7..88ebd681d7eb 100644
--- a/drivers/gpu/drm/vc4/vc4_bo.c
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -40,7 +40,7 @@ static bool is_user_label(int label)
 	return label >= VC4_BO_TYPE_COUNT;
 }
 
-static void vc4_bo_stats_dump(struct vc4_dev *vc4)
+static void vc4_bo_stats_print(struct drm_printer *p, struct vc4_dev *vc4)
 {
 	int i;
 
@@ -48,58 +48,35 @@ static void vc4_bo_stats_dump(struct vc4_dev *vc4)
 		if (!vc4->bo_labels[i].num_allocated)
 			continue;
 
-		DRM_INFO("%30s: %6dkb BOs (%d)\n",
-			 vc4->bo_labels[i].name,
-			 vc4->bo_labels[i].size_allocated / 1024,
-			 vc4->bo_labels[i].num_allocated);
+		drm_printf(p, "%30s: %6dkb BOs (%d)\n",
+			   vc4->bo_labels[i].name,
+			   vc4->bo_labels[i].size_allocated / 1024,
+			   vc4->bo_labels[i].num_allocated);
 	}
 
 	mutex_lock(&vc4->purgeable.lock);
 	if (vc4->purgeable.num)
-		DRM_INFO("%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
-			 vc4->purgeable.size / 1024, vc4->purgeable.num);
+		drm_printf(p, "%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
+			   vc4->purgeable.size / 1024, vc4->purgeable.num);
 
 	if (vc4->purgeable.purged_num)
-		DRM_INFO("%30s: %6zdkb BOs (%d)\n", "total purged BO",
-			 vc4->purgeable.purged_size / 1024,
-			 vc4->purgeable.purged_num);
+		drm_printf(p, "%30s: %6zdkb BOs (%d)\n", "total purged BO",
+			   vc4->purgeable.purged_size / 1024,
+			   vc4->purgeable.purged_num);
 	mutex_unlock(&vc4->purgeable.lock);
 }
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_bo_stats_debugfs(struct seq_file *m, void *unused)
+static int vc4_bo_stats_debugfs(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	mutex_lock(&vc4->bo_lock);
-	for (i = 0; i < vc4->num_labels; i++) {
-		if (!vc4->bo_labels[i].num_allocated)
-			continue;
-
-		seq_printf(m, "%30s: %6dkb BOs (%d)\n",
-			   vc4->bo_labels[i].name,
-			   vc4->bo_labels[i].size_allocated / 1024,
-			   vc4->bo_labels[i].num_allocated);
-	}
-	mutex_unlock(&vc4->bo_lock);
-
-	mutex_lock(&vc4->purgeable.lock);
-	if (vc4->purgeable.num)
-		seq_printf(m, "%30s: %6zdkb BOs (%d)\n", "userspace BO cache",
-			   vc4->purgeable.size / 1024, vc4->purgeable.num);
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	if (vc4->purgeable.purged_num)
-		seq_printf(m, "%30s: %6zdkb BOs (%d)\n", "total purged BO",
-			   vc4->purgeable.purged_size / 1024,
-			   vc4->purgeable.purged_num);
-	mutex_unlock(&vc4->purgeable.lock);
+	vc4_bo_stats_print(&p, vc4);
 
 	return 0;
 }
-#endif
 
 /* Takes ownership of *name and returns the appropriate slot for it in
  * the bo_labels[] array, extending it as necessary.
@@ -201,8 +178,6 @@ static void vc4_bo_destroy(struct vc4_bo *bo)
 		bo->validated_shader = NULL;
 	}
 
-	reservation_object_fini(&bo->_resv);
-
 	drm_gem_cma_free_object(obj);
 }
 
@@ -427,8 +402,6 @@ struct drm_gem_object *vc4_create_object(struct drm_device *dev, size_t size)
 	vc4->bo_labels[VC4_BO_TYPE_KERNEL].num_allocated++;
 	vc4->bo_labels[VC4_BO_TYPE_KERNEL].size_allocated += size;
 	mutex_unlock(&vc4->bo_lock);
-	bo->resv = &bo->_resv;
-	reservation_object_init(bo->resv);
 
 	return &bo->base.base;
 }
@@ -479,8 +452,9 @@ struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t unaligned_size,
 	}
 
 	if (IS_ERR(cma_obj)) {
+		struct drm_printer p = drm_info_printer(vc4->dev->dev);
 		DRM_ERROR("Failed to allocate from CMA:\n");
-		vc4_bo_stats_dump(vc4);
+		vc4_bo_stats_print(&p, vc4);
 		return ERR_PTR(-ENOMEM);
 	}
 	bo = to_vc4_bo(&cma_obj->base);
@@ -684,13 +658,6 @@ static void vc4_bo_cache_time_timer(struct timer_list *t)
 	schedule_work(&vc4->bo_cache.time_work);
 }
 
-struct reservation_object *vc4_prime_res_obj(struct drm_gem_object *obj)
-{
-	struct vc4_bo *bo = to_vc4_bo(obj);
-
-	return bo->resv;
-}
-
 struct dma_buf *
 vc4_prime_export(struct drm_device *dev, struct drm_gem_object *obj, int flags)
 {
@@ -822,14 +789,12 @@ vc4_prime_import_sg_table(struct drm_device *dev,
 			  struct sg_table *sgt)
 {
 	struct drm_gem_object *obj;
-	struct vc4_bo *bo;
 
 	obj = drm_gem_cma_prime_import_sg_table(dev, attach, sgt);
 	if (IS_ERR(obj))
 		return obj;
 
-	bo = to_vc4_bo(obj);
-	bo->resv = attach->dmabuf->resv;
+	obj->resv = attach->dmabuf->resv;
 
 	return obj;
 }
@@ -1038,6 +1003,8 @@ int vc4_bo_cache_init(struct drm_device *dev)
 
 	mutex_init(&vc4->bo_lock);
 
+	vc4_debugfs_add_file(dev, "bo_stats", vc4_bo_stats_debugfs, NULL);
+
 	INIT_LIST_HEAD(&vc4->bo_cache.time_list);
 
 	INIT_WORK(&vc4->bo_cache.time_work, vc4_bo_cache_time_work);
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 1baa10e94484..5e09389e1514 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -35,6 +35,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_atomic_uapi.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 #include <linux/clk.h>
 #include <drm/drm_fb_cma_helper.h>
@@ -67,67 +68,22 @@ to_vc4_crtc_state(struct drm_crtc_state *crtc_state)
 #define CRTC_WRITE(offset, val) writel(val, vc4_crtc->regs + (offset))
 #define CRTC_READ(offset) readl(vc4_crtc->regs + (offset))
 
-#define CRTC_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} crtc_regs[] = {
-	CRTC_REG(PV_CONTROL),
-	CRTC_REG(PV_V_CONTROL),
-	CRTC_REG(PV_VSYNCD_EVEN),
-	CRTC_REG(PV_HORZA),
-	CRTC_REG(PV_HORZB),
-	CRTC_REG(PV_VERTA),
-	CRTC_REG(PV_VERTB),
-	CRTC_REG(PV_VERTA_EVEN),
-	CRTC_REG(PV_VERTB_EVEN),
-	CRTC_REG(PV_INTEN),
-	CRTC_REG(PV_INTSTAT),
-	CRTC_REG(PV_STAT),
-	CRTC_REG(PV_HACT_ACT),
+static const struct debugfs_reg32 crtc_regs[] = {
+	VC4_REG32(PV_CONTROL),
+	VC4_REG32(PV_V_CONTROL),
+	VC4_REG32(PV_VSYNCD_EVEN),
+	VC4_REG32(PV_HORZA),
+	VC4_REG32(PV_HORZB),
+	VC4_REG32(PV_VERTA),
+	VC4_REG32(PV_VERTB),
+	VC4_REG32(PV_VERTA_EVEN),
+	VC4_REG32(PV_VERTB_EVEN),
+	VC4_REG32(PV_INTEN),
+	VC4_REG32(PV_INTSTAT),
+	VC4_REG32(PV_STAT),
+	VC4_REG32(PV_HACT_ACT),
 };
 
-static void vc4_crtc_dump_regs(struct vc4_crtc *vc4_crtc)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 crtc_regs[i].reg, crtc_regs[i].name,
-			 CRTC_READ(crtc_regs[i].reg));
-	}
-}
-
-#ifdef CONFIG_DEBUG_FS
-int vc4_crtc_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	int crtc_index = (uintptr_t)node->info_ent->data;
-	struct drm_crtc *crtc;
-	struct vc4_crtc *vc4_crtc;
-	int i;
-
-	i = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (i == crtc_index)
-			break;
-		i++;
-	}
-	if (!crtc)
-		return 0;
-	vc4_crtc = to_vc4_crtc(crtc);
-
-	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   crtc_regs[i].name, crtc_regs[i].reg,
-			   CRTC_READ(crtc_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id,
 			     bool in_vblank_irq, int *vpos, int *hpos,
 			     ktime_t *stime, ktime_t *etime,
@@ -434,8 +390,10 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	bool debug_dump_regs = false;
 
 	if (debug_dump_regs) {
-		DRM_INFO("CRTC %d regs before:\n", drm_crtc_index(crtc));
-		vc4_crtc_dump_regs(vc4_crtc);
+		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
+		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs before:\n",
+			 drm_crtc_index(crtc));
+		drm_print_regset32(&p, &vc4_crtc->regset);
 	}
 
 	if (vc4_crtc->channel == 2) {
@@ -476,8 +434,10 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	vc4_crtc_lut_load(crtc);
 
 	if (debug_dump_regs) {
-		DRM_INFO("CRTC %d regs after:\n", drm_crtc_index(crtc));
-		vc4_crtc_dump_regs(vc4_crtc);
+		struct drm_printer p = drm_info_printer(&vc4_crtc->pdev->dev);
+		dev_info(&vc4_crtc->pdev->dev, "CRTC %d regs after:\n",
+			 drm_crtc_index(crtc));
+		drm_print_regset32(&p, &vc4_crtc->regset);
 	}
 }
 
@@ -834,6 +794,14 @@ static void vc4_crtc_handle_page_flip(struct vc4_crtc *vc4_crtc)
 		drm_crtc_send_vblank_event(crtc, vc4_crtc->event);
 		vc4_crtc->event = NULL;
 		drm_crtc_vblank_put(crtc);
+
+		/* Wait for the page flip to unmask the underrun to ensure that
+		 * the display list was updated by the hardware. Before that
+		 * happens, the HVS will be using the previous display list with
+		 * the CRTC and encoder already reconfigured, leading to
+		 * underruns. This can be seen when reconfiguring the CRTC.
+		 */
+		vc4_hvs_unmask_underrun(dev, vc4_crtc->channel);
 	}
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
@@ -1075,6 +1043,7 @@ static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
 
 static const struct vc4_crtc_data pv0_data = {
 	.hvs_channel = 0,
+	.debugfs_name = "crtc0_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI0,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_DPI,
@@ -1083,6 +1052,7 @@ static const struct vc4_crtc_data pv0_data = {
 
 static const struct vc4_crtc_data pv1_data = {
 	.hvs_channel = 2,
+	.debugfs_name = "crtc1_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DSI] = VC4_ENCODER_TYPE_DSI1,
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_SMI,
@@ -1091,6 +1061,7 @@ static const struct vc4_crtc_data pv1_data = {
 
 static const struct vc4_crtc_data pv2_data = {
 	.hvs_channel = 1,
+	.debugfs_name = "crtc2_regs",
 	.encoder_types = {
 		[PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI] = VC4_ENCODER_TYPE_HDMI,
 		[PV_CONTROL_CLK_SELECT_VEC] = VC4_ENCODER_TYPE_VEC,
@@ -1169,11 +1140,16 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
 	if (!match)
 		return -ENODEV;
 	vc4_crtc->data = match->data;
+	vc4_crtc->pdev = pdev;
 
 	vc4_crtc->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(vc4_crtc->regs))
 		return PTR_ERR(vc4_crtc->regs);
 
+	vc4_crtc->regset.base = vc4_crtc->regs;
+	vc4_crtc->regset.regs = crtc_regs;
+	vc4_crtc->regset.nregs = ARRAY_SIZE(crtc_regs);
+
 	/* For now, we create just the primary and the legacy cursor
 	 * planes.  We should be able to stack more planes on easily,
 	 * but to do that we would need to compute the bandwidth
@@ -1247,6 +1223,9 @@ static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
 
 	platform_set_drvdata(pdev, vc4_crtc);
 
+	vc4_debugfs_add_regset32(drm, vc4_crtc->data->debugfs_name,
+				 &vc4_crtc->regset);
+
 	return 0;
 
 err_destroy_planes:
diff --git a/drivers/gpu/drm/vc4/vc4_debugfs.c b/drivers/gpu/drm/vc4/vc4_debugfs.c
index 7a0003de71ab..f9dec08267dc 100644
--- a/drivers/gpu/drm/vc4/vc4_debugfs.c
+++ b/drivers/gpu/drm/vc4/vc4_debugfs.c
@@ -15,26 +15,82 @@
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-static const struct drm_info_list vc4_debugfs_list[] = {
-	{"bo_stats", vc4_bo_stats_debugfs, 0},
-	{"dpi_regs", vc4_dpi_debugfs_regs, 0},
-	{"dsi1_regs", vc4_dsi_debugfs_regs, 0, (void *)(uintptr_t)1},
-	{"hdmi_regs", vc4_hdmi_debugfs_regs, 0},
-	{"vec_regs", vc4_vec_debugfs_regs, 0},
-	{"txp_regs", vc4_txp_debugfs_regs, 0},
-	{"hvs_regs", vc4_hvs_debugfs_regs, 0},
-	{"crtc0_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)0},
-	{"crtc1_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)1},
-	{"crtc2_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)2},
-	{"v3d_ident", vc4_v3d_debugfs_ident, 0},
-	{"v3d_regs", vc4_v3d_debugfs_regs, 0},
+struct vc4_debugfs_info_entry {
+	struct list_head link;
+	struct drm_info_list info;
 };
 
