[PATCH] Add test for pthread_setname_np and pthread_getname_np.

[Allan McRae]

On 05/02/14 13:19, Carlos O'Donell wrote:
> On 02/04/2014 09:18 AM, Joseph S. Myers wrote:
>> On Mon, 3 Feb 2014, Carlos O'Donell wrote:
>>
>>> diff --git a/nptl/sysdeps/unix/sysv/linux/tst-setgetname.c b/nptl/sysdeps/unix/sysv/linux/tst-setgetname.c
>>> index 2aceba5..0235d37 100644
>>> --- a/nptl/sysdeps/unix/sysv/linux/tst-setgetname.c
>>> +++ b/nptl/sysdeps/unix/sysv/linux/tst-setgetname.c
>>> @@ -99,6 +99,17 @@ do_test (int argc, char **argv)
>>>    if (res == 0)
>>>      {
>>>        res = get_self_comm (gettid (), name_check, TASK_COMM_LEN);
>>> +
>>> +      /* On this first test we look for ENOENT to be returned from
>>> +         get_self_comm to indicate that the kernel is older than
>>> +         2.6.33 and doesn't contain comm within the proc structure.
>>> +         In that case we skip the entire test.  */
>>> +      if (res == ENOENT)
>>> +       {
>>> +         printf ("SKIP: The kernel does not have /proc/self/task/%%lu/comm.\n");
>>> +         return 0;
>>> +       }
>>
>> I'd like it to be obvious that this can be removed when we require kernel 
>> 2.6.33 or later (maybe in a couple of years' time).  Could you condition 
>> this on __LINUX_KERNEL_VERSION < 0x020621, or (more cleanly) a 
>> corresponding __ASSUME_PROC_PID_TASK_COMM defined in kernel-features.h for 
>> __LINUX_KERNEL_VERSION >= 0x020621 (with an include of kernel-features.h 
>> in this test, of course)?
> 
> Allan,
> 
> The Linux-specific test for pthread_setname_np and pthread_getname_np
> fails on systems running the Linux kernel 2.6.32 or older. The fix
> below makes us skip this test when the kernel is older than 2.6.33
> (when /proc/self/task/%lu/comm was added, which is required by the test).
> 
> I'd like to check this in if you don't object to cleanup the testsuite
> for results for Adhemerval who is testing on systems with older kernels.
> 
> Joseph,
> 
> Absolutely. That's a great idea.
> 
> Simulated testing by using invalid path and rebuilding with
> arch_minimum_kernel set to 2.6.33.
> 
> Tested on x86-64 with no regressions.
> 
> OK to checkin?

Yes,

Allan