This is the mail archive of the java-patches@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: 3.0.1 PATCH: Allow java/net/natPlainDatagramSocketImpl.cc to compile on Tru64 UNIX V5.1


>>>>> "Rainer" == Rainer Orth <ro@TechFak.Uni-Bielefeld.DE> writes:

Rainer> The second problem is fixed by
Rainer> 	 http://gcc.gnu.org/ml/java/2001-07/msg00163.html

This just went in on the branch.

Rainer> * gcj seems to use cc1 for compilation!?
Rainer> FAIL: PR209 -O compilation from source
Rainer> cc1: Unrecognized option `-fencoding=UTF-8'

When linking (using --main), we do invoke cc1 to build `main' (the
function).

I just saw a PR about this (using -fno-check-bounds as the option) and
closed it since it didn't seem serious.  This one is more serious
since the test suite passes --encoding=UTF-8 to all compilations,
including links.

One fix would be to find out why the code in jvspec.c isn't omitting
-fencoding from the cc1 command line.  This spec entry:

	%{<fencoding*}

should do it.  Perhaps the %{<...} support in gcc.c is broken in 3.0?
I haven't looked.

Another fix would be to change the test suite so that --encoding isn't
used when linking.

Rainer> Do you think that the patch also qualifies for the branch now?

Yes, definitely.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]