Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: pass (herse.apache.org: domain of mike.fursov@gmail.com
 designates 64.233.184.224 as permitted sender)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references;
        b=K0rSWp4LdxudHuYZcrY5q6QHmqnssSlGhx03uUZ7fvD7uEQWHoNOpTdSz0NMmrX2rv3o/LagYuH51TnNiNk5scweWgdWRJwVEnDCns1cczX8BEzHHKXolj1TuHKGuOcSo7GGGx8RdpVigrN1TVf6SQCFbiSiMIHAj4bBVXr5Bic=
Message-ID: <bc79dd600612120422p66f80348u64ac04e5abb3f5b4@mail.gmail.com>
Date: Tue, 12 Dec 2006 18:22:47 +0600
From: "Mikhail Fursov" <mike.fursov@gmail.com>
To: dev@harmony.apache.org
Subject: Re: [drlvm] [interpreter] What is wrong?
In-Reply-To: <12385bbd0612101624x1b8ae6f6ta78a87632cf34318@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_Part_130190_1008414.1165926167498"
References: <el9d58$hes$1@sea.gmane.org> <umz5zio75.fsf@gmail.com>
	 <f944d83d0612072118x3f6005efub98d37f7e3037262@mail.gmail.com>
	 <12385bbd0612101624x1b8ae6f6ta78a87632cf34318@mail.gmail.com>

------=_Part_130190_1008414.1165926167498
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

The reason of the problem is VM properties refactoring.
Details: Interpreter is ON when vm.use_interpreter property is true
interpreter.emconf sets up this property using common EM mechanism to set up
any VM property: set the property from emconf unless it was already set from
command line.
After the last refactoring VM sets this property to 'false' before running
EM. So EM thinks that this is cmd-line property and does not override it.
Can we revert the original behaviour back: do not set this property to
'false' if this is not interpreter mode?
In this case the patch will be simple:

Index: src/init/vm_properties.cpp
===================================================================
--- src/init/vm_properties.cpp  (revision 484646)
+++ src/init/vm_properties.cpp  (working copy)
@@ -265,7 +265,7 @@
         properties.set("vm.crash_handler", "false");
         properties.set("vm.finalize", "true");
         properties.set("vm.jit_may_inline_sync", "true");
-        properties.set("vm.use_interpreter", "false");
+//        properties.set("vm.use_interpreter", "false");
         properties.set("vm.use_verifier", "true");
         properties.set("vm.jvmti.enabled", "false");
         properties.set("vm.cleanupOnExit", "false");


On 12/11/06, Ivan Volosyuk <ivan.volosyuk@gmail.com> wrote:
>
> It can be a problem is some platform specific initialization code.
> Make sure that interpreter_enabled() returns true. Obviously it is
> not.
> --
> Ivan
>
> On 12/8/06, Alexey Varlamov <alexey.v.varlamov@gmail.com> wrote:
> > 07 Dec 2006 23:57:18 +0600, Egor Pasko <egor.pasko@gmail.com>:
> > > On the 0x238 day of Apache Harmony Igor V. Chebykin wrote:
> > > > When I run java with -Xint, all is OK - the interpreter is running,
> > > > HelloWorld passed.
> > [snip]
> >
> > > > But when I use -Xem:interpreter, java aborted with following log:
> >
> > Guys, there is HARMONY-2399 about this (no patch yet, did not have
> > time to look).
> >
> > [snip]
> > > > My platform is Linux.ia64
> > >
> > > OK, looks like interpreter is broken on IPF :)
> >
> > Nope, interpreter is out of business here. Just note the first
> > sentence in the thread ;)
> >
> > --
> > Alexey
> >
>
>
> --
> Ivan
> Intel Enterprise Solutions Software Division
>


-- 
Mikhail Fursov

------=_Part_130190_1008414.1165926167498--