media: intel-ipu3: cio2: fixup off-by-one bug in cio2_vb2_buf_init

With "pages" initialized to vb length + 1 pages, the condition
check if(!pages--) will break at one more page than intended,
this can result in out-of-bound access to b->lop[i][j] when setting
the last dummy page.

Fixes: c7cbef1fdb54 ("media: intel-ipu3: cio2: fix a crash with out-of-bounds access")
Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Signed-off-by: Cao Bing Bu <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>

diff --git a/drivers/media/pci/intel/ipu3/ipu3-cio2.c b/drivers/media/pci/intel/ipu3/ipu3-cio2.c
index 724cd8d9d573335d825d2348ad93c9ecd8414511..6c4444b31f4b9f54ccc4af0c5e228a5d7d63907a 100644 (file)
--- a/drivers/media/pci/intel/ipu3/ipu3-cio2.c
+++ b/drivers/media/pci/intel/ipu3/ipu3-cio2.c
@@ -839,9 +839,8 @@ static int cio2_vb2_buf_init(struct vb2_buffer *vb)
                container_of(vb, struct cio2_buffer, vbb.vb2_buf);
        static const unsigned int entries_per_page =
                CIO2_PAGE_SIZE / sizeof(u32);
-       unsigned int pages = DIV_ROUND_UP(vb->planes[0].length,
-                                         CIO2_PAGE_SIZE) + 1;
-       unsigned int lops = DIV_ROUND_UP(pages, entries_per_page);
+       unsigned int pages = DIV_ROUND_UP(vb->planes[0].length, CIO2_PAGE_SIZE);
+       unsigned int lops = DIV_ROUND_UP(pages + 1, entries_per_page);
        struct sg_table *sg;
        struct sg_page_iter sg_iter;
        int i, j;