[LTP] [PATCH 3/3] mbind01: add more tests for MPOL_LOCAL

[Li Wang]

On Fri, Jul 30, 2021 at 6:35 PM Jan Stancek <jstancek@redhat.com> wrote:

>
>
> On Fri, Jul 30, 2021 at 10:03 AM Li Wang <liwang@redhat.com> wrote:
>
>>
>>
>> On Thu, Jul 29, 2021 at 10:20 PM Jan Stancek <jstancek@redhat.com> wrote:
>>
>>>
>>>
>>> On Thu, Jul 29, 2021 at 3:25 PM Li Wang <liwang@redhat.com> wrote:
>>>
>>>> Signed-off-by: Li Wang <liwang@redhat.com>
>>>> ---
>>>>  testcases/kernel/syscalls/mbind/mbind01.c | 25 ++++++++++++++++++++---
>>>>  1 file changed, 22 insertions(+), 3 deletions(-)
>>>>
>>>> diff --git a/testcases/kernel/syscalls/mbind/mbind01.c
>>>> b/testcases/kernel/syscalls/mbind/mbind01.c
>>>> index d2cf13c8f..b5c1e948d 100644
>>>> --- a/testcases/kernel/syscalls/mbind/mbind01.c
>>>> +++ b/testcases/kernel/syscalls/mbind/mbind01.c
>>>> @@ -34,7 +34,7 @@ static struct bitmask *nodemask, *getnodemask,
>>>> *empty_nodemask;
>>>>  static void test_default(unsigned int i, char *p);
>>>>  static void test_none(unsigned int i, char *p);
>>>>  static void test_invalid_nodemask(unsigned int i, char *p);
>>>> -static void check_policy_pref_no_target(int);
>>>> +static void check_policy_pref_or_local(int);
>>>>
>>>>  struct test_case {
>>>>         int policy;
>>>> @@ -92,7 +92,7 @@ static struct test_case tcase[] = {
>>>>                 .ret = 0,
>>>>                 .err = 0,
>>>>                 .test = test_none,
>>>> -               .check_policy = check_policy_pref_no_target,
>>>> +               .check_policy = check_policy_pref_or_local,
>>>>         },
>>>>         {
>>>>                 POLICY_DESC(MPOL_PREFERRED),
>>>> @@ -101,6 +101,20 @@ static struct test_case tcase[] = {
>>>>                 .test = test_default,
>>>>                 .exp_nodemask = &nodemask,
>>>>         },
>>>> +       {
>>>> +               POLICY_DESC(MPOL_LOCAL),
>>>> +               .ret = 0,
>>>> +               .err = 0,
>>>> +               .test = test_none,
>>>> +               .exp_nodemask = &empty_nodemask,
>>>> +               .check_policy = check_policy_pref_or_local,
>>>>
>>>
>>> This is a bit more permissive, it allows for MPOL_LOCAL to return also
>>> MPOL_PREFERRED.
>>> Shouldn't that still be treated as error?
>>>
>>
>> To strictly this should be an error.
>>
>> But I slightly think that it's acceptable to get 'MPOL_PREFERRED' on the
>> old
>> kernel (i.e. 4.18.0, v5.13) because 'MPOL_LOCAL' is not treated as a
>> real policy.
>> And the situation exists for quite a long time.
>>
>
> You're right, on older kernel it failed in similar way for MPOL_LOCAL
> as it failed for MPOL_PREFERRED on latest one.
>

Or, If we want something more precise, just cancel the
check_policy_pref_or_local on kernel >= 5.14.

Is this sound better?

-       if ((tst_kvercmp(3, 8, 0)) < 0 && (tc->policy == MPOL_LOCAL)) {
-               tst_res(TCONF, "%s is not supported",
tst_mempolicy_mode_name(tc->policy));
-               return;

+       if (tc->policy == MPOL_LOCAL) {
+               if ((tst_kvercmp(3, 8, 0)) < 0) {
+                       tst_res(TCONF, "%s is not supported",
tst_mempolicy_mode_name(tc->policy));
+                       return;
+               }
+
+               if ((tst_kvercmp(5, 14, 0)) >= 0)
+                       tc->check_policy = NULL;


>
> Acked-by: Jan Stancek <jstancek@redhat.com>
>
>

-- 
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20210730/aa0582ad/attachment-0001.htm>