Re: invokedynamic and subclasses
I've gotten this working with a simplified version of the inlining cache in
Remi's cookbook. Thanks guys.
Performance is about twice as slow as invokevirtual, which I am really
pleased with.
On Wed, Jun 24, 2015 at 9:53 AM, Rémi Forax fo...@univ-mlv.fr wrote:
Hi Mike,
i've compiled a small list of patterns,
https://code.google.com/p/jsr292-cookbook/
take a look to the first inlining cache example.
cheers,
Rémi
Le 24 juin 2015 14:19:32 CEST, Mike Jarmy mja...@gmail.com a écrit :
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having
difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D,
E,
etc), some of which will have a field called foo, and some of which
won't.
The
class A will never have any fields -- its sole purpose is to be the
base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo()
of an
instance of B, and then reroute the call inside B's bootstrap method
via
MethodHandles.Lookup.findGetter(), and finally return the current value
of
foo.
So far, so good.
However, at the place where I am compiling the invokedynamic
instruction, I
have no idea if the target is a B, C, D, or what. All I know is that
it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call
to an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want.
I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to
do it
anyway. I've tried all kinds of different approaches (making yet
another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by
generating
interfaces for each class that has e.g. a foo field, and casting to
that
interface every time I compile a get_foo() invocation. This works, and
its
reasonably performant, but using invokedynamic seems like it would be a
more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
Re: invokedynamic and subclasses
Okay, this is just the sort of thing invokeDynamic is designed for.
Where you want to call get_foo() you should use an invokeDynaimc instruction
that will create a MutableCallSite. You should set the target of this to be a
lookup method that can find the appropriate get_foo() method, and possibly
install that as a target. I highly recommend looking at Remi’s JSR 292 cookbook
(https://code.google.com/p/jsr292-cookbook/) for some simple examples of how to
do this sort of thing.
Hope this helps set you off in the right direction,
Duncan.
From: mlvm-dev
mlvm-dev-boun...@openjdk.java.netmailto:mlvm-dev-boun...@openjdk.java.net
on behalf of Mike Jarmy mja...@gmail.commailto:mja...@gmail.com
Reply-To: Da Vinci Machine Project
mlvm-dev@openjdk.java.netmailto:mlvm-dev@openjdk.java.net
Date: Wednesday, 24 June 2015 13:19
To: Da Vinci Machine Project
mlvm-dev@openjdk.java.netmailto:mlvm-dev@openjdk.java.net
Subject: invokedynamic and subclasses
I've been experimenting with invokedynamic in a compiler for a dynamic language
that I'm working on, and I've run into a problem that I'm having difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C, D, E,
etc), some of which will have a field called foo, and some of which won't. The
class A will never have any fields -- its sole purpose is to be the base class
of everything else.
I have been able to successfully use a static bootstrap method in B, so that I
can compile a call to invokedynamic on the imaginary method get_foo() of an
instance of B, and then reroute the call inside B's bootstrap method via
MethodHandles.Lookup.findGetter(), and finally return the current value of foo.
So far, so good.
However, at the place where I am compiling the invokedynamic instruction, I
have no idea if the target is a B, C, D, or what. All I know is that it must
be an A.
So what I really want to be able to do (I think) is to use a static bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call to an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want. I
understand why that is -- foo doesn't exist in A, so there is no way to create
a findGetter() call. But I'm hoping there might be some clever way to do it
anyway. I've tried all kinds of different approaches (making yet another
invokedynamic call from inside A, etc, etc) but I can't come up with anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by generating
interfaces for each class that has e.g. a foo field, and casting to that
interface every time I compile a get_foo() invocation. This works, and its
reasonably performant, but using invokedynamic seems like it would be a more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
Re: invokedynamic and subclasses
Hi Mike,
i've compiled a small list of patterns,
https://code.google.com/p/jsr292-cookbook/
take a look to the first inlining cache example.
cheers,
Rémi
Le 24 juin 2015 14:19:32 CEST, Mike Jarmy mja...@gmail.com a écrit :
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having
difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D,
E,
etc), some of which will have a field called foo, and some of which
won't.
The
class A will never have any fields -- its sole purpose is to be the
base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo()
of an
instance of B, and then reroute the call inside B's bootstrap method
via
MethodHandles.Lookup.findGetter(), and finally return the current value
of
foo.
So far, so good.
However, at the place where I am compiling the invokedynamic
instruction, I
have no idea if the target is a B, C, D, or what. All I know is that
it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call
to an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want.
