[PATCH 3/4] cr_tests/ptree1: wait for background tasks to exit

[Nathan Lynch]

On Fri, 2010-02-26 at 12:56 -0800, Matt Helsley wrote:
> On Fri, Feb 26, 2010 at 01:51:53PM -0600, Nathan Lynch wrote:
> > It is more reliable and usually faster to wait for the shell's
> > notification that background tasks have exited than it is to assume
> > they have exited after three seconds have passed.
> 
> Yes, a wait should normally do the trick here.
> 
> What happens if thaw fails and the task to wait for is still frozen?
> We'd need a "wait or timeout" here to gracefully handle that. However,
> I suppose it's arguable whether thaw failing is worth worrying about
> here...

I think it's not, unless this scenario could occur in the absence of
kernel or testcase bugs?