From mmertens@akam.be Mon Jun 1 07:52:00 1998 Received: from chekov.Belgium.eu.net (relay.eunet.be [192.92.130.25]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA08570 for ; Mon, 1 Jun 1998 07:51:59 -0700 (PDT) Received: from (marc@idialup250.antwerp.eunet.be [193.121.237.250]) by chekov.Belgium.eu.net with ESMTP id QAA08096 for ; Mon, 1 Jun 1998 16:57:03 +0200 (MET DST) Sender: marc@chekov.Belgium.eu.net Message-ID: <3572C22F.8E4E4102@akam.be> Date: Mon, 01 Jun 1998 17:01:03 +0200 From: Marc Mertens X-Mailer: Mozilla 4.05 [en] (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 To: clisp-list@clisp.cons.org Subject: socket interface in clisp Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello , I have the following question concerning CLISP on LINUX and sockets . 1. Does there exist a function simular to 'listen' which works on the stream returned by (socket-connect port) , so that the following piece of code works without blocking (setf s (socket-connect 9734) (if (listen s) (princ (read s)) (princ "No input on socket")) 2. Is it possible to connect a handler function to a socket-stream (like in cmucl) . What I want is a way to link a lisp function to a socket stream . The function should be executed if input is available on the socket stream. 3. Does the read-eval-print loop has some hooks where we can call a function when the user is no busy typing something in. The raison I ask this questing is that I'm writing a client server application using lisp (actually a lisp debugger) where the lisp system can ask something from the server (easy , just send something to the server using the socket stream) and if the lisp system is idle waiting on keystrokes from the user , commands from the server could be send for execution to the lisp interpretor. Thanks a lot in advance mmertens@akam.be From haible@ilog.fr Mon Jun 1 13:12:51 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id NAA10203 for ; Mon, 1 Jun 1998 13:12:49 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.29]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id WAA08425 for ; Mon, 1 Jun 1998 22:17:06 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id WAA02188; Mon, 1 Jun 1998 22:18:15 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id WAA25227; Mon, 1 Jun 1998 22:18:03 +0200 (MET DST) Date: Mon, 1 Jun 1998 22:18:03 +0200 (MET DST) Message-Id: <199806012018.WAA25227@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: Re: socket interface in clisp In-Reply-To: <3572C22F.8E4E4102@akam.be> References: <3572C22F.8E4E4102@akam.be> Hello Marc, > 1. Does there exist a function simular to 'listen' which works on the > stream returned by > (socket-connect port) , so that the following piece of code works > without blocking > (setf s (socket-connect 9734) > (if (listen s) > (princ (read s)) > (princ "No input on socket")) In my experience, the function `listen' works on socket clients. > 2. Is it possible to connect a handler function to a socket-stream > (like in cmucl) . What I want is a way to link a lisp function to a > socket stream . The function should be executed if input is > available on the socket stream. This is not possible. > 3. Does the read-eval-print loop has some hooks where we can call a > function when the user is no busy typing something in. This is not possible either; the thing that comes closest to it is to redefine the sys::read-eval-print function so that it calls socket-wait on the *terminal-io* file descriptor and your socket connection, and passes control to the original sys::read-eval-print function when some input has been seen on *terminal-io*. Bruno From mmertens@akam.be Tue Jun 2 12:10:19 1998 Received: from chekov.Belgium.eu.net (relay.eunet.be [192.92.130.25]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id MAA16683 for ; Tue, 2 Jun 1998 12:10:09 -0700 (PDT) Received: from (marc@idialup179.antwerp.eunet.be [193.121.237.179]) by chekov.Belgium.eu.net with ESMTP id VAA11973 for ; Tue, 2 