Hi Eric,
On 06/27/2018 03:16 PM, Eric Dumazet wrote:
>
>
> On 06/27/2018 02:59 PM, Jesus Sanchez-Palencia wrote:
>> From: Richard Cochran <rcoch...@linutronix.de>
>>
>> This patch introduces SO_TXTIME. User space enables this option in
>> order to pass a desired future transmit time in a CMSG when calling
>> sendmsg(2). The argument to this socket option is a 6-bytes long struct
>> defined as:
>>
>> struct sock_txtime {
>> clockid_t clockid;
>> u16 flags;
>> };
>
> Note that sizeof(struct sock_txtime) is 8, not 6, because of alignments.
Oh yeah, sure.
>
> This means that your implementation of getsockopt(... SO_TXTIME )
> is probably leaking two bytes of kernel stack to user space.
I'm failing to see how... There is a memset() in sock.c:1147 clearing all the 8
bytes that we later use to (explicitly) assign each member of the struct. Aren't
the 2 extra bytes sanitized, then? What have I missed?
Thanks,
Jesus
