[LTP] [PATCH v2 0/2] Rewrite the gettid() testing suite

Andrea Cervesato andrea.cervesato@suse.com
Thu Oct 5 16:39:01 CEST 2023


Hi!

On 10/5/23 16:24, Marius Kittler wrote:
> Am Donnerstag, 5. Oktober 2023, 15:49:09 CEST schrieb Andrea Cervesato:
>> From: Andrea Cervesato <andrea.cervesato@suse.com>
>>
>> Rewrite gettid01 and create a new gettid02 test in order to check all
>> gettid() basic functionalities, according with its manual.
>>
>> Andrea Cervesato (2):
>>    Rewrite gettid01 test
>>    Add gettid02 test
>>
>>   runtest/syscalls                            |   1 +
>>   testcases/kernel/syscalls/gettid/.gitignore |   1 +
>>   testcases/kernel/syscalls/gettid/Makefile   |   4 +-
>>   testcases/kernel/syscalls/gettid/gettid01.c | 105 ++++----------------
>>   testcases/kernel/syscalls/gettid/gettid02.c |  68 +++++++++++++
>>   5 files changed, 93 insertions(+), 86 deletions(-)
>>   create mode 100644 testcases/kernel/syscalls/gettid/gettid02.c
> Looks generally good to merge.
>
> I'm only wondering whether the goto in the 2nd test could be avoided. After
> all it might even be a benefit to get an info for all the wrongly equivalent
> thread IDs (and not just the first).
>
> Then it would also make sense to use `int j = i + 1;` in the inner loop to
> avoid duplicate/redundant commutative comparisons. And then you could actually
> also drop the `i == j` check.
>
>
>
>
It makes sense, I update the code.

Andrea



More information about the ltp mailing list