Merge tag 'drm-forgot-about-tegra-for-v4.12-rc1' of git://people.freedesktop.org/~airlied/linux

Pull drm tegra updates from Dave Airlie:
 "I missed a pull request from Thierry, this stuff has been in
  linux-next for a while anyways.

  It does contain a branch from the iommu tree, but Thierry said it
  should be fine"

* tag 'drm-forgot-about-tegra-for-v4.12-rc1' of git://people.freedesktop.org/~airlied/linux:
  gpu: host1x: Fix host1x driver shutdown
  gpu: host1x: Support module reset
  gpu: host1x: Sort includes alphabetically
  drm/tegra: Add VIC support
  dt-bindings: Add bindings for the Tegra VIC
  drm/tegra: Add falcon helper library
  drm/tegra: Add Tegra DRM allocation API
  drm/tegra: Add tiling FB modifiers
  drm/tegra: Don't leak kernel pointer to userspace
  drm/tegra: Protect IOMMU operations by mutex
  drm/tegra: Enable IOVA API when IOMMU support is enabled
  gpu: host1x: Add IOMMU support
  gpu: host1x: Fix potential out-of-bounds access
  iommu/iova: Fix compile error with CONFIG_IOMMU_IOVA=m
  iommu: Add dummy implementations for !IOMMU_IOVA
  MAINTAINERS: Add related headers to IOMMU section
  iommu/iova: Consolidate code for adding new node to iovad domain rbtree

25 files changed, 1517 insertions, 202 deletions

diff --git a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.txt b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.txt
index 0fad7ed2ea19..74e1e8add5a1 100644
--- a/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.txt
+++ b/Documentation/devicetree/bindings/display/tegra/nvidia,tegra20-host1x.txt
@@ -249,6 +249,19 @@ of the following host1x client modules:
   See ../pinctrl/nvidia,tegra124-dpaux-padctl.txt for information
   regarding the DPAUX pad controller bindings.
 
+- vic: Video Image Compositor
+  - compatible : "nvidia,tegra<chip>-vic"
+  - reg: Physical base address and length of the controller's registers.
+  - interrupts: The interrupt outputs from the controller.
+  - clocks: Must contain an entry for each entry in clock-names.
+    See ../clocks/clock-bindings.txt for details.
+  - clock-names: Must include the following entries:
+    - vic: clock input for the VIC hardware
+  - resets: Must contain an entry for each entry in reset-names.
+    See ../reset/reset.txt for details.
+  - reset-names: Must include the following entries:
+    - vic
+
 Example:
 
 / {
diff --git a/MAINTAINERS b/MAINTAINERS
index ac0ce262150c..8944b472b90f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6843,6 +6843,8 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu.git
 S:	Maintained
 F:	Documentation/devicetree/bindings/iommu/
 F:	drivers/iommu/
+F:	include/linux/iommu.h
+F:	include/linux/iova.h
 
 IP MASQUERADING
 M:	Juanjo Ciarlante <jjciarla@raiz.uncu.edu.ar>
diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig
index bbf5a4b7e0b6..2db29d67193d 100644
--- a/drivers/gpu/drm/tegra/Kconfig
+++ b/drivers/gpu/drm/tegra/Kconfig
@@ -7,6 +7,7 @@ config DRM_TEGRA
 	select DRM_MIPI_DSI
 	select DRM_PANEL
 	select TEGRA_HOST1X
+	select IOMMU_IOVA if IOMMU_SUPPORT
 	help
 	  Choose this option if you have an NVIDIA Tegra SoC.
 
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
index 2c66a8db9da4..6af3a9ad6565 100644
--- a/drivers/gpu/drm/tegra/Makefile
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -13,6 +13,8 @@ tegra-drm-y := \
 	sor.o \
 	dpaux.o \
 	gr2d.o \
-	gr3d.o
+	gr3d.o \
+	falcon.o \
+	vic.o
 
 obj-$(CONFIG_DRM_TEGRA) += tegra-drm.o
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index dba4e090d3df..9a1e34e48f64 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -1,13 +1,15 @@
 /*
  * Copyright (C) 2012 Avionic Design GmbH
- * Copyright (C) 2012-2013 NVIDIA CORPORATION.  All rights reserved.
+ * Copyright (C) 2012-2016 NVIDIA CORPORATION.  All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
 
