netdev - Re: [PATCH v2 5/5] io_uring/netcmd: add tx timestamping cmd support

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

Message-ID: <db7e43e3-89bf-4ec4-a225-0c352b6fc306@gmail.com>
Date: Wed, 4 Jun 2025 13:33:17 +0100
From: Pavel Begunkov <asml.silence@...il.com>
To: Jens Axboe <axboe@...nel.dk>, io-uring@...r.kernel.org,
 Vadim Fedorenko <vadim.fedorenko@...ux.dev>
Cc: netdev@...r.kernel.org, Eric Dumazet <edumazet@...gle.com>,
 Kuniyuki Iwashima <kuniyu@...zon.com>, Paolo Abeni <pabeni@...hat.com>,
 Willem de Bruijn <willemb@...gle.com>, "David S . Miller"
 <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>,
 Richard Cochran <richardcochran@...il.com>
Subject: Re: [PATCH v2 5/5] io_uring/netcmd: add tx timestamping cmd support

On 6/4/25 13:04, Jens Axboe wrote:
> On 6/4/25 2:42 AM, Pavel Begunkov wrote:
>> Add a new socket command which returns tx time stamps to the user. It
>> provide an alternative to the existing error queue recvmsg interface.
>> The command works in a polled multishot mode, which means io_uring will
>> poll the socket and keep posting timestamps until the request is
>> cancelled or fails in any other way (e.g. with no space in the CQ). It
>> reuses the net infra and grabs timestamps from the socket's error queue.
>>
>> The command requires IORING_SETUP_CQE32. All non-final CQEs (marked with
>> IORING_CQE_F_MORE) have cqe->res set to the tskey, and the upper 16 bits
>> of cqe->flags keep tstype (i.e. offset by IORING_CQE_BUFFER_SHIFT). The
>> timevalue is store in the upper part of the extended CQE. The final
>> completion won't have IORING_CQR_F_MORE and will have cqe->res storing
>                           ^^
> 
> CQE_F_MORE
> 
> Minor nit below.
> 
>> +static bool io_process_timestamp_skb(struct io_uring_cmd *cmd, struct sock *sk,
>> +				     struct sk_buff *skb, unsigned issue_flags)
>> +{
>> +	struct sock_exterr_skb *serr = SKB_EXT_ERR(skb);
>> +	struct io_uring_cqe cqe[2];
>> +	struct io_timespec *iots;
>> +	struct timespec64 ts;
>> +	u32 tskey;
>> +
>> +	BUILD_BUG_ON(sizeof(struct io_uring_cqe) != sizeof(struct io_timespec));
>> +
>> +	if (!skb_get_tx_timestamp(skb, sk, &ts))
>> +		return false;
>> +
>> +	tskey = serr->ee.ee_data;
>> +
>> +	cqe->user_data = 0;
>> +	cqe->res = tskey;
>> +	cqe->flags = IORING_CQE_F_MORE;
>> +	cqe->flags |= (u32)serr->ee.ee_info << IORING_CQE_BUFFER_SHIFT;
> 
> Get rid of the tskey variable?

Why? I named it specifically so that it's obvious what the field
means, and "cqe->res = serr->ee.ee_data" hardly tells the meaning
of the fields without extra digging.

And I think this would be more easily
> readable if it used:
> 
> 	cqe[0].user_data = 0;

I don't see how it'd be easier to read by cluttering the code
with "[0]" in multiple statements. Combined with that it's
one extended cqe and not some array of them.
  > etc.
> 
>> +	iots = (struct io_timespec *)&cqe[1];
>> +	iots->tv_sec = ts.tv_sec;
>> +	iots->tv_nsec = ts.tv_nsec;
>> +	return io_uring_cmd_post_mshot_cqe32(cmd, issue_flags, cqe);
>> +}
> 
> A bit of a shame we can't just get the double CQE and fill it in, rather
> than fill it on stack and copy it. But probably doesn't matter much.
> 

-- 
Pavel Begunkov