Jun 1998 21:15:01 +0200 (MET DST) Sender: marc@chekov.Belgium.eu.net Message-ID: <35745026.A6AFFDCF@akam.be> Date: Tue, 02 Jun 1998 21:19:03 +0200 From: Marc Mertens X-Mailer: Mozilla 4.05 [en] (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 To: clisp-list@clisp.cons.org Subject: Re: socket interface in clisp References: <199806012018.WAA25227@halles.ilog.fr> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello Bruno, Thanks for your reply on my questions about CLISP and about the testsuite . I looked in the tar file of CLISP and found the test suite you mention , concerning sockets , 1. I found that (listen s) returns T even if nothing is waiting on the stream s so that a following read wil block. 2. - 3. It is a pitty but you can't expect everything from what in my opinion is a very good lisp implementation. As for the read-eval-print loop , can you give me the arguments I have to give to this functions ? Thanks a lot for your help. Marc Mertens mmertens@akam.be From haible@ilog.fr Tue Jun 2 12:29:36 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id MAA16933 for ; Tue, 2 Jun 1998 12:29:35 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.29]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id VAA19643 for ; Tue, 2 Jun 1998 21:33:56 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id VAA21702; Tue, 2 Jun 1998 21:35:06 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id VAA07505; Tue, 2 Jun 1998 21:34:54 +0200 (MET DST) Date: Tue, 2 Jun 1998 21:34:54 +0200 (MET DST) Message-Id: <199806021934.VAA07505@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: Re: socket interface in clisp In-Reply-To: <35745026.A6AFFDCF@akam.be> References: <35745026.A6AFFDCF@akam.be> > concerning sockets , > 1. I found that (listen s) returns T even if nothing is waiting on the > stream s so that a following read wil block. (listen s) /= nil should guarantee that at least one character can be read (not necessarily one form or one line). Can you post some code which shows that this guarantee is not fulfilled? > 2. - 3. As for the read-eval-print loop , can you give me the > arguments I have to give to this functions ? (sys::read-eval-print prompt &optional command-list) reads a form from *standard-input*, evaluates it and prints the values to *standard-output*. The prompt is a string and goes to *standard-output*. Bruno From mmertens@akam.be Wed Jun 3 22:37:05 1998 Received: from chekov.Belgium.eu.net (relay.eunet.be [192.92.130.25]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id WAA24584 for ; Wed, 3 Jun 1998 22:37:04 -0700 (PDT) Received: from (marc@idialup204.antwerp.eunet.be [193.121.237.204]) by chekov.Belgium.eu.net with ESMTP id HAA22889 for ; Thu, 4 Jun 1998 07:42:27 +0200 (MET DST) Sender: marc@chekov.Belgium.eu.net Message-ID: <357634B5.CF479A72@akam.be> Date: Thu, 04 Jun 1998 07:46:29 +0200 From: Marc Mertens X-Mailer: Mozilla 4.05 [en] (X11; I; Linux 2.0.30 i586) MIME-Version: 1.0 To: clisp-list@clisp.cons.org Subject: Re: socket interface in clisp References: <199806012018.WAA25227@halles.ilog.fr> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello Bruno, thanks for your prompt reply , concerning listen you are correct , I found that a read-char was not blocking while a read was (a bug in my interface program) .. Marc From sds@goems.com Tue Jun 16 12:09:36 1998 Received: from uhura.concentric.net (uhura.concentric.net [206.173.119.93]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id MAA17369 for ; Tue, 16 Jun 1998 12:09:35 -0700 (PDT) Received: from cliff.concentric.net (cliff [206.173.119.90]) by uhura.concentric.net (8.8.8/(98/05/18 5.10)) id PAA18091; Tue, 16 Jun 1998 15:16:55 -0400 (EDT) [1-800-745-2747 The Concentric Network] Errors-To: Received: from mute.eaglets.com (ts008d38.phe-pa.concentric.net [209.31.155.146]) by cliff.concentric.net (8.8.8) id PAA16397; Tue, 16 Jun 1998 15:16:52 -0400 (EDT) Received: (from sds@localhost) by mute.eaglets.com (8.8.7/8.8.7) id PAA02059; Tue, 16 Jun 1998 15:16:41 -0400 To: clisp-devel@seagull.cons.org, clisp-list@clisp.cons.org Subject: where are we running? Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes From: Sam Steingold Date: 16 Jun 1998 15:16:41 -0400 Message-ID: Lines: 9 X-Mailer: Gnus v5.5/Emacs 20.2 Is there a way to determine whether the lisp is running under Emacs or under an xterm or a console? The only thing that comes to mind is (equalp (getenv "TERM") "dumb"). -- Sam Steingold, running RedHat5.1 GNU/Linux (http://www.linux.org) Micros**t is not the answer. Micros**t is a question, and the answer is Linux, the choice of the GNU (http://www.gnu.org) generation. All extremists should be taken out and shot. From gadbois@cyc.com Tue Jun 16 13:34:18 1998 Received: from catbert.cyc.com (catbert.cyc.com [207.207.8.5]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id NAA18032 for ; Tue, 16 Jun 1998 13:33:48 -0700 (PDT) Received: from lagavulin.cyc.com (gadbois@lagavulin [207.207.8.13]) by catbert.cyc.com (8.8.8/8.8.8) with ESMTP id PAA09050; Tue, 16 Jun 1998 15:39:49 -0500 (CDT) Received: (from gadbois@localhost) by lagavulin.cyc.com (8.8.8/8.8.8) id PAA10700; Tue, 16 Jun 1998 15:41:18 -0500 Date: Tue, 16 Jun 1998 15:41:18 -0500 Message-Id: <199806162041.PAA10700@lagavulin.cyc.com> From: David Gadbois To: clisp-list@clisp.cons.org In-reply-to: (message from Sam Steingold on Tue, 16 Jun 1998 12:11:08 -0700 (PDT)) Subject: Re: where are we running? References: Date: Tue, 16 Jun 1998 12:11:08 -0700 (PDT) Reply-To: clisp-list@clisp.cons.org Is there a way to determine whether the lisp is running under Emacs or under an xterm or a console? The only thing that comes to mind is (equalp (getenv "TERM") "dumb"). I did: m-x shell bash$ env | grep -i emacs EMACS=t VISUAL=emacs EDITOR=emacs And, indeed, comint-exec-1 in comint.el does: (if (getenv "EMACS") nil (list "EMACS=t")) --David Gadbois From hoehle@zeus.gmd.de Thu Jun 18 07:21:23 1998 Received: from mail.gmd.de (mail.gmd.de [129.26.8.90]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA28371 for ; Thu, 18 Jun 1998 07:21:10 -0700 (PDT) Received: from zeus.gmd.de (zeus.gmd.de [129.26.145.1]) by mail.gmd.de (8.8.8/8.8.8) with SMTP id QAA30785 for ; Thu, 18 Jun 1998 16:28:36 +0200 (MET DST) Received: by zeus.gmd.de id AA29964 (5.67b8/IDA-1.5 for clisp-list@clisp.cons.org); Thu, 18 Jun 1998 16:28:05 +0200 Date: Thu, 18 Jun 1998 16:28:05 +0200 Message-Id: <199806181428.AA29964@zeus.gmd.de> From: Joerg.Hoehle@gmd.de (Joerg Hoehle) To: clisp-list@clisp.cons.org Subject: what to do CLISP in presence of Lisp stack overflow Content-Type: text Hi, while I'm extremely satisfied with CLISP's behaviour w.r.t. stability, I'm wondering if the current behaviour under a stack overflow can be controlled in some obvious way. Currently, CLISP seems to return to the last REPL, which I think is embodied by SYS::DRIVER. Of course, I know how to write such a driver function on my own (and especially how to exit it), but I'm wondering if there shouldn't be another, "better" way of executing a computation and returning to a known state. The condition system comes to mind, as well as CLISP's non-CLCS LISP:*ERROR-HANDLER. Consider a CLISP running non-interactively, e.g. as a server, executing some remote procedure calls, however (loop (ignore-errors (eval (remote (getthing))))) can be broken out. Should a stack overflow be handled logically by some HANDLER-CASE/BIND like any other serious condition or is there some reason to special case it? (defvar *depth* 0) (defun go-down2 (n) (let ((*depth* n)) (go-down2 (1+ n)))) (defun maybe-safe () (ignore-errors (go-down2 0)) 'ok) CMUCL17f/Solaris exists [Recursive no mapping fault (stack overflow?)] ACL4.2/Solaris returns OK [condition type: SYNCHRONOUS-OPERATING-SYSTEM-SIGNAL] [(lisp-implementation-version) -> "1994-10-26 (October 1994)"] What do you think about this issue? Jo"rg Ho"hle. Joerg.Hoehle@gmd.de http://zeus.gmd.de/~hoehle/amiga-clisp.html From