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: <352s5hjotedb3lfgy5hajbljjjzbsksr7tvuid6b6fes5s7wqg@nan3d5n64tgv>
Date: Fri, 13 Dec 2024 12:58:28 +0100
From: Stefano Garzarella <sgarzare@...hat.com>
To: Michal Luczaj <mhal@...x.co>
Cc: netdev@...r.kernel.org
Subject: Re: [PATCH net-next 0/4] vsock/test: Tests for memory leaks

On Fri, Dec 13, 2024 at 01:41:56AM +0100, Michal Luczaj wrote:
>On 12/10/24 17:25, Stefano Garzarella wrote:
>>> [...]
>>> I initially considered triggering (and parsing) a kmemleak scan after each
>>> test, but ultimately concluded that the slowdown and the required
>>> privileges would be too much.
>>
>> Yeah, what about adding something in the README to suggest using
>> kmemleak and how to check that everything is okay after a run?
>
>Something like this?
>
>diff --git a/tools/testing/vsock/README b/tools/testing/vsock/README
>index 84ee217ba8ee..0d6e73ecbf4d 100644
>--- a/tools/testing/vsock/README
>+++ b/tools/testing/vsock/README
>@@ -36,6 +36,21 @@ Invoke test binaries in both directions as follows:
>                        --control-port=1234 \
>                        --peer-cid=3
>
>+Some tests are designed to produce kernel memory leaks. Leaks detection,
>+however, is deferred to Kernel Memory Leak Detector. It is recommended to enable
>+kmemleak (CONFIG_DEBUG_KMEMLEAK=y) and explicitly trigger a scan after each test
>+run, e.g.
>+
>+  # echo clear > /sys/kernel/debug/kmemleak
>+  # $TEST_BINARY ...
>+  # echo "wait for any grace periods" && sleep 2
>+  # echo scan > /sys/kernel/debug/kmemleak
>+  # echo "wait for kmemleak" && sleep 5
>+  # echo scan > /sys/kernel/debug/kmemleak
>+  # cat /sys/kernel/debug/kmemleak
>+
>+For more information see Documentation/dev-tools/kmemleak.rst.
>+

Yep, this would be great!

> vsock_perf utility
> -------------------
> 'vsock_perf' is a simple tool to measure vsock performance. It works in
>
>> I'd suggest also to add something about that in each patch that
>> introduce tests where we expects the user to check kmemleak,
>> at least with a comment on top of the test functions, and maybe
>> also in the commit description.
>
>Sure, will do.

Thanks,
Stefano


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