+#include <linux/bitops.h>
 #include <linux/host1x.h>
+#include <linux/idr.h>
 #include <linux/iommu.h>
 
 #include <drm/drm_atomic.h>
@@ -23,8 +25,11 @@
 #define DRIVER_MINOR 0
 #define DRIVER_PATCHLEVEL 0
 
+#define CARVEOUT_SZ SZ_64M
+
 struct tegra_drm_file {
-	struct list_head contexts;
+	struct idr contexts;
+	struct mutex lock;
 };
 
 static void tegra_atomic_schedule(struct tegra_drm *tegra,
@@ -126,8 +131,9 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 		return -ENOMEM;
 
 	if (iommu_present(&platform_bus_type)) {
+		u64 carveout_start, carveout_end, gem_start, gem_end;
 		struct iommu_domain_geometry *geometry;
-		u64 start, end;
+		unsigned long order;
 
 		tegra->domain = iommu_domain_alloc(&platform_bus_type);
 		if (!tegra->domain) {
@@ -136,12 +142,26 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 		}
 
 		geometry = &tegra->domain->geometry;
-		start = geometry->aperture_start;
-		end = geometry->aperture_end;
-
-		DRM_DEBUG_DRIVER("IOMMU aperture initialized (%#llx-%#llx)\n",
-				 start, end);
-		drm_mm_init(&tegra->mm, start, end - start + 1);
+		gem_start = geometry->aperture_start;
+		gem_end = geometry->aperture_end - CARVEOUT_SZ;
+		carveout_start = gem_end + 1;
+		carveout_end = geometry->aperture_end;
+
+		order = __ffs(tegra->domain->pgsize_bitmap);
+		init_iova_domain(&tegra->carveout.domain, 1UL << order,
+				 carveout_start >> order,
+				 carveout_end >> order);
+
+		tegra->carveout.shift = iova_shift(&tegra->carveout.domain);
+		tegra->carveout.limit = carveout_end >> tegra->carveout.shift;
+
+		drm_mm_init(&tegra->mm, gem_start, gem_end - gem_start + 1);
+		mutex_init(&tegra->mm_lock);
+
+		DRM_DEBUG("IOMMU apertures:\n");
+		DRM_DEBUG("  GEM: %#llx-%#llx\n", gem_start, gem_end);
+		DRM_DEBUG("  Carveout: %#llx-%#llx\n", carveout_start,
+			  carveout_end);
 	}
 
 	mutex_init(&tegra->clients_lock);
@@ -161,6 +181,8 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 	drm->mode_config.max_width = 4096;
 	drm->mode_config.max_height = 4096;
 
+	drm->mode_config.allow_fb_modifiers = true;
+
 	drm->mode_config.funcs = &tegra_drm_mode_funcs;
 
 	err = tegra_drm_fb_prepare(drm);
@@ -208,6 +230,8 @@ config:
 	if (tegra->domain) {
 		iommu_domain_free(tegra->domain);
 		drm_mm_takedown(&tegra->mm);
+		mutex_destroy(&tegra->mm_lock);
+		put_iova_domain(&tegra->carveout.domain);
 	}
 free:
 	kfree(tegra);
@@ -232,6 +256,8 @@ static void tegra_drm_unload(struct drm_device *drm)
 	if (tegra->domain) {
 		iommu_domain_free(tegra->domain);
 		drm_mm_takedown(&tegra->mm);
+		mutex_destroy(&tegra->mm_lock);
+		put_iova_domain(&tegra->carveout.domain);
 	}
 
 	kfree(tegra);
@@ -245,7 +271,8 @@ static int tegra_drm_open(struct drm_device *drm, struct drm_file *filp)
 	if (!fpriv)
 		return -ENOMEM;
 
-	INIT_LIST_HEAD(&fpriv->contexts);
+	idr_init(&fpriv->contexts);
+	mutex_init(&fpriv->lock);
 	filp->driver_priv = fpriv;
 
 	return 0;
@@ -424,21 +451,16 @@ fail:
 
