Re: [LTP] nanosleep testcase problem

[Garrett C. <yan...@gm...>]

On Mon, Jan 4, 2010 at 6:55 AM, Michal Simek <mic...@pe...> wrote:
> Mitani wrote:
>> Hi,
>>
>> I want to wish you happy new year.
>>
>>
>> ------
>>
>> By the way, I found a new problem with the "nanosleep02" testcase.
>>
>> "main()" function executes fork(), and child process calls "do_child()"
>> function.
>> "do_child()" function must be ended by "exit()". But in "nanosleep02"
>> testcase, there is a case in which "exit()" isn't executed.
>> Therefore, the child process lives after "do_child()" function,
>> and it executes "wait()" function which must be executed in parent
>> process only.
>
> On Microblaze this test failed too. The patch fix it.
>
> Michal
>
>>
>> Here is the patch fixes these problem:
>>
>> ============
>> --- nanosleep02.c     2009-12-23 02:35:13.000000000 +0900
>> +++ nanosleep02.c.new 2010-01-04 10:15:08.000000000 +0900
>> @@ -277,6 +277,9 @@
>>               tst_resm(TPASS, "call succeeded");
>>               exit(0);
>>       }
>> +
>> +     /* Everything is fine, exit normally */
>> +     exit(0);
>>  }
>>
>>  /*

    Committed a change on CVS which forces the child to exit at the
end of do_child, like it should with tst_exit().
Thanks!
-Garrett