netdev - Re: [BUG] kernel panic 4.14.95 in kmem_cache_alloc / build

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: <418df4d9-1f32-b7cd-f903-ff29751e3a22@gmail.com>
Date:   Wed, 6 Feb 2019 10:48:13 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Frank Wunderlich <frank-w@...lic-files.de>, netdev@...r.kernel.org
Subject: Re: [BUG] kernel panic 4.14.95 in kmem_cache_alloc / build_skb



On 02/06/2019 09:36 AM, Frank Wunderlich wrote:
> Hi,
> 
> linux-net-mailinglist for reporting bugs seems not existing anymore and i have not found out the right one..."scripts/get_maintainer.pl -f ./include/linux/skbuff.h" shows only linux-kernel as mailing list which is not specific for network, so sorry, if i'm in the wrong mailinglist.
> 
> i've recently noticed panics on my bananapi-r2 with kernel 4.14.95, but can't locate them in detail...all is pointing to kmem_cache_alloc and build_skb, but over complete different ways. so i guess it's network-related
> 
> i currently don't know what is triggering this crashes, so i cannot reproduce them on my test-machine :(
> 
> it's not only an oops...the machine stays still and have to be resettet, sometimes after an half hour running
> 
> i had kernel 4.14.78 running before, so i think it's a change between these 2 Kernel-versions and no OS-related (debian-stretch) or Hardware-issue
> 
> currently i test with 4.14.90, if it has same behaviour...at the moment it seems stable...
> 
> i cannot test it with another kernel-branch (4.19+), because i need to test it on my productive machine till i find a way to reproduce it....maybe you can see a hint in stacktrace to guide me...at the moment crash occours unexpectely
> 
> i looked through git-tree but did not found anything related
> 
> regards Frank
> 
> here my 3 recorded crash-logs...maybe you can point me in the right direction (reproduce/debug/fix):

Make sure to try 4.14.98, as I am guessing this has been fixed by :

dc489ad6a2f2fcebdaecb7b8532e0d02d272ac6a Fix "net: ipv4: do not handle duplicate fragments as overlapping"