From frodef at cs.uit.no Tue May 2 18:39:07 2006 From: frodef at cs.uit.no (Frode Vatvedt Fjeld) Date: 02 May 2006 20:39:07 +0200 Subject: [movitz-devel] Trac Message-ID: <2hlktk6zlg.fsf@vserver.cs.uit.no> The Trac project management system has been set up for Movitz at http://trac.common-lisp.net/movitz Please use it to submit and comment tickets! :-) -- Frode Vatvedt Fjeld From kruhft at gmail.com Mon May 15 04:40:47 2006 From: kruhft at gmail.com (Burton Samograd) Date: Sun, 14 May 2006 22:40:47 -0600 Subject: [movitz-devel] compile problem with current cvs + patch Message-ID: <873bfbvr1s.fsf@gmail.com> Hi all, I just grabbed all the latest sources from cvs for the first time and had a slight problem when I tried to build the boot image. It seems that the function funobj-name was nowhere to be found, except in the procfs-image.lisp file, which I assume is not in the file list of the default bootable image. I moved the function over to compiler.lisp, which made the most sense to me, since there was another funobj- function defined there. After the move the image built, dumped and booted just fine. Here's a patch (cvs diff -u), although there's not much to it: Index: compiler.lisp =================================================================== RCS file: /project/movitz/cvsroot/movitz/compiler.lisp,v retrieving revision 1.170 diff -u -r1.170 compiler.lisp --- compiler.lisp 5 May 2006 18:37:32 -0000 1.170 +++ compiler.lisp 15 May 2006 04:30:16 -0000 @@ -3392,6 +3392,11 @@ (defun operands (x) (if (symbolp x) nil (cdr x))) +(defun funobj-name (x) + (typecase x + (movitz-funobj + (movitz-funobj-name x)))) + (defun funobj-assign-bindings (code env &optional (stack-frame-position 1) (frame-map (make-binding-map))) "This wrapper around assign-bindings checks if the first instructions of CODE Index: procfs-image.lisp =================================================================== RCS file: /project/movitz/cvsroot/movitz/procfs-image.lisp,v retrieving revision 1.25 diff -u -r1.25 procfs-image.lisp --- procfs-image.lisp 10 Apr 2006 11:46:25 -0000 1.25 +++ procfs-image.lisp 15 May 2006 04:30:16 -0000 @@ -238,11 +238,12 @@ do (format t "~& => "))) (values)) +#| (defun funobj-name (x) (typecase x (movitz-funobj (movitz-funobj-name x)))) - +|# (defun stack-frame (image) (do-stack-frame (image-register32 image :ebp) 0)) -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com From kruhft at gmail.com Mon May 15 05:28:47 2006 From: kruhft at gmail.com (Burton Samograd) Date: Sun, 14 May 2006 23:28:47 -0600 Subject: [movitz-devel] bochs and enter key Message-ID: <87ves7zwj4.fsf@gmail.com> Hi again, I was very impressed when I found that movitz booted just fine and worked on my PC, after that I tried booting with bochs. It booted fine and everything seemed to work except the key, thus not allowing me to enter any s-exps. I googled a bit and found a chat log stating that the solution to this problem is 'in one of the web pages', which I have unfortunately been unable to find (the solution, not the pages). Any pointers on how to fix to this problem? TIA -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com From sabetts at vcn.bc.ca Mon May 15 07:09:54 2006 From: sabetts at vcn.bc.ca (Shawn Betts) Date: Mon, 15 May 2006 03:09:54 -0400 Subject: [movitz-devel] bochs and enter key In-Reply-To: <87ves7zwj4.fsf@gmail.com> (Burton Samograd's message of "Sun, 14 May 2006 23:28:47 -0600") References: <87ves7zwj4.fsf@gmail.com> Message-ID: <86iro7wypp.fsf@shitbender.gagrod> Burton Samograd writes: > Hi again, > > I was very impressed when I found that movitz booted just fine and > worked on my PC, after that I tried booting with bochs. It booted > fine and everything seemed to work except the key, thus not > allowing me to enter any s-exps. I googled a bit and found a chat log > stating that the solution to this problem is 'in one of the web > pages', which I have unfortunately been unable to find (the solution, > not the pages). Any pointers on how to fix to this problem? Could it be your bochs keymap? You could also try qemu. -Shawn From kruhft at gmail.com Mon May 15 14:13:32 2006 From: kruhft at gmail.com (Burton Samograd) Date: Mon, 15 May 2006 08:13:32 -0600 Subject: [movitz-devel] Re: bochs and enter key References: <87ves7zwj4.fsf@gmail.com> <86iro7wypp.fsf@shitbender.gagrod> Message-ID: <87r72vz88j.fsf@gmail.com> Shawn Betts writes: > Burton Samograd writes: > >> Hi again, >> >> I was very impressed when I found that movitz booted just fine and >> worked on my PC, after that I tried booting with bochs. It booted >> fine and everything seemed to work except the key, thus not >> allowing me to enter any s-exps. I googled a bit and found a chat log >> stating that the solution to this problem is 'in one of the web >> pages', which I have unfortunately been unable to find (the