netdev - Re: [PATCH] gianfar: fix jumbo packets+napi+rx overrun crash

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:   Sun, 07 Mar 2021 21:51:59 +0100
From:   michael-dev <michael-dev@...i-braun.de>
To:     Claudiu Manoil <claudiu.manoil@....com>
Cc:     netdev@...r.kernel.org
Subject: Re: [PATCH] gianfar: fix jumbo packets+napi+rx overrun crash

Hi,

sorry I missed the mail.

Am 04.03.2021 11:05, schrieb Claudiu Manoil:
> Could you help provide some context, e.g.:
> On what board/soc were you able to trigger this issue?

I have an OpenWRT running on P1020WLAN boards and have IPsec for some 
gretap tunnel configured.
I run iperf3 -s on the AP and iperf3 -c --udp -b 1000M on some server 
and then the AP reliably crashes.
It also crashed sometimes during normal operations when doing some fast 
download over the IPsec tunnel, but that was hard to reproduce.

> How often does the overrun occur?
Reliably within seconds with the above test.

> What's the use case? Is the issue triggered with smaller packets than 
> 9600B?
It cannot be triggered with 1500 Byte  MTU packets as these have FIRST 
and LAST set in one.
I use jumbo frames to speed up ipsec.

> Increasing the Rx ring size does significantly reduce ring overruns?

I did not test this.

Regards,
Michael Braun