Re: [Bugme-new] [Bug 12562] New: High overhead while switching orsynchronizing threads on different cores

[Mike Galbraith]

On Mon, 2009-02-02 at 09:33 +0100, Thomas Pilarski wrote:
> Am Montag, den 02.02.2009, 09:19 +0100 schrieb Peter Zijlstra:
> > I suppose you'll have to go bug the glibc people about their random()
> > implementation.
> 
> Yes, I will.

Finding the below was easy enough...

/* POSIX.1c requires that there is mutual exclusion for the `rand' and
   `srand' functions to prevent concurrent calls from modifying common
   data.  */
__libc_lock_define_initialized (static, lock)

...

long int
__random ()
{
  int32_t retval;

  __libc_lock_lock (lock);

  (void) __random_r (&unsafe_state, &retval);

  __libc_lock_unlock (lock);

  return retval;
}

...but finding the plumbing leading to __lll_lock_wait_private()
over-taxed my attention span.

	-Mike

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/