haible@ilog.fr Tue Jun 23 07:13:41 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA28145 for ; Tue, 23 Jun 1998 07:13:38 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id QAA01750 for ; Tue, 23 Jun 1998 16:20:19 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id QAA25800; Tue, 23 Jun 1998 16:21:42 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id QAA17179; Tue, 23 Jun 1998 16:21:15 +0200 (MET DST) Date: Tue, 23 Jun 1998 16:21:15 +0200 (MET DST) Message-Id: <199806231421.QAA17179@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: Re: what to do CLISP in presence of Lisp stack overflow In-Reply-To: <199806181428.AA29964@zeus.gmd.de> References: <199806181428.AA29964@zeus.gmd.de> Joerg Hoehle writes: > Should a stack overflow be handled logically by some HANDLER-CASE/BIND > like any other serious condition or is there some reason to special > case it? It's purely an implementation issue. In order to call some handlers when a stack overflow has occurred, you need some room on the stack in order to execute the handlers. Such room is not currently kept ready by clisp. Bruno From haible@ilog.fr Tue Jun 23 08:51:47 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA28822 for ; Tue, 23 Jun 1998 08:51:45 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id RAA05736; Tue, 23 Jun 1998 17:58:19 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id RAA01739; Tue, 23 Jun 1998 17:59:34 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id RAA04859; Tue, 23 Jun 1998 17:59:07 +0200 (MET DST) Date: Tue, 23 Jun 1998 17:59:07 +0200 (MET DST) Message-Id: <199806231559.RAA04859@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: win95 and winnt clisp crashes Cc: ddc@mailhost.techone.com, vipul@mailhost.techone.com Hi, Those who have had clisp crashing on Windows NT and Windows 95 can try new binaries. The new clisp snapshot of Pierpaolo Bernardi contains a more robust memory management. The snapshot and the new binaries are on ftp://clisp.cons.org/pub/lisp/clisp/snapshots/. clisp-win32-fast.zip was built with -DNO_TYPECODES and should be more robust than the previous release from September 1997. clisp-win32-safe.zip was built with -DNO_TYPECODES -DNO_GENERATIONAL_GC -DNO_SELFMADE_MMAP and should be even more robust than clisp-win32-fast.zip. If you download these binaries and had problems with earlier binaries, please mail me (whether they work for you or not). I would like to know whether these binaries are indeed as robust as I expect. Greetings, Bruno From hoehle@zeus.gmd.de Tue Jun 23 14:09:21 1998 Received: from mail.gmd.de (mail.gmd.de [129.26.8.90]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id OAA01279 for ; Tue, 23 Jun 1998 14:09:08 -0700 (PDT) Received: from zeus.gmd.de (zeus.gmd.de [129.26.145.1]) by mail.gmd.de (8.8.8/8.8.8) with SMTP id XAA13160 for ; Tue, 23 Jun 1998 23:17:12 +0200 (MET DST) Received: from amiga ([141.12.144.207]) by zeus.gmd.de with SMTP id AA09631 (5.67b8/IDA-1.5 for ); Tue, 23 Jun 1998 23:16:39 +0200 Date: Tue, 23 Jun 1998 23:16:39 +0200 Message-Id: <199806232116.AA09631@zeus.gmd.de> From: Joerg.Hoehle@gmd.de (Jorg-Cyril Hohle) To: clisp-list@clisp.cons.org Subject: Re: what to do CLISP in presence of Lisp stack overflow In-Reply-To: <199806231421.QAA17179@halles.ilog.fr> References: <199806231421.QAA17179@halles.ilog.fr> Bruno Haible wrote: > Joerg Hoehle writes: > > Should a stack overflow be handled logically by some HANDLER-CASE/BIND > It's purely an implementation issue. In order to call some handlers when > a stack overflow has occurred, you need some room on the stack in order > to execute the handlers. Such room is not currently kept ready by clisp. Silly me. I should have avoided HANDLER-BIND and mentioned HANDLER-CASE only, as one can presumed enough stack is unwound then, whereas HANDLER-BIND piles up even more. > (ignore-errors (unwind-protect (foo 0) (print "hi"))) *** - De'bordement de pile de programme : RAZ "hi" > The above shows that