From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58081) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dUGPe-0000ec-QT for qemu-devel@nongnu.org; Sun, 09 Jul 2017 13:54:31 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dUGPb-0002mb-Ov for qemu-devel@nongnu.org; Sun, 09 Jul 2017 13:54:30 -0400 Received: from orth.archaic.org.uk ([2001:8b0:1d0::2]:37500) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1dUGPb-0002lb-IE for qemu-devel@nongnu.org; Sun, 09 Jul 2017 13:54:27 -0400 From: Peter Maydell Date: Sun, 9 Jul 2017 18:54:22 +0100 Message-Id: <20170709175422.30185-1-peter.maydell@linaro.org> Subject: [Qemu-devel] [PATCH] slirp: fork_exec(): Don't close() a negative number in fork_exec() List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: patches@linaro.org, Samuel Thibault , Jan Kiszka In a fork_exec() error path we try to closesocket(s) when s might be a negative number because the thing that failed was the qemu_socket() call. Add a guard so we don't do this. (Spotted by Coverity: CID 1005727 issue 1 of 2.) Signed-off-by: Peter Maydell --- Issue 2 of 2 in CID 1005727 is trickier -- we need to move as much as possible of the client-end connect/accept out of the child process and into the parent as possible. I'm not sure if it's safe to do it all in the parent without deadlocking... slirp/misc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/slirp/misc.c b/slirp/misc.c index 88e9d94197..260187b6b6 100644 --- a/slirp/misc.c +++ b/slirp/misc.c @@ -112,7 +112,9 @@ fork_exec(struct socket *so, const char *ex, int do_pty) bind(s, (struct sockaddr *)&addr, addrlen) < 0 || listen(s, 1) < 0) { error_report("Error: inet socket: %s", strerror(errno)); - closesocket(s); + if (s >= 0) { + closesocket(s); + } return 0; } -- 2.11.0