 
 #ifdef CONFIG_DRM_TEGRA_STAGING
-static struct tegra_drm_context *tegra_drm_get_context(__u64 context)
-{
-	return (struct tegra_drm_context *)(uintptr_t)context;
-}
-
-static bool tegra_drm_file_owns_context(struct tegra_drm_file *file,
-					struct tegra_drm_context *context)
+static struct tegra_drm_context *
+tegra_drm_file_get_context(struct tegra_drm_file *file, u32 id)
 {
-	struct tegra_drm_context *ctx;
+	struct tegra_drm_context *context;
 
-	list_for_each_entry(ctx, &file->contexts, list)
-		if (ctx == context)
-			return true;
+	mutex_lock(&file->lock);
+	context = idr_find(&file->contexts, id);
+	mutex_unlock(&file->lock);
 
-	return false;
+	return context;
 }
 
 static int tegra_gem_create(struct drm_device *drm, void *data,
@@ -519,6 +541,28 @@ static int tegra_syncpt_wait(struct drm_device *drm, void *data,
 				  &args->value);
 }
 
+static int tegra_client_open(struct tegra_drm_file *fpriv,
+			     struct tegra_drm_client *client,
+			     struct tegra_drm_context *context)
+{
+	int err;
+
+	err = client->ops->open_channel(client, context);
+	if (err < 0)
+		return err;
+
+	err = idr_alloc(&fpriv->contexts, context, 0, 0, GFP_KERNEL);
+	if (err < 0) {
+		client->ops->close_channel(context);
+		return err;
+	}
+
+	context->client = client;
+	context->id = err;
+
+	return 0;
+}
+
 static int tegra_open_channel(struct drm_device *drm, void *data,
 			      struct drm_file *file)
 {
@@ -533,19 +577,22 @@ static int tegra_open_channel(struct drm_device *drm, void *data,
 	if (!context)
 		return -ENOMEM;
 
+	mutex_lock(&fpriv->lock);
+
 	list_for_each_entry(client, &tegra->clients, list)
 		if (client->base.class == args->client) {
-			err = client->ops->open_channel(client, context);
-			if (err)
+			err = tegra_client_open(fpriv, client, context);
+			if (err < 0)
 				break;
 
-			list_add(&context->list, &fpriv->contexts);
-			args->context = (uintptr_t)context;
-			context->client = client;
-			return 0;
+			args->context = context->id;
+			break;
 		}
 
-	kfree(context);
+	if (err < 0)
+		kfree(context);
+
+	mutex_unlock(&fpriv->lock);
 	return err;
 }
 
@@ -555,16 +602,22 @@ static int tegra_close_channel(struct drm_device *drm, void *data,
 	struct tegra_drm_file *fpriv = file->driver_priv;
 	struct drm_tegra_close_channel *args = data;
 	struct tegra_drm_context *context;
+	int err = 0;
 
-	context = tegra_drm_get_context(args->context);
+	mutex_lock(&fpriv->lock);
 
-	if (!tegra_drm_file_owns_context(fpriv, context))
-		return -EINVAL;
+	context = tegra_drm_file_get_context(fpriv, args->context);
+	if (!context) {
+		err = -EINVAL;
+		goto unlock;
+	}
 
-	list_del(&context->list);
+	idr_remove(&fpriv->contexts, context->id);
 	tegra_drm_context_free(context);
 
-	return 0;
+unlock:
+	mutex_unlock(&fpriv->lock);
+	return err;
 }
 
 static int tegra_get_syncpt(struct drm_device *drm, void *data,
@@ -574,19 +627,27 @@ static int tegra_get_syncpt(struct drm_device *drm, void *data,
 	struct drm_tegra_get_syncpt *args = data;
 	struct tegra_drm_context *context;
 	struct host1x_syncpt *syncpt;
+	int err = 0;
 
-	context = tegra_drm_get_context(args->context);
+	mutex_lock(&fpriv->lock);
 
-	if (!tegra_drm_file_owns_context(fpriv, context))
-		return -ENODEV;
+	context = tegra_drm_file_get_context(fpriv, args->context);
+	if (!context) {
+		err = -ENODEV;
+		goto unlock;
+	}
 
-	if (args->index >= context->client->base.num_syncpts)
-		return -EINVAL;
+	if (args->index >= context->client->base.num_syncpts) {
+		err = -EINVAL;
+		goto unlock;
+	}
 
 	syncpt = context->client->base.syncpts[args->index];
 	args->id = host1x_syncpt_id(syncpt);
 
