From: Ørjan Eide <[email protected]>

When mapping buffers through the PRIME DMA-buf mmap path we might be
given an offset which has to be respected. The DRM GEM mmap path already
takes care of zeroing out the fake mmap offset, so we can just make the
IOMMU mmap implementation always respect the offset.

TEST=graphics_GLBench

Signed-off-by: rjan Eide <[email protected]>
Signed-off-by: Tomasz Figa <[email protected]>
Signed-off-by: Mark Yao <[email protected]>
Reviewed-on: https://chromium-review.googlesource.com/386477
Reviewed-by: Daniel Kurtz <[email protected]>
---
 drivers/gpu/drm/rockchip/rockchip_drm_gem.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c 
b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
index 1daa531..1769146 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
@@ -221,12 +221,16 @@ static int rockchip_drm_gem_object_mmap_iommu(struct 
drm_gem_object *obj,
        unsigned int i, count = obj->size >> PAGE_SHIFT;
        unsigned long user_count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
        unsigned long uaddr = vma->vm_start;
+       unsigned long offset = vma->vm_pgoff;
+       unsigned long end = user_count + offset;
        int ret;
 
-       if (user_count == 0 || user_count > count)
+       if (user_count == 0)
+               return -ENXIO;
+       if (end > count)
                return -ENXIO;
 
-       for (i = 0; i < user_count; i++) {
+       for (i = offset; i < end; i++) {
                ret = vm_insert_page(vma, uaddr, rk_obj->pages[i]);
                if (ret)
                        return ret;
-- 
1.9.1


Reply via email to