emacs-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Latest 21.3 CVS and MS Windows


From: Jason Rumney
Subject: Re: Latest 21.3 CVS and MS Windows
Date: 06 Feb 2003 20:02:56 +0000
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50

Frank Schmitt <address@hidden> writes:

> Shouldn't this be mentioned in section Building of nt/INSTALL?

nt/INSTALL is intended for the released version of Emacs, where this
step is unnecessary.  CVS users are expected to either read
INSTALL-CVS or cope with it when they get that error the first time
they build from CVS. Subsequent builds will not require a make
bootstrap (but you'll need to "make recompile" in the lisp subdir to
generate any new .elc files).


> > Is not broken, it's simply not supported, because most Windows users
> > don't have the tools to build info files. You can get a working makeinfo
> > from http://www.simtel.net/pub/djgpp/v2gnu/ (get txi42b.zip) and
> > manually build the info files (that's what I do). If you get make errors
> > try taking Cygwin bin directory from the PATH.
> 
> OK, I'll do so. However, wouldn't it be nice if configure.bat checked if
> Makeinfo is available and if it is change the Makefile accordingly?

The intent is to assume that it is available and try to build the
info files.  It is only an issue for CVS users, so it is not too much
to expect them to be able to install it.  Configure tests are
difficult to get right on Windows, because of the variety of shells
that need to be catered for.

> Yes, I had optimization enabled which resulted in a crash whenever I
> tried to display a png or tiff file. If I build without optimization, the
> crashes are gone.

Can one of you try to investigate, perhaps by disabling optimization
options one by one (I think we use -O2, but it is possible to have
finer control).  Once we narrow down the type of optimization that is
causing this, it may become apparent what sort of problem to look for.
The MS compiler I have does not do optimization, so I cannot debug this
directly myself.






reply via email to

[Prev in Thread] Current Thread [Next in Thread]