Now that the core works I thought I would take a quick look at the
execution speed. For this I
executed the same code in our current environment (OS/2 and a mid 90's
Smalltalk) and again using
our invokeDynamic based vm translation. The code gets a string, parses in
into commands and
generates
It's astonishing how many people want to generate dead code :)
Rémi
On 04/30/2011 01:03 AM, Charles Nutter wrote:
Rémi and I confirmed that it is in fact dead code (JRuby's fault)
which is causing a problem for ASM 4 (ASM's fault). He's looking into
a fix, and I'll try to eliminate the dead co
On 04/30/2011 11:13 PM, Charles Oliver Nutter wrote:
> Well it seems like the GWT-based PIC could be a *great* success, at
> least compared to falling back on a single-entry IC. I will shortly
> commit another patch to JRuby master that provides a configurable GWT
> chain size before giving up and
Hi Daniel,
On 04/30/2011 09:51 AM, Daniel Latrémolière wrote:
I suggest a syntax for InvokeDynamic.
Why ?
motivations ?
It is created like a field and has a name (for referencing and
bootstrapping it), then i suggest to create a new modifier "dynamic"
and define InvokeDynamic like a mix o
I suggest a syntax for InvokeDynamic.
Why ?
motivations ?
Because it seems that syntax for InvokeDynamic will not be available in
JDK7 and I dislike generating bytecode.
I have used ASM, some years ago, and have nothing against it, but now I
do not like Java bytecode (stack machine) and the
With respect to chaining gwt handles.
Any thoughts on at what depth the chain should be converted to some other
form of dispatch?
I am currently just letting them build
In smalltalk it seems that around 10-20 inline tests are about the same as
a hash lookup but I
would think it could be diffe
On Sun, May 1, 2011 at 1:01 PM, Mark Roos wrote:
> With respect to chaining gwt handles.
>
> Any thoughts on at what depth the chain should be converted to some other
> form of dispatch?
> I am currently just letting them build
>
> In smalltalk it seems that around 10-20 inline tests are about the
Not me, but the Smalltalk compiler I inherited seems to see this as an
easy way to convert block sends into inline code for some cases.
My fix is ugly ( post process to look for dead code and remove it) so when
I get around to it I plan to rewrite the compiler. But then
it works so its at the
On 04/29/2011 09:59 PM, Ola Bini wrote:
> Hi,
>
> Given that creating GWTs are expensive, is it a really bad idea to
> create them and bind them on a cache miss then? My current logic for
> call sites look something like this:
>
> invoke call site
> if fallback, check if current morphi
In the Smalltalk I am porting the solution they use is to just drop the
entire chain and let it
reform. The assumption would be that for any short time period in the
life of a program no sites
are megamorphic. the metric they use is the total amount of active code.
Since they actually
drop al
10 matches
Mail list logo
