headers
Aleksej Saushev | 2 Oct 2008 18:48
Picon

Re: pretest for 2.47 #4

Sam Steingold <sds <at> gnu.org> writes:

> Sorry about such a long pretest process.
> Let us hope that it bodes well for the quality of the release.
> Please stress-test clisp cvs head on your favorite platform.

wildcard.o: In function `module__wildcard__init_function_2':
wildcard.c:(.text+0x66): undefined reference to `gnu_fnmatch'

Where do I find it?
How does it differ from POSIX-standard 4.4BSD-derived fnmatch?

--

-- 
HE CE3OH...
Permalink | Reply | 4 comments |
headers
Sam Steingold | 2 Oct 2008 19:13
Picon

Re: pretest for 2.47 #4

Aleksej Saushev wrote:
> Sam Steingold <sds <at> gnu.org> writes:
> 
>> Sorry about such a long pretest process.
>> Let us hope that it bodes well for the quality of the release.
>> Please stress-test clisp cvs head on your favorite platform.
> 
> wildcard.o: In function `module__wildcard__init_function_2':
> wildcard.c:(.text+0x66): undefined reference to `gnu_fnmatch'

does this mean that the spvw_sigsegv.d problem has been resolved?
how?

> Where do I find it?
> How does it differ from POSIX-standard 4.4BSD-derived fnmatch?

this is a replacement for missing/broken fnmatch.
it should be in clisp/src/gllib.
please take a look at modules/wildcard/configure.in: gl_INIT calls
gl_FUNC_FNMATCH_POSIX which checks whether fnmatch is compliant.

what is ac_cv_func_fnmatch_gnu in config.cache?
is fnmatch.h present in build/wildcard?
what is
diff -u modules/wildcard/link.sh.in build/wildcard/link.sh
Permalink | Reply | 3 comments |
headers
Aleksej Saushev | 2 Oct 2008 22:35
Picon

Re: pretest for 2.47 #4

Sam Steingold <sds <at> gnu.org> writes:

> does this mean that the spvw_sigsegv.d problem has been resolved?
> how?

libsigsegv 2.5 breaks 3 of 4 its own tests on Dragonfly.
I don't want to dig into that right now, I've filed problem report
and am waiting for Dragonfly people.

Are there any plans to support libsigsegv 2.6?

--

-- 
HE CE3OH...
Permalink | Reply | 2 comments |
headers
Sam Steingold | 2 Oct 2008 23:02
Picon

Re: pretest for 2.47 #4

Aleksej Saushev wrote:
> Sam Steingold <sds <at> gnu.org> writes:
> 
>> does this mean that the spvw_sigsegv.d problem has been resolved?
>> how?
> 
> libsigsegv 2.5 breaks 3 of 4 its own tests on Dragonfly.

Bruno maintains libsigsegv, I hope he will address this issue.
you might also want to report it to
http://savannah.gnu.org/projects/libsigsegv/
OTOH, given that libsigsegv is at 2.6, I doubt that anything at all need to be 
done - especially if 2.6 passes the tests on Dragonfly (does it?)

> I don't want to dig into that right now, I've filed problem report
> and am waiting for Dragonfly people.

providing the link might help.

> Are there any plans to support libsigsegv 2.6?

yes, sooner or later :-)
we are in pretest now, so I am trying to suppress the urge to make changes.
OTOH, if libsigsegv 2.6 support is required for Dragonfly, I will make the changes.

Sam.
Permalink | Reply | 1 comment |
headers
SourceForge.net | 4 Oct 2008 11:41
Picon
Favicon

[ clisp-Bugs-2146126 ] name conflict with libmysqlclient

Bugs item #2146126, was opened at 2008-10-04 13:41
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=2146126&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: ffi
Group: segfault
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Dmitry Petrov (dmitry_petroff)
Assigned to: Jrg Hhle (hoehle)
Summary: name conflict with libmysqlclient

Initial Comment:
I've tried mysql part of clsql binding and got random segfaults.
After playing with gdb I've found that the reason is than function "my_alloc" exists in both
libmysqlclient.so and lisp.run and mysql code called clisp's my_alloc instead of it's own.
As a temporary solution I've patched clisp's source to set visibility attribute to hidden so my_malloc and
my_realloc won't appear in dynamic symbols table of executable. This perfectly solved problem for me.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=2146126&group_id=1355
(Continue reading)