-	return 0;
+unlock:
+	mutex_unlock(&fpriv->lock);
+	return err;
 }
 
 static int tegra_submit(struct drm_device *drm, void *data,
@@ -595,13 +656,21 @@ static int tegra_submit(struct drm_device *drm, void *data,
 	struct tegra_drm_file *fpriv = file->driver_priv;
 	struct drm_tegra_submit *args = data;
 	struct tegra_drm_context *context;
+	int err;
 
-	context = tegra_drm_get_context(args->context);
+	mutex_lock(&fpriv->lock);
+
+	context = tegra_drm_file_get_context(fpriv, args->context);
+	if (!context) {
+		err = -ENODEV;
+		goto unlock;
+	}
 
-	if (!tegra_drm_file_owns_context(fpriv, context))
-		return -ENODEV;
+	err = context->client->ops->submit(context, args, drm, file);
 
-	return context->client->ops->submit(context, args, drm, file);
+unlock:
+	mutex_unlock(&fpriv->lock);
+	return err;
 }
 
 static int tegra_get_syncpt_base(struct drm_device *drm, void *data,
@@ -612,24 +681,34 @@ static int tegra_get_syncpt_base(struct drm_device *drm, void *data,
 	struct tegra_drm_context *context;
 	struct host1x_syncpt_base *base;
 	struct host1x_syncpt *syncpt;
+	int err = 0;
 
-	context = tegra_drm_get_context(args->context);
+	mutex_lock(&fpriv->lock);
 
-	if (!tegra_drm_file_owns_context(fpriv, context))
-		return -ENODEV;
+	context = tegra_drm_file_get_context(fpriv, args->context);
+	if (!context) {
+		err = -ENODEV;
+		goto unlock;
+	}
 
-	if (args->syncpt >= context->client->base.num_syncpts)
-		return -EINVAL;
+	if (args->syncpt >= context->client->base.num_syncpts) {
+		err = -EINVAL;
+		goto unlock;
+	}
 
 	syncpt = context->client->base.syncpts[args->syncpt];
 
 	base = host1x_syncpt_get_base(syncpt);
-	if (!base)
-		return -ENXIO;
+	if (!base) {
+		err = -ENXIO;
+		goto unlock;
+	}
 
 	args->id = host1x_syncpt_base_id(base);
 
-	return 0;
+unlock:
+	mutex_unlock(&fpriv->lock);
+	return err;
 }
 
 static int tegra_gem_set_tiling(struct drm_device *drm, void *data,
@@ -804,14 +883,25 @@ static const struct file_operations tegra_drm_fops = {
 	.llseek = noop_llseek,
 };
 
+static int tegra_drm_context_cleanup(int id, void *p, void *data)
+{
+	struct tegra_drm_context *context = p;
+
+	tegra_drm_context_free(context);
+
+	return 0;
+}
+
 static void tegra_drm_preclose(struct drm_device *drm, struct drm_file *file)
 {
 	struct tegra_drm_file *fpriv = file->driver_priv;
-	struct tegra_drm_context *context, *tmp;
 
-	list_for_each_entry_safe(context, tmp, &fpriv->contexts, list)
-		tegra_drm_context_free(context);
+	mutex_lock(&fpriv->lock);
+	idr_for_each(&fpriv->contexts, tegra_drm_context_cleanup, NULL);
+	mutex_unlock(&fpriv->lock);
 
+	idr_destroy(&fpriv->contexts);
+	mutex_destroy(&fpriv->lock);
 	kfree(fpriv);
 }
 
@@ -844,7 +934,9 @@ static int tegra_debugfs_iova(struct seq_file *s, void *data)
 	struct tegra_drm *tegra = drm->dev_private;
 	struct drm_printer p = drm_seq_file_printer(s);
 
+	mutex_lock(&tegra->mm_lock);
 	drm_mm_print(&tegra->mm, &p);
+	mutex_unlock(&tegra->mm_lock);
 
 	return 0;
 }
@@ -919,6 +1011,84 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
 	return 0;
 }
 
