linux-kernel - Re: [PATCH net-next] tg3: extract GRXRINGS from .get

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] [day] [month] [year] [list]

Message-ID: <CACKFLi=W3ra-sQyAAwekJc6u9aGZ1vV2g44EUF1XYyhCDbBmZw@mail.gmail.com>
Date: Wed, 5 Nov 2025 22:00:50 -0800
From: Michael Chan <michael.chan@...adcom.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Breno Leitao <leitao@...ian.org>, Pavan Chebbi <pavan.chebbi@...adcom.com>, 
	Michael Chan <mchan@...adcom.com>, Andrew Lunn <andrew+netdev@...n.ch>, 
	"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org, linux-kernel@...r.kernel.org, 
	kernel-team@...a.com
Subject: Re: [PATCH net-next] tg3: extract GRXRINGS from .get_rxnfc

On Wed, Nov 5, 2025 at 5:01 PM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Wed, 5 Nov 2025 11:05:34 -0800 Michael Chan wrote:
> > The existing code to use num_online_cpus() is actually not correct.
> > This is more correct:
> >
> > return min(netif_get_num_default_rss_queues(), tp->rxq_max);
> >
> > I think when netif_get_num_default_rss_queues() was used to replace
> > num_online_cpus(), tg3_get_rxnfc() was not properly converted.
>
> All true, but perhaps we want to do that change as a follow up?
> Someone may show up later insisting that fewer queues cases
> a regression for their workload..

Sure, we can fix this separately.  This returned value here when
!netif_running() really won't change the ring count when it becomes
netif_running().  tg3_open() will call tg3_enable_msix() which will
set it to netif_get_num_default_rss_queues() regardless of what value
we return here.  IOW, fixing it should not cause any change in the
number of default rings.

Anyway,
Reviewed-by: Michael Chan <michael.chan@...adcom.com>


>
> The sensitivity to default queue count was why we didn't change most
> of the drivers when netif_get_num_default_rss_queues() got reworked
> to a more sane default than 8.

Download attachment "smime.p7s" of type "application/pkcs7-signature" (5469 bytes)