drm/tegra: Add support for tiled buffer objects

The gr2d and gr3d engines work more efficiently on buffers with a tiled
memory layout. Allow created buffers to be marked as tiled so that the
display controller can scan them out properly.

Signed-off-by: Thierry Reding <thierry.reding@avionic-design.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index 054ca1b..c51aaf7 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -47,6 +47,7 @@
 	window.dst.h = crtc_h;
 	window.format = tegra_dc_format(fb->pixel_format);
 	window.bits_per_pixel = fb->bits_per_pixel;
+	window.tiled = tegra_fb_is_tiled(fb);
 
 	for (i = 0; i < drm_format_num_planes(fb->pixel_format); i++) {
 		struct tegra_bo *bo = tegra_fb_get_plane(fb, i);
@@ -157,6 +158,16 @@
 	tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE);
 	tegra_dc_writel(dc, format, DC_WIN_COLOR_DEPTH);
 
+	if (tegra_fb_is_tiled(fb)) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
 	value = GENERAL_UPDATE | WIN_A_UPDATE;
 	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
 
@@ -509,6 +520,16 @@
 	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
 	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
 
+	if (window->tiled) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
 	value = WIN_ENABLE;
 
 	if (yuv) {
diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h
index 79eaec9..e0b94c2 100644
--- a/drivers/gpu/drm/tegra/dc.h
+++ b/drivers/gpu/drm/tegra/dc.h
@@ -365,6 +365,10 @@
 #define DC_WIN_BUF_STRIDE			0x70b
 #define DC_WIN_UV_BUF_STRIDE			0x70c
 #define DC_WIN_BUFFER_ADDR_MODE			0x70d
+#define DC_WIN_BUFFER_ADDR_MODE_LINEAR		(0 <<  0)
+#define DC_WIN_BUFFER_ADDR_MODE_TILE		(1 <<  0)
+#define DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV	(0 << 16)
+#define DC_WIN_BUFFER_ADDR_MODE_TILE_UV		(1 << 16)
 #define DC_WIN_DV_CONTROL			0x70e
 
 #define DC_WIN_BLEND_NOKEY			0x70f
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 26a2903..ecb6ec7 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -262,7 +262,7 @@
 	struct drm_tegra_gem_create *args = data;
 	struct tegra_bo *bo;
 
-	bo = tegra_bo_create_with_handle(file, drm, args->size,
+	bo = tegra_bo_create_with_handle(file, drm, args->size, args->flags,
 					 &args->handle);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index b599fd4..e07a10ee 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -148,6 +148,7 @@
 	unsigned int format;
 	unsigned int stride[2];
 	unsigned long base[3];
+	bool tiled;
 };
 
 /* from dc.c */
@@ -254,6 +255,7 @@
 /* from fb.c */
 struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer,
 				    unsigned int index);
+bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer);
 extern int tegra_drm_fb_init(struct drm_device *drm);
 extern void tegra_drm_fb_exit(struct drm_device *drm);
 extern void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 1fd4e19..fdb00e0 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -34,6 +34,16 @@
 	return fb->planes[index];
 }
 
+bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer)
+{
+	struct tegra_fb *fb = to_tegra_fb(framebuffer);
+
+	if (fb->planes[0]->flags & TEGRA_BO_TILED)
+		return true;
+
+	return false;
+}
+
 static void tegra_fb_destroy(struct drm_framebuffer *framebuffer)
 {
 	struct tegra_fb *fb = to_tegra_fb(framebuffer);
@@ -188,7 +198,7 @@
 
 	size = cmd.pitches[0] * cmd.height;
 
-	bo = tegra_bo_create(drm, size);
+	bo = tegra_bo_create(drm, size, 0);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index 267c0c2..d851ec1 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -18,6 +18,8 @@
  * GNU General Public License for more details.
  */
 
+#include <drm/tegra_drm.h>
+
 #include "gem.h"
 
 static inline struct tegra_bo *host1x_to_tegra_bo(struct host1x_bo *bo)
@@ -97,7 +99,8 @@
 	dma_free_writecombine(drm->dev, bo->gem.size, bo->vaddr, bo->paddr);
 }
 
-struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size)
+struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size,
+				 unsigned long flags)
 {
 	struct tegra_bo *bo;
 	int err;
@@ -126,6 +129,9 @@
 	if (err)
 		goto err_mmap;
 
+	if (flags & DRM_TEGRA_GEM_CREATE_TILED)
+		bo->flags |= TEGRA_BO_TILED;
+
 	return bo;
 
 err_mmap:
@@ -142,12 +148,13 @@
 struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
 					     struct drm_device *drm,
 					     unsigned int size,
+					     unsigned long flags,
 					     unsigned int *handle)
 {
 	struct tegra_bo *bo;
 	int ret;
 
-	bo = tegra_bo_create(drm, size);
+	bo = tegra_bo_create(drm, size, flags);
 	if (IS_ERR(bo))
 		return bo;
 
@@ -187,7 +194,7 @@
 	if (args->size < args->pitch * args->height)
 		args->size = args->pitch * args->height;
 
-	bo = tegra_bo_create_with_handle(file, drm, args->size,
+	bo = tegra_bo_create_with_handle(file, drm, args->size, 0,
 					 &args->handle);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
diff --git a/drivers/gpu/drm/tegra/gem.h b/drivers/gpu/drm/tegra/gem.h
index 2b54f14..c4993fc 100644
--- a/drivers/gpu/drm/tegra/gem.h
+++ b/drivers/gpu/drm/tegra/gem.h
@@ -24,9 +24,12 @@
 #include <drm/drm.h>
 #include <drm/drmP.h>
 
+#define TEGRA_BO_TILED (1 << 0)
+
 struct tegra_bo {
 	struct drm_gem_object gem;
 	struct host1x_bo base;
+	unsigned long flags;
 	dma_addr_t paddr;
 	void *vaddr;
 };
@@ -38,11 +41,13 @@
 
 extern const struct host1x_bo_ops tegra_bo_ops;
 
-struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size);
+struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size,
+				 unsigned long flags);
 struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
-					    struct drm_device *drm,
-					    unsigned int size,
-					    unsigned int *handle);
+					     struct drm_device *drm,
+					     unsigned int size,
+					     unsigned long flags,
+					     unsigned int *handle);
 void tegra_bo_free_object(struct drm_gem_object *gem);
 int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm,
 			 struct drm_mode_create_dumb *args);
diff --git a/include/uapi/drm/tegra_drm.h b/include/uapi/drm/tegra_drm.h
index 73bde4e..6b42002 100644
--- a/include/uapi/drm/tegra_drm.h
+++ b/include/uapi/drm/tegra_drm.h
@@ -19,6 +19,8 @@
 
 #include <drm/drm.h>
 
+#define DRM_TEGRA_GEM_CREATE_TILED (1 << 0)
+
 struct drm_tegra_gem_create {
 	__u64 size;
 	__u32 flags;