solution, >> not the pages). Any pointers on how to fix to this problem? > > Could it be your bochs keymap? You could also try qemu. Maybe? I just used to the configuration file that came with the cvs sources and there's nothing about the keymap in there. I'll check the bochs site for an answer, it just seems that this problem has cropped up before and there was an easy solution. qemu doesn't ever work on the console for me, but i'll try in the future when i fire up x. -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com From frodef at cs.uit.no Mon May 15 15:04:43 2006 From: frodef at cs.uit.no (Frode Vatvedt Fjeld) Date: 15 May 2006 17:04:43 +0200 Subject: [movitz-devel] Re: bochs and enter key References: <87ves7zwj4.fsf@gmail.com> <86iro7wypp.fsf@shitbender.gagrod> <87r72vz88j.fsf@gmail.com> Message-ID: <2hzmhj8h2s.fsf@vserver.cs.uit.no> Hi Burton, I hope you'll enjoy some Movitz hacking! Shawn Betts writes: > Could it be your bochs keymap? You could also try qemu. Burton Samograd writes: > Maybe? I just used to the configuration file that came with the cvs > sources and there's nothing about the keymap in there. I'll check > the bochs site for an answer, it just seems that this problem has > cropped up before and there was an easy solution. I found the irc logs you referred to earlier, but I don't really remember what the web-page info was. However, probably it was this: If you look in the file losp/los0.lisp (the file that by default is the first file to get loaded and run at bootup), in the function "genesis", theres a line that is commented out like this: ;; (muerte.toplevel:invoke-toplevel-command :mapkey #\newline) (This is equivalent to entering ":mapkey #\newline" at the REPL, but the problem obviously is that without enter you cannot even do this.) If you remove the semi-colons, recompile the genesis function, dump the image and boot it, you should be prompted for a keypress that will from then on be interpreted as #\newline. This is a quick and dirty solution to your problem. The real problem is I suppose that the keyboard driver is rather primitive and doesn't really take into account the variants of keyboard hardware as it should. BTW: There is a keyboard issue with qemu also, I have to start it with "-k en-us" for it to work properly. Regards, -- Frode Vatvedt Fjeld From frodef at cs.uit.no Mon May 15 19:55:59 2006 From: frodef at cs.uit.no (Frode Vatvedt Fjeld) Date: 15 May 2006 21:55:59 +0200 Subject: [movitz-devel] Re: compile problem with current cvs + patch References: <873bfbvr1s.fsf@gmail.com> Message-ID: <2hves783lc.fsf@vserver.cs.uit.no> Thanks for the patch Burton. I think I'll prefer to put the function in storage-types.lisp. I'll check in the fix to CVS right away. -- Frode Vatvedt Fjeld From kruhft at gmail.com Tue May 16 16:04:17 2006 From: kruhft at gmail.com (Burton Samograd) Date: Tue, 16 May 2006 10:04:17 -0600 Subject: [movitz-devel] Re: compile problem with current cvs + patch References: <873bfbvr1s.fsf@gmail.com> <2hves783lc.fsf@vserver.cs.uit.no> Message-ID: <878xp2yn0e.fsf@gmail.com> Frode Vatvedt Fjeld writes: > Thanks for the patch Burton. I think I'll prefer to put the function > in storage-types.lisp. I'll check in the fix to CVS right away. No problem. I'm still learning the layout of the source, so I was just proud of myself getting the image to build without asking for help :) -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com From kruhft at gmail.com Tue May 16 16:19:37 2006 From: kruhft at gmail.com (Burton Samograd) Date: Tue, 16 May 2006 10:19:37 -0600 Subject: [movitz-devel] movitz-mode.el and fi:: functions Message-ID: <874pzqymau.fsf@gmail.com> Hi, I'm trying out movitz-mode.el just for fun, but when I switch to movitz-mode I get the following error: Symbol's function definition is void: fi::initialize-mode-map I've never heard of the fi:: package and didn't realize that emacs even had namespaces at all. Am I missing a required package somewhere? -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com From kruhft at gmail.com Tue May 16 16:55:01 2006 From: kruhft at gmail.com (Burton Samograd) Date: Tue, 16 May 2006 10:55:01 -0600 Subject: [movitz-devel] Re: movitz-mode.el and fi:: functions References: <874pzqymau.fsf@gmail.com> Message-ID: <87wtclyknu.fsf@gmail.com> Burton Samograd writes: > I'm trying out movitz-mode.el just for fun, but when I switch to > movitz-mode I get the following error: > > Symbol's function definition is void: fi::initialize-mode-map > > I've never heard of the fi:: package and didn't realize that emacs > even had namespaces at all. Am I missing a required package > somewhere? Nevermind. I've been reading through the list and found that movitz-mode was tied to Franz. I'll go off and try the slime extension... -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com