[PATCH] Add UNSUPPORTED check in elf/tst-pldd.

[Stefan Liebler]

On 9/10/19 3:32 PM, Adhemerval Zanella wrote:
> 
> 
> On 10/09/2019 05:46, Stefan Liebler wrote:
>> On 9/6/19 5:21 AM, Carlos O'Donell wrote:
>>> On 9/3/19 9:34 AM, Adhemerval Zanella wrote:
>>>> Yes, my initial suggestion was just to make it as UNSUPPORTED for
>>>> ptrace_scope >= 1. But I do not oppose adjusting it to run on
>>>> ptrace_scope 1, it is just that the required hackery lead to make it
>>>> somewhat as complex than the test itself.
>>>
>>> The flip side of the coin is that the more "UNSUPPORTED" results we
>>> add *implies* there is "one valid way" to setup a glibc test run
>>> and we don't clearly document how to turn all the "UNSUPPORTED"
>>> entries into supported tests?
>>>
>>> Stefan's code can at least be refactored into support/ if we need
>>> to do the same thing again in another test.
>>>
>>
>> PING.
>>
>> As I have already posted multiple versions of the patch, how to proceed?
>> 1) UNSUPPORTED if support_ptrace_scope() >= 2;
>> Support support_ptrace_scope() == 1
>> by adjusting the process tree;
>> (see https://www.sourceware.org/ml/libc-alpha/2019-09/msg00024.html)
>>
>> 2) UNSUPPORTED if support_ptrace_scope() >= 2;
>> Support support_ptrace_scope() == 1
>> by calling support_ptrace_process_set_ptracer_any();
>> (see https://www.sourceware.org/ml/libc-alpha/2019-08/msg00722.html)
>>
>> 3) UNSUPPORTED if support_ptrace_scope() != 0
>> (patch would use support_ptrace_scope() of one of the patches above in order to trigger FAIL_UNSUPPORTED)
> 
> My view is although 2) is way complex that I would like, I think it should
> the more complete solution. Does still need review or is it ready to land?
> 
You've already reviewed the support part on a previous version of "patch 
2)" (the support part was not changed in the latest version; see 
https://www.sourceware.org/ml/libc-alpha/2019-08/msg00703.html).

But it needs review for the synchronization part between the 
target_process and do_test in tst-pldd.c.