Re: bug in wait code.

[Alex <garbanzo at hooked dot net>]

On Sun, 2 Nov 1997, Chris Faylor wrote:

> >(C:\UNIX\BIN\SH.EXE 1061) wait_subproc: wait failed. nchildren 1, wait
> >-1, err 6

Haven't seen that one quite yet.
 
> It is also possible that there is a problem with Win98 since I haven't
> seen reports of this on Windows 95 or Windows NT.

I too would have to go with a problem on Win98, although there are a few
buggies left that will GPF stuff under Win95 (try running ./configure;make
on the cdk source for one).  I'd really suggest that you _don't_ run this
on Win98 at all.  The last build I tried was 1602 (most current CD so
far), and it fixed some stuff, but really really had major stability
problems (bcppbuilder froze it solid).  If you _must_ run on 98, try using
the Beta 2 refresh cd (~1591) or so, otherwise don't be suprised if things
start breaking.

> >then seems to go into an infinate loop, and keeps printing the same
> >error.  I havent been able to find the exact problem, and it seems
> >pretty random.  I think i know why its going into a loop tho.  The
> >following patch should fix this, but i havent had a chance to test it
> >out.  It does seem to make sence to break out of the for(;;) loop if
> >the wait fails..

As I said, this probably belongs on the Apps.Compat group on
betanews.microsoft.com (if you haven't pirated it).  Too bad MS doesn't
really seem to care that 1602 breaks a lot of stuff (including their very
own bug reporting stuff *laugh*.)

[Patch snipped]
> This will certainly break out of the loop but the program will not
> function properly from then on.  waits will no longer work.  I'm not
> sure that's exactly the right solution.

The more I dive into using programming stuff on Windows, the more I wonder
how the heck this stuff manages to run at all.

- alex

-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".