RE: [Gcl-devel] Windows issues

[Mike Thomas]

Hi all.

I just realised something I should have noticed the other day.

Jacques has grabbed a copy of the experimental ANSI package I put up for Sam
G. recently.  It was never meant for wide release as it is inherently
unstable.  Those "Unrecoverable error: bind stack overflow" errors are
typical of that bad package - you should really be getting an error message
about the number of arguments to (gc$).

Unfortunately the ANSI package has still not been replaced by newer packages
on the GCL Savannah file page as our retrieval system has gone AWOL.

Camm - I've built a 2.6.1 package today:

        gcl_2.6.1_mingw32_japi_xdr_20030919.zip

for retrieval to your new "temporary!" distribution site.

Jacques - I suggest you preferably either get a copy of that package once
Camm has been able to put it up or failing that, some other one (eg 2.5.2,
2.5.3) which does not contain the keyword "ansi" in the file name.  I've
built ACL2 this morning and "make certify-books" and ":mini-proveall" seemed
to be alright.

In any event, please let me know how you go with it.

Cheers

Mike Thomas.



| -----Original Message-----
| From: address@hidden
| [mailto:address@hidden
| Behalf Of Matt Kaufmann
| Sent: Wednesday, September 17, 2003 10:52 PM
| To: address@hidden
| Cc: address@hidden; address@hidden
| Subject: [Gcl-devel] Windows issues
|
|
| Hi, Camm --
|
| I've received some questions from an ACL2 user who is having
| trouble with GCL
| on Windows.  He tells me that he has recently obtained GCL via
| cvs, which he
| says is providing a recent flavor of 2.5.3.  If there are
| previous versions
| that don't have the problem below he would like to know what they
| are.  Anyhow,
| here are those questions, which I forward with his permission.
| By the way, the
| function gc$ mentioned below is just an interface si::gbc; he
| tried (si::gbc)
| and got the same problem.
|
| Thanks --
| -- Matt
|   Date: Wed, 17 Sep 2003 08:59:22 +0200
|   From: Jacques Mequin <address@hidden>
|   Organization: JSM software infrastructure, Texas Instruments
| France (Nice)
|   X-Accept-Language: en
|   To: Matt Kaufmann <address@hidden>, Jacques Mequin
| <address@hidden>
|   Subject: Control-C
|   Content-Type: text/plain; charset=us-ascii
|
|
|   On windows XP, I have ran a "make large" after the installation of the
|   cvs gcl-2_5_3-ansi-japi-xdr_20030701_mingw32
|
|   So, the "garbage collection" bug is gone and I can restart my work
|   under my Xemacs environment
|
|   However, you need to know the following
|
|     * the install has now to take care of a GCL path access to a
|       "dll" named "oncrpc.dll"
|
|     * calling (gc$) does produce now the following
|
|        Unrecoverable error: bind stack overflow
|        This application has requested the Runtime to terminate it in an
|        unusual way
|        Please contact the application's support team for more information
|        Process acl2 exited abnormally with code 3
|
|      * but, the most annoying issue is the "control-c" that I used to do
|        when a proof was running out of control
|
|        This now results to the same above message
|
|        Unrecoverable error: bind stack overflow
|        This application has requested the Runtime to terminate it in an
|        unusual way
|        Please contact the application's support team for more information
|
|        I do not know if this is a problem occurring only on
| recent GCL release
|        but I am back to a situation where my chance to lose my session is
|        as great as before
|
|   Regards,
|     Jacques
|
|
| _______________________________________________
| Gcl-devel mailing list
| address@hidden
| http://mail.gnu.org/mailman/listinfo/gcl-devel
|
|