+void *tegra_drm_alloc(struct tegra_drm *tegra, size_t size,
+			      dma_addr_t *dma)
+{
+	struct iova *alloc;
+	void *virt;
+	gfp_t gfp;
+	int err;
+
+	if (tegra->domain)
+		size = iova_align(&tegra->carveout.domain, size);
+	else
+		size = PAGE_ALIGN(size);
+
+	gfp = GFP_KERNEL | __GFP_ZERO;
+	if (!tegra->domain) {
+		/*
+		 * Many units only support 32-bit addresses, even on 64-bit
+		 * SoCs. If there is no IOMMU to translate into a 32-bit IO
+		 * virtual address space, force allocations to be in the
+		 * lower 32-bit range.
+		 */
+		gfp |= GFP_DMA;
+	}
+
+	virt = (void *)__get_free_pages(gfp, get_order(size));
+	if (!virt)
+		return ERR_PTR(-ENOMEM);
+
+	if (!tegra->domain) {
+		/*
+		 * If IOMMU is disabled, devices address physical memory
+		 * directly.
+		 */
+		*dma = virt_to_phys(virt);
+		return virt;
+	}
+
+	alloc = alloc_iova(&tegra->carveout.domain,
+			   size >> tegra->carveout.shift,
+			   tegra->carveout.limit, true);
+	if (!alloc) {
+		err = -EBUSY;
+		goto free_pages;
+	}
+
+	*dma = iova_dma_addr(&tegra->carveout.domain, alloc);
+	err = iommu_map(tegra->domain, *dma, virt_to_phys(virt),
+			size, IOMMU_READ | IOMMU_WRITE);
+	if (err < 0)
+		goto free_iova;
+
+	return virt;
+
+free_iova:
+	__free_iova(&tegra->carveout.domain, alloc);
+free_pages:
+	free_pages((unsigned long)virt, get_order(size));
+
+	return ERR_PTR(err);
+}
+
+void tegra_drm_free(struct tegra_drm *tegra, size_t size, void *virt,
+		    dma_addr_t dma)
+{
+	if (tegra->domain)
+		size = iova_align(&tegra->carveout.domain, size);
+	else
+		size = PAGE_ALIGN(size);
+
+	if (tegra->domain) {
+		iommu_unmap(tegra->domain, dma, size);
+		free_iova(&tegra->carveout.domain,
+			  iova_pfn(&tegra->carveout.domain, dma));
+	}
+
+	free_pages((unsigned long)virt, get_order(size));
+}
+
 static int host1x_drm_probe(struct host1x_device *dev)
 {
 	struct drm_driver *driver = &tegra_drm_driver;
@@ -1003,11 +1173,13 @@ static const struct of_device_id host1x_drm_subdevs[] = {
 	{ .compatible = "nvidia,tegra124-sor", },
 	{ .compatible = "nvidia,tegra124-hdmi", },
 	{ .compatible = "nvidia,tegra124-dsi", },
+	{ .compatible = "nvidia,tegra124-vic", },
 	{ .compatible = "nvidia,tegra132-dsi", },
 	{ .compatible = "nvidia,tegra210-dc", },
 	{ .compatible = "nvidia,tegra210-dsi", },
 	{ .compatible = "nvidia,tegra210-sor", },
 	{ .compatible = "nvidia,tegra210-sor1", },
+	{ .compatible = "nvidia,tegra210-vic", },
 	{ /* sentinel */ }
 };
 
@@ -1029,6 +1201,7 @@ static struct platform_driver * const drivers[] = {
 	&tegra_sor_driver,
 	&tegra_gr2d_driver,
 	&tegra_gr3d_driver,
+	&tegra_vic_driver,
 };
 
 static int __init host1x_drm_init(void)
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 5747accb2271..85aa2e3d9d4e 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -12,6 +12,7 @@
 
 #include <uapi/drm/tegra_drm.h>
 #include <linux/host1x.h>
+#include <linux/iova.h>
 #include <linux/of_gpio.h>
 
 #include <drm/drmP.h>
@@ -42,8 +43,15 @@ struct tegra_drm {
 	struct drm_device *drm;
 
 	struct iommu_domain *domain;
+	struct mutex mm_lock;
 	struct drm_mm mm;
 
+	struct {
+		struct iova_domain domain;
+		unsigned long shift;
+		unsigned long limit;
+	} carveout;
+
 	struct mutex clients_lock;
 	struct list_head clients;
 
