netdev - RE: [PATCH 2/2] be2net: drop non-tso frames longer than mtu

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: <1381931680.2045.127.camel@edumazet-glaptop.roam.corp.google.com>
Date:	Wed, 16 Oct 2013 06:54:40 -0700
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Sathya Perla <Sathya.Perla@...lex.Com>
Cc:	"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: RE: [PATCH 2/2] be2net: drop non-tso frames longer than mtu

On Wed, 2013-10-16 at 12:08 +0000, Sathya Perla wrote:

> Won't the mss (gso_size) value for TCP pkts be derived from the mtu of the out-going interface?
> In that case when will gso_size + headers > mtu?


Not on forwarding :(

Let see we have two ethernet devices, with different mtu : 1500 and 1400

GRO could aggregate GRO packets with MSS=1460 (gso_size=1460), then
we hapilly send this GSO packet to the other device, and apparently we
have no check against the lower mtu.

In the best case GRO packet is lost, but in some other cases, the device
hangs...

> In any case, for TSO pkts, the HW just drops pkts when frames after segmentation
> are longer than 9018bytes (or in recent FW versions 9200b)


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html