Yaakov (Cygwin/X) wrote:
> Then on a wild hunch, since I had rebased my libstdc++6, I reinstalled
> it from tarball and suddenly everything worked again, including the
> ilmbase test. And sure enough, when I rebased it again, things broke.
>
> So for some reason, libstdc++6 is not rebase-able. Does that give you
> something to work with?
Great detective work there Yaakov. Yes, it's not currently rebaseable, and
I had a look and found out what's going on: the operator new/delete overide
array still points at the old addresses, and that's down to a bad interaction
between weak symbols and dll base relocations.
Basically, this is the exact opposite of the problem we had a while ago:
> 2008-07-31 Takashi Ono
>
> * pe-dll.c (generate_relocs): Don't generate relocs for
> undefined weak symbols.
when an unresolved undefined weak external (hence default value zero) was
getting relocated to a non-zero value; now a resolved (non-zero) weak external
is ending up /not/ getting relocated.
This is going to need fixing in the linker I'm afraid :-( Thanks for
finding it though!
cheers,
DaveK
------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Cygwin-ports-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cygwin-ports-general