@@ -67,7 +75,7 @@ struct tegra_drm_client;
 struct tegra_drm_context {
 	struct tegra_drm_client *client;
 	struct host1x_channel *channel;
-	struct list_head list;
+	unsigned int id;
 };
 
 struct tegra_drm_client_ops {
@@ -105,6 +113,10 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
 int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
 int tegra_drm_exit(struct tegra_drm *tegra);
 
+void *tegra_drm_alloc(struct tegra_drm *tegra, size_t size, dma_addr_t *iova);
+void tegra_drm_free(struct tegra_drm *tegra, size_t size, void *virt,
+		    dma_addr_t iova);
+
 struct tegra_dc_soc_info;
 struct tegra_output;
 
@@ -283,5 +295,6 @@ extern struct platform_driver tegra_dpaux_driver;
 extern struct platform_driver tegra_sor_driver;
 extern struct platform_driver tegra_gr2d_driver;
 extern struct platform_driver tegra_gr3d_driver;
+extern struct platform_driver tegra_vic_driver;
 
 #endif /* HOST1X_DRM_H */
diff --git a/drivers/gpu/drm/tegra/falcon.c b/drivers/gpu/drm/tegra/falcon.c
new file mode 100644
index 000000000000..f685e72949d1
--- /dev/null
+++ b/drivers/gpu/drm/tegra/falcon.c
@@ -0,0 +1,259 @@
+/*
+ * Copyright (c) 2015, NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/firmware.h>
+#include <linux/pci_ids.h>
+#include <linux/iopoll.h>
+
+#include "falcon.h"
+#include "drm.h"
+
+enum falcon_memory {
+	FALCON_MEMORY_IMEM,
+	FALCON_MEMORY_DATA,
+};
+
+static void falcon_writel(struct falcon *falcon, u32 value, u32 offset)
+{
+	writel(value, falcon->regs + offset);
+}
+
+int falcon_wait_idle(struct falcon *falcon)
+{
+	u32 value;
+
+	return readl_poll_timeout(falcon->regs + FALCON_IDLESTATE, value,
+				  (value == 0), 10, 100000);
+}
+
+static int falcon_dma_wait_idle(struct falcon *falcon)
+{
+	u32 value;
+
+	return readl_poll_timeout(falcon->regs + FALCON_DMATRFCMD, value,
+				  (value & FALCON_DMATRFCMD_IDLE), 10, 100000);
+}
+
+static int falcon_copy_chunk(struct falcon *falcon,
+			     phys_addr_t base,
+			     unsigned long offset,
+			     enum falcon_memory target)
+{
+	u32 cmd = FALCON_DMATRFCMD_SIZE_256B;
+
+	if (target == FALCON_MEMORY_IMEM)
+		cmd |= FALCON_DMATRFCMD_IMEM;
+
+	falcon_writel(falcon, offset, FALCON_DMATRFMOFFS);
+	falcon_writel(falcon, base, FALCON_DMATRFFBOFFS);
+	falcon_writel(falcon, cmd, FALCON_DMATRFCMD);
+
+	return falcon_dma_wait_idle(falcon);
+}
+
+static void falcon_copy_firmware_image(struct falcon *falcon,
+				       const struct firmware *firmware)
+{
+	u32 *firmware_vaddr = falcon->firmware.vaddr;
+	dma_addr_t daddr;
+	size_t i;
+	int err;
+
+	/* copy the whole thing taking into account endianness */
+	for (i = 0; i < firmware->size / sizeof(u32); i++)
+		firmware_vaddr[i] = le32_to_cpu(((u32 *)firmware->data)[i]);
+
+	/* ensure that caches are flushed and falcon can see the firmware */
+	daddr = dma_map_single(falcon->dev, firmware_vaddr,
+			       falcon->firmware.size, DMA_TO_DEVICE);
+	err = dma_mapping_error(falcon->dev, daddr);
+	if (err) {
+		dev_err(falcon->dev, "failed to map firmware: %d\n", err);
+		return;
+	}
+	dma_sync_single_for_device(falcon->dev, daddr,
+				   falcon->firmware.size, DMA_TO_DEVICE);
+	dma_unmap_single(falcon->dev, daddr, falcon->firmware.size,