-#define VC4_DEBUGFS_ENTRIES ARRAY_SIZE(vc4_debugfs_list)
-
+/**
+ * Called at drm_dev_register() time on each of the minors registered
+ * by the DRM device, to attach the debugfs files.
+ */
 int
 vc4_debugfs_init(struct drm_minor *minor)
 {
-	return drm_debugfs_create_files(vc4_debugfs_list, VC4_DEBUGFS_ENTRIES,
-					minor->debugfs_root, minor);
+	struct vc4_dev *vc4 = to_vc4_dev(minor->dev);
+	struct vc4_debugfs_info_entry *entry;
+	struct dentry *dentry;
+
+	dentry = debugfs_create_bool("hvs_load_tracker", S_IRUGO | S_IWUSR,
+				     minor->debugfs_root,
+				     &vc4->load_tracker_enabled);
+	if (!dentry)
+		return -ENOMEM;
+
+	list_for_each_entry(entry, &vc4->debugfs_list, link) {
+		int ret = drm_debugfs_create_files(&entry->info, 1,
+						   minor->debugfs_root, minor);
+
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int vc4_debugfs_regset32(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct debugfs_regset32 *regset = node->info_ent->data;
+	struct drm_printer p = drm_seq_file_printer(m);
+
+	drm_print_regset32(&p, regset);
+
+	return 0;
+}
+
+/**
+ * Registers a debugfs file with a callback function for a vc4 component.
+ *
+ * This is like drm_debugfs_create_files(), but that can only be
+ * called a given DRM minor, while the various VC4 components want to
+ * register their debugfs files during the component bind process.  We
+ * track the request and delay it to be called on each minor during
+ * vc4_debugfs_init().
+ */
+void vc4_debugfs_add_file(struct drm_device *dev,
+			  const char *name,
+			  int (*show)(struct seq_file*, void*),
+			  void *data)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	struct vc4_debugfs_info_entry *entry =
+		devm_kzalloc(dev->dev, sizeof(*entry), GFP_KERNEL);
+
+	if (!entry)
+		return;
+
+	entry->info.name = name;
+	entry->info.show = show;
+	entry->info.data = data;
+
+	list_add(&entry->link, &vc4->debugfs_list);
+}
+
+void vc4_debugfs_add_regset32(struct drm_device *drm,
+			      const char *name,
+			      struct debugfs_regset32 *regset)
+{
+	vc4_debugfs_add_file(drm, name, vc4_debugfs_regset32, regset);
 }
diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c
index 169521e547ba..34f90ca8f479 100644
--- a/drivers/gpu/drm/vc4/vc4_dpi.c
+++ b/drivers/gpu/drm/vc4/vc4_dpi.c
@@ -101,6 +101,8 @@ struct vc4_dpi {
 
 	struct clk *pixel_clock;
 	struct clk *core_clock;
+
+	struct debugfs_regset32 regset;
 };
 
 #define DPI_READ(offset) readl(dpi->regs + (offset))
@@ -118,37 +120,11 @@ to_vc4_dpi_encoder(struct drm_encoder *encoder)
 	return container_of(encoder, struct vc4_dpi_encoder, base.base);
 }
 
-#define DPI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} dpi_regs[] = {
-	DPI_REG(DPI_C),
-	DPI_REG(DPI_ID),
+static const struct debugfs_reg32 dpi_regs[] = {
+	VC4_REG32(DPI_C),
+	VC4_REG32(DPI_ID),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_dpi_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_dpi *dpi = vc4->dpi;
-	int i;
-
-	if (!dpi)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(dpi_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   dpi_regs[i].name, dpi_regs[i].reg,
-			   DPI_READ(dpi_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static const struct drm_encoder_funcs vc4_dpi_encoder_funcs = {
 	.destroy = drm_encoder_cleanup,
 };
@@ -314,6 +290,9 @@ static int vc4_dpi_bind(struct device *dev, struct device *master, void *data)
 	dpi->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(dpi->regs))
 		return PTR_ERR(dpi->regs);
+	dpi->regset.base = dpi->regs;
+	dpi->regset.regs = dpi_regs;
+	dpi->regset.nregs = ARRAY_SIZE(dpi_regs);
 
 	if (DPI_READ(DPI_ID) != DPI_ID_VALUE) {
 		dev_err(dev, "Port returned 0x%08x for ID instead of 0x%08x\n",
@@ -352,6 +331,8 @@ static int vc4_dpi_bind(struct device *dev, struct device *master, void *data)
 
 	vc4->dpi = dpi;
 
+	vc4_debugfs_add_regset32(drm, "dpi_regs", &dpi->regset);
+
 	return 0;
 
 err_destroy_encoder:
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index 5fcd2f0da7f7..6d9be20a32be 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -72,30 +72,30 @@ static int vc4_get_param_ioctl(struct drm_device *dev, void *data,
 	if (args->pad != 0)
 		return -EINVAL;
 
+	if (!vc4->v3d)
+		return -ENODEV;
+
 	switch (args->param) {
 	case DRM_VC4_PARAM_V3D_IDENT0:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT0);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_V3D_IDENT1:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT1);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_V3D_IDENT2:
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0)
+		ret = vc4_v3d_pm_get(vc4);
+		if (ret)
 			return ret;
 		args->value = V3D_READ(V3D_IDENT2);
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+		vc4_v3d_pm_put(vc4);
 		break;
 	case DRM_VC4_PARAM_SUPPORTS_BRANCHES:
 	case DRM_VC4_PARAM_SUPPORTS_ETC1:
@@ -200,7 +200,6 @@ static struct drm_driver vc4_drm_driver = {
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 	.gem_prime_import = drm_gem_prime_import,
 	.gem_prime_export = vc4_prime_export,
-	.gem_prime_res_obj = vc4_prime_res_obj,
 	.gem_prime_get_sg_table	= drm_gem_cma_prime_get_sg_table,
 	.gem_prime_import_sg_table = vc4_prime_import_sg_table,
 	.gem_prime_vmap = vc4_prime_vmap,
@@ -252,6 +251,7 @@ static int vc4_drm_bind(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct drm_device *drm;
 	struct vc4_dev *vc4;
+	struct device_node *node;
 	int ret = 0;
 
 	dev->coherent_dma_mask = DMA_BIT_MASK(32);
@@ -260,12 +260,19 @@ static int vc4_drm_bind(struct device *dev)
 	if (!vc4)
 		return -ENOMEM;
 
+	/* If VC4 V3D is missing, don't advertise render nodes. */
+	node = of_find_matching_node_and_match(NULL, vc4_v3d_dt_match, NULL);
+	if (!node || !of_device_is_available(node))
+		vc4_drm_driver.driver_features &= ~DRIVER_RENDER;
+	of_node_put(node);
+
 	drm = drm_dev_alloc(&vc4_drm_driver, dev);
 	if (IS_ERR(drm))
 		return PTR_ERR(drm);
 	platform_set_drvdata(pdev, drm);
 	vc4->dev = drm;
 	drm->dev_private = vc4;
+	INIT_LIST_HEAD(&vc4->debugfs_list);
 
 	ret = vc4_bo_cache_init(drm);
 	if (ret)
@@ -281,13 +288,15 @@ static int vc4_drm_bind(struct device *dev)
 
 	drm_fb_helper_remove_conflicting_framebuffers(NULL, "vc4drmfb", false);
 
-	ret = drm_dev_register(drm, 0);
+	ret = vc4_kms_load(drm);
 	if (ret < 0)
 		goto unbind_all;
 
-	vc4_kms_load(drm);
+	ret = drm_dev_register(drm, 0);
+	if (ret < 0)
+		goto unbind_all;
 
-	drm_fbdev_generic_setup(drm, 32);
+	drm_fbdev_generic_setup(drm, 16);
 
 	return 0;
 
@@ -312,6 +321,7 @@ static void vc4_drm_unbind(struct device *dev)
 
 	drm_mode_config_cleanup(drm);
 
+	drm_atomic_private_obj_fini(&vc4->load_tracker);
 	drm_atomic_private_obj_fini(&vc4->ctm_manager);
 
 	drm_dev_put(drm);
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 2c635f001c71..4f13f6262491 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -7,7 +7,6 @@
  */
 
 #include <linux/mm_types.h>
-#include <linux/reservation.h>
 #include <drm/drmP.h>
 #include <drm/drm_util.h>
 #include <drm/drm_encoder.h>
@@ -185,10 +184,20 @@ struct vc4_dev {
 	/* Bitmask of the current bin_alloc used for overflow memory. */
 	uint32_t bin_alloc_overflow;
 
+	/* Incremented when an underrun error happened after an atomic commit.
+	 * This is particularly useful to detect when a specific modeset is too
+	 * demanding in term of memory or HVS bandwidth which is hard to guess
+	 * at atomic check time.
+	 */
+	atomic_t underrun;
+
 	struct work_struct overflow_mem_work;
 
 	int power_refcount;
 
+	/* Set to true when the load tracker is active. */
+	bool load_tracker_enabled;
+
 	/* Mutex controlling the power refcount. */
 	struct mutex power_lock;
 
@@ -201,6 +210,12 @@ struct vc4_dev {
 
 	struct drm_modeset_lock ctm_state_lock;
 	struct drm_private_obj ctm_manager;
+	struct drm_private_obj load_tracker;
+
+	/* List of vc4_debugfs_info_entry for adding to debugfs once
+	 * the minor is available (after drm_dev_register()).
+	 */
+	struct list_head debugfs_list;
 };
 
 static inline struct vc4_dev *
@@ -240,10 +255,6 @@ struct vc4_bo {
 	 */
 	struct vc4_validated_shader_info *validated_shader;
 
-	/* normally (resv == &_resv) except for imported bo's */
-	struct reservation_object *resv;
-	struct reservation_object _resv;
-
 	/* One of enum vc4_kernel_bo_type, or VC4_BO_TYPE_COUNT + i
 	 * for user-allocated labels.
 	 */
@@ -290,6 +301,7 @@ struct vc4_v3d {
 	struct platform_device *pdev;
 	void __iomem *regs;
 	struct clk *clk;
+	struct debugfs_regset32 regset;
 };
 
 struct vc4_hvs {
@@ -306,6 +318,7 @@ struct vc4_hvs {
 	spinlock_t mm_lock;
 
 	struct drm_mm_node mitchell_netravali_filter;
+	struct debugfs_regset32 regset;
 };
 
 struct vc4_plane {
@@ -376,6 +389,16 @@ struct vc4_plane_state {
 	 * when async update is not possible.
 	 */
 	bool dlist_initialized;
+
+	/* Load of this plane on the HVS block. The load is expressed in HVS
+	 * cycles/sec.
+	 */
+	u64 hvs_load;
+
+	/* Memory bandwidth needed for this plane. This is expressed in
+	 * bytes/sec.
+	 */
+	u64 membus_load;
 };
 
 static inline struct vc4_plane_state *
@@ -411,10 +434,12 @@ struct vc4_crtc_data {
 	int hvs_channel;
 
 	enum vc4_encoder_type encoder_types[4];
+	const char *debugfs_name;
 };
 
 struct vc4_crtc {
 	struct drm_crtc base;
+	struct platform_device *pdev;
 	const struct vc4_crtc_data *data;
 	void __iomem *regs;
 
@@ -431,6 +456,8 @@ struct vc4_crtc {
 	u32 cob_size;
 
 	struct drm_pending_vblank_event *event;
+
+	struct debugfs_regset32 regset;
 };
 
 static inline struct vc4_crtc *
@@ -444,6 +471,8 @@ to_vc4_crtc(struct drm_crtc *crtc)
 #define HVS_READ(offset) readl(vc4->hvs->regs + offset)
 #define HVS_WRITE(offset, val) writel(val, vc4->hvs->regs + offset)
 
+#define VC4_REG32(reg) { .name = #reg, .offset = reg }
+
 struct vc4_exec_info {
 	/* Sequence number for this bin/render job. */
 	uint64_t seqno;
@@ -685,7 +714,6 @@ int vc4_label_bo_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
 vm_fault_t vc4_fault(struct vm_fault *vmf);
 int vc4_mmap(struct file *filp, struct vm_area_struct *vma);
-struct reservation_object *vc4_prime_res_obj(struct drm_gem_object *obj);
 int vc4_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma);
 struct drm_gem_object *vc4_prime_import_sg_table(struct drm_device *dev,
 						 struct dma_buf_attachment *attach,
@@ -693,7 +721,6 @@ struct drm_gem_object *vc4_prime_import_sg_table(struct drm_device *dev,
 void *vc4_prime_vmap(struct drm_gem_object *obj);
 int vc4_bo_cache_init(struct drm_device *dev);
 void vc4_bo_cache_destroy(struct drm_device *dev);
-int vc4_bo_stats_debugfs(struct seq_file *m, void *arg);
 int vc4_bo_inc_usecnt(struct vc4_bo *bo);
 void vc4_bo_dec_usecnt(struct vc4_bo *bo);
 void vc4_bo_add_to_purgeable_pool(struct vc4_bo *bo);
@@ -701,7 +728,6 @@ void vc4_bo_remove_from_purgeable_pool(struct vc4_bo *bo);
 
