[RFC][PATCH 0/3] dmabuf heaps: system uncached and cma uncached heaps

[RFC][PATCH 0/3] dmabuf heaps: system uncached and cma uncached heaps

[John Stultz]

After the last round submitting the system-uncached heap, I got
some feedback that Daniel would like to see it demonstrated with
a mesa based system. I'm still working on such a gralloc
implementation (using the db845c), but along with other work, so
I don't yet have something to share there. 

However, Bing Song reached out and was interested in having a
uncached variant for the CMA heap as well, and he shared this
patch providing an initial implementation.

This gave me some hesitation with regards to the earlier
discussion around what sort of attributes would be useful for
the flags field of the allocation IOCTL.

In earlier discussions, folks seemed happy to provide the
uncached system heap functionality as its own heap chardev, as
it seemed uncertain that the uncached attribute would truely be
generic across all heaps. 

But with Bing's patch, it seems like it may be generically useful,
and utilizing a flag might be a bit cleaner then adding lots of
duplicative heap names postfixed with "-uncached".

So I wanted to re-submit both of these together to reopen that
discussion on the question of if a BUF_FLAG_UNCACHED flag would
make sense, or if folks still think separate heap chardevs is
the way to go.

thanks
-john

Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Liam Mark <lmark@codeaurora.org>
Cc: Laura Abbott <labbott@kernel.org>
Cc: Brian Starkey <Brian.Starkey@arm.com>
Cc: Hridya Valsaraju <hridya@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Daniel Mentz <danielmentz@google.com>
Cc: Chris Goldsworthy <cgoldswo@codeaurora.org>
Cc: Ørjan Eide <orjan.eide@arm.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Ezequiel Garcia <ezequiel@collabora.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: James Jones <jajones@nvidia.com>
Cc: Bing Song <bing.song@nxp.com>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org

Bing Song (1):
  dma-buf: cma_heap: Add a cma-uncached heap re-using the cma heap

John Stultz (2):
  dma-buf: dma-heap: Keep track of the heap device struct
  dma-buf: system_heap: Add a system-uncached heap re-using the system
    heap

 drivers/dma-buf/dma-heap.c          |  33 ++++++--
 drivers/dma-buf/heaps/cma_heap.c    | 119 +++++++++++++++++++++++++---
 drivers/dma-buf/heaps/system_heap.c | 111 ++++++++++++++++++++++----
 include/linux/dma-heap.h            |   9 +++
 4 files changed, 236 insertions(+), 36 deletions(-)

-- 
2.17.1

[PATCH 1/3] dma-buf: dma-heap: Keep track of the heap device struct

[John Stultz]

Keep track of the heap device struct.

This will be useful for special DMA allocations
and actions.

Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Liam Mark <lmark@codeaurora.org>
Cc: Laura Abbott <labbott@kernel.org>
Cc: Brian Starkey <Brian.Starkey@arm.com>
Cc: Hridya Valsaraju <hridya@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Daniel Mentz <danielmentz@google.com>
Cc: Chris Goldsworthy <cgoldswo@codeaurora.org>
Cc: Ørjan Eide <orjan.eide@arm.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Ezequiel Garcia <ezequiel@collabora.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: James Jones <jajones@nvidia.com>
Cc: Bing Song <bing.song@nxp.com>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 drivers/dma-buf/dma-heap.c | 33 +++++++++++++++++++++++++--------
 include/linux/dma-heap.h   |  9 +++++++++
 2 files changed, 34 insertions(+), 8 deletions(-)

diff --git a/drivers/dma-buf/dma-heap.c b/drivers/dma-buf/dma-heap.c
index afd22c9dbdcf..72c746755d89 100644
--- a/drivers/dma-buf/dma-heap.c
+++ b/drivers/dma-buf/dma-heap.c
@@ -30,6 +30,7 @@
  * @heap_devt		heap device node
  * @list		list head connecting to list of heaps
  * @heap_cdev		heap char device
+ * @heap_dev		heap device struct
  *
  * Represents a heap of memory from which buffers can be made.
  */
@@ -40,6 +41,7 @@ struct dma_heap {
 	dev_t heap_devt;
 	struct list_head list;
 	struct cdev heap_cdev;
+	struct device *heap_dev;
 };
 
 static LIST_HEAD(heap_list);
@@ -190,10 +192,21 @@ void *dma_heap_get_drvdata(struct dma_heap *heap)
 	return heap->priv;
 }
 
