Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!uvaarpa!mmdf
From: prcrs!wrwalke@uunet.uu.net (William Walker)
Newsgroups: comp.lang.perl
Subject: perl3 pl15, hp/ux 3.10 9000/800
Message-ID: <3219@uvaarpa.virginia.edu>
Date: 15 Mar 90 06:40:40 GMT
Sender: mmdf@uvaarpa.virginia.edu
Reply-To: prcrs!wrwalke@uunet.uu.net
Lines: 25


ok larry, you were right, almost.  patchlevel 15 solved my problem
as far as the tests are concerned, but i did have to remove your
malloc().  with the system malloc, all tests passed.  with the
malloc included with perl, the man creation dumps core again.

since the project i am on has to do with memory allocation problems
under HP/ux on the 800 series, i can legitimately do this one on
the job.  i will send along my findings as soon as i trace it.

but for now --->

on hp/ux pre-7.0 on series 800 machines===>

-DJMPCLOBBER   ---   in Configure additional cc flags for syscall() bug
malloc()       ---   use system malloc(), or testmalloc distribution
memory         ---   have lots, or compile at night, eval.c eats up
			about 25Meg with -O.  if you want it, i can send
			out a uuencoded eval.o with full optimization
TOROL            ---   try out Randal's one liners  8-)


bill.

PS - thanks larry!!