linux-kernel - Re: [PATCH] selftests/mm: run_vmtests.sh: fix relative path handling

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: <3cd08409-8c6b-42ab-97d5-e4440efe4f6a@kernel.org>
Date: Fri, 9 Jan 2026 14:41:51 +0100
From: "David Hildenbrand (Red Hat)" <david@...nel.org>
To: sun jian <sun.jian.kdev@...il.com>,
 Andrew Morton <akpm@...ux-foundation.org>
Cc: Shuah Khan <shuah@...nel.org>, linux-kselftest@...r.kernel.org,
 linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] selftests/mm: run_vmtests.sh: fix relative path handling

On 1/9/26 03:08, sun jian wrote:
>> hm, why?  Is that a thing people actually do?
>>
>> Is anyone going to actually test this feature?
> 
> Yes — invoking selftests directly from the kernel root can easily happen in
> practice, for example::
> 
>    sudo tools/testing/selftests/mm/run_vmtests.sh
> 
> This currently results in false failures because relative paths being resolved
> against the caller's cwd instead of the script directory.
>>
>> Alternatively we could check that we're in the correct directory and
>> error out if not.
> 
> That would also be reasonable, but I slightly prefer auto-cd because it
> avoids an easy invocation pitfall and makes the runner more robust for
> wrappers/CI
> where the cwd is  not stable. That said, I'm happy to switch to a fail-fast cwd
> check if you prefer the behavior.

I'd prefer to just fail for the case that we never supported instead of 
adding support for it.

-- 
Cheers

David