[3.2] Can ada/6160 be cured from 3.3 or trunk?

[Christian Cornelssen]

Hello again,

I have got one step further.  I used STAGE1_CFLAGS="-march=i686 -O2",
which apparently caused either Debian's gcc-2.95.4 or the gnat1 as of
gnatgcc v2.8.1 to miscompile the stage1 gnat1 such that its *.ali
outputs each got just a single linefeed.  When I removed the
"-march=i686" option, everything went fine.

I'll report that to GNATS's ada/6160 too, as it could be related.

Regards,

Christian Cornelssen