netdev - Re: [PATCH v2] bnx2x: disable GSO where gso

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 16 Jan 2018 11:12:44 +1100
From:   Daniel Axtens <dja@...ens.net>
To:     David Miller <davem@...emloft.net>
Cc:     netdev@...r.kernel.org, tlfalcon@...ux.vnet.ibm.com,
        Yuval.Mintz@...ium.com
Subject: Re: [PATCH v2] bnx2x: disable GSO where gso_size is too big for hardware

David Miller <davem@...emloft.net> writes:

> From: Daniel Axtens <dja@...ens.net>
> Date: Fri, 12 Jan 2018 10:59:05 +1100
>
>> If a bnx2x card is passed a GSO packet with a gso_size larger than
>> ~9700 bytes, it will cause a firmware error that will bring the card
>> down:
>> 
>> bnx2x: [bnx2x_attn_int_deasserted3:4323(enP24p1s0f0)]MC assert!
>> bnx2x: [bnx2x_mc_assert:720(enP24p1s0f0)]XSTORM_ASSERT_LIST_INDEX 0x2
>> bnx2x: [bnx2x_mc_assert:736(enP24p1s0f0)]XSTORM_ASSERT_INDEX 0x0 = 0x00000000 0x25e43e47 0x00463e01 0x00010052
>> bnx2x: [bnx2x_mc_assert:750(enP24p1s0f0)]Chip Revision: everest3, FW Version: 7_13_1
>> ... (dump of values continues) ...
>> 
>> Detect when gso_size + header length is greater than the maximum
>> packet size (9700 bytes) and disable GSO. For simplicity and speed
>> this is approximated by comparing gso_size against 9200 and assuming
>> no-one will have more than 500 bytes of headers.
>
> What is the MTU size configured on the bnx2x device when these 9700
> byte packets are seen?
>
> If it's less than 9700, whatever is allowing your device (openvswitch,
> ibmveth, whatever) needs to be fixed.

Sure, I had an approach that checks the gso_size in is_skb_forwardable
and the equivalent openvswitch path - I'll send that.

Regards,
Daniel

>
> I don't like this at all, quite frankly.  We'll have one device now that
> has this special check, probably many others can run into this situation
> as well but they won't be used on these kinds of powerpc boxes and
> therefore nobody is going to notice.
>
> I'm not applying this without more information or better justification,
> sorry.