Re: [Gcl-devel] STABLE, WINDOWS: read_fasd1() and alloc_relblock()

[Camm Maguire]

Greetings, and thanks so much for this!  Can you please post the
result of objdump -d on clmacs.o produced with both gcc 3.3.1 and gcc
3.3.3?  

Your results are important especially as they appear to rule out any
corruption of gcl itself with gcc version.  

Would either of you know if you would need a CLEAR_CACHE macro
definition, like several risc architectures but unlike Linux x86?
Examples are in the h/ directory.  A simple call to mprotect as in the
amd64 example might suffice if it is indeed needed.

It would be nice if someone could post the gdb backtrace from running
with clmacs.o loaded via compiler::link as I mentioned in an earlier
email.  I can repeat the directions if needed.  

Mike, just a hunch, but can you see if this is affected by restoring
that inefficient extra space buffer that had been in sfasl.c?

These issues are complex enough that we really need a reproducible gdb
backtrace to continue, IMHO.

Take care,

"Billinghurst, David (CALCRTS)" <address@hidden> writes:

> > I played mix and match.  I have:
> >  - gcc-3.3.1 and gcc-3.3.3
> > - CVS gcl from last Friday built with both gcc's
> 
> I have now completed this task.  Comparing maxima builds
> using the two gcc version, there are three files that 
> give different results
>  - src/clmacs.lisp
>  - src/cpoly.lisp
>  - src/ronbrg.lisp
> 
> as discussed, clmacs.lisp causes a crash during the build
> cpoly and rombrg cause problems with the testsuite
> 
> With a gcc-3.3.3 compiled cpoly the testsuite finished with
> 
> Running tests in rtest13s.mac: 16/16 tests passed.
> Running tests in rtest14.mac: 
> Unrecoverable error: Segmentation violation..
> 
> This application has requested the Runtime to terminate it in an unusual way.
> Please contact the application's support team for more information.
> make[1]: *** [gcl-check] Error 3
> make[1]: Leaving directory `/usr/local/src/maxima/tests'
> make: *** [check-recursive] Error 1
> 
> 
> with a gcc-3.3.3 compiled rombrg.lisp the testsuite finished with
> 
> Running test suite with gcl...
> Maxima 5.9.0.1cvs http://maxima.sourceforge.net
> Using Lisp Kyoto Common Lisp GCL 2.6.1 (aka GCL)
> Distributed under the GNU Public License. See the file COPYING.
> Dedicated to the memory of William Schelter.
> This is a development version of Maxima. The function bug_report()
> provides bug reporting information.
> (C1)                            RUN_TESTSUITE()
> 
> Running tests in rtest1.mac: 
> real time       :     13.750 secs
> run-gbc time    :     11.483 secs
> child run time  :      0.000 secs
> gbc time        :      2.267 secs
> 
> 
> 
> 
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gcl-devel
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah