[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Configure bug under Win9x-Cygwin systems.
From: |
Jim . Fougeron |
Subject: |
Re: Configure bug under Win9x-Cygwin systems. |
Date: |
Wed, 24 Apr 2002 13:57:12 -0500 |
Thank you for the follow up,
I have not pursued this further than "getting it to work" for myself (hence
the comment about this is probably not what you would want to do for a
public distribution).
NOTE I do not do much work in the *nix/GNU arena, but have done so since I
work on a couple of things which require GMP (which of course requires, or
at least strongly wants a *nix environment). Up to this point, all of my
successful building had been done on a Win2k system (which is not my main
work dev system or my main home dev system, so it was not very convenient).
I had been have less than ideal success on any Win9x system. I simply
wanted to pass on the information and work around I found which made my
Win9x/Cygwin system able to build GMP successfully.
Note that I just updated Cygwin over the net yesterday, and without this
work-around, the latest release still had serious configuration problems.
If the bug has been fixed in the DJGPP's libc, what would it take to get a
working Cygwin? NOTE rebuilding Cygwin would be beyond my capabilities,
however, if I could replace a file or 2, then I can easily do that. If
this is the wrong forum to ask this, then simply tell me so. I am happy
with the ability to get it running in Win9x. NOTE Win9x is sub optimal to
build under, simply because all the process spawning which happens during a
configure or make is slow (compared with any NT variant), and there is a
LOT of process spawning which takes place.
Jim.
Richard Dawe
<address@hidden To: address@hidden
rve.co.uk> cc: address@hidden,
address@hidden
Sent by: Subject: Re: Configure bug
under Win9x-Cygwin systems.
address@hidden
ve.co.uk
04/23/02 03:34 PM
Hello.
address@hidden wrote:
[snip]
> I think I have found the problem. It is in the check for command line
> parameter length. I have changed .\configure from:
>
> case $host_os in
> msdosdjgpp*)
> # On DJGPP, this test can blow up pretty badly due to problems in
libc
> # (any single argument exceeding 2000 bytes causes a buffer overrun
> # during glob expansion). Even if it were fixed, the result of this
> # check would be larger than it should be.
> lt_cv_sys_max_cmd_len=12288; # 12K is about right
> ;;
This bug has been fixed in DJGPP's libc, but it's only in DJGPP CVS. I
think
this limit is still reasonable for DJGPP, to stop libtool using up large
amounts of memory. (The definitive answer lies in the mail archives for
djgpp-workers at http://www.delorie.com/djgpp/ . I can't remember all of
the
details right now.)
Perhaps Cygwin has some problems with its implementation of glob? Have you
tried asking on the Cygwin mailing lists whether anyone has had a similar
problem with libtool to the one mentioned above with DJGPP?
Regards,
--
Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]