drm/msm: add support for "stolen" mem

Add support to use the VRAM carveout (if specified in dtb) for fbdev
scanout buffer.  This allows drm/msm to take over a bootloader splash-
screen, and avoids corruption on screen that results if the kernel uses
memory that is still being scanned out for itself.

Signed-off-by: Rob Clark <robdclark@gmail.com>

drivers/gpu/drm/msm/msm_drv.c

@@ -182,21 +182,57 @@ static int get_mdp_ver(struct platform_device *pdev)
 	return 4;
 }
 
+#include <linux/of_address.h>
+
 static int msm_init_vram(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
+	unsigned long size = 0;
+	int ret = 0;
+
+#ifdef CONFIG_OF
+	/* In the device-tree world, we could have a 'memory-region'
+	 * phandle, which gives us a link to our "vram".  Allocating
+	 * is all nicely abstracted behind the dma api, but we need
+	 * to know the entire size to allocate it all in one go. There
+	 * are two cases:
+	 *  1) device with no IOMMU, in which case we need exclusive
+	 *     access to a VRAM carveout big enough for all gpu
+	 *     buffers
+	 *  2) device with IOMMU, but where the bootloader puts up
+	 *     a splash screen.  In this case, the VRAM carveout
+	 *     need only be large enough for fbdev fb.  But we need
+	 *     exclusive access to the buffer to avoid the kernel
+	 *     using those pages for other purposes (which appears
+	 *     as corruption on screen before we have a chance to
+	 *     load and do initial modeset)
+	 */
+	struct device_node *node;
+
+	node = of_parse_phandle(dev->dev->of_node, "memory-region", 0);
+	if (node) {
+		struct resource r;
+		ret = of_address_to_resource(node, 0, &r);
+		if (ret)
+			return ret;
+		size = r.end - r.start;
+		DRM_INFO("using VRAM carveout: %lx@%08x\n", size, r.start);
+	} else
+#endif
 
 	/* if we have no IOMMU, then we need to use carveout allocator.
 	 * Grab the entire CMA chunk carved out in early startup in
 	 * mach-msm:
 	 */
 	if (!iommu_present(&platform_bus_type)) {
+		DRM_INFO("using %s VRAM carveout\n", vram);
+		size = memparse(vram, NULL);
+	}
+
+	if (size) {
 		DEFINE_DMA_ATTRS(attrs);
-		unsigned long size;
 		void *p;
 
-		DBG("using %s VRAM carveout", vram);
-		size = memparse(vram, NULL);
 		priv->vram.size = size;
 
 		drm_mm_init(&priv->vram.mm, 0, (size >> PAGE_SHIFT) - 1);
@@ -220,7 +256,7 @@ static int msm_init_vram(struct drm_device *dev)
 				(uint32_t)(priv->vram.paddr + size));
 	}
 
-	return 0;
+	return ret;
 }
 
 static int msm_load(struct drm_device *dev, unsigned long flags)

drivers/gpu/drm/msm/msm_fbdev.c

@@ -110,7 +110,8 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	size = mode_cmd.pitches[0] * mode_cmd.height;
 	DBG("allocating %d bytes for fb %d", size, dev->primary->index);
 	mutex_lock(&dev->struct_mutex);
-	fbdev->bo = msm_gem_new(dev, size, MSM_BO_SCANOUT | MSM_BO_WC);
+	fbdev->bo = msm_gem_new(dev, size, MSM_BO_SCANOUT |
+			MSM_BO_WC | MSM_BO_STOLEN);
 	mutex_unlock(&dev->struct_mutex);
 	if (IS_ERR(fbdev->bo)) {
 		ret = PTR_ERR(fbdev->bo);

drivers/gpu/drm/msm/msm_gem.c

@@ -32,6 +32,12 @@ static dma_addr_t physaddr(struct drm_gem_object *obj)
 			priv->vram.paddr;
 }
 
+static bool use_pages(struct drm_gem_object *obj)
+{
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	return !msm_obj->vram_node;
+}
+
 /* allocate pages from VRAM carveout, used when no IOMMU: */
 static struct page **get_pages_vram(struct drm_gem_object *obj,
 		int npages)
@@ -72,7 +78,7 @@ static struct page **get_pages(struct drm_gem_object *obj)
 		struct page **p;
 		int npages = obj->size >> PAGE_SHIFT;
 
-		if (iommu_present(&platform_bus_type))
+		if (use_pages(obj))
 			p = drm_gem_get_pages(obj);
 		else
 			p = get_pages_vram(obj, npages);
@@ -116,7 +122,7 @@ static void put_pages(struct drm_gem_object *obj)
 		sg_free_table(msm_obj->sgt);
 		kfree(msm_obj->sgt);
 
-		if (iommu_present(&platform_bus_type))
+		if (use_pages(obj))
 			drm_gem_put_pages(obj, msm_obj->pages, true, false);
 		else {
 			drm_mm_remove_node(msm_obj->vram_node);
@@ -580,6 +586,7 @@ static int msm_gem_new_impl(struct drm_device *dev,
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj;
 	unsigned sz;
+	bool use_vram = false;
 
 	switch (flags & MSM_BO_CACHE_MASK) {
 	case MSM_BO_UNCACHED:
@@ -592,15 +599,23 @@ static int msm_gem_new_impl(struct drm_device *dev,
 		return -EINVAL;
 	}
 
-	sz = sizeof(*msm_obj);
 	if (!iommu_present(&platform_bus_type))
+		use_vram = true;
+	else if ((flags & MSM_BO_STOLEN) && priv->vram.size)
+		use_vram = true;
+
+	if (WARN_ON(use_vram && !priv->vram.size))
+		return -EINVAL;
+
+	sz = sizeof(*msm_obj);
+	if (use_vram)
 		sz += sizeof(struct drm_mm_node);
 
 	msm_obj = kzalloc(sz, GFP_KERNEL);
 	if (!msm_obj)
 		return -ENOMEM;
 
-	if (!iommu_present(&platform_bus_type))
+	if (use_vram)
 		msm_obj->vram_node = (void *)&msm_obj[1];
 
 	msm_obj->flags = flags;
@@ -630,7 +645,7 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev,
 	if (ret)
 		goto fail;
 
-	if (iommu_present(&platform_bus_type)) {
+	if (use_pages(obj)) {
 		ret = drm_gem_object_init(dev, obj, size);
 		if (ret)
 			goto fail;

drivers/gpu/drm/msm/msm_gem.h

@@ -21,6 +21,9 @@
 #include <linux/reservation.h>
 #include "msm_drv.h"
 
+/* Additional internal-use only BO flags: */
+#define MSM_BO_STOLEN        0x10000000    /* try to use stolen/splash memory */
+
 struct msm_gem_object {
 	struct drm_gem_object base;
 
@@ -59,7 +62,7 @@ struct msm_gem_object {
 	struct reservation_object _resv;
 
 	/* For physically contiguous buffers.  Used when we don't have
-	 * an IOMMU.
+	 * an IOMMU.  Also used for stolen/splashscreen buffer.
 	 */
 	struct drm_mm_node *vram_node;
 };