[PATCH] io_uring: use kvfree() in io_sqe_buffer_register()

[Denis Efremov]

Use kvfree() to free the pages and vmas, since they are allocated by
kvmalloc_array() in a loop.

Fixes: d4ef647510b1 ("io_uring: avoid page allocation warnings")
Cc: sta...@vger.kernel.org
Signed-off-by: Denis Efremov <efre...@linux.com>
---
I checked the v1 d4ef647510b1 discussion and these lines are using
kvfree() https://lkml.org/lkml/2019/5/1/254. This was somehow missed
in v2.

 fs/io_uring.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 9d4bd0d3a080..defb8a3538fc 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -7160,8 +7160,8 @@ static int io_sqe_buffer_register(struct io_ring_ctx 
*ctx, void __user *arg,
 
                ret = 0;
                if (!pages || nr_pages > got_pages) {
-                       kfree(vmas);
-                       kfree(pages);
+                       kvfree(vmas);
+                       kvfree(pages);
                        pages = kvmalloc_array(nr_pages, sizeof(struct page *),
                                                GFP_KERNEL);
                        vmas = kvmalloc_array(nr_pages,
-- 
2.26.2