thinking about proper implementation of tail calls

[Alexander Turner]

The is a command line flag to control the maximum size for the jitter . I
blogged about it some years ago.

On Monday, 3 September 2012, Mark Roos wrote:

>
> So then the issue is making the methods small enough for HotSpot to
> compile and not fall back to the interpreter.
> If that is the case why not find the means to have Hotspot not do that?
>  It seems that would be less effort than
> adding hard tail calls to the jvm.  In other words is this really the test
> case that John is looking for?
>
> by the way what do you use as the byte code size limit to prevent this
> today?
>
> thanks
> mark
>
>
>
> From:        Charles Oliver Nutter <headius at headius.com<javascript:_e({}, 'cvml', 'headius at headius.com');>
> >
> To:        Da Vinci Machine Project <mlvm-dev at openjdk.java.net<javascript:_e({}, 'cvml', 'mlvm-dev at openjdk.java.net');>
> >
> Date:        09/02/2012 09:37 AM
> Subject:        Re: thinking about proper implementation of tail calls
> Sent by:        mlvm-dev-bounces at openjdk.java.net <javascript:_e({},
> 'cvml', 'mlvm-dev-bounces at openjdk.java.net');>
> ------------------------------
>
>
>
> The classfile limit is only part of the problem. In JRuby we frequently
> have larger Ruby methods fall over in the compilation process because
> Hotspot bails out on the whole thing. If we could easily break it up, we
> would at least get things to compile, even if they needed a CALL here and
> there to stitch the bits together. But except for trivial cases (chaining
> along root statements) we need tail calls for this to work well.
>
> - Charlie (mobile)
>
> On Sep 1, 2012 6:31 PM, "Mark Roos" <*mroos at roos.com* <javascript:_e({},
> 'cvml', 'mroos at roos.com');>> wrote:
> I am looking to learn something here that I haven't seen in my code yet.
>
> John mentioned
>         Suppose you are compiling your favorite high-level language to the
> JVM, and you start
>         running into the various size limits in class files
>
> To which there seemed to be some agreement that this was an issue.
>  Running over my Smalltalk
> code base my largest method is 12000 bytes with only about a dozen that
> are more than
> 10k bytes.  The corresponding class file is 16k bytes ( I only do one
> method + blocks per class file)
>
> So my question is what is causing these mega methods.  Is it just an
> artifact of the language being
> implemented or is it from some language side optimization (such as trace
> optimization)?  Perhaps
> I am just lucky to not see it yet.
>
> thanks
> mark
> _______________________________________________
> mlvm-dev mailing list*
> **mlvm-dev at openjdk.java.net* <javascript:_e({}, 'cvml',
> 'mlvm-dev at openjdk.java.net');>*
> **http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev*<http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev>
> _______________________________________________
> mlvm-dev mailing list
> mlvm-dev at openjdk.java.net <javascript:_e({}, 'cvml',
> 'mlvm-dev at openjdk.java.net');>
> http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/mlvm-dev/attachments/20120903/9fd40818/attachment.html