mit-scheme-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MIT-Scheme-devel] GTk+ in MIT-Scheme?


From: bnewbold
Subject: Re: [MIT-Scheme-devel] GTk+ in MIT-Scheme?
Date: Thu, 2 Apr 2009 21:46:01 -0400 (EDT)
User-agent: Alpine 2.00 (DEB 1167 2008-08-23)


As a datapoint, it works for me. Needed to rerun autoconf in a couple directories and upgrade gtk+/atk/pango/glib by hand, but (scm-layout-demo) works. This is on debian "etchnhalf", ix86. I used an existing mit-scheme to first build the FFI microcode, then that machine to build the full FFI codebase with the GTK patch applied.

A tiny bug (?), for some reason building the documentation fails at:

@verbatiminclude ../../src/ffi/prhello.scm

even though that relative path is correct from the appropriate directory. Not sure what's wrong, tex logs just say they can't find the file. Could be my mangled setup?

-bryan

On Thu, 19 Mar 2009, Matt Birkholz wrote:

I finally have my FFI and GTK systems working properly.  Turning up
the timer interrupts to 200/sec really turns up the interrupt holes!
And I learned never to poll forever.  A timer interrupt aught to wake
you from such a poll, but it won't.  It will sneak into a corner and
hide, leaving you permanently "polled".

The FFI system did not change, like, at all.  The GTK system got much
simpler, due to a more complete understanding of runtime/thread.scm,
which was due to hours poring over trace output.  By introducing a
"toolkit thread", I now arrange for Scheme to "run" the toolkit in the
same round-robin fashion as its other threads, AND there is always a
thread to run callbacks.

The works:

        http://birkholz.chandler.az.us/~matt/Scheme/


Questions from last time:

Do I need to sign something?

Furthermore: what copyright notice should I attach?  I am tempted to
assign the work to MIT, since most (all?) files name MIT as the
copyright holder.

The GNU Coding Standards and Maintainer Guide
(http://www.gnu.org/prep/maintain/) suggest the FSF would like to see
my signature if I assign the work to them.

[...] Would y'all rather see these diffs on a branch?

[...] Is a shockingly simple representation of C pointers worth [the
consing]?


_______________________________________________
MIT-Scheme-devel mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/mit-scheme-devel





reply via email to

[Prev in Thread] Current Thread [Next in Thread]