What's the benefit of doing the GCJ AOT-compilation steps as defined in https://fedoraproject.org/wiki/Packaging/GCJGuidelines? Doesn't everyone just use OpenJDK these days? It adds a bunch of extra stuff to every Java spec file and makes all java packages arch-specific instead of noarch, which always feels a bit weird ...
MEF
On Mon, Nov 16, 2009 at 4:33 PM, Mary Ellen Foster wrote:
What's the benefit of doing the GCJ AOT-compilation steps as defined in https://fedoraproject.org/wiki/Packaging/GCJGuidelines? Doesn't everyone just use OpenJDK these days? It adds a bunch of extra stuff to every Java spec file and makes all java packages arch-specific instead of noarch, which always feels a bit weird ...
MEF
Ah the famous topic :)
I think the latest explanation for the AOT bits were that they are still needed in ppc* architectures because openjdk is still slow on those machines.
I don't know if this has changed since. But from F-13 on, I expect that we will drop the AOT bit requirement since ppc* won't be primary architectures anymore.
Orcan
On 16:41 Mon 16 Nov , Orcan Ogetbil wrote:
On Mon, Nov 16, 2009 at 4:33 PM, Mary Ellen Foster wrote:
What's the benefit of doing the GCJ AOT-compilation steps as defined in https://fedoraproject.org/wiki/Packaging/GCJGuidelines? Doesn't everyone just use OpenJDK these days? It adds a bunch of extra stuff to every Java spec file and makes all java packages arch-specific instead of noarch, which always feels a bit weird ...
MEF
Ah the famous topic :)
I think the latest explanation for the AOT bits were that they are still needed in ppc* architectures because openjdk is still slow on those machines.
I don't know if this has changed since. But from F-13 on, I expect that we will drop the AOT bit requirement since ppc* won't be primary architectures anymore.
Orcan
-- fedora-devel-java-list mailing list fedora-devel-java-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-java-list
AOT is still beneficial with some applications, even on non-ppc architectures -- ecj being the prime example. -- Andrew :)
Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com)
Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint = F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
Orcan Ogetbil wrote:
On Mon, Nov 16, 2009 at 4:33 PM, Mary Ellen Foster wrote:
What's the benefit of doing the GCJ AOT-compilation steps as defined in https://fedoraproject.org/wiki/Packaging/GCJGuidelines? Doesn't everyone just use OpenJDK these days? It adds a bunch of extra stuff to every Java spec file and makes all java packages arch-specific instead of noarch, which always feels a bit weird ...
Ah the famous topic :)
I think the latest explanation for the AOT bits were that they are still needed in ppc* architectures because openjdk is still slow on those machines.
I don't know if this has changed since. But from F-13 on, I expect that we will drop the AOT bit requirement since ppc* won't be primary architectures anymore.
And Shark, the LLVM-based JIT for OpenJDK, is getting close to prime time. See http://icedtea.classpath.org/wiki/ZeroSharkFaq
Andrew.
On Tue, 2009-11-17 at 11:36 +0000, Andrew Haley wrote:
And Shark, the LLVM-based JIT for OpenJDK, is getting close to prime time. See http://icedtea.classpath.org/wiki/ZeroSharkFaq
Very nice. Out of curiosity, are there plans to try to get Shark to pass the Java SE 6 TCK?
Best, Ismael
Ismael Juma wrote:
On Tue, 2009-11-17 at 11:36 +0000, Andrew Haley wrote:
And Shark, the LLVM-based JIT for OpenJDK, is getting close to prime time. See http://icedtea.classpath.org/wiki/ZeroSharkFaq
Very nice. Out of curiosity, are there plans to try to get Shark to pass the Java SE 6 TCK?
Yes, as soon as it's done.
Andrew.
java-devel@lists.fedoraproject.org
-
Andrew Haley -
Andrew Hughes -
Ismael Juma -
Mary Ellen Foster -
Orcan Ogetbil