> From: Marius Vollmer <[EMAIL PROTECTED]>
> Date: 11 Mar 2000 18:06:29 +0100
>
> The segmentation fault only occurs when libguile has been compiled
> with optimizations.
>
> When libguile is compiled without optimizations, it produces a clean
> `abort' in scm_ithrow:642 in this situation becaus
> > It's quite interesting, it seems like a bug in gcc.
> >
> > Please compile this with -O2 and -O2 -fno-gcse and check my sanity:
>
> You are sane:
>
> % gcc -o bug bug.c
> % ./bug
> Aborted (core dumped)
I sent it to gcc-bugs (the problem is still there in 2.95.2) and
got the f
> From: Mikael Djurfeldt <[EMAIL PROTECTED]>
> Date: 29 Apr 2000 01:03:24 +0200
>
> Marius Vollmer <[EMAIL PROTECTED]> writes:
>
> > Guile is in a loop, calling take_signal, marking the signal_async, and
> > then returning from the signal handler only to get the sig11 again,
> > without making a
> From: "Richard Y. Kim" <[EMAIL PROTECTED]>
> Date: Thu, 6 Jul 2000 14:42:02 -0700 (PDT)
>
> I got around to compiling guile 1.4 on Windows 2000 with the latest
> update of cygwin tools. Two problems had to be overcome.
>
> One is the name of "and-let*.scm" file where "*" is illegal under NTFS
> From: Bill Schottstaedt <[EMAIL PROTECTED]>
> Date: Mon, 7 Aug 2000 05:19:10 -0700
>
> Guile 1.4.1 builds and works fine in Mac OS-X (developer
> version), but there's one small glitch in the configure
> process -- it doesn't recognize the name "Darwin" as the
> OS name (I think -- I'm no Mac
> From: "Jacques A. Vidrine" <[EMAIL PROTECTED]>
> Date: Wed, 27 Sep 2000 13:48:50 -0500
>
> Hi,
>
> (setgrent) calls the C library's `endgrent', as does (endgrent).
> This is due to a typo in posix.scm.
>
> --- ice-9/posix.scm.orig Wed Sep 27 13:44:32 2000
> +++ ice-9/posix.scm Wed
> From: Han-Wen Nienhuys <[EMAIL PROTECTED]>
> Date: Sun, 1 Oct 2000 00:42:23 +0200
>
> I get this.
>
> gcc -DHAVE_CONFIG_H -I.. -I./.. -I../libltdl -g -O2 -Wall
> -Wmissing-prototypes -c net_db.c -fPIC -DPIC -o .libs/net_db.lo
> net_db.c:85: conflicting types for `inet_aton'
> From: Han-Wen Nienhuys <[EMAIL PROTECTED]>
> Date: Sun, 22 Oct 2000 01:01:38 +0200
>
> is there a reason why (port?) is not in guile-1.4?
Inherited from R5RS. You can make your own:
(define (port? obj) (or (input-port? obj) (output-port? obj)))
_
> From: Han-Wen Nienhuys <[EMAIL PROTECTED]>
> Date: Sun, 22 Oct 2000 11:46:30 +0200
>
> I don't understand. On page 6 R5RS mentions port?, suggesting it
> should be in.
Thanks for pointing this out, I guess you mean this (from the Info version:)
Disjointness of types
=
No
> From: George Caswell <[EMAIL PROTECTED]>
> Date: Wed, 8 Nov 2000 13:02:00 -0500
>
>OK, yeah, CR/LF is lame - but I'm on a Win32 system here and am lucky
> enough to have installed Cygwin and Guile. The problem is I need to use
> Guile to write some binary files with uniform-vector-write, e
> From: bernard URBAN <[EMAIL PROTECTED]>
> Date: 20 Dec 2000 14:38:54 +0100
>
> The following causes a crash when using continuations for
> simulating a threaded application:
> scm_dowinds (continuation->dynenv,
> !scm_ilength (scm_dynwinds) - continuation->dynenv);
>
Than
> The error I obtain with guile is:
>
> polymer:/farm/rossi/bin$ ./guile-config
> ERROR: In procedure scm_ungetc:
> ERROR: Memory allocation error
I thought I'd seen this before and I found it in the mailing list archives.
Are you trying to install an old version of guile, like 1.3.2 perhaps?
I
> From: Marius Vollmer <[EMAIL PROTECTED]>
> Date: 14 Feb 2001 18:10:45 +0100
>
> What does configure say about inet_aton? Normally, when your system
> has inet_aton (RH should have it), the prototype in net_db.c should
> not be processed.
Only true since 2000-10-01, i.e., the problem is fixed
> From: Mikael Djurfeldt <[EMAIL PROTECTED]>
> Date: Fri, 23 Feb 2001 09:28:35 -0500
>
> The node "C Port Interface" in the reference manual contains the
> following paragraph:
>
> Port basics
> ---
>
> There are two main data structures. A port type object (ptob) is of
> typ
> From: Martin Grabmueller <[EMAIL PROTECTED]>
> Date: Thu, 1 Mar 2001 10:28:55 +0100 (MET)
>
> Hello list,
>
> Guile segfaults when a Unix domain address with a very long pathname is
> passed to calls like `connect', `bind' or `sendto'.
>
> Example:
>
> guile> (define s (socket AF_UNIX SOCK_S
> Date: Sat, 3 Mar 2001 14:50:18 +0100 (MET)
> From: Dirk Herrmann <[EMAIL PROTECTED]>
>
> > soka->sun_family = AF_UNIX;
> > - SCM_ASSERT (SCM_STRINGP (address), address, which_arg, proc);
> > memcpy (soka->sun_path, SCM_STRING_CHARS (address),
> > - 1 + SCM_ST
> Undefined first referenced
> symbol in file
> SUN_LEN .libs/libguile.so
Thanks, I'll add something soon for the case that SUN_LEN is undefined.
