Re: [Gcl-devel] 2.6.2

[Camm Maguire]

Greetings!

"Mike Thomas" <address@hidden> writes:

> Hi there.
> 
> For some reason gcl 2.6.2 (and current 2.6.1) has gone backwards re gcc
> 3.4.0 on Windows, not even being able to compile.  This is apart from the
> need to patch o/makefile to workaround the -I hanging directory name
> separator bug as follows:
> 
> ====================================================================
> $ cvs diff o/makefile

Mike, I'm confused as to why this is needed -- we've always defined
directory variables with a trailing slash.  What exactly is the ' -I
hanging directory name separator' bug and its symptoms?

As for the 3.4.0, the way forward is in two directions:

1) Add -g to the CFLAGS, and give me a gdb backtrace when loading
   gcl_setf.lisp. 

2) Take dated CVS snapshots going back sometime in the last month to
   narrow down the change in question.  I'm asusming the gcc/lib
   install hasn't changed in the intervening time.

and a question:

3) Does the 3.4.0 compile go through with --enable-debug?

Take care,

> Enter passphrase for key '/home/miketh/.ssh/id_dsa':
> Index: o/makefile
> ===================================================================
> RCS file: /cvsroot/gcl/gcl/o/makefile,v
> retrieving revision 1.27.6.9
> diff -r1.27.6.9 makefile
> 7c7
> < HDIR  = ../h/
> ---
> > HDIR  = ../h
> 22c22
> < DECL := $(HDIR)new_decl.h
> ---
> > DECL := $(HDIR)/new_decl.h
> 47c47
> < MP    = $(HDIR)mp.h
> ---
> > MP    = $(HDIR)/mp.h
> 49,52c49,52
> < CMPINCLUDE_FILES=${HDIR}compbas.h ${HDIR}enum.h ${HDIR}object.h
> ${HDIR}vs.h \
> <       ${HDIR}bds.h ${HDIR}frame.h \
> <       ${HDIR}lex.h ${HDIR}eval.h    ${HDIR}funlink.h \
> <       ${HDIR}att_ext.h ${HDIR}compbas2.h ${HDIR}/compat.h
> ${HDIR}/cmponly.h
> ---
> > CMPINCLUDE_FILES=${HDIR}/compbas.h ${HDIR}/enum.h ${HDIR}/object.h
> ${HDIR}/vs.
> h \
> >       ${HDIR}/bds.h ${HDIR}/frame.h \
> >       ${HDIR}/lex.h ${HDIR}/eval.h    ${HDIR}/funlink.h \
> >       ${HDIR}/att_ext.h ${HDIR}/compbas2.h ${HDIR}/compat.h
> ${HDIR}/cmponly.h
> 54c54
> < OTHERS=${HDIR}notcomp.h ${HDIR}rgbc.h ${HDIR}stacks.h
> ---
> > OTHERS=${HDIR}/notcomp.h ${HDIR}/rgbc.h ${HDIR}/stacks.h
> 56c56
> < HFILES        = $(HDIR)include.h $(CMPINCLUDE_FILES) $(OTHERS)
> ---
> > HFILES        = $(HDIR)/include.h $(CMPINCLUDE_FILES) $(OTHERS)
> 118c118
> <       ../xbin/move-if-changed mv tmpx ${HDIR}new_decl.h
> ---
> >       ../xbin/move-if-changed mv tmpx ${HDIR}/new_decl.h
> 
> ============================================================================
> =====
> 
> | to the extent time permits:
> |
> | 1) checkout Version_2_6_2c1
> | 2) report the results of self building under the permutations of the
> |    major configuration and building options:
> |         a) linking:
> |                 1) dlopen
> |                 2) statsysbfd or locbfd (indicate which)
> |                 3) custreloc
> |         b) flavor: ansi or cltl1
> |         c) gbc: SGC or not
> |         d) gcc versions: (if there are several popular but differing
> |                                 versions in wide use)
> 
> MSYS 1.0.9(0.46/3/2) 2003-07-03 07:26 i686 unknown; targ=MINGW32
> GNU bash, version 2.04.0(1)-release (i686-pc-msys); ENV=.profile
> GNU Make version 3.79.1,Built for i686-pc-msys; MAKE_MODE=unix
> gcc.exe (GCC) 3.4.0 (mingw special); targ=MINGW32
> GNU ld version 2.15.90 20040222
>    789320 Thu Jul 03 08:17:28 2003 /bin/msys-1.0.dll
>    135680 Thu Jul 03 08:17:26 2003 /bin/make.exe
>     87552 Sat May 01 19:34:06 2004 /mingw/bin/gcc.exe
>    610304 Mon Feb 23 13:48:22 2004 /mingw/bin/ld.exe
> 
> $ time ./configure --prefix=c:/lang/gcl-2.6.2-cltl1-gcc-3.4.0 >configure.log
> 2>
> &1
> 
> real    4m44.500s
> user    3m22.248s
> sys     2m5.217s
> 
> $ time make >make.log 2>&1
