This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Qt on Linux-AXP
- To: jochen at uni-duesseldorf dot de, Quant-X Alpha Linux Support <support at quant-x dot com>, Quant-X Alpha Linux Support <stefan at sime dot com>
- Subject: Re: Qt on Linux-AXP
- From: Richard Henderson <rth at cygnus dot com>
- Date: Wed, 26 Aug 1998 12:07:09 -0700
- Cc: qt-interest at troll dot no, axp-list at redhat dot com, egcs at cygnus dot com, egcs-bugs at cygnus dot com
- References: <199808251523.RAA00632@stefan.sime.com> <98082609025504.01454@bacchus.pc1.uni-duesseldorf.de>
- Reply-To: Richard Henderson <rth at cygnus dot com>
On Wed, Aug 26, 1998 at 08:55:47AM +0200, Jochen Kuepper wrote:
> #0 0x48c6d4 in ?? ()
> #1 0x15555779fd0 in QColor::QColor (this=0x15555b0dab0, r=0, g=0, b=0) \
> at kernel/qcolor.h:132
> #2 0x15555779bac in global constructors keyed to color0 () \
> at kernel/qcolor.cpp:87
> #3 0x155559c1018 in __do_global_ctors_aux () \
> at ../sysdeps/alpha/elf/crtend.S:75
> #4 0x1555574c6ec in _init ()
> (gdb)
>
> The constructor QColor::QColor( r, g, b ) just calls QColor::setRgb(r,g,b),
> but it looks like we don't get there ?
It sounds more like an ld or ld.so bug. Can you step through
QColor::QColor instruction by instruction to see exactly where
things go wrong? Perhaps a vtable was not relocated properly,
or a mistake in .got generation or something. Depends on the
nature of setRgb...
I'll also want version info on the binutils and libc involved.
r~