[Gcl-devel] New GMP configuration avoidance and general communication notes

[Mike Thomas]

Hi Camm.

I had the luxury of two whole hours of pseudo-(I played Mister Potato Head
with Rosie while waiting for builds)-uninterrupted GCL programming on
Sunday, and just wanted to say thanks for your rework of the GMP
configuration - it cuts configuration down from 4 to 1 minute on my 800 MHz
PIII and enabled me to cover much more ground than usual.

I'm pursuing three different avenues with GCL at present:

1. Cleaning up the Windows linking - now moved into separate sfaslpecoff.c
and rel_pecoff.c files.  I'm hoping in the short term to make the code more
readable, to allow more flexibility with respect to the location of the text
section raw data and in the longer term, to use the appropriate structures
and macros provided in the Windows system headers for PECOFF objects rather
than the non-standard hand coded equivalents provided in the rats nest of
GCL sfasl code and related headers.  Testing these changes with complete
rebuilds of GCL and Maxima really benefitied from the time savings I
mentioned above.

2. Automating the setting of C stack size and related items.  This has been
working well for almost a month now with the default stack size, but I'm
stuck on how to pass the configurable stack size (CSSIZE) into the
configure.in test for the C Stack address (CSTACK_ADDRESS) - the final
connection.

3. I have had mixed success with Axiom on Windows - I am now back to
tracking a crash in the build which mysteriously went away and then
returned.  This is a lower priority because of the sheer amount of time it
takes to build.

I am also curious as to the state of your 2.7.0 planning.

Cheers

Mike Thomas.