Netdev List
 help / color / mirror / Atom feed
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
To: Vadim Fedorenko <vadim.fedorenko@linux.dev>,
	 Willem de Bruijn <willemdebruijn.kernel@gmail.com>,
	 Vadim Fedorenko <vadfed@meta.com>,
	 Willem de Bruijn <willemb@google.com>,
	 Jakub Kicinski <kuba@kernel.org>,
	 Paolo Abeni <pabeni@redhat.com>,
	 David Ahern <dsahern@kernel.org>,
	 Jason Xing <kerneljasonxing@gmail.com>,
	 Simon Horman <horms@kernel.org>
Cc: netdev@vger.kernel.org
Subject: Re: [PATCH net-next v3 1/4] net_tstamp: add SCM_TS_OPT_ID to provide OPT_ID in control message
Date: Thu, 05 Sep 2024 09:29:23 -0400	[thread overview]
Message-ID: <66d9b23322f2f_18ac212944a@willemb.c.googlers.com.notmuch> (raw)
In-Reply-To: <1946af56-9f6f-439d-b954-6bcb82367741@linux.dev>

Vadim Fedorenko wrote:
> On 04/09/2024 21:54, Willem de Bruijn wrote:
> > Vadim Fedorenko wrote:
> >> SOF_TIMESTAMPING_OPT_ID socket option flag gives a way to correlate TX
> >> timestamps and packets sent via socket. Unfortunately, there is no way
> >> to reliably predict socket timestamp ID value in case of error returned
> >> by sendmsg. For UDP sockets it's impossible because of lockless
> >> nature of UDP transmit, several threads may send packets in parallel. In
> >> case of RAW sockets MSG_MORE option makes things complicated. More
> >> details are in the conversation [1].
> >> This patch adds new control message type to give user-space
> >> software an opportunity to control the mapping between packets and
> >> values by providing ID with each sendmsg for UDP sockets.
> >> The documentation is also added in this patch.
> >>
> >> [1] https://lore.kernel.org/netdev/CALCETrU0jB+kg0mhV6A8mrHfTE1D1pr1SD_B9Eaa9aDPfgHdtA@mail.gmail.com/
> >>
> >> Signed-off-by: Vadim Fedorenko <vadfed@meta.com>

> >> @@ -1325,8 +1330,11 @@ static int ip_setup_cork(struct sock *sk, struct inet_cork *cork,
> >>   	cork->mark = ipc->sockc.mark;
> >>   	cork->priority = ipc->priority;
> >>   	cork->transmit_time = ipc->sockc.transmit_time;
> >> +	cork->ts_opt_id = ipc->sockc.ts_opt_id;
> >>   	cork->tx_flags = 0;
> >>   	sock_tx_timestamp(sk, ipc->sockc.tsflags, &cork->tx_flags);
> >> +	if (ipc->sockc.tsflags & SOCKCM_FLAG_TS_OPT_ID)
> >> +		cork->flags |= IPCORK_TS_OPT_ID;
> > 
> > We can move initialization of ts_opt_id into the branch.
> > 
> > Or conversely avoid the branch with some convoluted shift operations
> > to have the rval be either 1 << 1 or 0 << 1. But let's do the simpler
> > thing.
> 
> What is the reason to move initialization behind the flag? We are not
> doing this for transmit_time even though it's also used with flag only.
> 
> It's not a big deal to change, I just wonder what are the benefits?

Just avoid one assignment in the hot path that does not use this
feature.

cork->ts_opt_id is only valuid if cork->flags & IPCORK_TS_OPT_ID.



  reply	other threads:[~2024-09-05 13:29 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-04 11:31 [PATCH net-next v3 0/4] Add option to provide OPT_ID value via cmsg Vadim Fedorenko
2024-09-04 11:31 ` [PATCH net-next v3 1/4] net_tstamp: add SCM_TS_OPT_ID to provide OPT_ID in control message Vadim Fedorenko
2024-09-04 13:56   ` Jason Xing
2024-09-04 14:01     ` Vadim Fedorenko
2024-09-04 20:54   ` Willem de Bruijn
2024-09-05 10:10     ` Vadim Fedorenko
2024-09-05 13:29       ` Willem de Bruijn [this message]
2024-09-05  8:24   ` Jason Xing
2024-09-05  8:34     ` Vadim Fedorenko
2024-09-05  8:49       ` Jason Xing
2024-09-06  9:22   ` kernel test robot
2024-09-06 12:58   ` kernel test robot
2024-09-04 11:31 ` [PATCH net-next v3 2/4] net_tstamp: add SCM_TS_OPT_ID for TCP sockets Vadim Fedorenko
2024-09-05 14:58   ` Vadim Fedorenko
2024-09-05 15:37     ` Jason Xing
2024-09-05 16:39     ` Willem de Bruijn
2024-09-06 12:20       ` Vadim Fedorenko
2024-09-06 15:25         ` Willem de Bruijn
2024-09-06 16:33           ` Willem de Bruijn
2024-09-06 17:27             ` Vadim Fedorenko
2024-09-06 23:50               ` Willem de Bruijn
2024-09-06 17:27           ` Vadim Fedorenko
2024-09-06 23:48             ` Willem de Bruijn
2024-09-07 21:55               ` Vadim Fedorenko
2024-09-08 19:19                 ` Willem de Bruijn
2024-09-08 19:55                   ` Vadim Fedorenko
2024-09-04 11:31 ` [PATCH net-next v3 3/4] net_tstamp: add SCM_TS_OPT_ID for RAW sockets Vadim Fedorenko
2024-09-04 11:31 ` [PATCH net-next v3 4/4] selftests: txtimestamp: add SCM_TS_OPT_ID test Vadim Fedorenko
2024-09-04 11:36 ` [PATCH net-next v3 0/4] Add option to provide OPT_ID value via cmsg Vadim Fedorenko
2024-09-04 21:04   ` Willem de Bruijn
2024-09-04 22:50     ` Vadim Fedorenko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=66d9b23322f2f_18ac212944a@willemb.c.googlers.com.notmuch \
    --to=willemdebruijn.kernel@gmail.com \
    --cc=dsahern@kernel.org \
    --cc=horms@kernel.org \
    --cc=kerneljasonxing@gmail.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=vadfed@meta.com \
    --cc=vadim.fedorenko@linux.dev \
    --cc=willemb@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox