[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: ar cru piecewise linking of same named objects
From: |
Robert Boehne |
Subject: |
Re: ar cru piecewise linking of same named objects |
Date: |
Mon, 08 Oct 2001 08:11:19 -0500 |
Kevin Ryde wrote:
>
> I think I've struck a bit of a problem trying the cvs libtool with
> GMP. GMP has some object files with the same name, like mpz/random.lo
> and mpn/random.lo. On vax-dec-ultrix4.5, libtool decides the command
> line limit is 3073, and does the final libgmp library link in two
> pieces. Unfortunately one random.lo happens to be in the first piece
> and one in the second. The library construction uses "ar cru" for
> both pieces and the second therefore replaces the first, whereas I
> hoped to have both in the library.
>
> Leaving aside the question of whether same-named objects are a good
> idea in general, perhaps libtool should be using "ar q" or "ar qs" or
> some such for the second and subsequent pieces.
>
> _______________________________________________
> Libtool mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/libtool
Kevin:
Could you try changing AR_FLAGS in your libtool script
to see if that works?
This problem is really inherent in our design, or an inherent
problem in using objects with the same name in a single
library, depending on your point of view. ;)
I don't really see a way around this, unless some other
combination of ar flags is found to not produce any regressions
on any platform. I'm open to ideas though...
Thanks,
Robert
--
Robert Boehne Software Engineer
Ricardo Software Chicago Technical Center
TEL: (630)789-0003 x. 238
FAX: (630)789-0127
email: address@hidden