From: Dan Carpenter <dan.carpenter@xxxxxxxxxx> [ Upstream commit 4b408c74ee5a0b74fc9265c2fe39b0e7dec7c056 ] The concern here is that "gup->size" is a u64 and "nr_pages" is unsigned long. On 32 bit systems we could trick the kernel into allocating fewer pages than expected. Link: http://lkml.kernel.org/r/20181025061546.hnhkv33diogf2uis@kili.mountain Fixes: 64c349f4ae78 ("mm: add infrastructure for get_user_pages_fast() benchmarking") Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx> Acked-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> Reviewed-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Cc: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> Cc: Keith Busch <keith.busch@xxxxxxxxx> Cc: "Michael S. Tsirkin" <mst@xxxxxxxxxx> Cc: Kees Cook <keescook@xxxxxxxxxxxx> Cc: YueHaibing <yuehaibing@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> --- mm/gup_benchmark.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/mm/gup_benchmark.c b/mm/gup_benchmark.c index 7405c9d89d65..7e6f2d2dafb5 100644 --- a/mm/gup_benchmark.c +++ b/mm/gup_benchmark.c @@ -23,6 +23,9 @@ static int __gup_benchmark_ioctl(unsigned int cmd, int nr; struct page **pages; + if (gup->size > ULONG_MAX) + return -EINVAL; + nr_pages = gup->size / PAGE_SIZE; pages = kvcalloc(nr_pages, sizeof(void *), GFP_KERNEL); if (!pages) -- 2.17.1
