Re: DWARF EH for PPC

[Dale Scheetz <dwarf@polaris.net>]

On Tue, 8 Dec 1998, H.J. Lu wrote:

> > 
> > On Tue, 8 Dec 1998, H.J. Lu wrote:
> > 
> > > > > 
> > > > > As far as I know, libm.so in glibc 2.0.7 still exports register_frame:
> > > > 
> > > > The Debian release called 2.0.7u was constructed from the 2.0.7-pre6
> > > > tarballs and originally had the __register_frame_info symbol defined in
> > > > libm. Starting with release 2.0.7u-6 this symbol was removed using the
> > > > patch that was applied to the 2.1 pre-release.
> > > > 
> > > 
> > > So Debian applied a patch for glibc 2.1 on glibc 2.0? That explains
> > > why Debian's glibc 2.0 is different from the real glibc 2.0.
> > > 
> > The "real" glibc 2.0 is different than any of the other "real" glibc 2.0
> > pre-releases. Are you suggesting that this behaviour is correct?
> > 
> 
> I am not suggesting anything. I got different rerults on glibc 2.0
> than Debian since Debian uses some glibc 2.1 patches on glibc 2.0.
> That is all. I was trying to explain why those frame functionas were
> in my libstdc++.so under my glibc 2.0.
> 
My appologies for being so argumentative. As the Debian maintainer who
applied Zack's patch to resolve a bug report, I have been dealing with the
consequences ever since, leaving me in a somewhat frustrated state.

When I get frustrated it is typically because I don't understand what is
happening, which is an adequate representation of my current state.

The static libgcc.a provided by gcc does not provide __register_frame_info
but the same library provided by egcs does provide that symbol.

Once I removed the symbols from the various libraries provided by the
glibc package, according to Zack's patch, the problem then seemed to
center around libstdc++. The versions which provide that symbol allow C++
programs compiled against the "true" 2.0.7-pre6 to run on systems with
the "modified" 2.0.7u, but I am still not convinced even yet that we have
it right.

I would appreciate any insites you can give me into the nature of this
situation. Particularly, should this symbol be static linked at compile
time from the libgcc.a, or is it supposed to be provided at link time,
from some shared library? If so, where?

Thanks for your time,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz                   Phone:   1 (850) 656-9769
      Flexible Software              11000 McCrackin Road
      e-mail:  dwarf@polaris.net     Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-