___
Bug-guile mailing list
> From: Mingshey <[EMAIL PROTECTED]>
> Date: Sat, 10 Mar 2001 22:21:28 -0800 (PST)
>
> I'm compiling Guile in Cygwin 1.1.8-2 on WinNY/PIII
> The guile library guile-1.4/libguile/.libs/guilelib.a does not recognize
> 'h_errno'
> Running make, I got following error message:
>
> .libs/libguile.a(ne
> Date: Sun, 11 Mar 2001 18:10:38 -0800 (PST)
> From: Mingshey <[EMAIL PROTECTED]>
> Thanks.
> I've tied as you suggested and got a different error message:
>
>
> PATH=.:/usr/local/bin:/usr/bin:/bin:/cygdrive/c/WINDOWS:/cygdrive/c/WINDOWS/COMM
> AND ./guile-doc-snarf net_db.c -DHAVE_CONFIG_H -I
> From: Gary Houston <[EMAIL PROTECTED]>
> Date: 11 Mar 2001 10:58:21 -
>
> > From: Mingshey <[EMAIL PROTECTED]>
> > Date: Sat, 10 Mar 2001 22:21:28 -0800 (PST)
> >
> > I'm compiling Guile in Cygwin 1.1.8-2 on WinNY/PIII
> > The gu
> Date: Sun, 18 Mar 2001 18:42:31 -0800 (PST)
> From: Mingshey <[EMAIL PROTECTED]>
>
> Er... another question:
> I'm doing this job having and-let*.scm removed from the target list from
> ice-9/Makefile. Because I get following error messages:
>
> Making all in ice-9
> make[1]: Entering dire
> From: Alexander Klimov <[EMAIL PROTECTED]>
> Date: Mon, 19 Mar 2001 13:29:43 +0200 (IST)
> with GUILE_LOAD_PATH: .
>
> ;;; WARNING (no code for module (ice-9 and-let-star))
> ;;; The module name (ice-9 and-let*) is deprecated.
> ;;; Use (ice-9 and-let-star) instead.
>
> ERROR: no such module
> From: Alexander Klimov <[EMAIL PROTECTED]>
> Date: Sun, 25 Mar 2001 12:50:33 +0200 (IST)
>
> Hi.
>
> According to `man test' on Solaris 2.6 there is no `-e' primitive for sh,
> but only for ksh. It also exists in bash.
>
> There is an error in check-guile:36
>
> if [ ! -e guile-procedures.tx
> From: Martin Grabmueller <[EMAIL PROTECTED]>
> Date: Thu, 19 Apr 2001 22:24:34 +0200
>
> Hello list,
>
> CVS Guile does not build here with the latest IPv6 changes:
Thanks, I am not surprised. I'm following rfc2553 (March 1999) and it
seems to work OK with recent linux/glibc.
I guess a conf
> From: "Dale P. Smith" <[EMAIL PROTECTED]>
> Date: Thu, 03 May 2001 15:01:43 -0400
>
> This is from cvs guile, around May 1st.
>
> First off, even though AF_INET6 is defined, I don't relly believe there
> is and real support for it. To get socket.c to compile, I #undef'ed
> AF_INET6. Also, uin
> From: "Dale P. Smith" <[EMAIL PROTECTED]>
> Date: Fri, 04 May 2001 15:40:02 -0400
>
> Here is where compiling guile stops with cvs guile, morning of May 4th
> (which contains your check).
> Note: I do a cvs update on my Linux latop running Debian Potato, build
> and install Guile there, then do
> From: Seth Alves <[EMAIL PROTECTED]>
> Date: Tue, 10 Jul 2001 10:52:32 -0700
>
> I feel confused. Can anyone help?
>
> -seth
>
>
> - procedure: load filename
> Load FILE and evaluate its contents in the top-level environment.
> The load paths are searched. If the variable
> From: "Dale P. Smith" <[EMAIL PROTECTED]>
> Date: Wed, 11 Jul 2001 12:31:47 -0400
>
> I was using a string port with guile-gd and the generated images were
> always invalid. The strncpy in st_write.c was stopping at the first NUL
> (as strncpy is supposed to do). Using memcpy seems to work ni
> From: Alex Schroeder <[EMAIL PROTECTED]>
> Date: Mon, 24 Sep 2001 15:07:13 +0200
>
> The documentation for drain-input is confusing -- or I am easily
> confused being a Scheme newbie. :)
>
> - primitive: drain-input port
> Drain PORT's read buffers (including any pushed-back characters)
> From: Alex Schroeder <[EMAIL PROTECTED]>
> Date: Tue, 25 Sep 2001 11:10:29 +0200
> Hm. Maybe the problem is that I don't understand what an "input and
> push-back buffer". Now that you've made the analogy to buffered
> output, it makes much more sense. :)
>
> Nevertheless, I suggest to add
> From: Bill Schottstaedt <[EMAIL PROTECTED]>
> Date: Wed, 10 Oct 2001 06:46:02 -0700
>
> I think that the SGI C compiler/library for Irix 6.3 does not
> include snprintf used in libguile/version.c around line 83.
> I changed it to use sprintf to continue compilation.
>
> Also, less certain abou
scheme-report-environment and null-environment are new features in 1.6
(see NEWS) for R5RS compatibility. However they are only available if
the (ice-9 r5rs) module is used. Shouldn't they be available by
default?
On the other hand, interaction-environment, which is only optional
in R5RS, is av
32 matches
Mail list logo