the stack is not completely dropped but still unwound properly (presumably if room is available), so CLISP may as well look for the next error handler frame and unwind down to that (or farther below until there's enough room). Regards, Jo"rg Ho"hle. Joerg.Hoehle@gmd.de http://zeus.gmd.de/~hoehle/amiga.html From DIGIORM@questdiagnostics.com Wed Jun 24 06:27:33 1998 Received: from tbrims01.questdiagnostics.com (tbrims01.questdiagnostics.com [208.148.238.5]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id GAA05769 for ; Wed, 24 Jun 1998 06:27:33 -0700 (PDT) Received: by tbrims01.questdiagnostics.com with SMTP (Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63) id <01BD9F53.FEB33020@tbrims01.questdiagnostics.com>; Wed, 24 Jun 1998 09:39:34 -0400 Message-ID: From: "DiGiorgi, Matthew" To: Multiple recipients of list , Bruno Haible Subject: RE: win95 and winnt clisp crashes Date: Wed, 24 Jun 1998 09:35:00 -0400 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit I tried both the update _fast and _safe binaries with the result that: 1 On my Compaq 566 both ran fine as did the previous 1997-09-19 set. 2 On my Dell LM133 the 1997-09-19 version died during memory expansion, the new sets both refused to load altogether. Error msg is: C:\>clisp Cannot reserve address range 0x1A370000-0x668DFFFF . GetLastError() = 0x8 (ERROR_NOT_ENOUGH_MEMORY): Not enough storage is available to process this command. D:\CLISP-~2\LISP.EXE: Not enough memory for Lisp. C:\> Matthew ---------- >From: Bruno Haible >To: Multiple recipients of list >Subject: win95 and winnt clisp crashes >Date: Tuesday, June 23, 1998 11:52AM > > > >Hi, > >Those who have had clisp crashing on Windows NT and Windows 95 can try >new binaries. > >The new clisp snapshot of Pierpaolo Bernardi contains a more robust >memory management. The snapshot and the new binaries are on >ftp://clisp.cons.org/pub/lisp/clisp/snapshots/. > > clisp-win32-fast.zip was built with -DNO_TYPECODES > and should be more robust than the previous > release from September 1997. > > clisp-win32-safe.zip was built with -DNO_TYPECODES > -DNO_GENERATIONAL_GC > -DNO_SELFMADE_MMAP and should be even more > robust than clisp-win32-fast.zip. > >If you download these binaries and had problems with earlier binaries, >please mail me (whether they work for you or not). I would like to know >whether these binaries are indeed as robust as I expect. > >Greetings, > >Bruno > From haible@ilog.fr Wed Jun 24 08:45:22 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA06876 for ; Wed, 24 Jun 1998 08:45:17 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id RAA18202 for ; Wed, 24 Jun 1998 17:51:54 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id RAA28281; Wed, 24 Jun 1998 17:53:18 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id RAA23484; Wed, 24 Jun 1998 17:52:50 +0200 (MET DST) Date: Wed, 24 Jun 1998 17:52:50 +0200 (MET DST) Message-Id: <199806241552.RAA23484@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: RE: win95 and winnt clisp crashes In-Reply-To: References: Matthew DiGiorgi writes: > I tried both the update _fast and _safe binaries with the result that: > > 2 On my Dell LM133 the 1997-09-19 version died during memory expansion, > the new sets both refused to load altogether. Error msg is: > > C:\>clisp > Cannot reserve address range 0x1A370000-0x668DFFFF . GetLastError() = > 0x8 (ERROR_NOT_ENOUGH_MEMORY): Not enough storage is available to > process this command. Thank you for testing. There is indeed one more variation possible. On ftp://clisp.cons.org/pub/lisp/clisp/snapshots/: clisp-win32-safest.zip built with -DNO_TYPECODES -DNO_GENERATIONAL_GC -DNO_SELFMADE_MMAP -DNO_TRIVIALMAP and should be even more robust than clisp-win32-safe.zip. How does this one behave? Bruno ! To unsubscribe from the clisp-list mailing list, send mail to ! ! listproc@clisp.cons.org ! ! including the two words "unsubscribe clisp-list" as message body. ! From DIGIORM@questdiagnostics.com