[PATCH] dma-buf: might_sleep() in dma_buf_unmap_attachment()

[Rob Clark]

From: Rob Clark <rob at ti.com>

We never really clarified if unmap could be done in atomic context.
But since mapping might require sleeping, this implies mutex in use
to synchronize mapping/unmapping, so unmap could sleep as well.  Add
a might_sleep() to clarify this.

Signed-off-by: Rob Clark <rob at ti.com>
Acked-by: Daniel Vetter <daniel.vetter at ffwll.ch>
---
 drivers/base/dma-buf.c |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c
index c30f3e1..877eacb 100644
--- a/drivers/base/dma-buf.c
+++ b/drivers/base/dma-buf.c
@@ -298,6 +298,8 @@ void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
 				struct sg_table *sg_table,
 				enum dma_data_direction direction)
 {
+	might_sleep();
+
 	if (WARN_ON(!attach || !attach->dmabuf || !sg_table))
 		return;
 
-- 
1.7.9.5