+/**
+ * dma_heap_get_dev() - get device struct for the heap
+ * @heap: DMA-Heap to retrieve device struct from
+ *
+ * Returns:
+ * The device struct for the heap.
+ */
+struct device *dma_heap_get_dev(struct dma_heap *heap)
+{
+	return heap->heap_dev;
+}
+
 struct dma_heap *dma_heap_add(const struct dma_heap_export_info *exp_info)
 {
 	struct dma_heap *heap, *h, *err_ret;
-	struct device *dev_ret;
 	unsigned int minor;
 	int ret;
 
@@ -247,16 +260,20 @@ struct dma_heap *dma_heap_add(const struct dma_heap_export_info *exp_info)
 		goto err1;
 	}
 
-	dev_ret = device_create(dma_heap_class,
-				NULL,
-				heap->heap_devt,
-				NULL,
-				heap->name);
-	if (IS_ERR(dev_ret)) {
+	heap->heap_dev = device_create(dma_heap_class,
+				       NULL,
+				       heap->heap_devt,
+				       NULL,
+				       heap->name);
+	if (IS_ERR(heap->heap_dev)) {
 		pr_err("dma_heap: Unable to create device\n");
-		err_ret = ERR_CAST(dev_ret);
+		err_ret = ERR_CAST(heap->heap_dev);
 		goto err2;
 	}
+
+	/* Make sure it doesn't disappear on us */
+	heap->heap_dev = get_device(heap->heap_dev);
+
 	/* Add heap to the list */
 	mutex_lock(&heap_list_lock);
 	list_add(&heap->list, &heap_list);
diff --git a/include/linux/dma-heap.h b/include/linux/dma-heap.h
index 454e354d1ffb..82857e096910 100644
--- a/include/linux/dma-heap.h
+++ b/include/linux/dma-heap.h
@@ -50,6 +50,15 @@ struct dma_heap_export_info {
  */
 void *dma_heap_get_drvdata(struct dma_heap *heap);
 
+/**
+ * dma_heap_get_dev() - get device struct for the heap
+ * @heap: DMA-Heap to retrieve device struct from
+ *
+ * Returns:
+ * The device struct for the heap.
+ */
+struct device *dma_heap_get_dev(struct dma_heap *heap);
+
 /**
  * dma_heap_add - adds a heap to dmabuf heaps
  * @exp_info:		information needed to register this heap
-- 
2.17.1

[PATCH 2/3] dma-buf: system_heap: Add a system-uncached heap re-using the system heap

[John Stultz]

This adds a heap that allocates non-contiguous buffers that are
marked as writecombined, so they are not cached by the CPU.

This is useful, as most graphics buffers are usually not touched
by the CPU or only written into once by the CPU. So when mapping
the buffer over and over between devices, we can skip the CPU
syncing, which saves a lot of cache management overhead, greatly
improving performance.

For folk using ION, there was a ION_FLAG_CACHED flag, which
signaled if the returned buffer should be CPU cacheable or not.
With DMA-BUF heaps, we do not yet have such a flag, and by default
the current heaps (system and cma) produce CPU cachable buffers.
So for folks transitioning from ION to DMA-BUF Heaps, this fills
in some of that missing functionality.

There has been a suggestion to make this functionality a flag
(DMAHEAP_FLAG_UNCACHED?) on the system heap, similar to how
ION used the ION_FLAG_CACHED. But I want to make sure an
_UNCACHED flag would truely be a generic attribute across all
heaps. So far that has been unclear, so having it as a separate
heap seemes better for now. (But I'm open to discussion on this
point!)

This is a rework of earlier efforts to add a uncached system heap,
done utilizing the exisitng system heap, adding just a bit of
logic to handle the uncached case.

Feedback would be very welcome!

Many thanks to Liam Mark for his help to get this working.