Thu Jun 25 07:57:06 1998 Received: from tbrims01.questdiagnostics.com (tbrims01.questdiagnostics.com [208.148.238.5]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id HAA13422 for ; Thu, 25 Jun 1998 07:57:05 -0700 (PDT) Received: by tbrims01.questdiagnostics.com with SMTP (Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63) id <01BDA029.BC6D87C0@tbrims01.questdiagnostics.com>; Thu, 25 Jun 1998 11:09:35 -0400 Message-ID: From: "DiGiorgi, Matthew" To: Multiple recipients of list , Bruno Haible Subject: RE: win95 and winnt clisp crashes Date: Thu, 25 Jun 1998 11:15:00 -0400 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hello there. Well I tried the _safest updates on both machines and they both failed identically. C:\>clisp *** - Program stack overflow. RESET *** - Program stack overflow. RESET *** - Program stack overflow. RESET ... (continues until ^c segfaults) No problem testing, I would really like this to work. Matthew ---------- >From: Bruno Haible >To: Multiple recipients of list >Subject: RE: win95 and winnt clisp crashes >Date: Wednesday, June 24, 1998 11:48AM > > >Matthew DiGiorgi writes: > >> I tried both the update _fast and _safe binaries with the result that: >> >> 2 On my Dell LM133 the 1997-09-19 version died during memory expansion, >> the new sets both refused to load altogether. Error msg is: >> >> C:\>clisp >> Cannot reserve address range 0x1A370000-0x668DFFFF . GetLastError() = >> 0x8 (ERROR_NOT_ENOUGH_MEMORY): Not enough storage is available to >> process this command. > >Thank you for testing. There is indeed one more variation possible. On >ftp://clisp.cons.org/pub/lisp/clisp/snapshots/: > > clisp-win32-safest.zip built with -DNO_TYPECODES -DNO_GENERATIONAL_GC > -DNO_SELFMADE_MMAP -DNO_TRIVIALMAP and should > be even more robust than clisp-win32-safe.zip. > >How does this one behave? > > Bruno > > >! To unsubscribe from the clisp-list mailing list, send mail to ! >! listproc@clisp.cons.org ! >! including the two words "unsubscribe clisp-list" as message body. ! > From zbyszek@mimuw.edu.pl Thu Jun 25 08:49:42 1998 Received: from ghost.mimuw.edu.pl (ghost.mimuw.edu.pl [148.81.13.1]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA14035 for ; Thu, 25 Jun 1998 08:49:18 -0700 (PDT) Received: from [148.81.12.171] (pc4390a.mimuw.edu.pl [148.81.12.171]) by ghost.mimuw.edu.pl (8.7.5/8.7.3) with SMTP id RAA14323 for ; Thu, 25 Jun 1998 17:51:50 +0200 From: "Zbyszek Jurkiewicz" Date: Thu, 25 Jun 98 17:55:51 CST Message-Id: <726.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> X-POPmail-Charset: English To: clisp-list@clisp.cons.org Subject: Special variables It looks like in CLISP it is enough to declare variable special where it is bound. Then any use of it in the extent (I mean dynamic scope) of this declaration is special, even if variable is not declared as such. Is it intended effect, because it seems to me that CLtL does not prescribe this behavior? Example (in CLISP): > (setq x 5) 5 > (defun f1 (x) (declare (special x) (f2 x 5) (print x) 'ok) F1 > (defun f2 (y z) (setq x 7)) F2 > (f1 3) 3 'OK > x 5 Best regards, Zbyszek Jurkiewicz University of Warsaw From haible@ilog.fr Sat Jun 27 09:49:09 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA01431 for ; Sat, 27 Jun 1998 09:49:06 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id SAA26908 for ; Sat, 27 Jun 1998 18:56:14 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id SAA21935; Sat, 27 Jun 1998 18:57:42 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id SAA28693; Sat, 27 Jun 1998 18:57:10 +0200 (MET DST) Date: Sat, 27 Jun 1998 18:57:10 +0200 (MET DST) Message-Id: <199806271657.SAA28693@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: Re: Special variables In-Reply-To: <726.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> References: <726.