+			 DMA_TO_DEVICE);
+}
+
+static int falcon_parse_firmware_image(struct falcon *falcon)
+{
+	struct falcon_fw_bin_header_v1 *bin = (void *)falcon->firmware.vaddr;
+	struct falcon_fw_os_header_v1 *os;
+
+	/* endian problems would show up right here */
+	if (bin->magic != PCI_VENDOR_ID_NVIDIA) {
+		dev_err(falcon->dev, "incorrect firmware magic\n");
+		return -EINVAL;
+	}
+
+	/* currently only version 1 is supported */
+	if (bin->version != 1) {
+		dev_err(falcon->dev, "unsupported firmware version\n");
+		return -EINVAL;
+	}
+
+	/* check that the firmware size is consistent */
+	if (bin->size > falcon->firmware.size) {
+		dev_err(falcon->dev, "firmware image size inconsistency\n");
+		return -EINVAL;
+	}
+
+	os = falcon->firmware.vaddr + bin->os_header_offset;
+
+	falcon->firmware.bin_data.size = bin->os_size;
+	falcon->firmware.bin_data.offset = bin->os_data_offset;
+	falcon->firmware.code.offset = os->code_offset;
+	falcon->firmware.code.size = os->code_size;
+	falcon->firmware.data.offset = os->data_offset;
+	falcon->firmware.data.size = os->data_size;
+
+	return 0;
+}
+
+int falcon_read_firmware(struct falcon *falcon, const char *name)
+{
+	int err;
+
+	/* request_firmware prints error if it fails */
+	err = request_firmware(&falcon->firmware.firmware, name, falcon->dev);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+int falcon_load_firmware(struct falcon *falcon)
+{
+	const struct firmware *firmware = falcon->firmware.firmware;
+	int err;
+
+	falcon->firmware.size = firmware->size;
+
+	/* allocate iova space for the firmware */
+	falcon->firmware.vaddr = falcon->ops->alloc(falcon, firmware->size,
+						    &falcon->firmware.paddr);
+	if (!falcon->firmware.vaddr) {
+		dev_err(falcon->dev, "dma memory mapping failed\n");
+		return -ENOMEM;
+	}
+
+	/* copy firmware image into local area. this also ensures endianness */
+	falcon_copy_firmware_image(falcon, firmware);
+
+	/* parse the image data */
+	err = falcon_parse_firmware_image(falcon);
+	if (err < 0) {
+		dev_err(falcon->dev, "failed to parse firmware image\n");
+		goto err_setup_firmware_image;
+	}
+
+	release_firmware(firmware);
+	falcon->firmware.firmware = NULL;
+
+	return 0;
+
+err_setup_firmware_image:
+	falcon->ops->free(falcon, falcon->firmware.size,
+			  falcon->firmware.paddr, falcon->firmware.vaddr);
+
+	return err;
+}
+
+int falcon_init(struct falcon *falcon)
+{
+	/* check mandatory ops */
+	if (!falcon->ops || !falcon->ops->alloc || !falcon->ops->free)
+		return -EINVAL;
+
+	falcon->firmware.vaddr = NULL;
+
+	return 0;
+}
+
+void falcon_exit(struct falcon *falcon)
+{
+	if (falcon->firmware.firmware) {
+		release_firmware(falcon->firmware.firmware);
+		falcon->firmware.firmware = NULL;
+	}
+
+	if (falcon->firmware.vaddr) {
+		falcon->ops->free(falcon, falcon->firmware.size,
+				  falcon->firmware.paddr,
+				  falcon->firmware.vaddr);
+		falcon->firmware.vaddr = NULL;
+	}
+}
+
+int falcon_boot(struct falcon *falcon)
+{
+	unsigned long offset;
+	int err;
+
+	if (!falcon->firmware.vaddr)
+		return -EINVAL;
+
+	falcon_writel(falcon, 0, FALCON_DMACTL);
+
+	/* setup the address of the binary data so Falcon can access it later */
