On Fri, Oct 19, 2001 at 05:12:49PM +0100, Neil Jerram wrote:
> >>>>> "Michael" == Michael Carmack <[EMAIL PROTECTED]> writes:
>
> Michael> On versions 1.3.2a, 1.4, 1.5.4, compiles always fail
> Michael> because usleep is apparently de
On Sat, Oct 20, 2001 at 09:46:21AM +0100, Neil Jerram wrote:
> >>>>> "Michael" == Michael Carmack <[EMAIL PROTECTED]> writes:
>
> Michael> On Fri, Oct 19, 2001 at 05:12:49PM +0100, Neil Jerram
> Michael> wrote:
> >>
&
On Sat, Oct 20, 2001 at 08:13:29PM +0200, Marius Vollmer wrote:
> Michael Carmack <[EMAIL PROTECTED]> writes:
>
> > Can the unistd.h bit be altered in the configure.in script on your end,
> > so I don't have to remember to make the change myself in the future.
>
This affects 1.5.4, but not 1.4 or 1.3.2a. Apparently in some sort of
recursive configure, the original configure attempts to pass environment
variables, but during 'libltdl' this seems to cause a problem. Here is
the output:
configure: configuring in libltdl
configure: running /bin/sh './confi
On Mon, Nov 12, 2001 at 11:07:41PM +0100, Marius Vollmer wrote:
> Michael Carmack <[EMAIL PROTECTED]> writes:
>
> > This affects 1.5.4, but not 1.4 or 1.3.2a. Apparently in some sort of
> > recursive configure, the original configure attempts to pass environment
&
On versions 1.3.2a, 1.4, 1.5.4, compiles always fail because usleep is
apparently detected as an int instead of a void. This is a gnu/linux
system, kernel 2.4.10, glibc 2.1.3, gcc 2.95.3, binutils 2.11.2.
Additional information can be found in the build logs attached to this
mail. Lines that sta
Ok, it appears someone has put an absolute path to a header file in
the 'configure.in' script. This occurs on line 275. On systems that
don't store unistd.h under /usr/include the test program will fail and
cause configure to assume the usleep return type is 'int'. This breaks
the compile later
On Mon, Oct 29, 2001 at 08:50:16PM +, Michael Carmack wrote:
Hello bug-guile.
The problem below still seems to be present in a recent CVS checkout.
It seems that the environment variables that I have set are causing
'configure' to fail once it enters libltdl. Apparently the manne
On Mon, Nov 12, 2001 at 11:07:41PM +0100, Marius Vollmer wrote:
Short followup to the previous message: The following is the way that
I originally called 'configure'. Note that I did not explicitly call
configure with the environment variables, but they show up in the
recursive calls:
./configur
On Wed, Jan 02, 2002 at 11:05:06PM +0100, Marius Vollmer wrote:
> Michael Carmack <[EMAIL PROTECTED]> writes:
>
> > The problem below still seems to be present in a recent CVS checkout.
> > It seems that the environment variables that I have set are causing
> >
Guile-oops/1.0.2 doesn't seem to build against guile/1.4.1, apparently
because the order of the CPP arguments has changed for bin/guile-snarf.
(Guile-oops/1.0.2 did build successfully against guile/1.4 for me.)
In 1.4.1, 'guile-snarf --help' reports:
Usage: guile-snarf [-d | -D] [-o OUTFIL
11 matches
Mail list logo
