qemu-stable
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] qemu-sockets: fix unix socket path copy (again)


From: Marc-André Lureau
Subject: Re: [PATCH] qemu-sockets: fix unix socket path copy (again)
Date: Tue, 31 Aug 2021 23:21:43 +0400

Hi

On Tue, Aug 31, 2021 at 10:26 PM Michael Tokarev <mjt@tls.msk.ru> wrote:
We test whenever the path of unix-domain socket
address is non-empty and strictly-less than
the length of the path buffer. Both these
conditions are wrong: the socket can be unnamed,
with empty path, or socket can have pathname
null-terminated _after_ the sun_path buffer,
since we provided more room when asking kernel.

So on one side, allow empty, unnamed sockets
(and adjust the test for abstract socket too -
only do that if the socket is not unnamed),
and on another side, allow path length to be
up to our own maximum, - we have up to size
of sockaddr_storage there.

While at it, fix the duplication of regular
pathname socket to not require trailing \0
(since it can be missing for unnamed sockets).

Fixes: 4cfd970ec188558daa6214f26203fe553fb1e01f (first in 6.1.0)
Fixes: http://bugs.debian.org/993145
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: qemu-stable@nongnu.org
--
Two questions.
1. Why do we store the name of the socket to start with?
2. The code in the abstract socket case should not use
   g_strndup but g_memdup instead, since the whole thing
   is a blob of the given length, not a \0-terminated string.
---
 util/qemu-sockets.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c
index f2f3676d1f..7c83d81792 100644
--- a/util/qemu-sockets.c
+++ b/util/qemu-sockets.c
@@ -1345,13 +1345,20 @@ socket_sockaddr_to_address_unix(struct sockaddr_storage *sa,
     SocketAddress *addr;
     struct sockaddr_un *su = (struct sockaddr_un *)sa;

-    assert(salen >= sizeof(su->sun_family) + 1 &&
-           salen <= sizeof(struct sockaddr_un));
+    /* there's a corner case when trailing \0 does not fit into
+     * sockaddr_un. Compare length with sizeof(sockaddr_storage),
+     * not with sizeof(sockaddr_un), since this is what we actually
+     * provide, to ensure we had no truncation and a room for
+     * the trailing \0 which we add below.
+     * When salen == sizeof(sun_family) it is unnamed socket,
+     * and when first byte of sun_path is \0, it is abstract. */
+    assert(salen >= sizeof(su->sun_family) &&
+           salen <= sizeof(struct sockaddr_storage));

Seems right to me, however there are some notes in libc bits/socket.h
/* Structure large enough to hold any socket address (with the historical
   exception of AF_UNIX).  */

And also this
https://idea.popcount.org/2019-12-06-addressing/#fn:sockaddr_storage

I must say I feel confused by those comments :) Is it large enough or not??


     addr = g_new0(SocketAddress, 1);
     addr->type = SOCKET_ADDRESS_TYPE_UNIX;
 #ifdef CONFIG_LINUX
-    if (!su->sun_path[0]) {
+    if (salen > sizeof(su->sun_family) && !su->sun_path[0]) {
         /* Linux abstract socket */
         addr->u.q_unix.path = g_strndup(su->sun_path + 1,
                                         salen - sizeof(su->sun_family) - 1);
@@ -1363,7 +1370,7 @@ socket_sockaddr_to_address_unix(struct sockaddr_storage *sa,
     }
 #endif

-    addr->u.q_unix.path = g_strndup(su->sun_path, sizeof(su->sun_path));
+    addr->u.q_unix.path = g_strndup(su->sun_path, salen - sizeof(su->sun_family));

looks good to me otherwise

     return addr;
 }
 #endif /* WIN32 */
--
2.30.2


reply via email to

[Prev in Thread] Current Thread [Next in Thread]