[maemo-developers] [maemo-developers] Java Question for the Nokia guys.

[From: Jesper Zuschlag]

Jep! The 770's ARM variant does indeed support Java acceleration  
through Jazelle technology. However, through their infinite wisdom  
ARM has chosen to keep the Jazelle-part of the processors programming  
model a secret for anybody else than their licensee. This in reality  
rules out any open-source implementation of a Jazelle-enabled JVM's  
(unless somebody will take up the challenge of reverse-engineer  
Jazelle)  , which means that we can only hope that Nokia will port  
one of their Jazelle-enabled JVM's to the 770. I must confessed that  
I have been *really* annoyed by ARM's attitude on this subject for a  
long time and fails to see the rational behind. Imagine a manufacture  
releasing a new processor while keeping the instruction set  
specification a secret!

/Jesper


On 19/01/2006, at 10.00, <monteslu at cox.net> wrote:

> Not sure if this is the right place to ask, but I've seen some  
> emails on this list from people with @nokia.com addresses.
>
> I've read the archives, and so far don't see anything conclusive  
> with regards to a working java implementation J2ME or J2SE.
>
>
> I noticed a post in the archives about this output from /proc/cpuinfo
>
>
> Processor	: ARM926EJ-Sid(wb) rev 3 (v5l)
> BogoMIPS	: 125.03
> Features	: swp half thumb fastmult edsp java
> CPU implementer	: 0x41
> CPU architecture: 5TEJ
> CPU variant	: 0x0
> CPU part	: 0x926
> CPU revision	: 3
> Cache type	: write-back
> Cache clean	: cp15 c7 ops
> Cache lockdown	: format C
> Cache format	: Harvard
> I size		: 32768
> I assoc		: 4
> I line length	: 32
> I sets		: 256
> D size		: 16384
> D assoc		: 4
> D line length	: 32
> D sets		: 128
>
> Hardware	: Generic OMAP1510/1610/1710
> Revision	: 17100016
> Serial		: 033302008b5f702f
>
>
>
>
>
> Does the Features line mean the 770's ARM chip has the potential  
> for hardware acceleration for java?
>
> If that's the case is Nokia working on a JVM for the device? I like  
> writing applications more than hacking at JVMs, so I don't want to  
> spend too much time doing hildon specific bindings or trying to get  
> a slim enough version of gnuclasspath and kaffe/gcj/jamvm working  
> efficiently.
>
>
> Thanks,
> Luis
>
> _______________________________________________
> maemo-developers mailing list
> maemo-developers at maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers