The big Zero deletion

[Andrew John Hughes]

On 17 June 2010 19:08, Andrew John Hughes <ahughes at redhat.com> wrote:
> On 17 June 2010 18:39, Gary Benson <gbenson at redhat.com> wrote:
>> Andrew John Hughes wrote:
>>> There also seem to be a few minor differences that haven't been
>>> upstreamed.
>>
>> There shouldn't be :(
>>> ports/hotspot/src/./cpu/zero/vm/globals_zero.hpp:
>>>
>>> Has the following differences left.
>>>
>>> (- is IcedTea6, + is OpenJDK6)
>>>
>>> @@ -35,7 +35,6 @@
>>>  define_pd_global(bool,  UncommonNullCast,     true);
>>>
>>>  define_pd_global(intx,  CodeEntryAlignment,   32);
>>> -define_pd_global(intx,  OptoLoopAlignment,    16);
>>>  define_pd_global(intx,  InlineFrequencyCount, 100);
>>>  define_pd_global(intx,  PreInflateSpin,       10);
>>>
>>> @@ -45,5 +44,3 @@
>>>
>>>  define_pd_global(bool,  RewriteBytecodes,     true);
>>>  define_pd_global(bool,  RewriteFrequentPairs, true);
>>> -
>>> -define_pd_global(intx,  InlineSmallCode,      1000);
>>>
>>> plus old Sun copyright.
>>
>> We should be able to ignore this one and just use whatever comes from
>> upstream.  If we're changing things like this to make it build then
>> presumably that means upstream doesn't build?
>>
>
> Upstream builds.  I wouldn't have pushed it otherwise.  But upstream
> doesn't have Shark.  Are these additions not needed for Shark?
>

Ok the IcedTea6 local changes actually break the build:

/mnt/builder/icedtea6-hg/openjdk/hotspot/src/share/vm/runtime/globals.hpp:50:1:
error: redefinition of 'const intx pd_InlineSmallCode'
/mnt/builder/icedtea6-hg/openjdk/hotspot/src/cpu/zero/vm/globals_zero.hpp:49:1:
error: 'const intx pd_InlineSmallCode' previously defined here

I presume they were added for hs14, when we're now on hs17.  I'm
dropping these changes.

>>> ports/hotspot/src/./cpu/zero/vm/disassembler_zero.hpp:
>>>
>>> Has the following differences left
>>>
>>> +// The disassembler prints out zero code annotated
>>> +// with Java specific information.
>>> +
>>>    static int pd_instruction_alignment() {
>>> -    return 1;
>>> +    ShouldNotCallThis();
>>>    }
>>>
>>>    static const char* pd_cpu_opts() {
>>> -    return "";
>>> +    ShouldNotCallThis();
>>
>> This change is being upstreamed as part of Shark.  It's probably best
>> to simply ignore this one and just use whatever comes from upstream.
>> It's only for a not-very-useful debug option that probably nobody even
>> knows about.
>>
>
> Ok, do you have this somewhere else if we delete it from IcedTea6?
>

I think I'll leave this in, but convert it to a patch (shark-debug-option.patch)

>> Cheers,
>> Gary
>>
>> --
>> http://gbenson.net/
>>
>
>
>
> --
> 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
>



-- 
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