netdev - Broken SOF_TIMESTAMPING_OPT_ID in linux-4.19.y and earlier stable branches

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20220324213954.3ln7kvl5utadnux6@skbuf>
Date:   Thu, 24 Mar 2022 23:39:54 +0200
From:   Vladimir Oltean <olteanv@...il.com>
To:     Willem de Bruijn <willemdebruijn.kernel@...il.com>
Cc:     netdev@...r.kernel.org, Paolo Abeni <pabeni@...hat.com>,
        Jakub Kicinski <kuba@...nel.org>,
        "David S. Miller" <davem@...emloft.net>
Subject: Broken SOF_TIMESTAMPING_OPT_ID in linux-4.19.y and earlier stable
 branches

Hello Willem,

I have an application which makes use of SOF_TIMESTAMPING_OPT_ID, and I
received reports from multiple users that all timestamps are delivered
with a tskey of 0 for all stable kernel branches earlier than, and
including, 4.19.

I bisected this issue down to:

| commit 8f932f762e7928d250e21006b00ff9b7718b0a64 (HEAD)
| Author: Willem de Bruijn <willemb@...gle.com>
| Date:   Mon Dec 17 12:24:00 2018 -0500
| 
|     net: add missing SOF_TIMESTAMPING_OPT_ID support
| 
|     SOF_TIMESTAMPING_OPT_ID is supported on TCP, UDP and RAW sockets.
|     But it was missing on RAW with IPPROTO_IP, PF_PACKET and CAN.
| 
|     Add skb_setup_tx_timestamp that configures both tx_flags and tskey
|     for these paths that do not need corking or use bytestream keys.
| 
|     Fixes: 09c2d251b707 ("net-timestamp: add key to disambiguate concurrent datagrams")
|     Signed-off-by: Willem de Bruijn <willemb@...gle.com>
|     Acked-by: Soheil Hassas Yeganeh <soheil@...gle.com>
|     Signed-off-by: David S. Miller <davem@...emloft.net>

and, interestingly, I found this discussion on the topic:
https://www.spinics.net/lists/netdev/msg540752.html
(copied here in case the link rots in the future)

| > Series applied.
| >
| > What is your opinion about -stable for this?
| 
| Thanks David. Since these are just missing features that no one has
| reported as actually having been missing a whole lot, I don't think
| that they are worth the effort or risk.

So I have 2 questions:

Is there a way for user space to validate functional kernel support for
SOF_TIMESTAMPING_OPT_ID? What I'm noticing is that (at least with
AF_PACKET sockets) the "level == SOL_PACKET && type == PACKET_TX_TIMESTAMP"
cmsg is _not_ missing, but instead contains a valid sock_err->ee_data
(tskey) of 0.

If it's not possible, could you please consider sending these fixes as
patches to linux-stable?