Hi Christoph,

On Feb 14, 2020, at 9:51 AM, Thiede, Christoph < Christoph.Thiede@student.hpi.uni-potsdam.de> wrote:



Hi Eliot,

Or provide a vm flag which controls whether special selector #class (the

bytecode for sending #class) is inlined or not. This flag can also control #== & #~~.

Could you give me some (code) pointers about this inlining? :-)

This is something I first did in VW 3.0 so that people working with proxies could avoid the inlining of #== & #class. See primitive 537 in VW. The blue book definition for special selector #class & special selector #== has them inlined. See pp 618 (640 in the online pdf) through 619 in the blue book, and methods specialSelectorPrimitiveResponse and commonSelectorPrimitiveResponse. In the VM sources see bytecodePrimClass, bytecodePrimIdentical et al in initializeBytecodeTableForSqueakV3PlusClosures, initializeBytecodeTableForSistaV1 and the implementations in StackInterpreter.

The other special selector send bytecodes either apply only to a specific set of classes (e.g. arithmetic is inlined for any combination of SmallInteger and float) or simply function as a space saver (1 byte for the bytecode instead of 1 byte for the send bytecode plus 4 or 8 bytes for the literal selector). But #class, #== & #~~ are inlined. They apply to any and all objects. This is how the system has been defined. IMO it is a mistake and no longer justified on performance grounds. But t is the reality on the ground. And changing it will take some effort.

self systemNavigation browseAllSelect: [:m| (m selector beginsWith: 'bytecodePrim') and: [(m messages includes: #normalSend) not]] localTo: StackInterpreter

(and in the Cogit see implementations of genSpecialSelectorEqualsEquals, genSpecialSelectorClass, genSpecialSelectorNotEqualsEquals).

What I propose is adding a flag bit that persists in the image header, queryable/settable via a vmParameterAt: that turns off this inlining, and turns them into true sends. I think we'll see slow downs in the 5% to 10% range because of the cost of looking up #== (used heavily in ifNotNil: et al) and we can reduce this by e.g. inlining for nil as a receiver and having the bytecode compiler generate pushNil pushFoo send #== for ifNotNil et al.

Note that Nicolas did something very similar recently to provide control over mixed mode arithmetic. He added the primitiveDoMixedArithmetic flag, controlled by flag bit 6 in the VM primitive flags (vm parameter 48). Something like noInlineCommonSelectors would be controlled by bit 7.

There is a good solution to this. Low-level Inspectors *should not* use

instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy" (MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Sounds good.

Hopefully I’ve answered your questions and given you pointers. I’m

interested in collaborating to make this a reality in the next major release.

You did. I'm too.

Cool. Let's do it :-) I'm of course busy until next week. But come Tuesday I should be able to have at it.

Best, Christoph

------------------------------ *Von:* Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Eliot Miranda eliot.miranda@gmail.com *Gesendet:* Freitag, 14. Februar 2020 18:27:14 *An:* The general-purpose Squeak developers list *Betreff:* Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Hi Christoph,

On Feb 14, 2020, at 5:18 AM, Thiede, Christoph < Christoph.Thiede@student.hpi.uni-potsdam.de> wrote:



Hi all! :-)

When we were fiddling around with some proxy implementations today, Marcel and I discovered an interesting set of issues regarding ProtoObjects.

Let's start with a concrete bug:

*Code to reproduce <do it>:*

ProtoObject new class.

*Expected behavior:*

DNU exception, because #class is implemented on Object only.

*Actual behavior:*

Outputs ProtoObject.

Even stranger: If you debug the method instead and simulate the #class send, you get a "simulated message class not understood" error.

If you subclass ProtoObject and override/forward #class, the results will deviate based on whether you are debugging or executing the code.

*Some thoughts:*

There are two options: Either to implement #class on ProtoObject (by moving the primitive definition up from Object), or to remove it from the specialObjectsArray so that a send to #class is not compiled differently than a send to any other regular message.

Or provide a vm flag which controls whether special selector #class (the bytecode for sending #class) is inlined or not. This flag can also control #== & #~~.

This leads us to the rather general question: *What special messages should an instance of ProtoObject understand?*

IMO *only* #doesNotUnderstand:

See below on the mirror protocol

We don't answer this question completely consistent at the moment: For example, #instVarAt: is implemented on Object, but #instVarsInclude: is implemented on ProtoObject. Again, it is a bit weird that the implementation of #instVarsInclude: calls #class which is implemented on Object only. And so it goes on:

- #someObject is implemented on Object, but #nextObject is available in ProtoObject. - #withArgs:executeMethod: is implemented on ProtoObject, whereas #perform:withArguments: is on Object. - #ifNotNil: is implemented on ProtoObject; however, #ifNotNilDo: is implemented on Object. (btw: does the latter still have any raison d'être?) - #flag: exists on ProtoObject, and #yourself exists on Object only. Isn't #yourself rather a syntactical element we would like to behave exactly the same way for every possible message receiver? (btw: Is there a good reason not to speed up #yourself via specialObjectsArray)? - (I don't claim for completeness)

And just some other problems regarding to ProtoObject (just collecting them here instead of forgetting them forever):

- ObjectTracer is broken due to several reasons. Will fix this soon. - Inspectors cannot inspect ProtoObjects correctly (for example: Inspector openOn: (ObjectTracer on: Display)). This is because #instVarAt: is implemented on Object only so this message is forwarded. Maybe we should replace these critical calls from Inspector by something like #tryPrimitive:173withArgs:? But should we do so in Inspector or rather in a new subclass named ProtoInspector or similar? Hm, then we would need to implement #inspectorClass on ProtoObject, which is bad either ... We will keep investigating this issue.

There is a good solution to this. Low-level Inspectors *should not* use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

You see hacks everywhere (the xxxFoo messages in serialization) because we don’t have a well-defined notion of what is in a proxy. Having one that is as minimal as possible is IMNERHO the clearest definition possible, and the most minimal is just having a #doesNotUnderstand: that prevents recursive does not understand vm exit and displays a stack trace without sending further messages to the proxy (because it uses the mirror primitives).

Note that the mirror primitives are what the execution machinery in Context use to ensure that bytecodes do what they are expected to do. An I et car access should *not* send instVarAt: to a proxy; it should access the inst var of the proxy, etc.

However, the crucial question is: What special messages should an instance of ProtoObject understand? On the one hand, the idea of ProtoObjects is to work as total proxies with a maximum amount of forwarding potential, which implies a minimum number of methods. On the other hand, certain aspects such as accessing instvars or executing messages are really identity-related. Looking forward to your comments :-)

Hopefully I’ve answered your questions and given you pointers. I’m interested in collaborating to make this a reality in the next major release.

Best, Christoph

Hi Jakob,

On Fri, Feb 14, 2020 at 1:14 PM Jakob Reschke forums.jakob@resfarm.de wrote:

Am Fr., 14. Feb. 2020 um 19:42 Uhr schrieb Eliot Miranda < eliot.miranda@gmail.com>:

...

Hi Christoph,

On Feb 14, 2020, at 9:51 AM, Thiede, Christoph < Christoph.Thiede@student.hpi.uni-potsdam.de> wrote:

...

There is a good solution to this. Low-level Inspectors *should not*

use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy" (MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Sounds good.

If inspectors et al would use this, my decorators might not have to carry

the following collection of methods/copies:

asExplorerString basicInspect basicSize class instVarAt: instVarAt:put: perform: perform:withArguments: primitiveFailed printString printStringLimitedTo: respondsTo: xxxClass

If DataStream et al did, add the following: readDataFrom:size: storeDataOn:

I'm not sure how to generally handle #yourself. I have this in my decorators as well because I don't want to undecorate when the decorator receives it.

typically use of the mirror protocol, at its most basic, looks like

thisContext object: foo instVarAt: index

so #yourself is handled by ending the cascade early, and answering the object (foo in the above). Yes that's a tedious refactor, but straightforward, no?

Example:

anObject := (MyDecorator on: anotherObject) doSomething; yourself.

I would like that anObject is the decorator and not anotherObject at the end.

Marcel told me the following tip for this scenario, which sounds quite interesting to me:

doesNotUnderstand: aMessage | result | result := aMessage sendTo: target. ^ target == result ifTrue: [self] ifFalse: [result]

:-)

________________________________ Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Jakob Reschke forums.jakob@resfarm.de Gesendet: Samstag, 15. Februar 2020 09:28:58 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Thanks Eliot, I understood that for the inspector use case, but my thought about #yourself was rather the following:

If I pass the decorator around and some domain code (not in Inspector or similar) uses the decorator in a cascade with yourself at the end, I do not want the result to be the undecorated object.

Example:

anObject := (MyDecorator on: anotherObject) doSomething; yourself.