Pending opensource users of this code include:
* AOSP HiKey960 gralloc:
  - https://android-review.googlesource.com/c/device/linaro/hikey/+/1399519
  - Visibly improves performance over the system heap
* AOSP Codec2:
  - https://android-review.googlesource.com/c/platform/frameworks/av/+/1543685

Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Liam Mark <lmark@codeaurora.org>
Cc: Laura Abbott <labbott@kernel.org>
Cc: Brian Starkey <Brian.Starkey@arm.com>
Cc: Hridya Valsaraju <hridya@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Daniel Mentz <danielmentz@google.com>
Cc: Chris Goldsworthy <cgoldswo@codeaurora.org>
Cc: Ørjan Eide <orjan.eide@arm.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Ezequiel Garcia <ezequiel@collabora.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: James Jones <jajones@nvidia.com>
Cc: Bing Song <bing.song@nxp.com>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
v4:
* Make sys_uncached_heap static, as
    Reported-by: kernel test robot <lkp@intel.com>
* Fix wrong return value, caught by smatch
    Reported-by: kernel test robot <lkp@intel.com>
    Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
* Ensure we call flush/invalidate_kernel_vmap_range() in the
  uncached cases to try to address feedback about VIVT caches
  from Christoph
* Reorder a few lines as suggested by BrianS
* Avoid holding the initial mapping for the lifetime of the buffer
  as suggested by BrianS
* Fix a unlikely race between allocate and updating the dma_mask
  that BrianS noticed.
---
 drivers/dma-buf/heaps/system_heap.c | 111 ++++++++++++++++++++++++----
 1 file changed, 95 insertions(+), 16 deletions(-)

diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c
index 17e0e9a68baf..3548b20cb98c 100644
--- a/drivers/dma-buf/heaps/system_heap.c
+++ b/drivers/dma-buf/heaps/system_heap.c
@@ -22,6 +22,7 @@
 #include <linux/vmalloc.h>
 
 static struct dma_heap *sys_heap;
+static struct dma_heap *sys_uncached_heap;
 
 struct system_heap_buffer {
 	struct dma_heap *heap;
@@ -31,6 +32,8 @@ struct system_heap_buffer {
 	struct sg_table sg_table;
 	int vmap_cnt;
 	void *vaddr;
+
+	bool uncached;
 };
 
 struct dma_heap_attachment {
@@ -38,6 +41,8 @@ struct dma_heap_attachment {
 	struct sg_table *table;
 	struct list_head list;
 	bool mapped;
+
+	bool uncached;
 };
 
 #define HIGH_ORDER_GFP  (((GFP_HIGHUSER | __GFP_ZERO | __GFP_NOWARN \
@@ -100,7 +105,7 @@ static int system_heap_attach(struct dma_buf *dmabuf,
 	a->dev = attachment->dev;
 	INIT_LIST_HEAD(&a->list);
 	a->mapped = false;
-
+	a->uncached = buffer->uncached;
 	attachment->priv = a;
 
 	mutex_lock(&buffer->lock);
@@ -130,9 +135,13 @@ static struct sg_table *system_heap_map_dma_buf(struct dma_buf_attachment *attac
 {
 	struct dma_heap_attachment *a = attachment->priv;
 	struct sg_table *table = a->table;
+	int attr = 0;
 	int ret;
 
-	ret = dma_map_sgtable(attachment->dev, table, direction, 0);
+	if (a->uncached)
+		attr = DMA_ATTR_SKIP_CPU_SYNC;
+
+	ret = dma_map_sgtable(attachment->dev, table, direction, attr);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -145,9 +154,12 @@ static void system_heap_unmap_dma_buf(struct dma_buf_attachment *attachment,
 				      enum dma_data_direction direction)
 {
 	struct dma_heap_attachment *a = attachment->priv;
+	int attr = 0;
 
+	if (a->uncached)
+		attr = DMA_ATTR_SKIP_CPU_SYNC;
 	a->mapped = false;
-	dma_unmap_sgtable(attachment->dev, table, direction, 0);
+	dma_unmap_sgtable(attachment->dev, table, direction, attr);
 }
 
 static int system_heap_dma_buf_begin_cpu_access(struct dma_buf *dmabuf,
@@ -161,10 +173,12 @@ static int system_heap_dma_buf_begin_cpu_access(struct dma_buf *dmabuf,
 	if (buffer->vmap_cnt)
 		invalidate_kernel_vmap_range(buffer->vaddr, buffer->len);
 
