diff mbox

import newer Libtool?

Message ID 20110112173553.GA26529@gmx.de
State New
Headers show

Commit Message

Ralf Wildenhues Jan. 12, 2011, 5:35 p.m. UTC 
Hello,

* Ralf Wildenhues wrote on Mon, Jan 10, 2011 at 07:24:26PM CET:
> * Richard Guenther wrote on Mon, Jan 10, 2011 at 04:24:49PM CET:
> > On Mon, Jan 10, 2011 at 4:16 PM, Paolo Bonzini <bonzini@gnu.org> wrote:
> > >
> > > The main pro is that allowing to build GCC with LTO "sounds as" if it
> > > would give more exposure/coverage to new features.  But without a list
> > > of PRs from Ralf I cannot really judge either.
> > 
> > Hm, but LTO bootstrap works for me right now.  Or is this about
> > things like installing a static libgfortran with LTO bytecode (the only
> > other useful LTO related thing when building GCC)?
> 
> Well, current in-tree libtool will not pass -flto*, -fwhopr*, or
> -fuse-linker-plugin through to the linker, when creating a shared
> library.

D'oh.  I keep forgetting existence of the libtool-ldflags script, so
sorry about that.  If the postdeps computation issues when
-fuse-linker-plugin is passed are not a problem inside GCC either, then
that makes a wholesale Libtool update indeed not a good idea at this
point.

> - linking/testing/installing on HP-UX should work again
>   <http://gcc.gnu.org/ml/gcc-patches/2010-12/msg00893.html>,

In that case I'd still like to ask for committing the above one-line
patch to trunk right away; it had been submitted in time but we simply
didn't get it analyzed in time.  I think that it may even be considered
a regression because due to libquadmath not existing in 4.5 GCC did not
have this issue before.  OK?

Thanks,
Ralf

Here's the patch:

2010-12-10  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>

	* ltmain.sh (relink): Use absolute path when hardcoding with -L.

Comments

Paolo Bonzini Jan. 13, 2011, 9:06 a.m. UTC | #1 
On 01/12/2011 06:35 PM, Ralf Wildenhues wrote:
>> >  - linking/testing/installing on HP-UX should work again
>> >     <http://gcc.gnu.org/ml/gcc-patches/2010-12/msg00893.html>,
> In that case I'd still like to ask for committing the above one-line
> patch to trunk right away; it had been submitted in time but we simply
> didn't get it analyzed in time.  I think that it may even be considered
> a regression because due to libquadmath not existing in 4.5 GCC did not
> have this issue before.  OK?

That's okay if the patch is in upstream Libtool.

Paolo
diff mbox

Patch

Index: ltmain.sh
===================================================================
--- ltmain.sh	(revision 167668)
+++ ltmain.sh	(working copy)
@@ -5928,7 +5928,7 @@ 
 	         test "$hardcode_direct_absolute" = no; then
 		add="$dir/$linklib"
 	      elif test "$hardcode_minus_L" = yes; then
-		add_dir="-L$dir"
+		add_dir="-L$absdir"
 		# Try looking first in the location we're being installed to.
 		if test -n "$inst_prefix_dir"; then
 		  case $libdir in