I would like that anObject is the decorator and not anotherObject at the end.

Am Fr., 14. Feb. 2020 um 22:39 Uhr schrieb Eliot Miranda <eliot.miranda@gmail.commailto:eliot.miranda@gmail.com>: Hi Jakob,

On Fri, Feb 14, 2020 at 1:14 PM Jakob Reschke <forums.jakob@resfarm.demailto:forums.jakob@resfarm.de> wrote: Am Fr., 14. Feb. 2020 um 19:42 Uhr schrieb Eliot Miranda <eliot.miranda@gmail.commailto:eliot.miranda@gmail.com>: Hi Christoph,

On Feb 14, 2020, at 9:51 AM, Thiede, Christoph <Christoph.Thiede@student.hpi.uni-potsdam.demailto:Christoph.Thiede@student.hpi.uni-potsdam.de> wrote:

There is a good solution to this. Low-level Inspectors *should not* use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy" (MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Sounds good.

If inspectors et al would use this, my decorators might not have to carry the following collection of methods/copies:

asExplorerString basicInspect basicSize class instVarAt: instVarAt:put: perform: perform:withArguments: primitiveFailed printString printStringLimitedTo: respondsTo: xxxClass

If DataStream et al did, add the following: readDataFrom:size: storeDataOn:

I'm not sure how to generally handle #yourself. I have this in my decorators as well because I don't want to undecorate when the decorator receives it.

typically use of the mirror protocol, at its most basic, looks like

thisContext object: foo instVarAt: index

so #yourself is handled by ending the cascade early, and answering the object (foo in the above). Yes that's a tedious refactor, but straightforward, no? -- _,,,^..^,,,_ best, Eliot

Not sendTo: instead of sentTo:? ;-)

This was the first mention of #sentTo: in this thread :)

And I suppose it is fine to ask the decorator programmer to implement an #undecorated method to return the inner object.

Sounds like a mirror problem again? Perfect decorators can be nested with arbitrary depth. What should be the return value of (EasterDecorator on: (ChristmasDecorator on: Object new)) undecorated? ________________________________ Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Jakob Reschke forums.jakob@resfarm.de Gesendet: Samstag, 15. Februar 2020 16:14:54 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Not sendTo: instead of sentTo:? ;-)

Agreed that solves it. And I suppose it is fine to ask the decorator programmer to implement an #undecorated method to return the inner object. If it makes sense for the particular decorator.

Am Sa., 15. Feb. 2020 um 15:59 Uhr schrieb Thiede, Christoph Christoph.Thiede@student.hpi.uni-potsdam.de:

...

Example:

...

anObject := (MyDecorator on: anotherObject) doSomething; yourself.

I would like that anObject is the decorator and not anotherObject at the end.

Marcel told me the following tip for this scenario, which sounds quite interesting to me:

doesNotUnderstand: aMessage | result | result := aMessage sendTo: target. ^ target == result ifTrue: [self] ifFalse: [result]

:-)

---

Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Jakob Reschke forums.jakob@resfarm.de Gesendet: Samstag, 15. Februar 2020 09:28:58 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Thanks Eliot, I understood that for the inspector use case, but my thought about #yourself was rather the following:

If I pass the decorator around and some domain code (not in Inspector or similar) uses the decorator in a cascade with yourself at the end, I do not want the result to be the undecorated object.

Example:

anObject := (MyDecorator on: anotherObject)
   doSomething;
   yourself.

I would like that anObject is the decorator and not anotherObject at the end.

Am Fr., 14. Feb. 2020 um 22:39 Uhr schrieb Eliot Miranda eliot.miranda@gmail.com:

...

Hi Jakob,

On Fri, Feb 14, 2020 at 1:14 PM Jakob Reschke forums.jakob@resfarm.de wrote:

...

Am Fr., 14. Feb. 2020 um 19:42 Uhr schrieb Eliot Miranda eliot.miranda@gmail.com:

...

Hi Christoph,

On Feb 14, 2020, at 9:51 AM, Thiede, Christoph Christoph.Thiede@student.hpi.uni-potsdam.de wrote:

...

There is a good solution to this. Low-level Inspectors *should not* use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy" (MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Sounds good.

