RE: Request for Obsoletion: DWARF

[Kean Johnston <jkj at caldera dot com>]

> I know this was true on SCO, but I think RJL decided to 
> switch away from
> DWARF due to the fact that we had lots of bugs there.  (Robert?)
While this is true it does indeed cause problems. Right now the most
reliable debugging stuff for SCO is still DWARF1, as the OSR5 CCS is
based on very old code. As it gets updated to something approaching
the mid-1990's this will be less of an issue, but I for one would
be sad to see it go. I have been hacking up a new sco5.h that removes
a lot of the old cruft I introduced for making it both COFf and ELF
aware and I find that when I have DWARF2 enabled, things fall apart.
Admittedly, I usually only test C to any degree, not C++.

One issue I have currently, and I realize this isnt this group's
problem, is that binutils' strip doesn't seem to work on OSR5. If
I forget to remove -g, I end up with HUGE binaries, as neither the
native ELF strip nor the binutils strip work. I need to fix binutils
at some point, and I need to make the native ELF strip DWARF2 aware.
Both will take more time than I have right now but I will work on
it slowly.

So plewase consider this a loud but not bellowing complaint to
keep the code there :-)

Kean