 /* vc4_crtc.c */
 extern struct platform_driver vc4_crtc_driver;
-int vc4_crtc_debugfs_regs(struct seq_file *m, void *arg);
 bool vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id,
 			     bool in_vblank_irq, int *vpos, int *hpos,
 			     ktime_t *stime, ktime_t *etime,
@@ -714,17 +740,37 @@ void vc4_crtc_get_margins(struct drm_crtc_state *state,
 
 /* vc4_debugfs.c */
 int vc4_debugfs_init(struct drm_minor *minor);
+#ifdef CONFIG_DEBUG_FS
+void vc4_debugfs_add_file(struct drm_device *drm,
+			  const char *filename,
+			  int (*show)(struct seq_file*, void*),
+			  void *data);
+void vc4_debugfs_add_regset32(struct drm_device *drm,
+			      const char *filename,
+			      struct debugfs_regset32 *regset);
+#else
+static inline void vc4_debugfs_add_file(struct drm_device *drm,
+					const char *filename,
+					int (*show)(struct seq_file*, void*),
+					void *data)
+{
+}
+
+static inline void vc4_debugfs_add_regset32(struct drm_device *drm,
+					    const char *filename,
+					    struct debugfs_regset32 *regset)
+{
+}
+#endif
 
 /* vc4_drv.c */
 void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index);
 
 /* vc4_dpi.c */
 extern struct platform_driver vc4_dpi_driver;
-int vc4_dpi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_dsi.c */
 extern struct platform_driver vc4_dsi_driver;
-int vc4_dsi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_fence.c */
 extern const struct dma_fence_ops vc4_fence_ops;
@@ -752,15 +798,12 @@ int vc4_gem_madvise_ioctl(struct drm_device *dev, void *data,
 
 /* vc4_hdmi.c */
 extern struct platform_driver vc4_hdmi_driver;
-int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_vec.c */
 extern struct platform_driver vc4_vec_driver;
-int vc4_vec_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_txp.c */
 extern struct platform_driver vc4_txp_driver;
-int vc4_txp_debugfs_regs(struct seq_file *m, void *unused);
 
 /* vc4_irq.c */
 irqreturn_t vc4_irq(int irq, void *arg);
@@ -772,7 +815,8 @@ void vc4_irq_reset(struct drm_device *dev);
 /* vc4_hvs.c */
 extern struct platform_driver vc4_hvs_driver;
 void vc4_hvs_dump_state(struct drm_device *dev);
-int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused);
+void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel);
+void vc4_hvs_mask_underrun(struct drm_device *dev, int channel);
 
 /* vc4_kms.c */
 int vc4_kms_load(struct drm_device *dev);
@@ -787,9 +831,10 @@ void vc4_plane_async_set_fb(struct drm_plane *plane,
 
 /* vc4_v3d.c */
 extern struct platform_driver vc4_v3d_driver;
-int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused);
-int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused);
+extern const struct of_device_id vc4_v3d_dt_match[];
 int vc4_v3d_get_bin_slot(struct vc4_dev *vc4);
+int vc4_v3d_pm_get(struct vc4_dev *vc4);
+void vc4_v3d_pm_put(struct vc4_dev *vc4);
 
 /* vc4_validate.c */
 int
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 11702e1d9011..9412709067f5 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -545,6 +545,8 @@ struct vc4_dsi {
 
 	struct completion xfer_completion;
 	int xfer_result;
+
+	struct debugfs_regset32 regset;
 };
 
 #define host_to_dsi(host) container_of(host, struct vc4_dsi, dsi_host)
@@ -605,113 +607,56 @@ to_vc4_dsi_encoder(struct drm_encoder *encoder)
 	return container_of(encoder, struct vc4_dsi_encoder, base.base);
 }
 
-#define DSI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} dsi0_regs[] = {
-	DSI_REG(DSI0_CTRL),
-	DSI_REG(DSI0_STAT),
-	DSI_REG(DSI0_HSTX_TO_CNT),
-	DSI_REG(DSI0_LPRX_TO_CNT),
-	DSI_REG(DSI0_TA_TO_CNT),
-	DSI_REG(DSI0_PR_TO_CNT),
-	DSI_REG(DSI0_DISP0_CTRL),
-	DSI_REG(DSI0_DISP1_CTRL),
-	DSI_REG(DSI0_INT_STAT),
-	DSI_REG(DSI0_INT_EN),
-	DSI_REG(DSI0_PHYC),
-	DSI_REG(DSI0_HS_CLT0),
-	DSI_REG(DSI0_HS_CLT1),
-	DSI_REG(DSI0_HS_CLT2),
-	DSI_REG(DSI0_HS_DLT3),
-	DSI_REG(DSI0_HS_DLT4),
-	DSI_REG(DSI0_HS_DLT5),
-	DSI_REG(DSI0_HS_DLT6),
-	DSI_REG(DSI0_HS_DLT7),
-	DSI_REG(DSI0_PHY_AFEC0),
-	DSI_REG(DSI0_PHY_AFEC1),
-	DSI_REG(DSI0_ID),
+static const struct debugfs_reg32 dsi0_regs[] = {
+	VC4_REG32(DSI0_CTRL),
+	VC4_REG32(DSI0_STAT),
+	VC4_REG32(DSI0_HSTX_TO_CNT),
+	VC4_REG32(DSI0_LPRX_TO_CNT),
+	VC4_REG32(DSI0_TA_TO_CNT),
+	VC4_REG32(DSI0_PR_TO_CNT),
+	VC4_REG32(DSI0_DISP0_CTRL),
+	VC4_REG32(DSI0_DISP1_CTRL),
+	VC4_REG32(DSI0_INT_STAT),
+	VC4_REG32(DSI0_INT_EN),
+	VC4_REG32(DSI0_PHYC),
+	VC4_REG32(DSI0_HS_CLT0),
+	VC4_REG32(DSI0_HS_CLT1),
+	VC4_REG32(DSI0_HS_CLT2),
+	VC4_REG32(DSI0_HS_DLT3),
+	VC4_REG32(DSI0_HS_DLT4),
+	VC4_REG32(DSI0_HS_DLT5),
+	VC4_REG32(DSI0_HS_DLT6),
+	VC4_REG32(DSI0_HS_DLT7),
+	VC4_REG32(DSI0_PHY_AFEC0),
+	VC4_REG32(DSI0_PHY_AFEC1),
+	VC4_REG32(DSI0_ID),
 };
 
-static const struct {
-	u32 reg;
-	const char *name;
-} dsi1_regs[] = {
-	DSI_REG(DSI1_CTRL),
-	DSI_REG(DSI1_STAT),
-	DSI_REG(DSI1_HSTX_TO_CNT),
-	DSI_REG(DSI1_LPRX_TO_CNT),
-	DSI_REG(DSI1_TA_TO_CNT),
-	DSI_REG(DSI1_PR_TO_CNT),
-	DSI_REG(DSI1_DISP0_CTRL),
-	DSI_REG(DSI1_DISP1_CTRL),
-	DSI_REG(DSI1_INT_STAT),
-	DSI_REG(DSI1_INT_EN),
-	DSI_REG(DSI1_PHYC),
-	DSI_REG(DSI1_HS_CLT0),
-	DSI_REG(DSI1_HS_CLT1),
-	DSI_REG(DSI1_HS_CLT2),
-	DSI_REG(DSI1_HS_DLT3),
-	DSI_REG(DSI1_HS_DLT4),
-	DSI_REG(DSI1_HS_DLT5),
-	DSI_REG(DSI1_HS_DLT6),
-	DSI_REG(DSI1_HS_DLT7),
-	DSI_REG(DSI1_PHY_AFEC0),
-	DSI_REG(DSI1_PHY_AFEC1),
-	DSI_REG(DSI1_ID),
+static const struct debugfs_reg32 dsi1_regs[] = {
+	VC4_REG32(DSI1_CTRL),
+	VC4_REG32(DSI1_STAT),
+	VC4_REG32(DSI1_HSTX_TO_CNT),
+	VC4_REG32(DSI1_LPRX_TO_CNT),
+	VC4_REG32(DSI1_TA_TO_CNT),
+	VC4_REG32(DSI1_PR_TO_CNT),
+	VC4_REG32(DSI1_DISP0_CTRL),
+	VC4_REG32(DSI1_DISP1_CTRL),
+	VC4_REG32(DSI1_INT_STAT),
+	VC4_REG32(DSI1_INT_EN),
+	VC4_REG32(DSI1_PHYC),
+	VC4_REG32(DSI1_HS_CLT0),
+	VC4_REG32(DSI1_HS_CLT1),
+	VC4_REG32(DSI1_HS_CLT2),
+	VC4_REG32(DSI1_HS_DLT3),
+	VC4_REG32(DSI1_HS_DLT4),
+	VC4_REG32(DSI1_HS_DLT5),
+	VC4_REG32(DSI1_HS_DLT6),
+	VC4_REG32(DSI1_HS_DLT7),
+	VC4_REG32(DSI1_PHY_AFEC0),
+	VC4_REG32(DSI1_PHY_AFEC1),
+	VC4_REG32(DSI1_ID),
 };
 
-static void vc4_dsi_dump_regs(struct vc4_dsi *dsi)
-{
-	int i;
-
-	if (dsi->port == 0) {
-		for (i = 0; i < ARRAY_SIZE(dsi0_regs); i++) {
-			DRM_INFO("0x%04x (%s): 0x%08x\n",
-				 dsi0_regs[i].reg, dsi0_regs[i].name,
-				 DSI_READ(dsi0_regs[i].reg));
-		}
-	} else {
-		for (i = 0; i < ARRAY_SIZE(dsi1_regs); i++) {
-			DRM_INFO("0x%04x (%s): 0x%08x\n",
-				 dsi1_regs[i].reg, dsi1_regs[i].name,
-				 DSI_READ(dsi1_regs[i].reg));
-		}
-	}
-}
-
-#ifdef CONFIG_DEBUG_FS
-int vc4_dsi_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *drm = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(drm);
-	int dsi_index = (uintptr_t)node->info_ent->data;
-	struct vc4_dsi *dsi = (dsi_index == 1 ? vc4->dsi1 : NULL);
-	int i;
-
-	if (!dsi)
-		return 0;
-
-	if (dsi->port == 0) {
-		for (i = 0; i < ARRAY_SIZE(dsi0_regs); i++) {
-			seq_printf(m, "0x%04x (%s): 0x%08x\n",
-				   dsi0_regs[i].reg, dsi0_regs[i].name,
-				   DSI_READ(dsi0_regs[i].reg));
-		}
-	} else {
-		for (i = 0; i < ARRAY_SIZE(dsi1_regs); i++) {
-			seq_printf(m, "0x%04x (%s): 0x%08x\n",
-				   dsi1_regs[i].reg, dsi1_regs[i].name,
-				   DSI_READ(dsi1_regs[i].reg));
-		}
-	}
-
-	return 0;
-}
-#endif
-
 static void vc4_dsi_encoder_destroy(struct drm_encoder *encoder)
 {
 	drm_encoder_cleanup(encoder);
@@ -900,8 +845,9 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 	}
 
 	if (debug_dump_regs) {
-		DRM_INFO("DSI regs before:\n");
-		vc4_dsi_dump_regs(dsi);
+		struct drm_printer p = drm_info_printer(&dsi->pdev->dev);
+		dev_info(&dsi->pdev->dev, "DSI regs before:\n");
+		drm_print_regset32(&p, &dsi->regset);
 	}
 
 	/* Round up the clk_set_rate() request slightly, since
@@ -1135,8 +1081,9 @@ static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
 	drm_bridge_enable(dsi->bridge);
 
 	if (debug_dump_regs) {
-		DRM_INFO("DSI regs after:\n");
-		vc4_dsi_dump_regs(dsi);
+		struct drm_printer p = drm_info_printer(&dsi->pdev->dev);
+		dev_info(&dsi->pdev->dev, "DSI regs after:\n");
+		drm_print_regset32(&p, &dsi->regset);
 	}
 }
 
@@ -1527,6 +1474,15 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(dsi->regs))
 		return PTR_ERR(dsi->regs);
 
+	dsi->regset.base = dsi->regs;
+	if (dsi->port == 0) {
+		dsi->regset.regs = dsi0_regs;
+		dsi->regset.nregs = ARRAY_SIZE(dsi0_regs);
+	} else {
+		dsi->regset.regs = dsi1_regs;
+		dsi->regset.nregs = ARRAY_SIZE(dsi1_regs);
+	}
+
 	if (DSI_PORT_READ(ID) != DSI_ID_VALUE) {
 		dev_err(dev, "Port returned 0x%08x for ID instead of 0x%08x\n",
 			DSI_PORT_READ(ID), DSI_ID_VALUE);
@@ -1662,6 +1618,11 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	 */
 	dsi->encoder->bridge = NULL;
 
+	if (dsi->port == 0)
+		vc4_debugfs_add_regset32(drm, "dsi0_regs", &dsi->regset);
+	else
+		vc4_debugfs_add_regset32(drm, "dsi1_regs", &dsi->regset);
+
 	pm_runtime_enable(dev);
 
 	return 0;
diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
index aea2b8dfec17..d9311be32a4f 100644
--- a/drivers/gpu/drm/vc4/vc4_gem.c
+++ b/drivers/gpu/drm/vc4/vc4_gem.c
@@ -74,6 +74,11 @@ vc4_get_hang_state_ioctl(struct drm_device *dev, void *data,
 	u32 i;
 	int ret = 0;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("VC4_GET_HANG_STATE with no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	spin_lock_irqsave(&vc4->job_lock, irqflags);
 	kernel_state = vc4->hang_state;
 	if (!kernel_state) {
@@ -536,7 +541,7 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno)
 		bo = to_vc4_bo(&exec->bo[i]->base);
 		bo->seqno = seqno;
 