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> Zbyszek Jurkiewicz writes: > It looks like in CLISP it is enough to declare variable special > where it is bound. Then any use of it in the extent (I mean dynamic > scope) of this declaration is special, even if variable is not declared > as such. No. This is not what ANSI CL specifies, and it is not what CLISP does. > Example (in CLISP): > > > (setq x 5) > 5 > > (defun f1 (x) (declare (special x) (f2 x 5) (print x) 'ok) > F1 > > (defun f2 (y z) (setq x 7)) > F2 > > (f1 3) > 3 > 'OK > > x > 5 I cannot reproduce this: > (setq x 5) 5 > (defun f1 (x) (declare (special x)) (f2 x 5) (print x) 'ok) F1 > (defun f2 (y z) (setq x 7)) F2 > (f1 3) 7 OK > x 5 Only if I remove the `(declare (special x))', does `print' print 3. But then the final value of x is the one set by f2: > (setq x 5) 5 > (defun f1 (x) (f2 x 5) (print x) 'ok) F1 > (defun f2 (y z) (setq x 7)) F2 > (f1 3) 3 OK > x 7 Bruno From zbyszek@mimuw.edu.pl Mon Jun 29 06:45:01 1998 Received: from ghost.mimuw.edu.pl (ghost.mimuw.edu.pl [148.81.13.1]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA12418 for ; Mon, 29 Jun 1998 06:44:55 -0700 (PDT) Received: from [148.81.12.171] (pc4390a.mimuw.edu.pl [148.81.12.171]) by ghost.mimuw.edu.pl (8.7.5/8.7.3) with SMTP id PAA14977 for ; Mon, 29 Jun 1998 15:47:59 +0200 From: "Zbyszek Jurkiewicz" Date: Mon, 29 Jun 98 15:52:26 CST Message-Id: <522.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> X-POPmail-Charset: English To: clisp-list@clisp.cons.org Subject: Re: Special variables On Sat, 27 Jun 1998 09:50:00 -0700 (PD, Bruno Haible wrote: >Zbyszek Jurkiewicz writes: >> It looks like in CLISP it is enough to declare variable special >> where it is bound. Then any use of it in the extent (I mean dynamic >> scope) of this declaration is special, even if variable is not declared >> as such. > >No. This is not what ANSI CL specifies, and it is not what CLISP does. >I cannot reproduce this: > > >> (setq x 5) >5 >> (defun f1 (x) (declare (special x)) (f2 x 5) (print x) 'ok) >F1 >> (defun f2 (y z) (setq x 7)) >F2 >> (f1 3) > >7 >OK >> x >5 > You are right, I should copy the dribble instead of writing by hand. However, my point is valid. Which X is being set by F2? The parameter of F1, because this one is printed by it. However F2 should have no access to the inside of F1 and should set the global X. But global X remains 5. Best regards Zbyszek From haible@ilog.fr Mon Jun 29 07:24:16 1998 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA12836 for ; Mon, 29 Jun 1998 07:24:15 -0700 (PDT) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.8.8/8.8.7) with ESMTP id QAA00951 for ; Mon, 29 Jun 1998 16:31:26 +0200 (MET DST) Received: from halles.ilog.fr (halles.ilog.fr [172.16.1.96]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id QAA16056; Mon, 29 Jun 1998 16:32:53 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by halles.ilog.fr (8.8.5/8.8.5) id QAA03703; Mon, 29 Jun 1998 16:32:20 +0200 (MET DST) Date: Mon, 29 Jun 1998 16:32:20 +0200 (MET DST) Message-Id: <199806291432.QAA03703@halles.ilog.fr> To: clisp-list@clisp.cons.org Subject: Re: Special variables In-Reply-To: <522.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> References: <522.zbyszek@ghost.mimuw.edu.pl_POPMail/PC_3.2.2> Zbyszek Jurkiewicz writes: > > > (setq x 5) > > 5 > > > (defun f1 (x) (declare (special x)) (f2 x 5) (print x) 'ok) > > F1 > > > (defun f2 (y z) (setq x 7)) > > F2 > > > (f1 3) > > > > 7 > > OK > > > x > > 5 > > However, my point is valid. Which X is being set by F2? The parameter > of F1, because this one is printed by it. However F2 should have > no access to the inside of F1 and should set the global X. > But global X remains 5. If X in F1 were a lexical variable, then F2 should not have access to the X in F1. But you have declared X in F1 special, which (according to the CL HyperSpec, section 3.1.2.1.1.2) makes it a "dynamic variable". Furthermore: A dynamic variable can be referenced outside the dynamic extent of a form that binds it. Such a variable is sometimes called a ``global variable'' but is still in all respects