netdev - Re: [PATCH net] net: diag: Don't double-free TCP_NEW_SYN_RECV sockets in tcp

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: <86b06140-40c0-7e55-f1d9-db6aec372302@cumulusnetworks.com>
Date:   Mon, 9 Jul 2018 09:00:39 -0600
From:   David Ahern <dsa@...ulusnetworks.com>
To:     Lorenzo Colitti <lorenzo@...gle.com>
Cc:     Eric Dumazet <eric.dumazet@...il.com>,
        David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
        Alistair Strachan <astrachan@...gle.com>,
        Subash Abhinov Kasiviswanathan <subashab@...eaurora.org>
Subject: Re: [PATCH net] net: diag: Don't double-free TCP_NEW_SYN_RECV sockets
 in tcp_abort

On 7/8/18 11:24 PM, Lorenzo Colitti wrote:
> On Sat, Jul 7, 2018 at 10:56 PM David Ahern <dsa@...ulusnetworks.com> wrote:
>>> Here is a packetdrill test :
>>
>> So I have to either learn how to use packetdrill or install a web server
>> and put load on it. If the Android tests are not publicly available then
>> the reference should be removed from the commit log.
> 
> It is publicly available. The test is at
> https://android.googlesource.com/kernel/tests/+/master/net/test/sock_diag_test.py
> . Documentation is at
> https://source.android.com/devices/architecture/kernel/network_tests .
> 
> Note that not all the tests pass on upstream kernels. We do try to
> keep them working, but because they are also run as conformance tests
> for Android devices they must test out-of-tree behaviour, such as the
> fact that unlike upstream, Android kernels use RFC-compliant SHA256
> truncation when the PF_KEY API is used.
> 
> Again, I don't think you'll see this particular issue on the ARCH=um
> tests that run by default. But if you run sock_diag_test.py on a VM or
> physical machine as root, you should.
> 

ok. thanks for reference.

I general I am always looking for more easy to run tests to catch
problems like this.