netdev - Re: [PATCH bpf v2] selftests/bpf: remove bash feature in test_xdp

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] [day] [month] [year] [list]

Message-ID: <36d7e0f5-e3e4-637e-d970-1dc69d1055c0@infradead.org>
Date:   Mon, 8 Feb 2021 22:54:49 -0800
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Björn Töpel <bjorn.topel@...el.com>,
        Andrii Nakryiko <andrii.nakryiko@...il.com>,
        Björn Töpel <bjorn.topel@...il.com>
Cc:     Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        u9012063@...il.com
Subject: Re: [PATCH bpf v2] selftests/bpf: remove bash feature in
 test_xdp_redirect.sh

On 2/8/21 10:41 PM, Björn Töpel wrote:
> On 2021-02-09 06:52, Andrii Nakryiko wrote:
>> On Sat, Feb 6, 2021 at 1:29 AM Björn Töpel <bjorn.topel@...il.com> wrote:
>>>
>>> From: Björn Töpel <bjorn.topel@...el.com>
>>>
>>> The test_xdp_redirect.sh script uses a bash redirect feature,
>>> '&>/dev/null'. Use '>/dev/null 2>&1' instead.
>>
>> We have plenty of explicit bash uses in selftest scripts, I'm not sure
>> it's a good idea to make scripts more verbose.
>>
> 
> $ cd tools/testing/selftests
> $ git grep '\#!/bin/bash'|wc -l
> 282
> $ git grep '\#!/bin/sh'|wc -l
> 164
> 
> Andrii/Randy, I'm fine with whatever. I just want to be able to run the
> test on Debian-derived systems. ;-)
> 
> 
>>>
>>> Also remove the 'set -e' since the script actually relies on that the
>>> return value can be used to determine pass/fail of the test.
>>
>> This sounds like a dubious decision. The script checks return results
>> only of last two commands, for which it's better to write and if
>> [<first command>] && [<second command>] check and leave set -e intact.
>>
> 
> Ok!
> 
> Please decide on the shell flavor, and I'll respin a v3.

In general shell scripts in the kernel try not to use bash (we have taken
several patches to convert from /bin/bash to /bin/sh scripts).
OTOH, perf and bpf seem to be large exceptions to this trend,
so it is apparently OK to use bash. :)
Sorry to sidetrack you.

-- 
~Randy