-	list_for_each_entry(a, &buffer->attachments, list) {
-		if (!a->mapped)
-			continue;
-		dma_sync_sgtable_for_cpu(a->dev, a->table, direction);
+	if (!buffer->uncached) {
+		list_for_each_entry(a, &buffer->attachments, list) {
+			if (!a->mapped)
+				continue;
+			dma_sync_sgtable_for_cpu(a->dev, a->table, direction);
+		}
 	}
 	mutex_unlock(&buffer->lock);
 
@@ -182,10 +196,12 @@ static int system_heap_dma_buf_end_cpu_access(struct dma_buf *dmabuf,
 	if (buffer->vmap_cnt)
 		flush_kernel_vmap_range(buffer->vaddr, buffer->len);
 
-	list_for_each_entry(a, &buffer->attachments, list) {
-		if (!a->mapped)
-			continue;
-		dma_sync_sgtable_for_device(a->dev, a->table, direction);
+	if (!buffer->uncached) {
+		list_for_each_entry(a, &buffer->attachments, list) {
+			if (!a->mapped)
+				continue;
+			dma_sync_sgtable_for_device(a->dev, a->table, direction);
+		}
 	}
 	mutex_unlock(&buffer->lock);
 
@@ -200,6 +216,9 @@ static int system_heap_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma)
 	struct sg_page_iter piter;
 	int ret;
 
+	if (buffer->uncached)
+		vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
+
 	for_each_sgtable_page(table, &piter, vma->vm_pgoff) {
 		struct page *page = sg_page_iter_page(&piter);
 
@@ -221,17 +240,21 @@ static void *system_heap_do_vmap(struct system_heap_buffer *buffer)
 	struct page **pages = vmalloc(sizeof(struct page *) * npages);
 	struct page **tmp = pages;
 	struct sg_page_iter piter;
+	pgprot_t pgprot = PAGE_KERNEL;
 	void *vaddr;
 
 	if (!pages)
 		return ERR_PTR(-ENOMEM);
 
+	if (buffer->uncached)
+		pgprot = pgprot_writecombine(PAGE_KERNEL);
+
 	for_each_sgtable_page(table, &piter, 0) {
 		WARN_ON(tmp - pages >= npages);
 		*tmp++ = sg_page_iter_page(&piter);
 	}
 
-	vaddr = vmap(pages, npages, VM_MAP, PAGE_KERNEL);
+	vaddr = vmap(pages, npages, VM_MAP, pgprot);
 	vfree(pages);
 
 	if (!vaddr)
@@ -331,10 +354,11 @@ static struct page *alloc_largest_available(unsigned long size,
 	return NULL;
 }
 
-static int system_heap_allocate(struct dma_heap *heap,
-				unsigned long len,
-				unsigned long fd_flags,
-				unsigned long heap_flags)
+static int system_heap_do_allocate(struct dma_heap *heap,
+				   unsigned long len,
+				   unsigned long fd_flags,
+				   unsigned long heap_flags,
+				   bool uncached)
 {
 	struct system_heap_buffer *buffer;
 	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
@@ -355,6 +379,7 @@ static int system_heap_allocate(struct dma_heap *heap,
 	mutex_init(&buffer->lock);
 	buffer->heap = heap;
 	buffer->len = len;
+	buffer->uncached = uncached;
 
 	INIT_LIST_HEAD(&pages);
 	i = 0;
@@ -404,6 +429,18 @@ static int system_heap_allocate(struct dma_heap *heap,
 		/* just return, as put will call release and that will free */
 		return ret;
 	}
+
+	/*
+	 * For uncached buffers, we need to initially flush cpu cache, since
+	 * the __GFP_ZERO on the allocation means the zeroing was done by the
+	 * cpu and thus it is likely cached. Map (and implicitly flush) and
+	 * unmap it now so we don't get corruption later on.
+	 */
+	if (buffer->uncached) {
+		dma_map_sgtable(dma_heap_get_dev(heap), table, DMA_BIDIRECTIONAL, 0);
+		dma_unmap_sgtable(dma_heap_get_dev(heap), table, DMA_BIDIRECTIONAL, 0);
+	}
+
 	return ret;
 
 free_pages:
@@ -421,10 +458,40 @@ static int system_heap_allocate(struct dma_heap *heap,
 	return ret;
 }
 
