Greetings, and thanks for the reports!
1) gcl self compile time is deliberately slower, but may be
mitigatable somewhat. Most of the core functions are now
implemented in lisp rather than C, which allows for automatic
source inlining. Unfortunately, this makes bootstrapping much
slo
Greetings! Would you accept the following patch:
--- ../nqthm-1992/make/compile.lisp 1994-01-07 16:59:22.0 -0500
+++ make/compile.lisp 2007-06-21 19:51:56.0 -0400
@@ -68,7 +68,7 @@
(IN-PACKAGE "USER")
-(PROGN
+(with-compilation-unit nil
; This function gets referenc
Greetings! OK, have committed my preliminary overhaul of the error
system together with the attempt to be abort proof. Closed many ansi
failures. Still need some cleanup, but thought it would be more time
efficient to let others see what is going on at the moment and provide
feedback.
Take car
Greetings! Just a quick note here -- the expansion is almost
completely due to running the compiler in the *final* raw image
initialization stage with si::*optimize-maximum-pages* set to t, the
default. Not only should this never happen, but we should build a
more compact image with the optimize
Greetings! OK, the latest changes are in -- hopefully you might see
some performance gains.
Take care,
--
Camm Maguire[EMAIL PROTECTED]
==
"The earth is but one country, and mankin
Greetings! Just a note that the following was just committed:
'length-expander cleanup,endp-expander cleanup,bind alll arrays in
aref expansion to avoid pass2 typing,inlinable=fn in
member-compiler-macro,some,every,notany,notevery compiler macros,wrap
functions in `(function ) at safety 3,let*->l
Greetings! Just a quick announcement that I've got maxima under 2.7.0
cvs snapshot passing all tests. Hope to commit tomorrow, after which
I'll be away for 1 week. sys-proclaim is not quite working yet, and
the efficiency still needs work, but it appears something quite
simple.
Take care,
--
C
Greetings! Just a brief heads up about the latest commit just made.
Quite a lot of changes, improvements hopefully :-).
1) si::proper-list is now propagated somewhat. This is basically
intended to allow sequence -> list simplification inside certain do
loops, etc. Check out the difference
Hi Camm.
With HEAD up again on Windows I've returned to the issue of the Tk
binding as the Axiom crew are waiting on a LISP GUI.
I decided that gcl-tk is too tricky in itself as the server relies on a
function not available in the Windows version of Tcl/Tk and I would
rather a system in which a
Greetings!
"Mike Thomas" <[EMAIL PROTECTED]> writes:
> | Greetings, and thanks! Should be fixed now.
>
> Thanks Camm. With that fix, HEAD now builds Maxima 5.9.2 on Windows,
> although there are a couple of test failures.
>
Yes, and alas there seems to be some slowness introduced into the te
I checked out GCL 2.7.0 this evening and built it successfully on a
Fedora Core 3 system, but failed on Red Hat 9.0.
...
gcc -c -Wall -DVOL=volatile -fsigned-char -pipe -O3
-fomit-frame-pointer -I/home/steve/dl/lisp/gcl-2.7.0/gcl/o -I../h
-I../gcl-tk plt.c
In file included from plt.c:30:
pltd
Camm Maguire writes:
Greetings! I'd like to poll the summer plans/schedules of some of the
key developers to see if they might be able to find some time this
summer for a burst of development to get 2.7.0 out roughly by summers'
end. Please let me know what your time availability looks like.
I'll
Greetings! I'd like to poll the summer plans/schedules of some of the
key developers to see if they might be able to find some time this
summer for a burst of development to get 2.7.0 out roughly by summers'
end. Please let me know what your time availability looks like.
Take care,
Camm Maguire
13 matches
Mail list logo
