linux-kernel - Re: [PATCH] selftests: lsm: Refactor `flags_overset_lsm_set_self

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: <c16b7517-e490-48d9-a2b6-f0077cbb0eba@schaufler-ca.com>
Date: Thu, 14 Nov 2024 08:55:08 -0800
From: Casey Schaufler <casey@...aufler-ca.com>
To: Shuah Khan <skhan@...uxfoundation.org>,
 Amit Vadhavana <av2082000@...il.com>, paul@...l-moore.com,
 jmorris@...ei.org, serge@...lyn.com, shuah@...nel.org
Cc: ricardo@...liere.net, linux-kernel-mentees@...ts.linux.dev,
 linux-security-module@...r.kernel.org, linux-kselftest@...r.kernel.org,
 linux-kernel@...r.kernel.org, Casey Schaufler <casey@...aufler-ca.com>
Subject: Re: [PATCH] selftests: lsm: Refactor
 `flags_overset_lsm_set_self_attr` test


On 11/14/2024 8:25 AM, Shuah Khan wrote:
> On 11/12/24 11:28, Amit Vadhavana wrote:
>> - Remove unnecessary `tctx` variable, use `ctx` directly.
>> - Simplified code with no functional changes.
>>
>
> I would rephrase the short to simply say Remove unused variable,
> as refactor implies more extensive changes than what this patch
> is actually doing.
>
> Please write complete sentences instead of bullet points in the
> change log.
>
> How did you find this problem? Do include the details on how
> in the change log.
>
>> Signed-off-by: Amit Vadhavana <av2082000@...il.com>
>> ---
>>   tools/testing/selftests/lsm/lsm_set_self_attr_test.c | 7 +++----
>>   1 file changed, 3 insertions(+), 4 deletions(-)
>>
>> diff --git a/tools/testing/selftests/lsm/lsm_set_self_attr_test.c
>> b/tools/testing/selftests/lsm/lsm_set_self_attr_test.c
>> index 66dec47e3ca3..732e89fe99c0 100644
>> --- a/tools/testing/selftests/lsm/lsm_set_self_attr_test.c
>> +++ b/tools/testing/selftests/lsm/lsm_set_self_attr_test.c
>> @@ -56,16 +56,15 @@ TEST(flags_zero_lsm_set_self_attr)
>>   TEST(flags_overset_lsm_set_self_attr)
>>   {
>>       const long page_size = sysconf(_SC_PAGESIZE);
>> -    char *ctx = calloc(page_size, 1);
>> +    struct lsm_ctx *ctx = calloc(page_size, 1);
>
> Why not name this tctx and avoid changes to the ASSERT_EQs
> below?

In the realm of linux security modules ctx is short for "context".
I used tctx here because I was lazy. It would be much better to
drop tctx, even if it means a tiny bit more change.

>
>>       __u32 size = page_size;
>> -    struct lsm_ctx *tctx = (struct lsm_ctx *)ctx;
>>         ASSERT_NE(NULL, ctx);
>>       if (attr_lsm_count()) {
>> -        ASSERT_LE(1, lsm_get_self_attr(LSM_ATTR_CURRENT, tctx, &size,
>> +        ASSERT_LE(1, lsm_get_self_attr(LSM_ATTR_CURRENT, ctx, &size,
>>                              0));
>>       }
>> -    ASSERT_EQ(-1, lsm_set_self_attr(LSM_ATTR_CURRENT |
>> LSM_ATTR_PREV, tctx,
>> +    ASSERT_EQ(-1, lsm_set_self_attr(LSM_ATTR_CURRENT |
>> LSM_ATTR_PREV, ctx,
>>                       size, 0));
>>         free(ctx);
>
> You have to change this tctx for sure.
>
> With these changes:
>
> Reviewed-by: Shuah Khan <skhan@...uxfoundation.org>
>
> Paul, James,
>
> Please do let me know if you would me to take this through
> kselftest tree.
>
> thanks,
> -- Shuah
>
>