I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to
do it
anyway. I've tried all kinds of different approaches (making yet
another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by
generating
interfaces for each class that has e.g. a foo field, and casting to
that
interface every time I compile a get_foo() invocation. This works, and
its
reasonably performant, but using invokedynamic seems like it would be a
more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
Re: invokedynamic and subclasses
Jochen -- that comes tantalizingly close to making sense to me :-). I
don't quite follow all of the terminology yet, but the basic idea that you
describe of chaining bootstrap calls is sort of like what I had vaguely
designed in my head. Once I am doing learning how all this works I should
be able to make my language work just the way I hoped it would.
On Wed, Jun 24, 2015 at 10:16 AM, Jochen Theodorou blackd...@gmx.org
wrote:
Hi Mike,
First of all, the bootstrap method does not have to be in the same class.
In Groovy we have for example one bootstrap method for all classes only. It
is part of the runtime then you could say.
We do it like this. We have two bootstrap methods in the Groovy runtime
(you don't need to have them in the class). The first one is normally
called and will produce a handle to call the second one. I tend to call the
first one the static bootstrapper and the second one the dynamic
bootstrapper. The dynamic bootstrapper will be called like a method (takes
an Object[]), thus I will have all the arguments of the call as well as the
receiver. This I can use to identify runtime classes and produce a new
handle that fits my needs and replaces the first one. This method also is a
fallback. The installed handle usually has to be guarded (like first call
with B, second with C), and we use the dynamic bootstrapper as fallback. Of
course you can easily check more cases and implement a polymorphic inline
cache.
Of course that implies, that if the dynamic bootstrapper is called it will
also have to do the first execution of the target (like getting the value
of the field foo). We use (part of) the produced method handle for this.
And of course the callsite will have to be mutable then
Does this help you out? It works quite good for Groovy.
bye blackdrag
Am 24.06.2015 14:19, schrieb Mike Jarmy:
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having
difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D, E,
etc), some of which will have a field called foo, and some of which
won't. The
class A will never have any fields -- its sole purpose is to be the base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo() of
an
instance of B, and then reroute the call inside B's bootstrap method via
MethodHandles.Lookup.findGetter(), and finally return the current value
of foo.
So far, so good.
However, at the place where I am compiling the invokedynamic instruction,
I
have no idea if the target is a B, C, D, or what. All I know is that it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call to
an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want. I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to do
it
anyway. I've tried all kinds of different approaches (making yet another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by generating
interfaces for each class that has e.g. a foo field, and casting to that
interface every time I compile a get_foo() invocation. This works, and
its
reasonably performant, but using invokedynamic seems like it would be a
more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
--
Jochen blackdrag Theodorou
blog: http://blackdragsview.blogspot.com/
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
Re: invokedynamic and subclasses
JRuby also chains the bootstraps, you might find inspiration here:
https://github.com/jruby/jruby/blob/master/core/src/main/java/org/jruby/runtime/invokedynamic/InvokeDynamicSupport.java
On Wed, Jun 24, 2015 at 11:25 AM, Mike Jarmy mja...@gmail.com wrote:
Jochen -- that comes tantalizingly close to making sense to me :-). I
don't quite follow all of the terminology yet, but the basic idea that you
describe of chaining bootstrap calls is sort of like what I had vaguely
designed in my head. Once I am doing learning how all this works I should
be able to make my language work just the way I hoped it would.
On Wed, Jun 24, 2015 at 10:16 AM, Jochen Theodorou blackd...@gmx.org
wrote:
Hi Mike,
First of all, the bootstrap method does not have to be in the same class.
In Groovy we have for example one bootstrap method for all classes only. It
is part of the runtime then you could say.
We do it like this. We have two bootstrap methods in the Groovy runtime
(you don't need to have them in the class). The first one is normally
called and will produce a handle to call the second one. I tend to call the
first one the static bootstrapper and the second one the dynamic
bootstrapper. The dynamic bootstrapper will be called like a method (takes
an Object[]), thus I will have all the arguments of the call as well as the
receiver. This I can use to identify runtime classes and produce a new
handle that fits my needs and replaces the first one. This method also is a
fallback. The installed handle usually has to be guarded (like first call
with B, second with C), and we use the dynamic bootstrapper as fallback. Of
course you can easily check more cases and implement a polymorphic inline
cache.
Of course that implies, that if the dynamic bootstrapper is called it
will also have to do the first execution of the target (like getting the
value of the field foo). We use (part of) the produced method handle for
this. And of course the callsite will have to be mutable then
Does this help you out? It works quite good for Groovy.
bye blackdrag
Am 24.06.2015 14:19, schrieb Mike Jarmy:
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having
difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D, E,
etc), some of which will have a field called foo, and some of which
won't. The
class A will never have any fields -- its sole purpose is to be the base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo() of
an
instance of B, and then reroute the call inside B's bootstrap method via
MethodHandles.Lookup.findGetter(), and finally return the current value
of foo.
So far, so good.