-		reservation_object_add_shared_fence(bo->resv, exec->fence);
+		reservation_object_add_shared_fence(bo->base.base.resv, exec->fence);
 	}
 
 	list_for_each_entry(bo, &exec->unref_list, unref_head) {
@@ -547,7 +552,7 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno)
 		bo = to_vc4_bo(&exec->rcl_write_bo[i]->base);
 		bo->write_seqno = seqno;
 
-		reservation_object_add_excl_fence(bo->resv, exec->fence);
+		reservation_object_add_excl_fence(bo->base.base.resv, exec->fence);
 	}
 }
 
@@ -559,7 +564,7 @@ vc4_unlock_bo_reservations(struct drm_device *dev,
 	int i;
 
 	for (i = 0; i < exec->bo_count; i++) {
-		struct vc4_bo *bo = to_vc4_bo(&exec->bo[i]->base);
+		struct drm_gem_object *bo = &exec->bo[i]->base;
 
 		ww_mutex_unlock(&bo->resv->lock);
 	}
@@ -581,13 +586,13 @@ vc4_lock_bo_reservations(struct drm_device *dev,
 {
 	int contended_lock = -1;
 	int i, ret;
-	struct vc4_bo *bo;
+	struct drm_gem_object *bo;
 
 	ww_acquire_init(acquire_ctx, &reservation_ww_class);
 
 retry:
 	if (contended_lock != -1) {
-		bo = to_vc4_bo(&exec->bo[contended_lock]->base);
+		bo = &exec->bo[contended_lock]->base;
 		ret = ww_mutex_lock_slow_interruptible(&bo->resv->lock,
 						       acquire_ctx);
 		if (ret) {
@@ -600,19 +605,19 @@ retry:
 		if (i == contended_lock)
 			continue;
 
-		bo = to_vc4_bo(&exec->bo[i]->base);
+		bo = &exec->bo[i]->base;
 
 		ret = ww_mutex_lock_interruptible(&bo->resv->lock, acquire_ctx);
 		if (ret) {
 			int j;
 
 			for (j = 0; j < i; j++) {
-				bo = to_vc4_bo(&exec->bo[j]->base);
+				bo = &exec->bo[j]->base;
 				ww_mutex_unlock(&bo->resv->lock);
 			}
 
 			if (contended_lock != -1 && contended_lock >= i) {
-				bo = to_vc4_bo(&exec->bo[contended_lock]->base);
+				bo = &exec->bo[contended_lock]->base;
 
 				ww_mutex_unlock(&bo->resv->lock);
 			}
@@ -633,7 +638,7 @@ retry:
 	 * before we commit the CL to the hardware.
 	 */
 	for (i = 0; i < exec->bo_count; i++) {
-		bo = to_vc4_bo(&exec->bo[i]->base);
+		bo = &exec->bo[i]->base;
 
 		ret = reservation_object_reserve_shared(bo->resv, 1);
 		if (ret) {
@@ -964,12 +969,7 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 	/* Release the reference we had on the perf monitor. */
 	vc4_perfmon_put(exec->perfmon);
 
-	mutex_lock(&vc4->power_lock);
-	if (--vc4->power_refcount == 0) {
-		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
-		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
-	}
-	mutex_unlock(&vc4->power_lock);
+	vc4_v3d_pm_put(vc4);
 
 	kfree(exec);
 }
@@ -1124,6 +1124,11 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 	struct dma_fence *in_fence;
 	int ret = 0;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("VC4_SUBMIT_CL with no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	if ((args->flags & ~(VC4_SUBMIT_CL_USE_CLEAR_COLOR |
 			     VC4_SUBMIT_CL_FIXED_RCL_ORDER |
 			     VC4_SUBMIT_CL_RCL_ORDER_INCREASING_X |
@@ -1143,17 +1148,11 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 		return -ENOMEM;
 	}
 
-	mutex_lock(&vc4->power_lock);
-	if (vc4->power_refcount++ == 0) {
-		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
-		if (ret < 0) {
-			mutex_unlock(&vc4->power_lock);
-			vc4->power_refcount--;
-			kfree(exec);
-			return ret;
-		}
+	ret = vc4_v3d_pm_get(vc4);
+	if (ret) {
+		kfree(exec);
+		return ret;
 	}
-	mutex_unlock(&vc4->power_lock);
 
 	exec->args = args;
 	INIT_LIST_HEAD(&exec->unref_list);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 88fd5df7e7dc..99fc8569e0f5 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -97,6 +97,9 @@ struct vc4_hdmi {
 
 	struct clk *pixel_clock;
 	struct clk *hsm_clock;
+
+	struct debugfs_regset32 hdmi_regset;
+	struct debugfs_regset32 hd_regset;
 };
 
 #define HDMI_READ(offset) readl(vc4->hdmi->hdmicore_regs + offset)
@@ -134,103 +137,69 @@ to_vc4_hdmi_connector(struct drm_connector *connector)
 	return container_of(connector, struct vc4_hdmi_connector, base);
 }
 
-#define HDMI_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} hdmi_regs[] = {
-	HDMI_REG(VC4_HDMI_CORE_REV),
-	HDMI_REG(VC4_HDMI_SW_RESET_CONTROL),
-	HDMI_REG(VC4_HDMI_HOTPLUG_INT),
-	HDMI_REG(VC4_HDMI_HOTPLUG),
-	HDMI_REG(VC4_HDMI_MAI_CHANNEL_MAP),
-	HDMI_REG(VC4_HDMI_MAI_CONFIG),
-	HDMI_REG(VC4_HDMI_MAI_FORMAT),
-	HDMI_REG(VC4_HDMI_AUDIO_PACKET_CONFIG),
-	HDMI_REG(VC4_HDMI_RAM_PACKET_CONFIG),
-	HDMI_REG(VC4_HDMI_HORZA),
-	HDMI_REG(VC4_HDMI_HORZB),
-	HDMI_REG(VC4_HDMI_FIFO_CTL),
-	HDMI_REG(VC4_HDMI_SCHEDULER_CONTROL),
-	HDMI_REG(VC4_HDMI_VERTA0),
-	HDMI_REG(VC4_HDMI_VERTA1),
-	HDMI_REG(VC4_HDMI_VERTB0),
-	HDMI_REG(VC4_HDMI_VERTB1),
-	HDMI_REG(VC4_HDMI_TX_PHY_RESET_CTL),
-	HDMI_REG(VC4_HDMI_TX_PHY_CTL0),
-
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_1),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_2),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_3),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_4),
-	HDMI_REG(VC4_HDMI_CEC_CNTRL_5),
-	HDMI_REG(VC4_HDMI_CPU_STATUS),
-	HDMI_REG(VC4_HDMI_CPU_MASK_STATUS),
-
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_1),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_2),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_3),
-	HDMI_REG(VC4_HDMI_CEC_RX_DATA_4),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_1),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_2),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_3),
-	HDMI_REG(VC4_HDMI_CEC_TX_DATA_4),
+static const struct debugfs_reg32 hdmi_regs[] = {
+	VC4_REG32(VC4_HDMI_CORE_REV),
+	VC4_REG32(VC4_HDMI_SW_RESET_CONTROL),
+	VC4_REG32(VC4_HDMI_HOTPLUG_INT),
+	VC4_REG32(VC4_HDMI_HOTPLUG),
+	VC4_REG32(VC4_HDMI_MAI_CHANNEL_MAP),
+	VC4_REG32(VC4_HDMI_MAI_CONFIG),
+	VC4_REG32(VC4_HDMI_MAI_FORMAT),
+	VC4_REG32(VC4_HDMI_AUDIO_PACKET_CONFIG),
+	VC4_REG32(VC4_HDMI_RAM_PACKET_CONFIG),
+	VC4_REG32(VC4_HDMI_HORZA),
+	VC4_REG32(VC4_HDMI_HORZB),
+	VC4_REG32(VC4_HDMI_FIFO_CTL),
+	VC4_REG32(VC4_HDMI_SCHEDULER_CONTROL),
+	VC4_REG32(VC4_HDMI_VERTA0),
+	VC4_REG32(VC4_HDMI_VERTA1),
+	VC4_REG32(VC4_HDMI_VERTB0),
+	VC4_REG32(VC4_HDMI_VERTB1),
+	VC4_REG32(VC4_HDMI_TX_PHY_RESET_CTL),
+	VC4_REG32(VC4_HDMI_TX_PHY_CTL0),
+
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_1),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_2),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_3),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_4),
+	VC4_REG32(VC4_HDMI_CEC_CNTRL_5),
+	VC4_REG32(VC4_HDMI_CPU_STATUS),
+	VC4_REG32(VC4_HDMI_CPU_MASK_STATUS),
+
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_1),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_2),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_3),
+	VC4_REG32(VC4_HDMI_CEC_RX_DATA_4),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_1),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_2),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_3),
+	VC4_REG32(VC4_HDMI_CEC_TX_DATA_4),
 };
 
-static const struct {
-	u32 reg;
-	const char *name;
-} hd_regs[] = {
-	HDMI_REG(VC4_HD_M_CTL),
-	HDMI_REG(VC4_HD_MAI_CTL),
-	HDMI_REG(VC4_HD_MAI_THR),
-	HDMI_REG(VC4_HD_MAI_FMT),
-	HDMI_REG(VC4_HD_MAI_SMP),
-	HDMI_REG(VC4_HD_VID_CTL),
-	HDMI_REG(VC4_HD_CSC_CTL),
-	HDMI_REG(VC4_HD_FRAME_COUNT),
+static const struct debugfs_reg32 hd_regs[] = {
+	VC4_REG32(VC4_HD_M_CTL),
+	VC4_REG32(VC4_HD_MAI_CTL),
+	VC4_REG32(VC4_HD_MAI_THR),
+	VC4_REG32(VC4_HD_MAI_FMT),
+	VC4_REG32(VC4_HD_MAI_SMP),
+	VC4_REG32(VC4_HD_VID_CTL),
+	VC4_REG32(VC4_HD_CSC_CTL),
+	VC4_REG32(VC4_HD_FRAME_COUNT),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hdmi_regs[i].name, hdmi_regs[i].reg,
-			   HDMI_READ(hdmi_regs[i].reg));
-	}
+	struct vc4_hdmi *hdmi = vc4->hdmi;
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hd_regs[i].name, hd_regs[i].reg,
-			   HD_READ(hd_regs[i].reg));
-	}
+	drm_print_regset32(&p, &hdmi->hdmi_regset);
+	drm_print_regset32(&p, &hdmi->hd_regset);
 
 	return 0;
 }
-#endif /* CONFIG_DEBUG_FS */
-
-static void vc4_hdmi_dump_regs(struct drm_device *dev)
-{
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hdmi_regs[i].reg, hdmi_regs[i].name,
-			 HDMI_READ(hdmi_regs[i].reg));
-	}
-	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hd_regs[i].reg, hd_regs[i].name,
-			 HD_READ(hd_regs[i].reg));
-	}
-}
 
 static enum drm_connector_status
 vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
@@ -561,8 +530,11 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
 	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0);
 
 	if (debug_dump_regs) {
-		DRM_INFO("HDMI regs before:\n");
-		vc4_hdmi_dump_regs(dev);
+		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+
+		dev_info(&hdmi->pdev->dev, "HDMI regs before:\n");
+		drm_print_regset32(&p, &hdmi->hdmi_regset);
+		drm_print_regset32(&p, &hdmi->hd_regset);
 	}
 
 	HD_WRITE(VC4_HD_VID_CTL, 0);
@@ -637,8 +609,11 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
 	HDMI_WRITE(VC4_HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
 
 	if (debug_dump_regs) {
-		DRM_INFO("HDMI regs after:\n");
-		vc4_hdmi_dump_regs(dev);
+		struct drm_printer p = drm_info_printer(&hdmi->pdev->dev);
+
+		dev_info(&hdmi->pdev->dev, "HDMI regs after:\n");
+		drm_print_regset32(&p, &hdmi->hdmi_regset);
+		drm_print_regset32(&p, &hdmi->hd_regset);
 	}
 
 	HD_WRITE(VC4_HD_VID_CTL,
@@ -1333,6 +1308,13 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(hdmi->hd_regs))
 		return PTR_ERR(hdmi->hd_regs);
 
+	hdmi->hdmi_regset.base = hdmi->hdmicore_regs;
+	hdmi->hdmi_regset.regs = hdmi_regs;
+	hdmi->hdmi_regset.nregs = ARRAY_SIZE(hdmi_regs);
+	hdmi->hd_regset.base = hdmi->hd_regs;
+	hdmi->hd_regset.regs = hd_regs;
+	hdmi->hd_regset.nregs = ARRAY_SIZE(hd_regs);
+
 	hdmi->pixel_clock = devm_clk_get(dev, "pixel");
 	if (IS_ERR(hdmi->pixel_clock)) {
 		DRM_ERROR("Failed to get pixel clock\n");
@@ -1448,6 +1430,8 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		goto err_destroy_encoder;
 
+	vc4_debugfs_add_file(drm, "hdmi_regs", vc4_hdmi_debugfs_regs, hdmi);
+
 	return 0;
 