+static int system_heap_allocate(struct dma_heap *heap,
+				unsigned long len,
+				unsigned long fd_flags,
+				unsigned long heap_flags)
+{
+	return system_heap_do_allocate(heap, len, fd_flags, heap_flags, false);
+}
+
 static const struct dma_heap_ops system_heap_ops = {
 	.allocate = system_heap_allocate,
 };
 
+static int system_uncached_heap_allocate(struct dma_heap *heap,
+					 unsigned long len,
+					 unsigned long fd_flags,
+					 unsigned long heap_flags)
+{
+	return system_heap_do_allocate(heap, len, fd_flags, heap_flags, true);
+}
+
+/* Dummy function to be used until we can call coerce_mask_and_coherent */
+static int system_uncached_heap_not_initialized(struct dma_heap *heap,
+						unsigned long len,
+						unsigned long fd_flags,
+						unsigned long heap_flags)
+{
+	return -EBUSY;
+}
+
+static struct dma_heap_ops system_uncached_heap_ops = {
+	/* After system_heap_create is complete, we will swap this */
+	.allocate = system_uncached_heap_not_initialized,
+};
+
 static int system_heap_create(void)
 {
 	struct dma_heap_export_info exp_info;
@@ -437,6 +504,18 @@ static int system_heap_create(void)
 	if (IS_ERR(sys_heap))
 		return PTR_ERR(sys_heap);
 
+	exp_info.name = "system-uncached";
+	exp_info.ops = &system_uncached_heap_ops;
+	exp_info.priv = NULL;
+
+	sys_uncached_heap = dma_heap_add(&exp_info);
+	if (IS_ERR(sys_uncached_heap))
+		return PTR_ERR(sys_uncached_heap);
+
+	dma_coerce_mask_and_coherent(dma_heap_get_dev(sys_uncached_heap), DMA_BIT_MASK(64));
+	mb(); /* make sure we only set allocate after dma_mask is set */
+	system_uncached_heap_ops.allocate = system_uncached_heap_allocate;
+
 	return 0;
 }
 module_init(system_heap_create);
-- 
2.17.1

[PATCH 3/3] dma-buf: cma_heap: Add a cma-uncached heap re-using the cma heap

[John Stultz]

From: Bing Song <bing.song@nxp.com>

This adds a heap that allocates CMA buffers that are
marked as writecombined, so they are not cached by the CPU.

Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Liam Mark <lmark@codeaurora.org>
Cc: Laura Abbott <labbott@kernel.org>
Cc: Brian Starkey <Brian.Starkey@arm.com>
Cc: Hridya Valsaraju <hridya@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Sandeep Patil <sspatil@google.com>
Cc: Daniel Mentz <danielmentz@google.com>
Cc: Chris Goldsworthy <cgoldswo@codeaurora.org>
Cc: Ørjan Eide <orjan.eide@arm.com>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Ezequiel Garcia <ezequiel@collabora.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: James Jones <jajones@nvidia.com>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Bing Song <bing.song@nxp.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 drivers/dma-buf/heaps/cma_heap.c | 119 +++++++++++++++++++++++++++----
 1 file changed, 107 insertions(+), 12 deletions(-)

diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c
index 364fc2f3e499..1b8c6eb0a8ea 100644
--- a/drivers/dma-buf/heaps/cma_heap.c
+++ b/drivers/dma-buf/heaps/cma_heap.c
@@ -38,6 +38,7 @@ struct cma_heap_buffer {
 	pgoff_t pagecount;
 	int vmap_cnt;
 	void *vaddr;
+	bool uncached;
 };
 
 struct dma_heap_attachment {
@@ -45,6 +46,7 @@ struct dma_heap_attachment {
 	struct sg_table table;
 	struct list_head list;
 	bool mapped;
+	bool uncached;
 };
 
 static int cma_heap_attach(struct dma_buf *dmabuf,
@@ -70,6 +72,7 @@ static int cma_heap_attach(struct dma_buf *dmabuf,
 	a->dev = attachment->dev;
 	INIT_LIST_HEAD(&a->list);
 	a->mapped = false;
+	a->uncached = buffer->uncached;
 
 	attachment->priv = a;
 
@@ -99,8 +102,12 @@ static struct sg_table *cma_heap_map_dma_buf(struct dma_buf_attachment *attachme
 {
 	struct dma_heap_attachment *a = attachment->priv;
 	struct sg_table *table = &a->table;
+	int attr = 0;
 	int ret;
 
