[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250624165039.5bf4b3d9@kernel.org>
Date: Tue, 24 Jun 2025 16:50:39 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: willemb@...gle.com, Joe Damato <joe@...a.to>
Cc: Samiullah Khawaja <skhawaja@...gle.com>, "David S . Miller "
<davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, Paolo Abeni
<pabeni@...hat.com>, almasrymina@...gle.com, mkarsten@...terloo.ca,
netdev@...r.kernel.org
Subject: Re: [PATCH net-next v9] Add support to set NAPI threaded for
individual NAPI
[swapping Joe's email for a working one]
On Mon, 23 Jun 2025 17:53:16 +0000 Samiullah Khawaja wrote:
> A net device has a threaded sysctl that can be used to enable threaded
> NAPI polling on all of the NAPI contexts under that device. Allow
> enabling threaded NAPI polling at individual NAPI level using netlink.
>
> Extend the netlink operation `napi-set` and allow setting the threaded
> attribute of a NAPI. This will enable the threaded polling on a NAPI
> context.
>
> Add a test in `nl_netdev.py` that verifies various cases of threaded
> NAPI being set at NAPI and at device level.
LGTM, but as we discussed many version ago my subjective preference
would be for the per-queue setting to be tri-state (unset=follow the
device setting, enabled, disabled). Rather than have the device level
act as an override when written. It sounded like Willem and Joe thought
that's too complicated and diverges from the existing behavior.
Which is fine by me. But I'd like to see some review tags from them
now :)
Gentlemen?
Powered by blists - more mailing lists