 #ifdef CONFIG_DRM_VC4_HDMI_CEC
diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
index 5d8c749c9749..f746e9a7a88c 100644
--- a/drivers/gpu/drm/vc4/vc4_hvs.c
+++ b/drivers/gpu/drm/vc4/vc4_hvs.c
@@ -22,58 +22,52 @@
  * each CRTC.
  */
 
+#include <drm/drm_atomic_helper.h>
 #include <linux/component.h>
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-#define HVS_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} hvs_regs[] = {
-	HVS_REG(SCALER_DISPCTRL),
-	HVS_REG(SCALER_DISPSTAT),
-	HVS_REG(SCALER_DISPID),
-	HVS_REG(SCALER_DISPECTRL),
-	HVS_REG(SCALER_DISPPROF),
-	HVS_REG(SCALER_DISPDITHER),
-	HVS_REG(SCALER_DISPEOLN),
-	HVS_REG(SCALER_DISPLIST0),
-	HVS_REG(SCALER_DISPLIST1),
-	HVS_REG(SCALER_DISPLIST2),
-	HVS_REG(SCALER_DISPLSTAT),
-	HVS_REG(SCALER_DISPLACT0),
-	HVS_REG(SCALER_DISPLACT1),
-	HVS_REG(SCALER_DISPLACT2),
-	HVS_REG(SCALER_DISPCTRL0),
-	HVS_REG(SCALER_DISPBKGND0),
-	HVS_REG(SCALER_DISPSTAT0),
-	HVS_REG(SCALER_DISPBASE0),
-	HVS_REG(SCALER_DISPCTRL1),
-	HVS_REG(SCALER_DISPBKGND1),
-	HVS_REG(SCALER_DISPSTAT1),
-	HVS_REG(SCALER_DISPBASE1),
-	HVS_REG(SCALER_DISPCTRL2),
-	HVS_REG(SCALER_DISPBKGND2),
-	HVS_REG(SCALER_DISPSTAT2),
-	HVS_REG(SCALER_DISPBASE2),
-	HVS_REG(SCALER_DISPALPHA2),
-	HVS_REG(SCALER_OLEDOFFS),
-	HVS_REG(SCALER_OLEDCOEF0),
-	HVS_REG(SCALER_OLEDCOEF1),
-	HVS_REG(SCALER_OLEDCOEF2),
+static const struct debugfs_reg32 hvs_regs[] = {
+	VC4_REG32(SCALER_DISPCTRL),
+	VC4_REG32(SCALER_DISPSTAT),
+	VC4_REG32(SCALER_DISPID),
+	VC4_REG32(SCALER_DISPECTRL),
+	VC4_REG32(SCALER_DISPPROF),
+	VC4_REG32(SCALER_DISPDITHER),
+	VC4_REG32(SCALER_DISPEOLN),
+	VC4_REG32(SCALER_DISPLIST0),
+	VC4_REG32(SCALER_DISPLIST1),
+	VC4_REG32(SCALER_DISPLIST2),
+	VC4_REG32(SCALER_DISPLSTAT),
+	VC4_REG32(SCALER_DISPLACT0),
+	VC4_REG32(SCALER_DISPLACT1),
+	VC4_REG32(SCALER_DISPLACT2),
+	VC4_REG32(SCALER_DISPCTRL0),
+	VC4_REG32(SCALER_DISPBKGND0),
+	VC4_REG32(SCALER_DISPSTAT0),
+	VC4_REG32(SCALER_DISPBASE0),
+	VC4_REG32(SCALER_DISPCTRL1),
+	VC4_REG32(SCALER_DISPBKGND1),
+	VC4_REG32(SCALER_DISPSTAT1),
+	VC4_REG32(SCALER_DISPBASE1),
+	VC4_REG32(SCALER_DISPCTRL2),
+	VC4_REG32(SCALER_DISPBKGND2),
+	VC4_REG32(SCALER_DISPSTAT2),
+	VC4_REG32(SCALER_DISPBASE2),
+	VC4_REG32(SCALER_DISPALPHA2),
+	VC4_REG32(SCALER_OLEDOFFS),
+	VC4_REG32(SCALER_OLEDCOEF0),
+	VC4_REG32(SCALER_OLEDCOEF1),
+	VC4_REG32(SCALER_OLEDCOEF2),
 };
 
 void vc4_hvs_dump_state(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct drm_printer p = drm_info_printer(&vc4->hvs->pdev->dev);
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
-		DRM_INFO("0x%04x (%s): 0x%08x\n",
-			 hvs_regs[i].reg, hvs_regs[i].name,
-			 HVS_READ(hvs_regs[i].reg));
-	}
+	drm_print_regset32(&p, &vc4->hvs->regset);
 
 	DRM_INFO("HVS ctx:\n");
 	for (i = 0; i < 64; i += 4) {
@@ -86,23 +80,17 @@ void vc4_hvs_dump_state(struct drm_device *dev)
 	}
 }
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_hvs_debugfs_underrun(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
+	struct drm_printer p = drm_seq_file_printer(m);
 
-	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   hvs_regs[i].name, hvs_regs[i].reg,
-			   HVS_READ(hvs_regs[i].reg));
-	}
+	drm_printf(&p, "%d\n", atomic_read(&vc4->underrun));
 
 	return 0;
 }
-#endif
 
 /* The filter kernel is composed of dwords each containing 3 9-bit
  * signed integers packed next to each other.
@@ -166,6 +154,67 @@ static int vc4_hvs_upload_linear_kernel(struct vc4_hvs *hvs,
 	return 0;
 }
 
+void vc4_hvs_mask_underrun(struct drm_device *dev, int channel)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	u32 dispctrl = HVS_READ(SCALER_DISPCTRL);
+
+	dispctrl &= ~SCALER_DISPCTRL_DSPEISLUR(channel);
+
+	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
+}
+
+void vc4_hvs_unmask_underrun(struct drm_device *dev, int channel)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	u32 dispctrl = HVS_READ(SCALER_DISPCTRL);
+
+	dispctrl |= SCALER_DISPCTRL_DSPEISLUR(channel);
+
+	HVS_WRITE(SCALER_DISPSTAT,
+		  SCALER_DISPSTAT_EUFLOW(channel));
+	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
+}
+
+static void vc4_hvs_report_underrun(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	atomic_inc(&vc4->underrun);
+	DRM_DEV_ERROR(dev->dev, "HVS underrun\n");
+}
+
+static irqreturn_t vc4_hvs_irq_handler(int irq, void *data)
+{
+	struct drm_device *dev = data;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	irqreturn_t irqret = IRQ_NONE;
+	int channel;
+	u32 control;
+	u32 status;
+
+	status = HVS_READ(SCALER_DISPSTAT);
+	control = HVS_READ(SCALER_DISPCTRL);
+
+	for (channel = 0; channel < SCALER_CHANNELS_COUNT; channel++) {
+		/* Interrupt masking is not always honored, so check it here. */
+		if (status & SCALER_DISPSTAT_EUFLOW(channel) &&
+		    control & SCALER_DISPCTRL_DSPEISLUR(channel)) {
+			vc4_hvs_mask_underrun(dev, channel);
+			vc4_hvs_report_underrun(dev);
+
+			irqret = IRQ_HANDLED;
+		}
+	}
+
+	/* Clear every per-channel interrupt flag. */
+	HVS_WRITE(SCALER_DISPSTAT, SCALER_DISPSTAT_IRQMASK(0) |
+				   SCALER_DISPSTAT_IRQMASK(1) |
+				   SCALER_DISPSTAT_IRQMASK(2));
+
+	return irqret;
+}
+
 static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -185,6 +234,10 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	if (IS_ERR(hvs->regs))
 		return PTR_ERR(hvs->regs);
 
+	hvs->regset.base = hvs->regs;
+	hvs->regset.regs = hvs_regs;
+	hvs->regset.nregs = ARRAY_SIZE(hvs_regs);
+
 	hvs->dlist = hvs->regs + SCALER_DLIST_START;
 
 	spin_lock_init(&hvs->mm_lock);
@@ -219,15 +272,40 @@ static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
 	dispctrl = HVS_READ(SCALER_DISPCTRL);
 
 	dispctrl |= SCALER_DISPCTRL_ENABLE;
+	dispctrl |= SCALER_DISPCTRL_DISPEIRQ(0) |
+		    SCALER_DISPCTRL_DISPEIRQ(1) |
+		    SCALER_DISPCTRL_DISPEIRQ(2);
 
 	/* Set DSP3 (PV1) to use HVS channel 2, which would otherwise
 	 * be unused.
 	 */
 	dispctrl &= ~SCALER_DISPCTRL_DSP3_MUX_MASK;
+	dispctrl &= ~(SCALER_DISPCTRL_DMAEIRQ |
+		      SCALER_DISPCTRL_SLVWREIRQ |
+		      SCALER_DISPCTRL_SLVRDEIRQ |
+		      SCALER_DISPCTRL_DSPEIEOF(0) |
+		      SCALER_DISPCTRL_DSPEIEOF(1) |
+		      SCALER_DISPCTRL_DSPEIEOF(2) |
+		      SCALER_DISPCTRL_DSPEIEOLN(0) |
+		      SCALER_DISPCTRL_DSPEIEOLN(1) |
+		      SCALER_DISPCTRL_DSPEIEOLN(2) |
+		      SCALER_DISPCTRL_DSPEISLUR(0) |
+		      SCALER_DISPCTRL_DSPEISLUR(1) |
+		      SCALER_DISPCTRL_DSPEISLUR(2) |
+		      SCALER_DISPCTRL_SCLEIRQ);
 	dispctrl |= VC4_SET_FIELD(2, SCALER_DISPCTRL_DSP3_MUX);
 
 	HVS_WRITE(SCALER_DISPCTRL, dispctrl);
 
+	ret = devm_request_irq(dev, platform_get_irq(pdev, 0),
+			       vc4_hvs_irq_handler, 0, "vc4 hvs", drm);
+	if (ret)
+		return ret;
+
+	vc4_debugfs_add_regset32(drm, "hvs_regs", &hvs->regset);
+	vc4_debugfs_add_file(drm, "hvs_underrun", vc4_hvs_debugfs_underrun,
+			     NULL);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vc4/vc4_irq.c b/drivers/gpu/drm/vc4/vc4_irq.c
index 4cd2ccfe15f4..ffd0a4388752 100644
--- a/drivers/gpu/drm/vc4/vc4_irq.c
+++ b/drivers/gpu/drm/vc4/vc4_irq.c
@@ -229,6 +229,9 @@ vc4_irq_preinstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return;
+
 	init_waitqueue_head(&vc4->job_wait_queue);
 	INIT_WORK(&vc4->overflow_mem_work, vc4_overflow_mem_work);
 
@@ -243,6 +246,9 @@ vc4_irq_postinstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return 0;
+
 	/* Enable both the render done and out of memory interrupts. */
 	V3D_WRITE(V3D_INTENA, V3D_DRIVER_IRQS);
 
@@ -254,6 +260,9 @@ vc4_irq_uninstall(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
+	if (!vc4->v3d)
+		return;
+
 	/* Disable sending interrupts for our driver's IRQs. */
 	V3D_WRITE(V3D_INTDIS, V3D_DRIVER_IRQS);
 
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index 91b8c72ff361..295dacc8bcb9 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -34,6 +34,18 @@ static struct vc4_ctm_state *to_vc4_ctm_state(struct drm_private_state *priv)
 	return container_of(priv, struct vc4_ctm_state, base);
 }
 
+struct vc4_load_tracker_state {
+	struct drm_private_state base;
+	u64 hvs_load;
+	u64 membus_load;
+};
+
+static struct vc4_load_tracker_state *
+to_vc4_load_tracker_state(struct drm_private_state *priv)
+{
+	return container_of(priv, struct vc4_load_tracker_state, base);
+}
+
 static struct vc4_ctm_state *vc4_get_ctm_state(struct drm_atomic_state *state,
 					       struct drm_private_obj *manager)
 {
@@ -138,6 +150,16 @@ vc4_atomic_complete_commit(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc;
+	int i;
+
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		if (!state->crtcs[i].ptr || !state->crtcs[i].commit)
+			continue;
+
+		vc4_crtc = to_vc4_crtc(state->crtcs[i].ptr);
+		vc4_hvs_mask_underrun(dev, vc4_crtc->channel);
+	}
 
 	drm_atomic_helper_wait_for_fences(dev, state, false);
 
@@ -385,6 +407,85 @@ vc4_ctm_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 	return 0;
 }
 
