netdev - Re: [RFC net-next 08/22] eth: bnxt: support setting size of agg buffers via ethtool

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: <20250425155849.0e3a6109@kernel.org>
Date: Fri, 25 Apr 2025 15:58:49 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Mina Almasry <almasrymina@...gle.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, edumazet@...gle.com,
 pabeni@...hat.com, andrew+netdev@...n.ch, horms@...nel.org,
 donald.hunter@...il.com, sdf@...ichev.me, dw@...idwei.uk,
 asml.silence@...il.com, ap420073@...il.com, jdamato@...tly.com,
 dtatulea@...dia.com, michael.chan@...adcom.com
Subject: Re: [RFC net-next 08/22] eth: bnxt: support setting size of agg
 buffers via ethtool

On Wed, 23 Apr 2025 14:00:01 -0700 Mina Almasry wrote:
> Where does the power of 2 limitation come from? bnxt itself? Or some mp issue?
> 
> dmabuf mp can do non-power of 2 rx_buf_len, I think. I haven't tested
> recently. It may be good to only validate here what bnxt can't do at
> all, and let a later check in the pp/mp let us know if the mp doesn't
> like the size.

I haven't actually tested anything else, but no real reason at this
point. I was wondering if it's worth trying to allow 64k - 1 since
that'd still fit on 16 bits. But left that for future work, cause
it will make all copy offsets funny-sized. We'd probably want something
like 64k - 4k ? Dunno, either way slightly unpleasant. There's probably
more that can be done from the NIC side.