[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANn89iK40zw8hqtut8u9Jp4+24eCM9m9S-kwdDNRq-P=jkF=ng@mail.gmail.com>
Date: Thu, 16 May 2024 10:04:05 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: "Subash Abhinov Kasiviswanathan (KS)" <quic_subashab@...cinc.com>
Cc: soheil@...gle.com, ncardwell@...gle.com, yyd@...gle.com, ycheng@...gle.com,
quic_stranche@...cinc.com, davem@...emloft.net, kuba@...nel.org,
netdev@...r.kernel.org
Subject: Re: Potential impact of commit dfa2f0483360 ("tcp: get rid of sysctl_tcp_adv_win_scale")
On Thu, May 16, 2024 at 9:16 AM Subash Abhinov Kasiviswanathan (KS)
<quic_subashab@...cinc.com> wrote:
> rmnet doesn't directly interface with HW. It is a virtual driver which
> attaches over real hardware drivers like MHI (PCIe), QMI_WWAN (USB), IPA
> to expose networking across different mobile APNs.
>
> As rmnet didn't have it's own NAPI struct, I added support for GRO using
> cells. I tried disabling GRO and I don't see a difference in download
> speeds or the receiver window either.
If you do not use native NAPI GRO but gro_cells instead, there is no
reordering avoidance.
Make sure that only one cpu is feeding packets to the gro_cell for a
given RX queue.
This is probably orthogonal, but could matter at high speeds.
Powered by blists - more mailing lists