[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <a43feb58-5fa2-4b39-8aa1-04085cd5a8d6@quicinc.com>
Date: Thu, 29 Feb 2024 15:15:04 -0800
From: "Abhishek Chauhan (ABC)" <quic_abchauha@...cinc.com>
To: Paolo Abeni <pabeni@...hat.com>, "David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>,
<netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
Andrew Halaney
<ahalaney@...hat.com>,
Willem de Bruijn <willemdebruijn.kernel@...il.com>,
Martin KaFai Lau <martin.lau@...nel.org>,
Martin KaFai Lau
<martin.lau@...ux.dev>
CC: <kernel@...cinc.com>
Subject: Re: [PATCH net-next v2] net: Modify mono_delivery_time with
clockid_delivery_time
On 2/29/2024 1:30 AM, Paolo Abeni wrote:
> On Tue, 2024-02-27 at 17:12 -0800, Abhishek Chauhan wrote:
>> Bridge driver today has no support to forward the userspace timestamp
>> packets and ends up resetting the timestamp. ETF qdisc checks the
>> packet coming from userspace and encounters to be 0 thereby dropping
>> time sensitive packets. These changes will allow userspace timestamps
>> packets to be forwarded from the bridge to NIC drivers.
>>
>> Existing functionality of mono_delivery_time is not altered here
>> instead just extended with userspace tstamp support for bridge
>> forwarding path.
>>
>> Signed-off-by: Abhishek Chauhan <quic_abchauha@...cinc.com>
>
> Since this is touching a quite sensitive code path, I think v3 would
> deserve some paired self-tests.
>
I did verify the existing so_txtime selftest for the same which does FQ
and ETF tests which sends/recieve packets with both mono and tai clocks.
OK. All tests passed (so_txtime.sh)
Sometimes i saw resource temporarily unavailable in RX which i believe is a
known issue
https://www.mail-archive.com/linux-kselftest@vger.kernel.org/msg06792.html
Along with this i also did mqprio + etf offload testing on internal platform.
1. iperf3 best effort testing UDP/TCP
2. Sent ETF packets from userspace application just like so_txtime.c but
with SO_PRIORITY and added dumpstack to check the flow of packets. I saw no
issues on internal platform.
3. I will mention all the testing in Tx/Rx path done for v3 patch.
> Thanks!
>
> Paolo
>
Powered by blists - more mailing lists