Regression checking vs. my -shared-libgcc patch

[Franz Sirl]

On Sunday 04 February 2001 01:22, Mark Mitchell wrote:
> The regression checker is very mad at me for the -shared-libgcc patch.
>
> I think it's probably something related to the stuff that Franz
> mentioned about the GNU/Linux PowerPC stuff whereby something is
> undefined in the shared version of libgcc.
>
> Unfortunately, I have no very good way of figuring this out, since the
> regression checker just tells me that the tests fail (not why), and
> since I don't have a suitable system for testing.  If anyone else
> would be willing to help out with this, I would be eternally grateful.

It's most probably the shared libgcc thing. I worked around that on linux by 
simply removing eabi.S from LIB2FUNCS_EXTRA, since it's not used on linux 
unless some -m switch is passed.
I wonder if we need a LIB2FUNCS_SHARED_EXTRA? It seems to me none of the 
stuff in eabi.S belongs in a shared libgcc... 

BTW, are the -B options honored while searching for libgcc_s? It seems to me 
that the libstdc++-v3 testsuite used /usr/lib/libgcc_s.so.0 even though a 
valid -B was passed.

Franz.