+static int vc4_load_tracker_atomic_check(struct drm_atomic_state *state)
+{
+	struct drm_plane_state *old_plane_state, *new_plane_state;
+	struct vc4_dev *vc4 = to_vc4_dev(state->dev);
+	struct vc4_load_tracker_state *load_state;
+	struct drm_private_state *priv_state;
+	struct drm_plane *plane;
+	int i;
+
+	priv_state = drm_atomic_get_private_obj_state(state,
+						      &vc4->load_tracker);
+	if (IS_ERR(priv_state))
+		return PTR_ERR(priv_state);
+
+	load_state = to_vc4_load_tracker_state(priv_state);
+	for_each_oldnew_plane_in_state(state, plane, old_plane_state,
+				       new_plane_state, i) {
+		struct vc4_plane_state *vc4_plane_state;
+
+		if (old_plane_state->fb && old_plane_state->crtc) {
+			vc4_plane_state = to_vc4_plane_state(old_plane_state);
+			load_state->membus_load -= vc4_plane_state->membus_load;
+			load_state->hvs_load -= vc4_plane_state->hvs_load;
+		}
+
+		if (new_plane_state->fb && new_plane_state->crtc) {
+			vc4_plane_state = to_vc4_plane_state(new_plane_state);
+			load_state->membus_load += vc4_plane_state->membus_load;
+			load_state->hvs_load += vc4_plane_state->hvs_load;
+		}
+	}
+
+	/* Don't check the load when the tracker is disabled. */
+	if (!vc4->load_tracker_enabled)
+		return 0;
+
+	/* The absolute limit is 2Gbyte/sec, but let's take a margin to let
+	 * the system work when other blocks are accessing the memory.
+	 */
+	if (load_state->membus_load > SZ_1G + SZ_512M)
+		return -ENOSPC;
+
+	/* HVS clock is supposed to run @ 250Mhz, let's take a margin and
+	 * consider the maximum number of cycles is 240M.
+	 */
+	if (load_state->hvs_load > 240000000ULL)
+		return -ENOSPC;
+
+	return 0;
+}
+
+static struct drm_private_state *
+vc4_load_tracker_duplicate_state(struct drm_private_obj *obj)
+{
+	struct vc4_load_tracker_state *state;
+
+	state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return NULL;
+
+	__drm_atomic_helper_private_obj_duplicate_state(obj, &state->base);
+
+	return &state->base;
+}
+
+static void vc4_load_tracker_destroy_state(struct drm_private_obj *obj,
+					   struct drm_private_state *state)
+{
+	struct vc4_load_tracker_state *load_state;
+
+	load_state = to_vc4_load_tracker_state(state);
+	kfree(load_state);
+}
+
+static const struct drm_private_state_funcs vc4_load_tracker_state_funcs = {
+	.atomic_duplicate_state = vc4_load_tracker_duplicate_state,
+	.atomic_destroy_state = vc4_load_tracker_destroy_state,
+};
+
 static int
 vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 {
@@ -394,7 +495,11 @@ vc4_atomic_check(struct drm_device *dev, struct drm_atomic_state *state)
 	if (ret < 0)
 		return ret;
 
-	return drm_atomic_helper_check(dev, state);
+	ret = drm_atomic_helper_check(dev, state);
+	if (ret)
+		return ret;
+
+	return vc4_load_tracker_atomic_check(state);
 }
 
 static const struct drm_mode_config_funcs vc4_mode_funcs = {
@@ -407,13 +512,20 @@ int vc4_kms_load(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_ctm_state *ctm_state;
+	struct vc4_load_tracker_state *load_state;
 	int ret;
 
+	/* Start with the load tracker enabled. Can be disabled through the
+	 * debugfs load_tracker file.
+	 */
+	vc4->load_tracker_enabled = true;
+
 	sema_init(&vc4->async_modeset, 1);
 
 	/* Set support for vblank irq fast disable, before drm_vblank_init() */
 	dev->vblank_disable_immediate = true;
 
+	dev->irq_enabled = true;
 	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to initialize vblank\n");
@@ -436,6 +548,15 @@ int vc4_kms_load(struct drm_device *dev)
 	drm_atomic_private_obj_init(dev, &vc4->ctm_manager, &ctm_state->base,
 				    &vc4_ctm_state_funcs);
 
+	load_state = kzalloc(sizeof(*load_state), GFP_KERNEL);
+	if (!load_state) {
+		drm_atomic_private_obj_fini(&vc4->ctm_manager);
+		return -ENOMEM;
+	}
+
+	drm_atomic_private_obj_init(dev, &vc4->load_tracker, &load_state->base,
+				    &vc4_load_tracker_state_funcs);
+
 	drm_mode_config_reset(dev);
 
 	drm_kms_helper_poll_init(dev);
diff --git a/drivers/gpu/drm/vc4/vc4_perfmon.c b/drivers/gpu/drm/vc4/vc4_perfmon.c
index 495150415020..f4aa75efd16b 100644
--- a/drivers/gpu/drm/vc4/vc4_perfmon.c
+++ b/drivers/gpu/drm/vc4/vc4_perfmon.c
@@ -100,12 +100,18 @@ void vc4_perfmon_close_file(struct vc4_file *vc4file)
 int vc4_perfmon_create_ioctl(struct drm_device *dev, void *data,
 			     struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_create *req = data;
 	struct vc4_perfmon *perfmon;
 	unsigned int i;
 	int ret;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Creating perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	/* Number of monitored counters cannot exceed HW limits. */
 	if (req->ncounters > DRM_VC4_MAX_PERF_COUNTERS ||
 	    !req->ncounters)
@@ -146,10 +152,16 @@ int vc4_perfmon_create_ioctl(struct drm_device *dev, void *data,
 int vc4_perfmon_destroy_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_destroy *req = data;
 	struct vc4_perfmon *perfmon;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Destroying perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	mutex_lock(&vc4file->perfmon.lock);
 	perfmon = idr_remove(&vc4file->perfmon.idr, req->id);
 	mutex_unlock(&vc4file->perfmon.lock);
@@ -164,11 +176,17 @@ int vc4_perfmon_destroy_ioctl(struct drm_device *dev, void *data,
 int vc4_perfmon_get_values_ioctl(struct drm_device *dev, void *data,
 				 struct drm_file *file_priv)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_file *vc4file = file_priv->driver_priv;
 	struct drm_vc4_perfmon_get_values *req = data;
 	struct vc4_perfmon *perfmon;
 	int ret;
 
+	if (!vc4->v3d) {
+		DRM_DEBUG("Getting perfmon no VC4 V3D probed\n");
+		return -ENODEV;
+	}
+
 	mutex_lock(&vc4file->perfmon.lock);
 	perfmon = idr_find(&vc4file->perfmon.idr, req->id);
 	vc4_perfmon_get(perfmon);
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index d098337c10e9..4d918d3e4858 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -488,6 +488,61 @@ static void vc4_write_scaling_parameters(struct drm_plane_state *state,
 	}
 }
 
+static void vc4_plane_calc_load(struct drm_plane_state *state)
+{
+	unsigned int hvs_load_shift, vrefresh, i;
+	struct drm_framebuffer *fb = state->fb;
+	struct vc4_plane_state *vc4_state;
+	struct drm_crtc_state *crtc_state;
+	unsigned int vscale_factor;
+
+	vc4_state = to_vc4_plane_state(state);
+	crtc_state = drm_atomic_get_existing_crtc_state(state->state,
+							state->crtc);
+	vrefresh = drm_mode_vrefresh(&crtc_state->adjusted_mode);
+
+	/* The HVS is able to process 2 pixels/cycle when scaling the source,
+	 * 4 pixels/cycle otherwise.
+	 * Alpha blending step seems to be pipelined and it's always operating
+	 * at 4 pixels/cycle, so the limiting aspect here seems to be the
+	 * scaler block.
+	 * HVS load is expressed in clk-cycles/sec (AKA Hz).
+	 */
+	if (vc4_state->x_scaling[0] != VC4_SCALING_NONE ||
+	    vc4_state->x_scaling[1] != VC4_SCALING_NONE ||
+	    vc4_state->y_scaling[0] != VC4_SCALING_NONE ||
+	    vc4_state->y_scaling[1] != VC4_SCALING_NONE)
+		hvs_load_shift = 1;
+	else
+		hvs_load_shift = 2;
+
+	vc4_state->membus_load = 0;
+	vc4_state->hvs_load = 0;
+	for (i = 0; i < fb->format->num_planes; i++) {
+		/* Even if the bandwidth/plane required for a single frame is
+		 *
+		 * vc4_state->src_w[i] * vc4_state->src_h[i] * cpp * vrefresh
+		 *
+		 * when downscaling, we have to read more pixels per line in
+		 * the time frame reserved for a single line, so the bandwidth
+		 * demand can be punctually higher. To account for that, we
+		 * calculate the down-scaling factor and multiply the plane
+		 * load by this number. We're likely over-estimating the read
+		 * demand, but that's better than under-estimating it.
+		 */
+		vscale_factor = DIV_ROUND_UP(vc4_state->src_h[i],
+					     vc4_state->crtc_h);
+		vc4_state->membus_load += vc4_state->src_w[i] *
+					  vc4_state->src_h[i] * vscale_factor *
+					  fb->format->cpp[i];
+		vc4_state->hvs_load += vc4_state->crtc_h * vc4_state->crtc_w;
+	}
+
+	vc4_state->hvs_load *= vrefresh;
+	vc4_state->hvs_load >>= hvs_load_shift;
+	vc4_state->membus_load *= vrefresh;
+}
+
 static int vc4_plane_allocate_lbm(struct drm_plane_state *state)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(state->plane->dev);
@@ -875,6 +930,8 @@ static int vc4_plane_mode_set(struct drm_plane *plane,
 	 */
 	vc4_state->dlist_initialized = 1;
 
+	vc4_plane_calc_load(state);
+
 	return 0;
 }
 
@@ -1082,7 +1139,7 @@ static int vc4_prepare_fb(struct drm_plane *plane,
 
 	bo = to_vc4_bo(&drm_fb_cma_get_gem_obj(state->fb, 0)->base);
 
-	fence = reservation_object_get_excl_rcu(bo->resv);
+	fence = reservation_object_get_excl_rcu(bo->base.base.resv);
 	drm_atomic_set_fence_for_plane(state, fence);
 
 	if (plane->state->fb == state->fb)
diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h
index 931088014272..c0c5fadaf7e3 100644
--- a/drivers/gpu/drm/vc4/vc4_regs.h
+++ b/drivers/gpu/drm/vc4/vc4_regs.h
@@ -212,11 +212,11 @@
 
 #define PV_HACT_ACT				0x30
 
+#define SCALER_CHANNELS_COUNT			3
+
 #define SCALER_DISPCTRL                         0x00000000
 /* Global register for clock gating the HVS */
 # define SCALER_DISPCTRL_ENABLE			BIT(31)
-# define SCALER_DISPCTRL_DSP2EISLUR		BIT(15)
-# define SCALER_DISPCTRL_DSP1EISLUR		BIT(14)
 # define SCALER_DISPCTRL_DSP3_MUX_MASK		VC4_MASK(19, 18)
 # define SCALER_DISPCTRL_DSP3_MUX_SHIFT		18
 
@@ -224,45 +224,25 @@
  * SCALER_DISPSTAT_IRQDISP0.  Note that short frame contributions are
  * always enabled.
  */
-# define SCALER_DISPCTRL_DSP0EISLUR		BIT(13)
-# define SCALER_DISPCTRL_DSP2EIEOLN		BIT(12)
-# define SCALER_DISPCTRL_DSP2EIEOF		BIT(11)
-# define SCALER_DISPCTRL_DSP1EIEOLN		BIT(10)
-# define SCALER_DISPCTRL_DSP1EIEOF		BIT(9)
+# define SCALER_DISPCTRL_DSPEISLUR(x)		BIT(13 + (x))
 /* Enables Display 0 end-of-line-N contribution to
  * SCALER_DISPSTAT_IRQDISP0
  */
-# define SCALER_DISPCTRL_DSP0EIEOLN		BIT(8)
+# define SCALER_DISPCTRL_DSPEIEOLN(x)		BIT(8 + ((x) * 2))
 /* Enables Display 0 EOF contribution to SCALER_DISPSTAT_IRQDISP0 */
-# define SCALER_DISPCTRL_DSP0EIEOF		BIT(7)
+# define SCALER_DISPCTRL_DSPEIEOF(x)		BIT(7 + ((x) * 2))
 
 # define SCALER_DISPCTRL_SLVRDEIRQ		BIT(6)
 # define SCALER_DISPCTRL_SLVWREIRQ		BIT(5)
 # define SCALER_DISPCTRL_DMAEIRQ		BIT(4)
-# define SCALER_DISPCTRL_DISP2EIRQ		BIT(3)
-# define SCALER_DISPCTRL_DISP1EIRQ		BIT(2)
 /* Enables interrupt generation on the enabled EOF/EOLN/EISLUR
  * bits and short frames..
  */
-# define SCALER_DISPCTRL_DISP0EIRQ		BIT(1)
+# define SCALER_DISPCTRL_DISPEIRQ(x)		BIT(1 + (x))
 /* Enables interrupt generation on scaler profiler interrupt. */
 # define SCALER_DISPCTRL_SCLEIRQ		BIT(0)
 
 #define SCALER_DISPSTAT                         0x00000004
-# define SCALER_DISPSTAT_COBLOW2		BIT(29)
-# define SCALER_DISPSTAT_EOLN2			BIT(28)
-# define SCALER_DISPSTAT_ESFRAME2		BIT(27)
-# define SCALER_DISPSTAT_ESLINE2		BIT(26)
-# define SCALER_DISPSTAT_EUFLOW2		BIT(25)
-# define SCALER_DISPSTAT_EOF2			BIT(24)
-
-# define SCALER_DISPSTAT_COBLOW1		BIT(21)
-# define SCALER_DISPSTAT_EOLN1			BIT(20)
-# define SCALER_DISPSTAT_ESFRAME1		BIT(19)
-# define SCALER_DISPSTAT_ESLINE1		BIT(18)
-# define SCALER_DISPSTAT_EUFLOW1		BIT(17)
-# define SCALER_DISPSTAT_EOF1			BIT(16)
-
 # define SCALER_DISPSTAT_RESP_MASK		VC4_MASK(15, 14)
 # define SCALER_DISPSTAT_RESP_SHIFT		14
 # define SCALER_DISPSTAT_RESP_OKAY		0
@@ -270,23 +250,26 @@
 # define SCALER_DISPSTAT_RESP_SLVERR		2
 # define SCALER_DISPSTAT_RESP_DECERR		3
 
