From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 896B82566 for ; Tue, 18 Apr 2023 12:32:12 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 0DD52C433D2; Tue, 18 Apr 2023 12:32:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1681821132; bh=kXPbYP1+xg4yHvywccsSVTkUEry5cFNid9V1NlULpbk=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=jV9kmdPOVUWYLy1wrdtWVk+Hc7Ec2sUbkmxZKsxCMQPBRXZN0RmTmwyRfZtGLyKpS qrw5QQnKTXcuAokwDpaIMSLh272yiLZC06vhfHPVWrnw8qHVAsrxHND8mp0/mT9DX+ 4B6H711GsZvXFAltImdjvQIM3bFhu3xAtYnpwu8k= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, syzbot+47c24ca20a2fa01f082e@syzkaller.appspotmail.com, Xin Long , "David S. Miller" , Sasha Levin Subject: [PATCH 5.10 011/124] sctp: check send stream number after wait_for_sndbuf Date: Tue, 18 Apr 2023 14:20:30 +0200 Message-Id: <20230418120310.022051669@linuxfoundation.org> X-Mailer: git-send-email 2.40.0 In-Reply-To: <20230418120309.539243408@linuxfoundation.org> References: <20230418120309.539243408@linuxfoundation.org> User-Agent: quilt/0.67 Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From: Xin Long [ Upstream commit 2584024b23552c00d95b50255e47bd18d306d31a ] This patch fixes a corner case where the asoc out stream count may change after wait_for_sndbuf. When the main thread in the client starts a connection, if its out stream count is set to N while the in stream count in the server is set to N - 2, another thread in the client keeps sending the msgs with stream number N - 1, and waits for sndbuf before processing INIT_ACK. However, after processing INIT_ACK, the out stream count in the client is shrunk to N - 2, the same to the in stream count in the server. The crash occurs when the thread waiting for sndbuf is awake and sends the msg in a non-existing stream(N - 1), the call trace is as below: KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f] Call Trace: sctp_cmd_send_msg net/sctp/sm_sideeffect.c:1114 [inline] sctp_cmd_interpreter net/sctp/sm_sideeffect.c:1777 [inline] sctp_side_effects net/sctp/sm_sideeffect.c:1199 [inline] sctp_do_sm+0x197d/0x5310 net/sctp/sm_sideeffect.c:1170 sctp_primitive_SEND+0x9f/0xc0 net/sctp/primitive.c:163 sctp_sendmsg_to_asoc+0x10eb/0x1a30 net/sctp/socket.c:1868 sctp_sendmsg+0x8d4/0x1d90 net/sctp/socket.c:2026 inet_sendmsg+0x9d/0xe0 net/ipv4/af_inet.c:825 sock_sendmsg_nosec net/socket.c:722 [inline] sock_sendmsg+0xde/0x190 net/socket.c:745 The fix is to add an unlikely check for the send stream number after the thread wakes up from the wait_for_sndbuf. Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations") Reported-by: syzbot+47c24ca20a2fa01f082e@syzkaller.appspotmail.com Signed-off-by: Xin Long Signed-off-by: David S. Miller Signed-off-by: Sasha Levin --- net/sctp/socket.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index e9b4ea3d934fa..3a68d65f7d153 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1831,6 +1831,10 @@ static int sctp_sendmsg_to_asoc(struct sctp_association *asoc, err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len); if (err) goto err; + if (unlikely(sinfo->sinfo_stream >= asoc->stream.outcnt)) { + err = -EINVAL; + goto err; + } } if (sctp_state(asoc, CLOSED)) { -- 2.39.2