[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <54a8a9fa-9717-435e-9253-40f3a0a7f779@davidwei.uk>
Date: Wed, 9 Apr 2025 19:06:13 -0700
From: David Wei <dw@...idwei.uk>
To: Jakub Kicinski <kuba@...nel.org>
Cc: netdev@...r.kernel.org, Andrew Lunn <andrew+netdev@...n.ch>,
Paolo Abeni <pabeni@...hat.com>, "David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>
Subject: Re: [PATCH net-next] io_uring/zcrx: enable tcp-data-split in selftest
On 2025-04-09 17:06, Jakub Kicinski wrote:
> On Wed, 9 Apr 2025 09:31:53 -0700 David Wei wrote:
>> For bnxt when the agg ring is used then tcp-data-split is automatically
>> reported to be enabled, but __net_mp_open_rxq() requires tcp-data-split
>> to be explicitly enabled by the user.
>
>> diff --git a/tools/testing/selftests/drivers/net/hw/iou-zcrx.py b/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
>> index 9f271ab6ec04..6a0378e06cab 100755
>> --- a/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
>> +++ b/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
>> @@ -35,6 +35,7 @@ def test_zcrx(cfg) -> None:
>> rx_ring = _get_rx_ring_entries(cfg)
>>
>> try:
>> + ethtool(f"-G {cfg.ifname} tcp-data-split on", host=cfg.remote)
>
> You should really use defer() to register the "undo" actions
> individually. Something like:
>
> ethtool(f"-G {cfg.ifname} tcp-data-split on", host=cfg.remote)
> defer(ethtool, f"-G {cfg.ifname} rx {rx_ring}", host=cfg.remote)
> ethtool(f"-G {cfg.ifname} rx 64", host=cfg.remote)
> defer(ethtool, f"-G {cfg.ifname} rx {rx_ring}", host=cfg.remote)
> ethtool(f"-X {cfg.ifname} equal {combined_chans - 1}", host=cfg.remote)
> defer(ethtool, f"-X {cfg.ifname} default", host=cfg.remote)
> ...
>
> This patch is fine. But could you follow up and convert the test fully?
I'll send a follow up, one to switch to defer(), then another to call
tcp-data-split on.
Powered by blists - more mailing lists