However, at the place where I am compiling the invokedynamic
instruction, I
have no idea if the target is a B, C, D, or what. All I know is that it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call to
an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want. I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to
do it
anyway. I've tried all kinds of different approaches (making yet another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by generating
interfaces for each class that has e.g. a foo field, and casting to that
interface every time I compile a get_foo() invocation. This works, and
its
reasonably performant, but using invokedynamic seems like it would be a
more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
--
Jochen blackdrag Theodorou
blog: http://blackdragsview.blogspot.com/
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
Re: invokedynamic and subclasses
Hi Mike,
First of all, the bootstrap method does not have to be in the same
class. In Groovy we have for example one bootstrap method for all
classes only. It is part of the runtime then you could say.
We do it like this. We have two bootstrap methods in the Groovy runtime
(you don't need to have them in the class). The first one is normally
called and will produce a handle to call the second one. I tend to call
the first one the static bootstrapper and the second one the dynamic
bootstrapper. The dynamic bootstrapper will be called like a method
(takes an Object[]), thus I will have all the arguments of the call as
well as the receiver. This I can use to identify runtime classes and
produce a new handle that fits my needs and replaces the first one. This
method also is a fallback. The installed handle usually has to be
guarded (like first call with B, second with C), and we use the dynamic
bootstrapper as fallback. Of course you can easily check more cases and
implement a polymorphic inline cache.
Of course that implies, that if the dynamic bootstrapper is called it
will also have to do the first execution of the target (like getting the
value of the field foo). We use (part of) the produced method handle for
this. And of course the callsite will have to be mutable then
Does this help you out? It works quite good for Groovy.
bye blackdrag
Am 24.06.2015 14:19, schrieb Mike Jarmy:
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D, E,
etc), some of which will have a field called foo, and some of which
won't. The
class A will never have any fields -- its sole purpose is to be the base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo() of an
instance of B, and then reroute the call inside B's bootstrap method via
MethodHandles.Lookup.findGetter(), and finally return the current value
of foo.
So far, so good.
However, at the place where I am compiling the invokedynamic instruction, I
have no idea if the target is a B, C, D, or what. All I know is that it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call to an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want. I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to do it
anyway. I've tried all kinds of different approaches (making yet another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by generating
interfaces for each class that has e.g. a foo field, and casting to that
interface every time I compile a get_foo() invocation. This works, and its
reasonably performant, but using invokedynamic seems like it would be a more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
--
Jochen blackdrag Theodorou
blog: http://blackdragsview.blogspot.com/
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
Re: invokedynamic and subclasses
Thanks guys, the cookbook is just the kind of thing I was looking for.
On Wed, Jun 24, 2015 at 9:53 AM, Rémi Forax fo...@univ-mlv.fr wrote:
Hi Mike,
i've compiled a small list of patterns,
https://code.google.com/p/jsr292-cookbook/
take a look to the first inlining cache example.
cheers,
Rémi
Le 24 juin 2015 14:19:32 CEST, Mike Jarmy mja...@gmail.com a écrit :
I've been experimenting with invokedynamic in a compiler for a dynamic
language
that I'm working on, and I've run into a problem that I'm having
difficulty
solving.
Let's say I have a class called A (Java syntax used for clarity):
public class A {
}
And another called B, that is a subclass of A:
public class B extends A {
public Value foo;
public Value bar;
}
There will be lots of other subclasses of A as well (let's call them C,
D,
E,
etc), some of which will have a field called foo, and some of which
won't.
The
class A will never have any fields -- its sole purpose is to be the
base
class
of everything else.
I have been able to successfully use a static bootstrap method in B, so
that I
can compile a call to invokedynamic on the imaginary method get_foo()
of an
instance of B, and then reroute the call inside B's bootstrap method
via
MethodHandles.Lookup.findGetter(), and finally return the current value
of
foo.
So far, so good.
However, at the place where I am compiling the invokedynamic
instruction, I
have no idea if the target is a B, C, D, or what. All I know is that
it
must
be an A.
So what I really want to be able to do (I think) is to use a static
bootstrap
method in A. I want get_foo() to succeed for every invokedynamic call
to an
instance of A who's *subclass* really does have a field called foo.
Unfortunately there doesn't seem to be a way to make A do what I want.
I
understand why that is -- foo doesn't exist in A, so there is no way to
create
a findGetter() call. But I'm hoping there might be some clever way to
do it
anyway. I've tried all kinds of different approaches (making yet
another
invokedynamic call from inside A, etc, etc) but I can't come up with
anything
that works.
Any ideas?
By the way, I've figured out how to do this the old way, by
generating
interfaces for each class that has e.g. a foo field, and casting to
that
interface every time I compile a get_foo() invocation. This works, and
its
reasonably performant, but using invokedynamic seems like it would be a
more
elegant and flexible solution.
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
___
mlvm-dev mailing list
mlvm-dev@openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/mlvm-dev
