[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <10085ec8-0109-444c-bce4-d0b0ef1a4164@gmail.com>
Date: Wed, 17 Apr 2024 15:46:44 +0100
From: Ivan Orlov <ivan.orlov0322@...il.com>
To: Andy Shevchenko <andy@...nel.org>
Cc: keescook@...omium.org, akpm@...ux-foundation.org,
linux-kernel@...r.kernel.org, linux-hardening@...r.kernel.org,
skhan@...uxfoundation.org
Subject: Re: [PATCH] string: Implement KUnit test for str*cmp functions
On 4/17/24 15:28, Andy Shevchenko wrote:
> On Wed, Apr 17, 2024 at 02:54:15PM +0100, Ivan Orlov wrote:
>> Currently, str*cmp functions (strcmp, strncmp, strcasecmp and
>> strncasecmp) are not covered with tests. Implement the test which
>> covers them all.
>>
>> The strcmp test suite consist of 8 test cases:
>>
>> 1) strcmp test
>> 2) strcmp test on long strings (2048 chars)
>> 3) strncmp test
>> 4) strncmp test on long strings (2048 chars)
>> 5) strcasecmp test
>> 6) strcasecmp test on long strings
>> 7) strncasecmp test
>> 8) strncasecmp test on long strings
>>
>> The test aims at covering as many edge cases as possible, including
>> the tests on empty strings, situations when the different symbol is
>> placed at the end of one of the strings, etc.
>
> ...
>
>> lib/strcmp_kunit.c | 170 +++++++++++++++++++++++++++++++++++++++++++++
>
> Why is not part of the existing string_kunit.c?
>
There are already 2 other KUnit tests in `lib/` covering different
groups of string functions separately (lib/strscpy_kunit.c,
lib/strcat_kunit.c), so this patch just follows this pattern. I believe
it makes sense: the tests are separated to cover one specific group of
string functions with a similar purpose
---
Kind regards,
Ivan Orlov
Powered by blists - more mailing lists