RTC Thumb2 JIT enhancements

[Andrew John Hughes]

On 17 March 2010 14:01, Xerxes Rånby <xerxes at zafena.se> wrote:
> Hi!
>
> On 2010-03-11 12:23, ed at camswl.com wrote:
>> Hi folks,
>>
>> The following are some enhancements I have made to the T2JIT over the past
>> month.
>>
>> I have tested the changes with various benchmarks, with Think Free Office,
>> and with the hotspot and lang portions of JTreg which all pass. The jdk
>> portion is still running.
>>
>> Unfortunately the diffs are quite large so I am submitting the diffs to
>> the list as 4 separate sets of diffs.
>>
>> Has the b18 release been split off yet? I would prefer these changes not
>> to go into the b18 release as they have not had sufficient testing yet.
>>
>>
>
> A bit strange to have them left in limbo.. surely they are designed and
> tested to work with the openjdk used in 1.7.x but they are also too new
> to be included in that release branch since icedtea6 1.7.x only accepts
> bugfixes and no new features....
> The new icedtea6 1.8 release using openjdk b18 that will branch any time
> now from icedtea6 trunk do already contain the first implementation of
> the t2jit and does also full bootstrap compile and runs with these four
> new patches applied when I tested it.
>
> If we do not include these new t2 fixes for the 1.8 branch then we have
> to play catchup with the upcoming openjdk b19 and potentially with a new
> hotspot hs17.
>

No, b19 will be hs16, which is already available in IcedTea6 1.7 and
1.8 but not by default; --with-hotspot-build=hs16
We'll make it default once 1.8 branches, ahead of the b19 import.

> Both icedtea6 1.7.x and icedtea6 1.8 uses the same hotspot, hs16, so
> they should both be ok I think.
>

No, they both use hs14.  I realise it's confusing because our revision
of hs14 is b16...

$ /usr/lib/jvm/icedtea6/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.7.1) (Gentoo build 1.6.0_17-b17)
OpenJDK 64-Bit Server VM (build 16.0-b13, mixed mode)

That's hs16, which as you can see, can be built from 1.7.1.

Just saw your correction; thanks :-)

> With all this said I think the patches looks great, they apply build and
> run using icedtea6 trunk, and I am in favour for them to go in so that
> they can be included in the next 1.8 release.
>

I guess this is Ed's call, but I think it makes sense to get them in
1.8 and perhaps stabilise them in that branch rather than shipping
them with hs16.

> Cheers
> Xerxes
>
>> Regards,
>> Ed.
>>
>> - Support for frameless compilation
>>
>>       Local variables and interpreter state are now addressed from Java SP
>>       rather than separate 'istate' (R8) register. R8 now free for register
>>       allocation giving 6 allocatable registers instead of 5.
>>
>> - Support for compilation of methods containing monitorenter / monitorexit
>>
>>       monitorenter / monitorexit moved from asm to C so they can be called
>>       by JIT code compilation of monitorenter / monitorexit containinng
>>       methods uses frame adressing because stack can expand dynamically to
>>       accomodate monitors.
>>
>> - Interpreter caches 'thread' in a register
>>
>>       Interpreter reworked to cache 'thread' in a register. Involved
>>       removal of one 'tmp' register in interpreter to make room for
>>       'thread' register. Interpreter and JIT are now consistent in register
>>       usage.
>>
>> - Compression of JIT 'header'
>>
>>       Removed the OSR entry from the JIT header. Previously, each JITted
>>       method had an individual piece of code in the header to load the
>>       appropriate registers from locals and the jump to the OSR entry
>>       point. This is changed so the the OSR caller (the interpreter), uses
>>       the 'regusage' array in the JIT header to load the correct registers
>>       and then just jumps directly to the entry point in the method.
>>       Saving is 24 bytes per method.
>>
>> - Backwards branch targets now doubleword aligned.
>>
>>       Pads with no ops before a backwards branch target to ensure backwards
>>       branch target is 64 bit aligned. Small performance benefit.
>>
>> - Addition of OSPACE flag
>>
>>       Setting environment variable T2_OSPACE will cause the JIT to optimise
>>       for space as opposed to time. OSPACE does the following.
>>
>>       o Tweaks the register allocation to allocate register based on space
>>       o Enables tailing of returns. IE does a branch to the return.
>>       o Disables doubleword alignment of backwards branch targets
>>       o Disables inlining invoke virtual/static/special wrappers
>>
>>       OSPACE has a small performance effect on benchmarks (~10% on ECM) but
>>       has no noticeable effect on large applications.
>>
>>       Default is compile for time (OSPACE disabled)
>>
>> - Removed a pass from the compiler
>>
>>       The JIT had an additional pass to detect 'zombie' code (ie code which
>>       could be, but has never been called). Zombie code removal was
>>       disabled in prior versions because it was unreliable, however the
>>       additional pass was still there.
>>
>> - Removed vestigal and debug code
>>
>>       Removed #if 0 code and debug code which was no longer functional
>>
>
>



-- 
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