Fix race in tst-mqueue5

[Adhemerval Zanella]

On 13-01-2016 12:11, Carlos O'Donell wrote:
> On 01/13/2016 09:00 AM, Adhemerval Zanella wrote:
>>> While I agree that any fix that makes tst-mqueue5 fail less spuriously is a good
>>> thing, I'm curious about your review of the test as a whole (now that I've looked
>>> at it again).
>>
>> I would say to just remove the spurious synchronization issues, specially the ones
>> that might arise with the use of 'sleep' in bad ways.
> 
> Does that mean you'd rather see a more comprehensive fix?
> 
> The test still uses sleep() in one place.

I would prefer a comprehensible test that do raise false positives, even by
a slight chance (such as the case). This is also similar to the nanosleep
regression testcase [1] that generate different opinions whether a possible
racy testcase yields any meaningful validation. 

At first I did no oppose, but recently I noted that potentially racy testcase
are mainly ignored when testers see spurious fails (the tst-mqueue5 is an
example, this issue exists for a long time).

[1] https://sourceware.org/ml/libc-alpha/2015-11/msg00199.html