This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: [java, 3.0.1]: BUG: Unresolved symbols with -O3 and higher
- To: Jeff Sturm <jsturm at one-point dot com>
- Subject: Re: [java, 3.0.1]: BUG: Unresolved symbols with -O3 and higher
- From: Tom Tromey <tromey at redhat dot com>
- Date: 25 Jul 2001 10:34:32 -0600
- Cc: Manfred Hollstein <manfred dot h at gmx dot net>, gcc-bugs at gcc dot gnu dot org, java at gcc dot gnu dot org
- References: <Pine.LNX.4.10.10107241259100.15490-100000@mars.deadcafe.org>
- Reply-To: tromey at redhat dot com
>>>>> "Jeff" == Jeff Sturm <jsturm@one-point.com> writes:
Jeff> I'll argue this should be the default for gcj, due to
Jeff> reflection.
This seems reasonable to me.
I think we should default to compatibility, even if it means we emit
code for extra functions.
Then we can provide options to let the user choose when to reduce
this. For instance, -fno-keep-inline-functions.
This has come up before:
http://gcc.gnu.org/ml/java/1999-q4/msg00364.html
Jeff> In fact the behavior of the program below changes when compiled
Jeff> with/without -fkeep-inline-functions:
This example is important, because I think it shows that if we keep
the current behavior then we can break serialization, which relies on
private functions and reflection.
Tom