+	if (a->uncached)
+		attr = DMA_ATTR_SKIP_CPU_SYNC;
+
 	ret = dma_map_sgtable(attachment->dev, table, direction, 0);
 	if (ret)
 		return ERR_PTR(-ENOMEM);
@@ -113,7 +120,10 @@ static void cma_heap_unmap_dma_buf(struct dma_buf_attachment *attachment,
 				   enum dma_data_direction direction)
 {
 	struct dma_heap_attachment *a = attachment->priv;
+	int attr = 0;
 
+	if (a->uncached)
+		attr = DMA_ATTR_SKIP_CPU_SYNC;
 	a->mapped = false;
 	dma_unmap_sgtable(attachment->dev, table, direction, 0);
 }
@@ -128,10 +138,12 @@ static int cma_heap_dma_buf_begin_cpu_access(struct dma_buf *dmabuf,
 		invalidate_kernel_vmap_range(buffer->vaddr, buffer->len);
 
 	mutex_lock(&buffer->lock);
-	list_for_each_entry(a, &buffer->attachments, list) {
-		if (!a->mapped)
-			continue;
-		dma_sync_sgtable_for_cpu(a->dev, &a->table, direction);
+	if (!buffer->uncached) {
+		list_for_each_entry(a, &buffer->attachments, list) {
+			if (!a->mapped)
+				continue;
+			dma_sync_sgtable_for_cpu(a->dev, &a->table, direction);
+		}
 	}
 	mutex_unlock(&buffer->lock);
 
@@ -148,10 +160,12 @@ static int cma_heap_dma_buf_end_cpu_access(struct dma_buf *dmabuf,
 		flush_kernel_vmap_range(buffer->vaddr, buffer->len);
 
 	mutex_lock(&buffer->lock);
-	list_for_each_entry(a, &buffer->attachments, list) {
-		if (!a->mapped)
-			continue;
-		dma_sync_sgtable_for_device(a->dev, &a->table, direction);
+	if (!buffer->uncached) {
+		list_for_each_entry(a, &buffer->attachments, list) {
+			if (!a->mapped)
+				continue;
+			dma_sync_sgtable_for_device(a->dev, &a->table, direction);
+		}
 	}
 	mutex_unlock(&buffer->lock);
 
@@ -183,6 +197,9 @@ static int cma_heap_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma)
 	if ((vma->vm_flags & (VM_SHARED | VM_MAYSHARE)) == 0)
 		return -EINVAL;
 
+	if (buffer->uncached)
+		vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
+
 	vma->vm_ops = &dma_heap_vm_ops;
 	vma->vm_private_data = buffer;
 
@@ -191,9 +208,13 @@ static int cma_heap_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma)
 
 static void *cma_heap_do_vmap(struct cma_heap_buffer *buffer)
 {
+	pgprot_t pgprot = PAGE_KERNEL;
 	void *vaddr;
 
-	vaddr = vmap(buffer->pages, buffer->pagecount, VM_MAP, PAGE_KERNEL);
+	if (buffer->uncached)
+		pgprot = pgprot_writecombine(PAGE_KERNEL);
+
+	vaddr = vmap(buffer->pages, buffer->pagecount, VM_MAP, pgprot);
 	if (!vaddr)
 		return ERR_PTR(-ENOMEM);
 
@@ -271,10 +292,11 @@ static const struct dma_buf_ops cma_heap_buf_ops = {
 	.release = cma_heap_dma_buf_release,
 };
 
-static int cma_heap_allocate(struct dma_heap *heap,
+static int cma_heap_do_allocate(struct dma_heap *heap,
 				  unsigned long len,
 				  unsigned long fd_flags,
-				  unsigned long heap_flags)
+				  unsigned long heap_flags,
+				  bool uncached)
 {
 	struct cma_heap *cma_heap = dma_heap_get_drvdata(heap);
 	struct cma_heap_buffer *buffer;
@@ -283,8 +305,9 @@ static int cma_heap_allocate(struct dma_heap *heap,
 	pgoff_t pagecount = size >> PAGE_SHIFT;
 	unsigned long align = get_order(size);
 	struct page *cma_pages;
+	struct sg_table table;
 	struct dma_buf *dmabuf;
-	int ret = -ENOMEM;
+	int ret = -ENOMEM, ret_sg_table;
 	pgoff_t pg;
 
