[Linux-cachefs] [PATCH] fs/cachefiles: Fix a possible sleep-in-atomic bug in cachefiles_printk_object

[Jia-Ju Bai]

The kernel may sleep under a write spinlock, and the function call path is:
cachefiles_mark_object_active (acquire the spinlock)
  cachefiles_printk_object
    kmalloc(GFP_NOIO) --> may sleep

To fix it, GFP_NOIO is replaced with GFP_ATOMIC.
This bug is found by my static analysis tool and my code review.

Signed-off-by: Jia-Ju Bai <baijiaju1990 at 163.com>
---
 fs/cachefiles/namei.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c
index 3978b32..39e1504 100644
--- a/fs/cachefiles/namei.c
+++ b/fs/cachefiles/namei.c
@@ -83,7 +83,7 @@ static noinline void cachefiles_printk_object(struct cachefiles_object *object,
 {
 	u8 *keybuf;
 
-	keybuf = kmalloc(CACHEFILES_KEYBUF_SIZE, GFP_NOIO);
+	keybuf = kmalloc(CACHEFILES_KEYBUF_SIZE, GFP_ATOMIC);
 	if (object)
 		__cachefiles_printk_object(object, "", keybuf);
 	if (xobject)
-- 
1.7.9.5