> 
> real    22m31.922s
> user    14m37.751s
> sys     10m38.859s
> 
> Build fails as follows:
> 
> >
> #<"SYSTEM" package>
> 
> SYSTEM>
> *COMMAND-ARGS*
> 
> SYSTEM>loading c:/cvs/stable/gcl-2.6.2-c1/unixport/../lsp/gcl_predlib.lsp
> loading c:/cvs/stable/gcl-2.6.2-c1/unixport/../lsp/gcl_setf.lsp
> 
> 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]:
> *** [saved_pre_gcl] Error 3
> rm init_pre_gcl.lsp
> make[1]: Leaving directory `/c/cvs/stable/gcl-2.6.2-c1/unixport'
> make: *** [unixport/saved_pre_gcl] Error 2
> 
> Cheers
> 
> Mike Thomas
> 
> 
> 
> 
> 
> | 3) report the results of the ansi test suite on the build using the
> |    configuration options you recommend
> | 4) in the ansi-tests directory of the CVS head branch, run the random
> |    tester on your recommended build for as long as practicable, and report
> |    the results, including the parameters specifying the size of the
> |    generated test forms
> | 5) compile maxima 5.9.0 using the cltl1 build (preferably) and report
> |    the results of 'time make check'
> | 6) compile maxima cvs with the ansi build and report the results of
> |    'time make check'
> | 7) compile acl2 2.8 and report the results of 'time make
> |    certify-books-fresh'
> | 8) retrieve axiom cvs, copy or link your build tree to lsp/gcl-2.6.2a;
> |    touch lsp/gcldir; run AXIOM=$(pwd)/mnt/foo make; run 'find int
> |    -name "*.lsp" -exec touch {} \;'; run 'find int -name "*.lisp"
> |    -exec touch {} \;'; run AXIOM=$(pwd)/mnt/foo make; rm
> |    int/input/*.output; run and report 'AXIOM=$(pwd)/mnt/foo time make'
> | 9) (very optional) retrieve nqthm-1992 and pc-nqthm-1992, build, and
> |    report the results of 'time make giant-tests'
> |
> | Please collect any special library or compiler tool versions you need,
> | OS version, etc. and report these as footnotes, or parenthetical
> | remarks.  Also please note any known bugs or issues (apart from ansi
> | test failures), indicating in as much detail as practicable what is
> | known about the problem, and how a would be reader could best help
> | resolve it.  I'm thinking all this should be under a screenful of
> | text.  I'm also hoping to summarize these in table format on the
> | website.  Once collected (whatever we end up collecting, that is),
> | we'll add these as README.foo files into the Version_2_6_2pre branch
> | and CVS head, and then tag (read-only) Version_2_6_2pre as
> | Version_2_6_2.  I'll upload a source tarball to ftp.gnu.org, and then
> | I'd like to request binary packages from the port maintainers as
> | practicable.
> |
> | Comments/suggestions of course always appreciated.
> |
> | Take care,
> |
> | "Vadim V. Zhytnikov" <address@hidden> writes:
> |
> | > Camm Maguire:
> | > > Greetings!
> | > > Just an idea to float:  what do we all think of punting for the time
> | > > being on the remaining windows issues and leaving them for 2.7?  This
> | > > suggestion is predicated on the assumption that fully functional
> | > > maxima, acl2 (and hopefully axiom) can be generated with at least some
> | > > settings/patches reliably and reproducibly with all recent gcc, which
> | > > I think is correct.  Thoughts?
> | > >
> | >
> | > Sounds good to me.  GCL 2.6.1 (2.6.2) is stable enough to be useful on
> | > Windows - I'm judging from the Maxima's point of view.
> | > After all as soon some crucial fix/fixes to Windows problems
> | > will be ready we can release GCL 2.6.3 as bug-fix release.
> | >
> | > --
> | >       Vadim V. Zhytnikov
> | >
> | >       <address@hidden>
> | >
> | >
> | > _______________________________________________
> | > 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
> |
> |
> | _______________________________________________
> | 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