udmabuf: fix general protection fault in udmabuf_create
PATCH 1/1 udmabuf: fix general protection fault in udmabuf_create
Syzbot reported general protection fault in udmabuf_create. The problem was in wrong error handling.
In commit 16c243e99d33 ("udmabuf: Add support for mapping hugepages (v4)") shmem_read_mapping_page() call was replaced with find_get_page_flags(), but find_get_page_flags() returns NULL on failure instead PTR_ERR().
Wrong error checking was causing GPF in get_page(), since passed page was equal to NULL. Fix it by changing if (IS_ER(!hpage)) to if (!hpage)
--- drivers/dma-buf/udmabuf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma-buf/udmabuf.c +2 -2
@@ -227,8 +227,8 @@ static long udmabuf_create(struct miscdevice *device,if (!hpage) { hpage =find_get_page_flags (mapping, pgoff,FGP_ACCESSED ); -if (IS_ERR (hpage)) { - ret =PTR_ERR (hpage); +if (!hpage) { + ret = -EINVAL ; goto err; } }
Signature · 1 line
2.32.0
Quoted 7 lines · Pavel Skripkin
Syzbot reported general protection fault in
udmabuf_create. The problem was in wrong error handling.In commit
16c243e99d33("udmabuf: Add support for mapping hugepages (v4)")shmem_read_mapping_page()call was replaced withfind_get_page_flags(), butfind_get_page_flags()returns NULL on failure insteadPTR_ERR().
Quoted 6 lines
Wrong error checking was causing GPF in
get_page(), since passed page was equal to NULL. Fix it by changing if (IS_ER(!hpage)) to if (!hpage)Reported-by [email protected]Fixes 16c243e99d33 ("udmabuf: Add support for mapping hugepages (v4)")Signed-off-by Pavel Skripkin <[email protected]>
Pushed to drm-misc-next.
thanks,
Gerd