-# define SCALER_DISPSTAT_COBLOW0		BIT(13)
+# define SCALER_DISPSTAT_COBLOW(x)		BIT(13 + ((x) * 8))
 /* Set when the DISPEOLN line is done compositing. */
-# define SCALER_DISPSTAT_EOLN0			BIT(12)
+# define SCALER_DISPSTAT_EOLN(x)		BIT(12 + ((x) * 8))
 /* Set when VSTART is seen but there are still pixels in the current
  * output line.
  */
-# define SCALER_DISPSTAT_ESFRAME0		BIT(11)
+# define SCALER_DISPSTAT_ESFRAME(x)		BIT(11 + ((x) * 8))
 /* Set when HSTART is seen but there are still pixels in the current
  * output line.
  */
-# define SCALER_DISPSTAT_ESLINE0		BIT(10)
+# define SCALER_DISPSTAT_ESLINE(x)		BIT(10 + ((x) * 8))
 /* Set when the the downstream tries to read from the display FIFO
  * while it's empty.
  */
-# define SCALER_DISPSTAT_EUFLOW0		BIT(9)
+# define SCALER_DISPSTAT_EUFLOW(x)		BIT(9 + ((x) * 8))
 /* Set when the display mode changes from RUN to EOF */
-# define SCALER_DISPSTAT_EOF0			BIT(8)
+# define SCALER_DISPSTAT_EOF(x)			BIT(8 + ((x) * 8))
+
+# define SCALER_DISPSTAT_IRQMASK(x)		VC4_MASK(13 + ((x) * 8), \
+							 8 + ((x) * 8))
 
 /* Set on AXI invalid DMA ID error. */
 # define SCALER_DISPSTAT_DMA_ERROR		BIT(7)
@@ -298,12 +281,10 @@
  * SCALER_DISPSTAT_RESP_ERROR is not SCALER_DISPSTAT_RESP_OKAY.
  */
 # define SCALER_DISPSTAT_IRQDMA			BIT(4)
-# define SCALER_DISPSTAT_IRQDISP2		BIT(3)
-# define SCALER_DISPSTAT_IRQDISP1		BIT(2)
 /* Set when any of the EOF/EOLN/ESFRAME/ESLINE bits are set and their
  * corresponding interrupt bit is enabled in DISPCTRL.
  */
-# define SCALER_DISPSTAT_IRQDISP0		BIT(1)
+# define SCALER_DISPSTAT_IRQDISP(x)		BIT(1 + (x))
 /* On read, the profiler interrupt.  On write, clear *all* interrupt bits. */
 # define SCALER_DISPSTAT_IRQSCL			BIT(0)
 
diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c
index 273984f71ae2..3c918eeaf56e 100644
--- a/drivers/gpu/drm/vc4/vc4_render_cl.c
+++ b/drivers/gpu/drm/vc4/vc4_render_cl.c
@@ -148,6 +148,12 @@ static void emit_tile(struct vc4_exec_info *exec,
 	}
 
 	if (setup->zs_read) {
+		if (setup->color_read) {
+			/* Exec previous load. */
+			vc4_tile_coordinates(setup, x, y);
+			vc4_store_before_load(setup);
+		}
+
 		if (args->zs_read.flags &
 		    VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
 			rcl_u8(setup, VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER);
@@ -156,12 +162,6 @@ static void emit_tile(struct vc4_exec_info *exec,
 						    &args->zs_read, x, y) |
 				VC4_LOADSTORE_FULL_RES_DISABLE_COLOR);
 		} else {
-			if (setup->color_read) {
-				/* Exec previous load. */
-				vc4_tile_coordinates(setup, x, y);
-				vc4_store_before_load(setup);
-			}
-
 			rcl_u8(setup, VC4_PACKET_LOAD_TILE_BUFFER_GENERAL);
 			rcl_u16(setup, args->zs_read.bits);
 			rcl_u32(setup, setup->zs_read->paddr +
@@ -291,16 +291,15 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 		}
 	}
 	if (setup->zs_read) {
+		if (setup->color_read) {
+			loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE;
+			loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE;
+		}
+
 		if (args->zs_read.flags &
 		    VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) {
 			loop_body_size += VC4_PACKET_LOAD_FULL_RES_TILE_BUFFER_SIZE;
 		} else {
-			if (setup->color_read &&
-			    !(args->color_read.flags &
-			      VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES)) {
-				loop_body_size += VC4_PACKET_TILE_COORDINATES_SIZE;
-				loop_body_size += VC4_PACKET_STORE_TILE_BUFFER_GENERAL_SIZE;
-			}
 			loop_body_size += VC4_PACKET_LOAD_TILE_BUFFER_GENERAL_SIZE;
 		}
 	}
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index aa279b5b0de7..c8b89a78f9f4 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -148,6 +148,7 @@ struct vc4_txp {
 	struct drm_writeback_connector connector;
 
 	void __iomem *regs;
+	struct debugfs_regset32 regset;
 };
 
 static inline struct vc4_txp *encoder_to_vc4_txp(struct drm_encoder *encoder)
@@ -160,40 +161,14 @@ static inline struct vc4_txp *connector_to_vc4_txp(struct drm_connector *conn)
 	return container_of(conn, struct vc4_txp, connector.base);
 }
 
-#define TXP_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} txp_regs[] = {
-	TXP_REG(TXP_DST_PTR),
-	TXP_REG(TXP_DST_PITCH),
-	TXP_REG(TXP_DIM),
-	TXP_REG(TXP_DST_CTRL),
-	TXP_REG(TXP_PROGRESS),
+static const struct debugfs_reg32 txp_regs[] = {
+	VC4_REG32(TXP_DST_PTR),
+	VC4_REG32(TXP_DST_PITCH),
+	VC4_REG32(TXP_DIM),
+	VC4_REG32(TXP_DST_CTRL),
+	VC4_REG32(TXP_PROGRESS),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_txp_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_txp *txp = vc4->txp;
-	int i;
-
-	if (!txp)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(txp_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   txp_regs[i].name, txp_regs[i].reg,
-			   TXP_READ(txp_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static int vc4_txp_connector_get_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
@@ -249,7 +224,6 @@ static int vc4_txp_connector_atomic_check(struct drm_connector *conn,
 					struct drm_connector_state *conn_state)
 {
 	struct drm_crtc_state *crtc_state;
-	struct drm_gem_cma_object *gem;
 	struct drm_framebuffer *fb;
 	int i;
 
@@ -275,8 +249,6 @@ static int vc4_txp_connector_atomic_check(struct drm_connector *conn,
 	if (i == ARRAY_SIZE(drm_fmts))
 		return -EINVAL;
 
-	gem = drm_fb_cma_get_gem_obj(fb, 0);
-
 	/* Pitch must be aligned on 16 bytes. */
 	if (fb->pitches[0] & GENMASK(3, 0))
 		return -EINVAL;
@@ -327,7 +299,7 @@ static void vc4_txp_connector_atomic_commit(struct drm_connector *conn,
 
 	TXP_WRITE(TXP_DST_CTRL, ctrl);
 
-	drm_writeback_queue_job(&txp->connector, conn_state->writeback_job);
+	drm_writeback_queue_job(&txp->connector, conn_state);
 }
 
 static const struct drm_connector_helper_funcs vc4_txp_connector_helper_funcs = {
@@ -413,6 +385,9 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data)
 	txp->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(txp->regs))
 		return PTR_ERR(txp->regs);
+	txp->regset.base = txp->regs;
+	txp->regset.regs = txp_regs;
+	txp->regset.nregs = ARRAY_SIZE(txp_regs);
 
 	drm_connector_helper_add(&txp->connector.base,
 				 &vc4_txp_connector_helper_funcs);
@@ -431,6 +406,8 @@ static int vc4_txp_bind(struct device *dev, struct device *master, void *data)
 	dev_set_drvdata(dev, txp);
 	vc4->txp = txp;
 
+	vc4_debugfs_add_regset32(drm, "txp_regs", &txp->regset);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vc4/vc4_v3d.c b/drivers/gpu/drm/vc4/vc4_v3d.c
index e47e29426078..a4b6859e3af6 100644
--- a/drivers/gpu/drm/vc4/vc4_v3d.c
+++ b/drivers/gpu/drm/vc4/vc4_v3d.c
@@ -22,129 +22,145 @@
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
-#ifdef CONFIG_DEBUG_FS
-#define REGDEF(reg) { reg, #reg }
-static const struct {
-	uint32_t reg;
-	const char *name;
-} vc4_reg_defs[] = {
-	REGDEF(V3D_IDENT0),
-	REGDEF(V3D_IDENT1),
-	REGDEF(V3D_IDENT2),
-	REGDEF(V3D_SCRATCH),
-	REGDEF(V3D_L2CACTL),
-	REGDEF(V3D_SLCACTL),
-	REGDEF(V3D_INTCTL),
-	REGDEF(V3D_INTENA),
-	REGDEF(V3D_INTDIS),
-	REGDEF(V3D_CT0CS),
-	REGDEF(V3D_CT1CS),
-	REGDEF(V3D_CT0EA),
-	REGDEF(V3D_CT1EA),
-	REGDEF(V3D_CT0CA),
-	REGDEF(V3D_CT1CA),
-	REGDEF(V3D_CT00RA0),
-	REGDEF(V3D_CT01RA0),
-	REGDEF(V3D_CT0LC),
-	REGDEF(V3D_CT1LC),
-	REGDEF(V3D_CT0PC),
-	REGDEF(V3D_CT1PC),
-	REGDEF(V3D_PCS),
-	REGDEF(V3D_BFC),
-	REGDEF(V3D_RFC),
-	REGDEF(V3D_BPCA),
-	REGDEF(V3D_BPCS),
-	REGDEF(V3D_BPOA),
-	REGDEF(V3D_BPOS),
-	REGDEF(V3D_BXCF),
-	REGDEF(V3D_SQRSV0),
-	REGDEF(V3D_SQRSV1),
-	REGDEF(V3D_SQCNTL),
-	REGDEF(V3D_SRQPC),
-	REGDEF(V3D_SRQUA),
-	REGDEF(V3D_SRQUL),
-	REGDEF(V3D_SRQCS),
-	REGDEF(V3D_VPACNTL),
-	REGDEF(V3D_VPMBASE),
-	REGDEF(V3D_PCTRC),
-	REGDEF(V3D_PCTRE),
-	REGDEF(V3D_PCTR(0)),
-	REGDEF(V3D_PCTRS(0)),
-	REGDEF(V3D_PCTR(1)),
-	REGDEF(V3D_PCTRS(1)),
-	REGDEF(V3D_PCTR(2)),
-	REGDEF(V3D_PCTRS(2)),
-	REGDEF(V3D_PCTR(3)),
-	REGDEF(V3D_PCTRS(3)),
-	REGDEF(V3D_PCTR(4)),
-	REGDEF(V3D_PCTRS(4)),
-	REGDEF(V3D_PCTR(5)),
-	REGDEF(V3D_PCTRS(5)),
-	REGDEF(V3D_PCTR(6)),
-	REGDEF(V3D_PCTRS(6)),
-	REGDEF(V3D_PCTR(7)),
-	REGDEF(V3D_PCTRS(7)),
-	REGDEF(V3D_PCTR(8)),
-	REGDEF(V3D_PCTRS(8)),
-	REGDEF(V3D_PCTR(9)),
-	REGDEF(V3D_PCTRS(9)),
-	REGDEF(V3D_PCTR(10)),
-	REGDEF(V3D_PCTRS(10)),
-	REGDEF(V3D_PCTR(11)),
-	REGDEF(V3D_PCTRS(11)),
-	REGDEF(V3D_PCTR(12)),
-	REGDEF(V3D_PCTRS(12)),
-	REGDEF(V3D_PCTR(13)),
-	REGDEF(V3D_PCTRS(13)),
-	REGDEF(V3D_PCTR(14)),
-	REGDEF(V3D_PCTRS(14)),
-	REGDEF(V3D_PCTR(15)),
-	REGDEF(V3D_PCTRS(15)),
-	REGDEF(V3D_DBGE),
-	REGDEF(V3D_FDBGO),
-	REGDEF(V3D_FDBGB),
-	REGDEF(V3D_FDBGR),
-	REGDEF(V3D_FDBGS),
-	REGDEF(V3D_ERRSTAT),
+static const struct debugfs_reg32 v3d_regs[] = {
+	VC4_REG32(V3D_IDENT0),
+	VC4_REG32(V3D_IDENT1),
+	VC4_REG32(V3D_IDENT2),
+	VC4_REG32(V3D_SCRATCH),
+	VC4_REG32(V3D_L2CACTL),
+	VC4_REG32(V3D_SLCACTL),
+	VC4_REG32(V3D_INTCTL),
+	VC4_REG32(V3D_INTENA),
+	VC4_REG32(V3D_INTDIS),
+	VC4_REG32(V3D_CT0CS),
+	VC4_REG32(V3D_CT1CS),
+	VC4_REG32(V3D_CT0EA),
+	VC4_REG32(V3D_CT1EA),
+	VC4_REG32(V3D_CT0CA),
+	VC4_REG32(V3D_CT1CA),
+	VC4_REG32(V3D_CT00RA0),
+	VC4_REG32(V3D_CT01RA0),
+	VC4_REG32(V3D_CT0LC),
+	VC4_REG32(V3D_CT1LC),
+	VC4_REG32(V3D_CT0PC),
+	VC4_REG32(V3D_CT1PC),
+	VC4_REG32(V3D_PCS),
+	VC4_REG32(V3D_BFC),
+	VC4_REG32(V3D_RFC),
+	VC4_REG32(V3D_BPCA),
+	VC4_REG32(V3D_BPCS),
+	VC4_REG32(V3D_BPOA),
+	VC4_REG32(V3D_BPOS),
+	VC4_REG32(V3D_BXCF),
+	VC4_REG32(V3D_SQRSV0),
+	VC4_REG32(V3D_SQRSV1),
+	VC4_REG32(V3D_SQCNTL),
+	VC4_REG32(V3D_SRQPC),
+	VC4_REG32(V3D_SRQUA),
+	VC4_REG32(V3D_SRQUL),
+	VC4_REG32(V3D_SRQCS),
+	VC4_REG32(V3D_VPACNTL),
+	VC4_REG32(V3D_VPMBASE),
+	VC4_REG32(V3D_PCTRC),
+	VC4_REG32(V3D_PCTRE),
+	VC4_REG32(V3D_PCTR(0)),
+	VC4_REG32(V3D_PCTRS(0)),
+	VC4_REG32(V3D_PCTR(1)),
+	VC4_REG32(V3D_PCTRS(1)),
+	VC4_REG32(V3D_PCTR(2)),
+	VC4_REG32(V3D_PCTRS(2)),
+	VC4_REG32(V3D_PCTR(3)),
+	VC4_REG32(V3D_PCTRS(3)),
+	VC4_REG32(V3D_PCTR(4)),
+	VC4_REG32(V3D_PCTRS(4)),
+	VC4_REG32(V3D_PCTR(5)),
+	VC4_REG32(V3D_PCTRS(5)),
+	VC4_REG32(V3D_PCTR(6)),
+	VC4_REG32(V3D_PCTRS(6)),
+	VC4_REG32(V3D_PCTR(7)),
+	VC4_REG32(V3D_PCTRS(7)),
+	VC4_REG32(V3D_PCTR(8)),
+	VC4_REG32(V3D_PCTRS(8)),
+	VC4_REG32(V3D_PCTR(9)),
+	VC4_REG32(V3D_PCTRS(9)),
+	VC4_REG32(V3D_PCTR(10)),
+	VC4_REG32(V3D_PCTRS(10)),
+	VC4_REG32(V3D_PCTR(11)),
+	VC4_REG32(V3D_PCTRS(11)),
+	VC4_REG32(V3D_PCTR(12)),
+	VC4_REG32(V3D_PCTRS(12)),
+	VC4_REG32(V3D_PCTR(13)),
+	VC4_REG32(V3D_PCTRS(13)),
+	VC4_REG32(V3D_PCTR(14)),
+	VC4_REG32(V3D_PCTRS(14)),
+	VC4_REG32(V3D_PCTR(15)),
+	VC4_REG32(V3D_PCTRS(15)),
+	VC4_REG32(V3D_DBGE),
+	VC4_REG32(V3D_FDBGO),
+	VC4_REG32(V3D_FDBGB),
+	VC4_REG32(V3D_FDBGR),
+	VC4_REG32(V3D_FDBGS),
+	VC4_REG32(V3D_ERRSTAT),
 };
 