just a dynamic variable whose binding happens to exist in the global environment rather than in some dynamic environment. This means that there are no two bindings of X in the above example. The value given by entering F1 _is_ the value in the global environment, and is therefore seen by F2. CLISP, CMUCL, GCL and ACL all agree on this issue. Bruno From sds@goems.com Mon Jun 29 14:42:14 1998 Received: from darius.concentric.net (darius.concentric.net [207.155.184.79]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id OAA15730 for ; Mon, 29 Jun 1998 14:42:11 -0700 (PDT) Received: from newman.concentric.net (newman.concentric.net [207.155.184.71]) by darius.concentric.net (8.8.8/(98/04/23 5.10)) id RAA00376; Mon, 29 Jun 1998 17:51:05 -0400 (EDT) [1-800-745-2747 The Concentric Network] Errors-To: Received: from mute.eaglets.com (ts008d23.phe-pa.concentric.net [209.31.155.131]) by newman.concentric.net (8.8.8) id RAA01501; Mon, 29 Jun 1998 17:51:02 -0400 (EDT) Received: (from sds@localhost) by mute.eaglets.com (8.8.7/8.8.7) id RAA01580; Mon, 29 Jun 1998 17:50:53 -0400 To: clisp-devel@seagull.cons.org, clisp-list@clisp.cons.org Subject: 1998-06-29-compile-file-values.diff Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes From: Sam Steingold Date: 29 Jun 1998 17:50:53 -0400 Message-ID: Lines: 36 X-Mailer: Gnus v5.5/Emacs 20.2 1998-06-29 Sam Steingold * compiler.lsp (compile-file): in ANSI CL, compile-file is required to return 3 values. -- Sam Steingold, running RedHat5.1 GNU/Linux (http://www.linux.org) Micros**t is not the answer. Micros**t is a question, and the answer is Linux, the choice of the GNU (http://www.gnu.org) generation. Those who can't write, write manuals. *** compiler.lsp.old Mon Jun 29 17:03:15 1998 --- compiler.lsp Mon Jun 29 17:43:52 1998 *************** *** 12970,12978 **** ) ) ) (c-comment "~%") ! (setq compilation-successful ! (zerop *error-count*) ; Wert T, falls Compilation erfolgreich ! ) ) ) (when new-output-stream (terpri *fasoutput-stream*) (close *fasoutput-stream*) (close *liboutput-stream*) --- 12970,12980 ---- ) ) ) (c-comment "~%") ! (setq compilation-successful (zerop *error-count*)) ! (values (if compilation-successful output-file nil) ! (if (zerop *warning-count*) nil *warning-count*) ! (if (zerop *error-count*) nil *error-count*)) ! ) ) (when new-output-stream (terpri *fasoutput-stream*) (close *fasoutput-stream*) (close *liboutput-stream*) From sds@goems.com Mon Jun 29 15:16:09 1998 Received: from darius.concentric.net (darius.concentric.net [207.155.184.79]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id PAA16123 for ; Mon, 29 Jun 1998 15:16:08 -0700 (PDT) Received: from mcfeely.concentric.net (mcfeely.concentric.net [207.155.184.83]) by darius.concentric.net (8.8.8/(98/04/23 5.10)) id SAA07297; Mon, 29 Jun 1998 18:25:03 -0400 (EDT) [1-800-745-2747 The Concentric Network] Errors-To: Received: from mute.eaglets.com (ts008d23.phe-pa.concentric.net [209.31.155.131]) by mcfeely.concentric.net (8.8.8) id SAA16844; Mon, 29 Jun 1998 18:25:01 -0400 (EDT) Received: (from sds@localhost) by mute.eaglets.com (8.8.7/8.8.7) id SAA01859; Mon, 29 Jun 1998 18:24:55 -0400 To: clisp-list@clisp.cons.org Subject: sorry!!! Re: 1998-06-29-compile-file-values.diff References: Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes From: Sam Steingold In-Reply-To: Sam Steingold's message of Mon, 29 Jun 1998 14:44:09 -0700 (PDT) X-Mailer: Gnus v5.5/Emacs 20.2 Date: 29 Jun 1998 18:24:54 -0400 Message-ID: Lines: 13 I can't believe I did it again! Sorry about sending this to the wrong list! >>>> In message >>>> Sent on Mon, 29 Jun 1998 14:44:09 -0700 (PDT) >>>> Honorable Sam Steingold writes >>>> on the subject of "1998-06-29-compile-file-values.diff": -- Sam Steingold, running RedHat5.1 GNU/Linux (http://www.linux.org) Micros**t is not the answer. Micros**t is a question, and the answer is Linux, the choice of the GNU (http://www.gnu.org) generation. There's always free cheese in a mousetrap.