make_writable errors with gcl-2.6.8pre




On Tue, 21 Apr 2009, Robert Dodier wrote:

< On Tue, Apr 21, 2009 at 9:01 AM, Leo Butler <l.butler at ed.ac.uk> wrote:
< 
< > I built 5.18.1 with gcl-2.6.8pre sans `batch' patch, and I had no
< > make_writable errors during `make check', too.
< >
< > However, I have a script that behaves quite oddly with this maxima:
< >
< > -When invoked in maxima via load("foo.mac"), the make_writable error appears
< >  predictably after about 5 minutes.
< >
< > -When the script is run from the command line via `maxima -b foo.mac' the
< > script runs fine.
< 
< Well, I wonder what's in this foo.mac ...
< 
< > (It has been running for about 18hours at the moment, and I expect it'll run for
< > 10+days, barring further errors).
< 
< Are you sure it's making progress? My experience with GCL is that it
< seems to allocate more memory than other Lisps and Maxima ends up
< swapping endlessly. Is your Maxima process cpu-limited or is it swapping?
 
It is making progress. I am running the same script using maxima-5.18.1
built with gcl and clisp-2.4.7. Here is the output of top:
PID   USER        CPU  MEM   TIME    
25073 lbutler     100  6.0   2100:17 maxima    <-- gcl
25020 lbutler     100  0.5   2097:44 lisp.run  <-- clisp

Gcl-maxima does use much more memory, but I haven't observed it
thrashing. Perhaps in later calculations where memory requirements are
larger...

Leo 

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.