Re: lynx-dev lynx2.8.1pre.11

[Bela Lubkin]

Tom Dickey wrote:

> * modify waitpid logic in nsl-fork support to handle a case where the child
>   process exits without reporting anything (reported by Serge Munhoven) - BL

That's not quite it.  It was a race condition where the child exited
normally, at exactly the moment *between* the parent checking whether he
had spoken yet, and whether he had exited yet.  i.e.:

  parent: sleeping in select
  parent: select times out, returns no data
  child:  writes data down pipe
  child:  exits normally
  parent: wakes up from select, with no data indicated
  parent: checks whether child has exited: yes

Serge reported a way to provoke this race condition; which,
interestingly, worked on two significantly different Unix systems,
though I can think of no reason for it to have done so.  The race should
have been entirely statistical.  Anyway:

  * modify nsl_fork logic to handle a case where child exits successfully
    right after parent's select timed out (reported by Serge Munhoven) - BL

>Bela<