In the case getsockopt() is called with PACKET_HDRLEN and optlen < 4
|val| remains uninitialized and the syscall may behave differently
depending on its value, and even copy garbage to userspace on certain
architectures. To fix this we now return -EINVAL if optlen is too small.

This bug has been detected with KMSAN.

Signed-off-by: Alexander Potapenko <gli...@google.com>

---
The previous versions of this patch were called "net/packet: initialize
val in packet_getsockopt()"

v3: - change patch summary, return -EINVAL for optlen < sizeof(int)
v2: - if len < sizeof(int), make it 0

---
 net/packet/af_packet.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 8489beff5c25..ea81ccf3c7d6 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -3836,6 +3836,8 @@ static int packet_getsockopt(struct socket *sock, int 
level, int optname,
        case PACKET_HDRLEN:
                if (len > sizeof(int))
                        len = sizeof(int);
+               if (len < sizeof(int))
+                       return -EINVAL;
                if (copy_from_user(&val, optval, len))
                        return -EFAULT;
                switch (val) {
-- 
2.13.0.rc0.306.g87b477812d-goog

Reply via email to