+	falcon_writel(falcon, (falcon->firmware.paddr +
+			       falcon->firmware.bin_data.offset) >> 8,
+		      FALCON_DMATRFBASE);
+
+	/* copy the data segment into Falcon internal memory */
+	for (offset = 0; offset < falcon->firmware.data.size; offset += 256)
+		falcon_copy_chunk(falcon,
+				  falcon->firmware.data.offset + offset,
+				  offset, FALCON_MEMORY_DATA);
+
+	/* copy the first code segment into Falcon internal memory */
+	falcon_copy_chunk(falcon, falcon->firmware.code.offset,
+			  0, FALCON_MEMORY_IMEM);
+
+	/* setup falcon interrupts */
+	falcon_writel(falcon, FALCON_IRQMSET_EXT(0xff) |
+			      FALCON_IRQMSET_SWGEN1 |
+			      FALCON_IRQMSET_SWGEN0 |
+			      FALCON_IRQMSET_EXTERR |
+			      FALCON_IRQMSET_HALT |
+			      FALCON_IRQMSET_WDTMR,
+		      FALCON_IRQMSET);
+	falcon_writel(falcon, FALCON_IRQDEST_EXT(0xff) |
+			      FALCON_IRQDEST_SWGEN1 |
+			      FALCON_IRQDEST_SWGEN0 |
+			      FALCON_IRQDEST_EXTERR |
+			      FALCON_IRQDEST_HALT,
+		      FALCON_IRQDEST);
+
+	/* enable interface */
+	falcon_writel(falcon, FALCON_ITFEN_MTHDEN |
+			      FALCON_ITFEN_CTXEN,
+		      FALCON_ITFEN);
+
+	/* boot falcon */
+	falcon_writel(falcon, 0x00000000, FALCON_BOOTVEC);
+	falcon_writel(falcon, FALCON_CPUCTL_STARTCPU, FALCON_CPUCTL);
+
+	err = falcon_wait_idle(falcon);
+	if (err < 0) {
+		dev_err(falcon->dev, "Falcon boot failed due to timeout\n");
+		return err;
+	}
+
+	return 0;
+}
+
+void falcon_execute_method(struct falcon *falcon, u32 method, u32 data)
+{
+	falcon_writel(falcon, method >> 2, FALCON_UCLASS_METHOD_OFFSET);
+	falcon_writel(falcon, data, FALCON_UCLASS_METHOD_DATA);
+}
diff --git a/drivers/gpu/drm/tegra/falcon.h b/drivers/gpu/drm/tegra/falcon.h
new file mode 100644
index 000000000000..4504ed5a199e
--- /dev/null
+++ b/drivers/gpu/drm/tegra/falcon.h
@@ -0,0 +1,127 @@
+/*
+ * Copyright (c) 2015, NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _FALCON_H_
+#define _FALCON_H_
+
+#include <linux/types.h>
+
+#define FALCON_UCLASS_METHOD_OFFSET		0x00000040
+
+#define FALCON_UCLASS_METHOD_DATA		0x00000044
+
+#define FALCON_IRQMSET				0x00001010
+#define FALCON_IRQMSET_WDTMR			(1 << 1)
+#define FALCON_IRQMSET_HALT			(1 << 4)
+#define FALCON_IRQMSET_EXTERR			(1 << 5)
+#define FALCON_IRQMSET_SWGEN0			(1 << 6)
+#define FALCON_IRQMSET_SWGEN1			(1 << 7)
+#define FALCON_IRQMSET_EXT(v)			(((v) & 0xff) << 8)
+
+#define FALCON_IRQDEST				0x0000101c
+#define FALCON_IRQDEST_HALT			(1 << 4)
+#define FALCON_IRQDEST_EXTERR			(1 << 5)
+#define FALCON_IRQDEST_SWGEN0			(1 << 6)
+#define FALCON_IRQDEST_SWGEN1			(1 << 7)
+#define FALCON_IRQDEST_EXT(v)			(((v) & 0xff) << 8)
+
+#define FALCON_ITFEN				0x00001048
+#define FALCON_ITFEN_CTXEN			(1 << 0)
+#define FALCON_ITFEN_MTHDEN			(1 << 1)
+
+#define FALCON_IDLESTATE			0x0000104c
+
+#define FALCON_CPUCTL				0x00001100
+#define FALCON_CPUCTL_STARTCPU			(1 << 1)
+
+#define FALCON_BOOTVEC				0x00001104
+
+#define FALCON_DMACTL				0x0000110c
+#define FALCON_DMACTL_DMEM_SCRUBBING		(1 << 1)
+#define FALCON_DMACTL_IMEM_SCRUBBING		(1 << 2)
+
+#define FALCON_DMATRFBASE			0x00001110
+
+#define FALCON_DMATRFMOFFS			0x00001114
+
+#define FALCON_DMATRFCMD			0x00001118
+#define FALCO