netdev - Re: [PATCH net v2] net: dsa: tag_rtl4

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: <20231030152325.qdpvv4nbczhal35c@skbuf>
Date: Mon, 30 Oct 2023 17:23:25 +0200
From: Vladimir Oltean <olteanv@...il.com>
To: Linus Walleij <linus.walleij@...aro.org>
Cc: Andrew Lunn <andrew@...n.ch>, Florian Fainelli <f.fainelli@...il.com>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net v2] net: dsa: tag_rtl4_a: Bump min packet size

On Mon, Oct 30, 2023 at 03:37:24PM +0100, Linus Walleij wrote:
> On Mon, Oct 30, 2023 at 3:16 PM Vladimir Oltean <olteanv@...il.com> wrote:
> 
> > Could you please try to revert the effect of commit 339133f6c318 ("net:
> > dsa: tag_rtl4_a: Drop bit 9 from egress frames") by setting that bit in
> > the egress tag again? Who knows, maybe it is the bit which tells the
> > switch to bypass the forwarding process.
> 
> I have already tried that, it was one of the first things I tried,
> just looking over the git log and looking for usual suspects.
> 
> Sadly it has no effect whatsoever, the problem persists :(
> 
> > Or maybe there is a bit in a
> > different position which does this. You could try to fill in all bits in
> > unknown positions, check that there are no regressions with packet sizes
> > < 1496, and then see if that made any changes to packet sizes >= 1496.
> > If it did, try to see which bit made the difference.
> 
> Hehe now we're talking :D
> 
> I did something similar before, I think just switching a different bit
> every 10 packets or so and running a persistent ping until it succeeds
> or not.
> 
> I'll see what I can come up with.
> 
> Yours,
> Linus Walleij

And the drop reason in ethtool -S also stays unchanged despite all the
extra bits set in the tag? It still behaves as if the packet goes
through the forwarding path?