Mixed success building gcc-2.95.3.test2

[Martin Buchholz]

Executive summary: gcc-2.95.3.test2 has a C++ regression on hppa2.0-hp-hpux10.20

>>>>> "Bernd" == Bernd Schmidt <bernds@redhat.com> writes:

Bernd> On Sat, 20 Jan 2001, Martin Buchholz wrote:
>> Wounded by my previous experiences building with objdir == srcdir, I
>> tried the recommended objdir != srcdir for subsequent builds.
>> 
>> Executive summary:
>> Success on i686-pc-linux-gnu, sparc-sun-solaris2.7, rs6000-ibm-aix4.3.0.0.
>> Failure on hppa2.0-hp-hpux10.20

Bernd> Please let me know if this is a regression, i.e. whether 2.95.2 works
Bernd> properly.  I think I've had a successful bootstrap on that target, so I
Bernd> believe there must be something strange about your system, but I'd like
Bernd> confirmation.

I tried to bootstrap gcc-2.95.2 on hppa2.0-hp-hpux10.20 in the same
way, with objdir != srcdir, and srcdir pristine.

It fails in compare:

Bootstrap comparison failure!
alias.o differs
alloca.o differs
bitmap.o differs
c-aux-info.o differs
...

Which is probably (relatively) harmless due to the usual reasons like
timestamps in object files or something.  But of course a bug in 2.95.2.

I see an apparently functional g++ binary.

(martin@hpux) ~cdxw/gcc-2.95.2 $ gcc/g++ --version
2.95.2

I imagine that make install will work, but I don't want to try it.