Permalink | Reply | 0 comments |
headers
Sam Steingold | 2 Oct 2008 03:09
Picon

Re: pretest for 2.47 #4

> * Aleksej Saushev <nfnh <at> vaobk.eh> [2008-09-30 00:18:47 +0400]:
>
>   Hello!
>
> Sam Steingold <sds <at> gnu.org> writes:
>
>> Sorry about such a long pretest process.
>> Let us hope that it bodes well for the quality of the release.
>> Please stress-test clisp cvs head on your favorite platform.
>> Specifically, build clisp in a fresh check out and do "make all check mod-check".
>> Thanks.
>
> I was asked to support DragonflyBSD, and immediatly ran into this:
>
> ln -s gllib/localcharset.h localcharset.h
> cc -I/home/asau/build/pkg/include -I/usr/include  -Igllib -O2 -I/home/asau/build/pkg/include
-I/usr/include -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type
-Wmissing-declarations -Wno-sign-compare -O2 -fexpensive-optimizations -falign-functions=4
-DNO_SINGLEMAP -DNO_TRIVIALMAP -DUNICODE -DDYNAMIC_FFI -I. -c spvw.c
> In file included from spvw.d:23:
> lispbibl.d:9143: warning: register used for two global register variables
> In file included from spvw.d:520:
> spvw_sigsegv.d: In function 'stackoverflow_handler_continuation':
> spvw_sigsegv.d:157: warning: dereferencing 'void *' pointer
> spvw_sigsegv.d:157: error: request for member 'sc_ebx' in something not a structure or union
> *** Error code 1
>
> CVS checkout as of today ("-Dtoday").

please take a look at spvw_sigsegv.d: stackoverflow_context_t is
(Continue reading)

Permalink | Reply | 0 comments |
headers
SourceForge.net | 2 Oct 2008 19:57
Picon
Favicon

[ clisp-Feature Requests-2142848 ] add OpenTURNS module

Feature Requests item #2142848, was opened at 2008-10-02 13:57
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351355&aid=2142848&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: External Modules
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Sam Steingold (sds)
Assigned to: Sam Steingold (sds)
Summary: add OpenTURNS module

Initial Comment:
add interface to OpenTURNS
http://trac.openturns.org/wiki/Documentation

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351355&aid=2142848&group_id=1355
Permalink | Reply | 0 comments |
headers
Sam Steingold | 3 Oct 2008 21:16
Picon

Re: pretest for 2.47 #4

> Aleksej Saushev wrote:
>> Are there any plans to support libsigsegv 2.6?

Bruno had added libsigsegv 2.6 support to CLISP on 2008-08-24.
Permalink | Reply | 0 comments |
headers
clisp-cvs-request | 4 Oct 2008 14:06
Picon

clisp-cvs Digest, Vol 30, Issue 1

Send clisp-cvs mailing list submissions to
	clisp-cvs <at> lists.sourceforge.net

To subscribe or unsubscribe via the World Wide Web, visit
	https://lists.sourceforge.net/lists/listinfo/clisp-cvs
or, via email, send a message with subject or body 'help' to
	clisp-cvs-request <at> lists.sourceforge.net

You can reach the person managing the list at
	clisp-cvs-owner <at> lists.sourceforge.net

When replying, please edit your Subject line so it is more specific
than "Re: Contents of clisp-cvs digest..."

CLISP CVS commits for today

Today's Topics:

   1. clisp/doc mop.xml,2.91,2.92 (Sam Steingold)

----------------------------------------------------------------------

Message: 1
Date: Fri, 03 Oct 2008 19:20:24 +0000
From: Sam Steingold <sds <at> users.sourceforge.net>
Subject: clisp/doc mop.xml,2.91,2.92
To: clisp-cvs <at> lists.sourceforge.net
Message-ID: <E1KlqCO-0003KB-Pj <at> ddv4jf1.ch3.sourceforge.com>

Update of /cvsroot/clisp/clisp/doc
(Continue reading)

Permalink | Reply | 0 comments |
headers
Sam Steingold | 5 Oct 2008 05:35
Picon

Re: pretest for 2.47 #3

> * Don Cohen <qba-fbheprsbetr-kkm <at> vfvf.pf3-vap.pbz> [2008-09-28 20:13:52 -0700]:
>
>  $ cat CVS/R*
>  ffcall
>  :pserver:anonymous <at> cvs.savannah.gnu.org:/sources/libffcall
> (is that what you expected?)
yes

>  > >  $ make check
>  > >   ...
>  > >  (SOCKET-CONNECT 12345 "localhost" :TIMEOUT 30)
>  > >  [SIMPLE-OS-ERROR]: UNIX error 111 (ECONNREFUSED): Connection refused
>  > >  *** - handle_fault error2 ! address = 0x0 not in
>  > >  [0x3341db000,0x334629a60) !
>  > >  SIGSEGV cannot be cured. Fault address = 0x0.
>  ...
>  > yes, this is puzzling to me just as well.
>  > I think some weird interaction of ffi with gc is to "blame"
> In this case, even stranger, an interaction with LACK OF ffi.

exactly!

>  > yep.
>  > this is a clear case for the cvs binary search I described earlier.
>  > please do not think bad of it - this is the way I discovered the ffcall
>  > bug I fixed today.
> I don't think bad of it, but it's more difficult when the same
> directions for building don't work, or when I get different errors.
> I'll try beginning of 2008 and see what happens.
(Continue reading)

Permalink | Reply | 0 comments |

Gmane