Re: [PATCH] selftests/livepatch: add test skip handling

From: Joe Lawrence <joe.lawrence@redhat.com>
To: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>,
	live-patching@vger.kernel.org, linux-kselftest@vger.kernel.org
Cc: shuah@kernel.org
Subject: Re: [PATCH] selftests/livepatch: add test skip handling
Date: Mon, 15 Jul 2019 10:09:58 -0400	[thread overview]
Message-ID: <b675ce40-40c8-69aa-0767-dc1af9fc7797@redhat.com> (raw)
In-Reply-To: <9de46fed-785c-d5c4-8a76-205674bd0912@linux.vnet.ibm.com>

On 7/15/19 2:50 AM, Kamalesh Babulal wrote:
> On 7/14/19 7:58 PM, Joe Lawrence wrote:
>> Before running a livpeatch self-test, first verify that we've built and
>> installed the livepatch self-test kernel modules by running a 'modprobe
>> --dry-run'.  This should catch a few environment issues, including
>> !CONFIG_LIVEPATCH and !CONFIG_TEST_LIVEPATCH.  In these cases, exit
>> gracefully with test-skip status rather than test-fail status.
>>
>> Reported-by: Jiri Benc <jbenc@redhat.com>
>> Suggested-by: Shuah Khan <shuah@kernel.org>
>> Signed-off-by: Joe Lawrence <joe.lawrence@redhat.com>
> 
> Reviewed-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>
> 
> [...]
>>
>> +function assert_mod() {
>> +	local mod="$1"
>> +
>> +	if ! modprobe --dry-run "$mod" &>/dev/null ; then
> 
> Just a preference comment, shorter version 'modprobe -q -n'
> can be used here.
> 

Hi Kamalesh,

Re: command line options: my preference has been to use the long form 
command switches inside scripts as they are more likely to be self 
documenting than their short counterparts.  e.g. I could have guessed 
that -q is --quiet, but not that -n is --dry-run.

Re: --quiet vs. command redirection: Another detail I don't have a 
strong opinion about.  I guess I very slightly prefer the redirect so I 
don't have to research various modprobe versions to determine if --quiet 
is universally supported (it probably is).

In both cases, I'll defer to whatever reviewers think is more 
readable/conventional for the self-tests.

-- Joe