[LTP] [PATCH v2 6/6] syscalls/set_thread_area01: Refactor into new API

[Andrea Cervesato]

Hi,

On 3/28/25 11:12, Ricardo B. Marlière wrote:
> On Fri Mar 28, 2025 at 6:50 AM -03, Andrea Cervesato wrote:
>> Hi!
>>
>> This test is weird. It stays inside set_thread_area folder, but it's
>> testing get_thread_area(2) as well. We should probably create a new
>> testing folder for get_thread_area(2), splitting test in two, doing the
>> exact same thing: verifying that syscalls can be executed correctly and
>> raises the correct errors.
> That is not the original intent of the test, which depends on what
> happens when calling set_thread_area with the correct value, which then
> changed what happens with the get_thread_area call.
Thanks for clarifying. Then the way it's implemented needs some 
attention, because the `struct test` is misleading. Test cases are 
independent from each others by definition.
We should put the right sequence of syscalls inside the run test 
function and to obtain the right output.

I still believe these two tests could be easily created and they would 
include the current testing scenario for free.

Andrea