-int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused)
+static int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *)m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(vc4_reg_defs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   vc4_reg_defs[i].name, vc4_reg_defs[i].reg,
-			   V3D_READ(vc4_reg_defs[i].reg));
+	int ret = vc4_v3d_pm_get(vc4);
+
+	if (ret == 0) {
+		uint32_t ident1 = V3D_READ(V3D_IDENT1);
+		uint32_t nslc = VC4_GET_FIELD(ident1, V3D_IDENT1_NSLC);
+		uint32_t tups = VC4_GET_FIELD(ident1, V3D_IDENT1_TUPS);
+		uint32_t qups = VC4_GET_FIELD(ident1, V3D_IDENT1_QUPS);
+
+		seq_printf(m, "Revision:   %d\n",
+			   VC4_GET_FIELD(ident1, V3D_IDENT1_REV));
+		seq_printf(m, "Slices:     %d\n", nslc);
+		seq_printf(m, "TMUs:       %d\n", nslc * tups);
+		seq_printf(m, "QPUs:       %d\n", nslc * qups);
+		seq_printf(m, "Semaphores: %d\n",
+			   VC4_GET_FIELD(ident1, V3D_IDENT1_NSEM));
+		vc4_v3d_pm_put(vc4);
 	}
 
 	return 0;
 }
 
-int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused)
+/**
+ * Wraps pm_runtime_get_sync() in a refcount, so that we can reliably
+ * get the pm_runtime refcount to 0 in vc4_reset().
+ */
+int
+vc4_v3d_pm_get(struct vc4_dev *vc4)
 {
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	uint32_t ident1 = V3D_READ(V3D_IDENT1);
-	uint32_t nslc = VC4_GET_FIELD(ident1, V3D_IDENT1_NSLC);
-	uint32_t tups = VC4_GET_FIELD(ident1, V3D_IDENT1_TUPS);
-	uint32_t qups = VC4_GET_FIELD(ident1, V3D_IDENT1_QUPS);
-
-	seq_printf(m, "Revision:   %d\n",
-		   VC4_GET_FIELD(ident1, V3D_IDENT1_REV));
-	seq_printf(m, "Slices:     %d\n", nslc);
-	seq_printf(m, "TMUs:       %d\n", nslc * tups);
-	seq_printf(m, "QPUs:       %d\n", nslc * qups);
-	seq_printf(m, "Semaphores: %d\n",
-		   VC4_GET_FIELD(ident1, V3D_IDENT1_NSEM));
+	mutex_lock(&vc4->power_lock);
+	if (vc4->power_refcount++ == 0) {
+		int ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
+
+		if (ret < 0) {
+			vc4->power_refcount--;
+			mutex_unlock(&vc4->power_lock);
+			return ret;
+		}
+	}
+	mutex_unlock(&vc4->power_lock);
 
 	return 0;
 }
-#endif /* CONFIG_DEBUG_FS */
+
+void
+vc4_v3d_pm_put(struct vc4_dev *vc4)
+{
+	mutex_lock(&vc4->power_lock);
+	if (--vc4->power_refcount == 0) {
+		pm_runtime_mark_last_busy(&vc4->v3d->pdev->dev);
+		pm_runtime_put_autosuspend(&vc4->v3d->pdev->dev);
+	}
+	mutex_unlock(&vc4->power_lock);
+}
 
 static void vc4_v3d_init_hw(struct drm_device *dev)
 {
@@ -354,6 +370,9 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 	v3d->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(v3d->regs))
 		return PTR_ERR(v3d->regs);
+	v3d->regset.base = v3d->regs;
+	v3d->regset.regs = v3d_regs;
+	v3d->regset.nregs = ARRAY_SIZE(v3d_regs);
 
 	vc4->v3d = v3d;
 	v3d->vc4 = vc4;
@@ -409,6 +428,9 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 	pm_runtime_set_autosuspend_delay(dev, 40); /* a little over 2 frames. */
 	pm_runtime_enable(dev);
 
+	vc4_debugfs_add_file(drm, "v3d_ident", vc4_v3d_debugfs_ident, NULL);
+	vc4_debugfs_add_regset32(drm, "v3d_regs", &v3d->regset);
+
 	return 0;
 }
 
@@ -452,7 +474,7 @@ static int vc4_v3d_dev_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id vc4_v3d_dt_match[] = {
+const struct of_device_id vc4_v3d_dt_match[] = {
 	{ .compatible = "brcm,bcm2835-v3d" },
 	{ .compatible = "brcm,cygnus-v3d" },
 	{ .compatible = "brcm,vc4-v3d" },
diff --git a/drivers/gpu/drm/vc4/vc4_vec.c b/drivers/gpu/drm/vc4/vc4_vec.c
index 858c3a483229..0a27e48fab31 100644
--- a/drivers/gpu/drm/vc4/vc4_vec.c
+++ b/drivers/gpu/drm/vc4/vc4_vec.c
@@ -176,6 +176,8 @@ struct vc4_vec {
 	struct clk *clock;
 
 	const struct vc4_vec_tv_mode *tv_mode;
+
+	struct debugfs_regset32 regset;
 };
 
 #define VEC_READ(offset) readl(vec->regs + (offset))
@@ -223,59 +225,33 @@ struct vc4_vec_tv_mode {
 	void (*mode_set)(struct vc4_vec *vec);
 };
 
-#define VEC_REG(reg) { reg, #reg }
-static const struct {
-	u32 reg;
-	const char *name;
-} vec_regs[] = {
-	VEC_REG(VEC_WSE_CONTROL),
-	VEC_REG(VEC_WSE_WSS_DATA),
-	VEC_REG(VEC_WSE_VPS_DATA1),
-	VEC_REG(VEC_WSE_VPS_CONTROL),
-	VEC_REG(VEC_REVID),
-	VEC_REG(VEC_CONFIG0),
-	VEC_REG(VEC_SCHPH),
-	VEC_REG(VEC_CLMP0_START),
-	VEC_REG(VEC_CLMP0_END),
-	VEC_REG(VEC_FREQ3_2),
-	VEC_REG(VEC_FREQ1_0),
-	VEC_REG(VEC_CONFIG1),
-	VEC_REG(VEC_CONFIG2),
-	VEC_REG(VEC_INTERRUPT_CONTROL),
-	VEC_REG(VEC_INTERRUPT_STATUS),
-	VEC_REG(VEC_FCW_SECAM_B),
-	VEC_REG(VEC_SECAM_GAIN_VAL),
-	VEC_REG(VEC_CONFIG3),
-	VEC_REG(VEC_STATUS0),
-	VEC_REG(VEC_MASK0),
-	VEC_REG(VEC_CFG),
-	VEC_REG(VEC_DAC_TEST),
-	VEC_REG(VEC_DAC_CONFIG),
-	VEC_REG(VEC_DAC_MISC),
+static const struct debugfs_reg32 vec_regs[] = {
+	VC4_REG32(VEC_WSE_CONTROL),
+	VC4_REG32(VEC_WSE_WSS_DATA),
+	VC4_REG32(VEC_WSE_VPS_DATA1),
+	VC4_REG32(VEC_WSE_VPS_CONTROL),
+	VC4_REG32(VEC_REVID),
+	VC4_REG32(VEC_CONFIG0),
+	VC4_REG32(VEC_SCHPH),
+	VC4_REG32(VEC_CLMP0_START),
+	VC4_REG32(VEC_CLMP0_END),
+	VC4_REG32(VEC_FREQ3_2),
+	VC4_REG32(VEC_FREQ1_0),
+	VC4_REG32(VEC_CONFIG1),
+	VC4_REG32(VEC_CONFIG2),
+	VC4_REG32(VEC_INTERRUPT_CONTROL),
+	VC4_REG32(VEC_INTERRUPT_STATUS),
+	VC4_REG32(VEC_FCW_SECAM_B),
+	VC4_REG32(VEC_SECAM_GAIN_VAL),
+	VC4_REG32(VEC_CONFIG3),
+	VC4_REG32(VEC_STATUS0),
+	VC4_REG32(VEC_MASK0),
+	VC4_REG32(VEC_CFG),
+	VC4_REG32(VEC_DAC_TEST),
+	VC4_REG32(VEC_DAC_CONFIG),
+	VC4_REG32(VEC_DAC_MISC),
 };
 
-#ifdef CONFIG_DEBUG_FS
-int vc4_vec_debugfs_regs(struct seq_file *m, void *unused)
-{
-	struct drm_info_node *node = (struct drm_info_node *)m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_vec *vec = vc4->vec;
-	int i;
-
-	if (!vec)
-		return 0;
-
-	for (i = 0; i < ARRAY_SIZE(vec_regs); i++) {
-		seq_printf(m, "%s (0x%04x): 0x%08x\n",
-			   vec_regs[i].name, vec_regs[i].reg,
-			   VEC_READ(vec_regs[i].reg));
-	}
-
-	return 0;
-}
-#endif
-
 static void vc4_vec_ntsc_mode_set(struct vc4_vec *vec)
 {
 	VEC_WRITE(VEC_CONFIG0, VEC_CONFIG0_NTSC_STD | VEC_CONFIG0_PDEN);
@@ -587,6 +563,9 @@ static int vc4_vec_bind(struct device *dev, struct device *master, void *data)
 	vec->regs = vc4_ioremap_regs(pdev, 0);
 	if (IS_ERR(vec->regs))
 		return PTR_ERR(vec->regs);
+	vec->regset.base = vec->regs;
+	vec->regset.regs = vec_regs;
+	vec->regset.nregs = ARRAY_SIZE(vec_regs);
 
 	vec->clock = devm_clk_get(dev, NULL);
 	if (IS_ERR(vec->clock)) {
@@ -612,6 +591,8 @@ static int vc4_vec_bind(struct device *dev, struct device *master, void *data)
 
 	vc4->vec = vec;
 
+	vc4_debugfs_add_regset32(drm, "vec_regs", &vec->regset);
+
 	return 0;
 
 err_destroy_encoder:
author	Linus Torvalds <torvalds@linux-foundation.org>	2019-05-08 21:35:19 -0700
committer	Linus Torvalds <torvalds@linux-foundation.org>	2019-05-08 21:35:19 -0700
commit	a2d635decbfa9c1e4ae15cb05b68b2559f7f827c (patch)
tree	1c3766c35215450ff9e4228efed578d5e6ba65d1 /drivers/gpu/drm/vc4
parent	89c3b37af87ec183b666d83428cb28cc421671a6 (diff)
parent	eb85d03e01c3e9f3b0ba7282b2e3515a635decb2 (diff)
download	linux-a2d635decbfa9c1e4ae15cb05b68b2559f7f827c.tar.gz