Re: [Boost-testing] [Boost.testing]Timeout code wierdness

From: K. Noel Belcourt (kbelco_at_[hidden])
Date: 2007-09-29 16:28:02

On Sep 29, 2007, at 1:08 PM, Rene Rivera wrote:

> K. Noel Belcourt wrote:
>> You may want to hold off a little bit more before releasing the new
>> Boost.Jam (3.1.15).  I think there's still a subtle failure pathway
>> that can permit subprocesses to fail to terminate correctly on unix.
>> I'm tracking it down now, sorry about the late notice.
>
> :-(
>
> ...Here's hoping that you can fix it quickly. As I really want to do a
> release this weekend.

Fixed.  I forgot to limit the select timeout interval to a positive  
number, really, really weak.

-- Noel