[Top][All Lists]

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

Re: [Gcl-devel] Logs of failed build (ANSI) on WinXP -- configure proces

From: Camm Maguire
Subject: Re: [Gcl-devel] Logs of failed build (ANSI) on WinXP -- configure process
Date: Fri, 03 Sep 2010 15:01:16 -0400
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.2 (gnu/linux)

Gabriel Dos Reis <address@hidden> writes:

> Camm Maguire <address@hidden> writes:
> | Greetings!
> | 
> | Gabriel Dos Reis <address@hidden> writes:
> | 
> | > Donald Winiecki <address@hidden> writes:
> | >
> | > [...]
> | >
> | > | checking build system type... i686-pc-mingw32
> | > | checking host system type... i686-pc-mingw32
> | > | host=i686-pc-mingw32
> | > | The host is canonicalised to i386-pc-mingw32
> | >
> | > Hi Camm,
> | >
> | > I have been wondering why GCL deliberately recategorizes a
> | > i686-pc-mingw32 host as a i386-pc-mingw32 host?  Later on, it uses that
> | > to configure GMP, which might lead to suboptimal code selection on a
> | > host with greater capabilities..
> | >
> | 
> | This was put in by Mike Thomas a long time ago, and arguably should be
> | removed.  He ran into a situation where he tried to move the binary
> | from one machine to another and got an Illegal instruction fault.  gmp
> | is built in (i.e. not a shared library) on these systems.  He put in a
> | configure switch --enable/--disable-common-binary governing this.
> Indeed, I think we can/should remove that now.

So in other words, no one moves windows binaries across machines

> | Even when this is disabled, host=mingw32 defaults to i386.  One can do
> | --host=i686-pc-mingw32 to get a higher capability build if desired.
> | This is of course cross compiling so no cpu detection is pertinent.
> | On a separate front, am a little puzzled why --build fails in place of
> | --host. 
> That is puzzling, because `--host' is supposed to default to `--build'.
> I think it would be welcome to remove the `recanonalization' to i386.

OK, I'll look into this.

> | I think there should be a better way to handle this, and it needs
> | extending to any system building the local gmp.  Suggestions?
> You mean like telling GMP configure to use a specific target triplet?
> --enable-gmp-target=xxx?

I think we already have this with --host=i386... if necessary at the
top level.  We're just relying on the user to figure it out if they
get a segfault when moving a i686 build to a lower machine.

> | BTW, where is open-axiom re: compiling under wine?
> I am looking into that right now and I hope to report back to you soon.
> [ first week of semester here, so things a bit in `panic/emergency' mode ]
> | I did ask about wine running win64 binaries, and it is very
> | experimental.  I'd need root on a amd64 box anyway to support it,
> | which alas I do not have.  If you can suggest a way I might try to
> | extend gcl support here before release, I'd love to hear it.
> I did try to build GCL under Windows 64-bit but configure miserably
> fails doing anything all.  I don't understand why.  I'm able to
> configure other software (including ECL) using standard Autoconf, but
> GCL's configure systematically crashes.  I'll post a log in a separate
> message. 

This would be really helpful.  This is the only win64 environment I'm
aware of, including all of ut, Tim's machines, Don's machines, etc.

What I'm unsure of is whether the PE32 format changes will require
modifications to out loader.  I think this is rather likely.  Luckily
I have a clear doc on this, just nowhere to test.  We might just
release without this, but I've been hearing from people that all they
use these days is 64bit.  Guess I'm a little on the slow side.

I've verified maxima and it looks like acl2 is going through on a
windows XP 32 bit machine.  So I guess Don's earlier XP error was
machine related.  I'm leaving the axiom testing up to you to as
previously discussed.  Please let me know if issues arise here.

So in summary, wine looks just as good as the real thing so far re:
gcl development, which is good news.

About to commit a small patch regarding the expansion of #p"~/"
pathnames.  Esentially use getenv("HOME") when no pwent is available.
This leaves out #p"~joe/", of course.

Am still a little unfamiliar with what standard windws people expect
with this stuff.  Do they run everything in an msys window?  I take it
this is a must to use gcc at all.

Take care,

> -- Gaby

Camm Maguire                                        address@hidden
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah

reply via email to

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