 	buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
@@ -294,6 +317,7 @@ static int cma_heap_allocate(struct dma_heap *heap,
 	INIT_LIST_HEAD(&buffer->attachments);
 	mutex_init(&buffer->lock);
 	buffer->len = size;
+	buffer->uncached = uncached;
 
 	if (align > CONFIG_CMA_ALIGNMENT)
 		align = CONFIG_CMA_ALIGNMENT;
@@ -356,6 +380,18 @@ static int cma_heap_allocate(struct dma_heap *heap,
 		return ret;
 	}
 
+	if (buffer->uncached) {
+		ret_sg_table = sg_alloc_table(&table, 1, GFP_KERNEL);
+		if (ret_sg_table)
+			return ret_sg_table;
+
+		sg_set_page(table.sgl, cma_pages, size, 0);
+
+		dma_map_sgtable(dma_heap_get_dev(heap), &table, DMA_BIDIRECTIONAL, 0);
+		dma_unmap_sgtable(dma_heap_get_dev(heap), &table, DMA_BIDIRECTIONAL, 0);
+		sg_free_table(&table);
+	}
+
 	return ret;
 
 free_pages:
@@ -368,14 +404,45 @@ static int cma_heap_allocate(struct dma_heap *heap,
 	return ret;
 }
 
+static int cma_heap_allocate(struct dma_heap *heap,
+				  unsigned long len,
+				  unsigned long fd_flags,
+				  unsigned long heap_flags)
+{
+	return cma_heap_do_allocate(heap, len, fd_flags, heap_flags, false);
+}
+
+static int cma_uncached_heap_allocate(struct dma_heap *heap,
+				  unsigned long len,
+				  unsigned long fd_flags,
+				  unsigned long heap_flags)
+{
+	return cma_heap_do_allocate(heap, len, fd_flags, heap_flags, true);
+}
+
+/* Dummy function to be used until we can call coerce_mask_and_coherent */
+static int cma_uncached_heap_not_initialized(struct dma_heap *heap,
+						unsigned long len,
+						unsigned long fd_flags,
+						unsigned long heap_flags)
+{
+	return -EBUSY;
+}
+
 static const struct dma_heap_ops cma_heap_ops = {
 	.allocate = cma_heap_allocate,
 };
 
+static struct dma_heap_ops cma_uncached_heap_ops = {
+	.allocate = cma_uncached_heap_not_initialized,
+};
+
 static int __add_cma_heap(struct cma *cma, void *data)
 {
 	struct cma_heap *cma_heap;
 	struct dma_heap_export_info exp_info;
+	const char *postfixed = "-uncached";
+	char *cma_name;
 
 	cma_heap = kzalloc(sizeof(*cma_heap), GFP_KERNEL);
 	if (!cma_heap)
@@ -394,6 +461,34 @@ static int __add_cma_heap(struct cma *cma, void *data)
 		return ret;
 	}
 
+	cma_heap = kzalloc(sizeof(*cma_heap), GFP_KERNEL);
+	if (!cma_heap)
+		return -ENOMEM;
+	cma_heap->cma = cma;
+
+	cma_name = kzalloc(strlen(cma_get_name(cma)) + strlen(postfixed) + 1, GFP_KERNEL);
+	if (!cma_name) {
+		kfree(cma_heap);
+		return -ENOMEM;
+	}
+
+	exp_info.name = strcat(strcpy(cma_name, cma_get_name(cma)), postfixed);
+	exp_info.ops = &cma_uncached_heap_ops;
+	exp_info.priv = cma_heap;
+
+	cma_heap->heap = dma_heap_add(&exp_info);
+	if (IS_ERR(cma_heap->heap)) {
+		int ret = PTR_ERR(cma_heap->heap);
+
+		kfree(cma_heap);
+		kfree(cma_name);
+		return ret;
+	}
+
+	dma_coerce_mask_and_coherent(dma_heap_get_dev(cma_heap->heap), DMA_BIT_MASK(64));
+	mb(); /* make sure we only set allocate after dma_mask is set */
+	cma_uncached_heap_ops.allocate = cma_uncached_heap_allocate;
+
 	return 0;
 }
 
-- 
2.17.1