[Top][All Lists]

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

Re: Emacs 21.3.50 on cygwin bootstrap failed.

From: Eric Hanchrow
Subject: Re: Emacs 21.3.50 on cygwin bootstrap failed.
Date: Thu, 12 Feb 2004 09:20:10 -0800
User-agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3.50 (gnu/linux)

    "Steven Wu" <address@hidden> writes:
    > I took the CVS src of Emacs 21.3.50 to cygwin and did make
    > bootstrap.

    "Harald Maier" <address@hidden> responds:
    I assume you need the mingw-make 3.79.1.

I'd like to understand this better -- Harald says that one needs the
MinGW version of `make' to build Emacs under Cygwin.  But I am not
sure that's completely correct.  My understanding is that

* the Emacs source tree is capable of building two *different* flavors
  of Emacs -- 1) the so-called "Win32" version (which does not require
  the Cygwin DLLs); and 2) the pure Cygwin version (which of course
  does require the Cygwin DLLs).

* To build the Win32 version, you first cd to the `nt' directory.  You
  then have a choice of software with which to do the build: 1)
  Microsoft Visual Studio 6; and 2) the MinGW versions of gcc, make,
  etc., *plus* `cp', `rm', and a few other Unix-style programs.  In
  particular, using Cygwin make doesn't work, nor does using the
  Cygwin version of gcc.

* To build the Cygwin version, you do not cd into `nt'; instead you
  simply do `configure && make bootstrap'.  However, this always fails
  for me; I assume that the failures are simple mistakes in the Emacs
  source, which will get fixed eventually.  I don't believe the
  failures are due to Emacs never having been ported to Cygwin;
  because some files refer to Cygwin (`configure.in', for example).  I
  also assume that one does *not* use the MinGW tools when building
  the Cygwin version.

So have I understood everything correctly?

Implicit in my question is a criticism of the file nt/INSTALL; that
file has cleared up my confusion, and seems out of date.  I'd be
delighted to update it if I only knew the facts.

In the practice of computing, where we have so much latitude for
making a mess of it, mathematical elegance is not a dispensable
luxury, but a matter of life and death.

        -- Edsger W. Dijkstra: My Hopes of Computing Science (EWD  709)

reply via email to

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