Re: FYI: duplicate removal patch [Was Re: ok, new libtool for cygwinupdates]

Wed, 04 Apr 2001 21:12:49 -0700

Title: RE: FYI: duplicate removal patch [Was Re: ok, new libtool for cygwinupdates]
forgive me if this has already been mentioned...
 
how about a sub-optimal (but at least an) improvement?
 
something along the lines of
 
set X $deplibs; shift
_last=$1; shift
_deplibs=
for _lib in $*; do
  if [ x$_lib != x$_last ]; then
    _deplibs="$_deplibs $_lib"
  fi
  _last=$_lib
done
deplibs=$_deplibs
 
before: deplibs="-lfoo -lbar -lfoo -lfoo -lbar -lbar -lfoo"
after: deplibs="-lbar -lfoo -lbar -lfoo"
 
This cuts down on all the extraneous stuff *safely*, i think.
 
edward
 
 
----- Original Message -----
Sent: Wednesday, April 04, 2001 11:12 AM
Subject: RE: FYI: duplicate removal patch [Was Re: ok, new libtool for cygwinupdates]


Hello,

I agree with Gary that the release of 1.4 should have duplicate deplibs
removed, I also agree with Michael that the patch should be reversed
in MLB Libtool.  Currently the patch has rendered Libtool unusable
for several of the libraries in my project, and I would prefer not
to ship a patched version of Libtool when I release later this month.
  In the long term there is probably a better solution, but for now
I would prefer it to simply be reversed.

Robert

-----Original Message-----
From: Michael Matz [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, April 03, 2001 6:36 PM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: FYI: duplicate removal patch [Was Re: ok, new libtool for
cygwinupdates]


Ho,

On Tue, 3 Apr 2001, Gary V. Vaughan wrote:
> > IMHO, any duplicate libraries should be removed on
> > the platforms where it is possible.
>
> Agreed.
>
> Unfortunately it is a surprising difficult thing to do properly.  My current
> thinking is that we should add a new configure time variable which will turn
> back on duplicate stripping (for *shared* libs only) on hosts where it is
> known to be harmless.

Which should be most hosts, so we would turn this on unless a platform is
known to misbehave.

> I don't know how we will recognize which libraries in the list are
> shared

At least for .la's we can tell easily.  For other's the code to check this
already is in place, but not run for pass_all platforms (I once wanted to
change libtool in exactly the now proposed way, but then noticed, that
sharedness wasn't readily available for pass_all's, and I wasn't sure if I
wanted to take the performance hit for doing the file_magic test
unconditionally).

> Considering it took 2 years for anyone to report the bug, but only 2 days for
> people to complain about the fix,

:-)

> I am beginning to think that we should deliberately ship 1.4 with the
> bug reintroduced.

For 1.4 at least I don't have an opinion.

> We can then fix it properly in the next release.
> Pragmaticallym, the half cocked fix in place right now seems to be
> more trouble than it is worth.  Comments?

If it would be restored in MLB it would at least make the diffs between
KDE's libtool and CVS-MLB some lines shorter ;-)


Ciao,
Michael.


_______________________________________________
Libtool-patches mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/libtool-patches

Reply via email to