If inspectors et al would use this, my decorators might not have to carry the following collection of methods/copies:

asExplorerString basicInspect basicSize class instVarAt: instVarAt:put: perform: perform:withArguments: primitiveFailed printString printStringLimitedTo: respondsTo: xxxClass

If DataStream et al did, add the following: readDataFrom:size: storeDataOn:

I'm not sure how to generally handle #yourself. I have this in my decorators as well because I don't want to undecorate when the decorator receives it.

typically use of the mirror protocol, at its most basic, looks like

thisContext object: foo instVarAt: index

so #yourself is handled by ending the cascade early, and answering the object (foo in the above). Yes that's a tedious refactor, but straightforward, no?

_,,,^..^,,,_ best, Eliot

Ideally I would not have had the need to undecorate anything, right?

At least it seems an interesting vision :) Also see #xxxUnTrace, an interesting approach. ________________________________ Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Jakob Reschke forums.jakob@resfarm.de Gesendet: Samstag, 15. Februar 2020 17:49:27 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Am Sa., 15. Feb. 2020 um 16:56 Uhr schrieb Thiede, Christoph Christoph.Thiede@student.hpi.uni-potsdam.de:

...

Not sendTo: instead of sentTo:? ;-)

This was the first mention of #sentTo: in this thread :)

Sorry, I misread. I was referring to the tip you got from Marcel, but now I see it uses sendTo: already. Lucky for me, we do not also have a sentDo: ...

Sounds like a mirror problem again? Perfect decorators can be nested with arbitrary depth. What should be the return value of (EasterDecorator on: (ChristmasDecorator on: Object new)) undecorated?

It depends of course. :-) In my case #undecorated used to strip them all off, answering the Object at the core. But I think that design was never perfect anyway. Ideally I would not have had the need to undecorate anything, right?

Hi Eliot.

P.S. I may have the selector wrong, but there is a mirror identical.  Whatever the selector is, I meant that.

It is Context >> #object:eqeq:.

Best, Marcel Am 16.02.2020 02:59:55 schrieb Eliot Miranda eliot.miranda@gmail.com:

On Feb 15, 2020, at 6:59 AM, Thiede, Christoph Christoph.Thiede@student.hpi.uni-potsdam.de wrote:



Example:       anObject := (MyDecorator on: anotherObject)        doSomething;        yourself.    I would like that anObject is the decorator and not anotherObject at the end.

Marcel told me the following tip for this scenario, which sounds quite interesting to me:

doesNotUnderstand: aMessage | result |

result := aMessage sendTo: target. ^ target == result ifTrue: [self] ifFalse: [result]

+1.  Everyone writing proxies should be aware of this.  BUT!! !Ot should be written as

^ (thisContext object: target identical: result)        ifTrue: [self]        ifFalse: [result]

to avoid sending a message if #== is a true message send (which is what we’re intending to arrange).

P.S. I may have the selector wrong, but there is a mirror identical.  Whatever the selector is, I meant that.

:-) Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Jakob Reschke forums.jakob@resfarm.de Gesendet: Samstag, 15. Februar 2020 09:28:58 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject   Thanks Eliot, I understood that for the inspector use case, but my thought about #yourself was rather the following:

If I pass the decorator around and some domain code (not in Inspector or similar) uses the decorator in a cascade with yourself at the end, I do not want the result to be the undecorated object.

Example:

anObject := (MyDecorator on: anotherObject)        doSomething;        yourself.

I would like that anObject is the decorator and not anotherObject at the end.

Am Fr., 14. Feb. 2020 um 22:39 Uhr schrieb Eliot Miranda <eliot.miranda@gmail.com [mailto:eliot.miranda@gmail.com]>:

Hi Jakob,

On Fri, Feb 14, 2020 at 1:14 PM Jakob Reschke <forums.jakob@resfarm.de [mailto:forums.jakob@resfarm.de]> wrote:

Am Fr., 14. Feb. 2020 um 19:42 Uhr schrieb Eliot Miranda <eliot.miranda@gmail.com [mailto:eliot.miranda@gmail.com]>:

Hi Christoph,

On Feb 14, 2020, at 9:51 AM, Thiede, Christoph <Christoph.Thiede@student.hpi.uni-potsdam.de [mailto:Christoph.Thiede@student.hpi.uni-potsdam.de]> wrote:

There is a good solution to this.  Low-level Inspectors *should not* use instVarAt: et al.  Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy"

(MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Sounds good.

If inspectors et al would use this, my decorators might not have to carry the following collection of methods/copies:

asExplorerString basicInspect basicSize class instVarAt: instVarAt:put: perform: perform:withArguments: primitiveFailed printString printStringLimitedTo: respondsTo: xxxClass

If DataStream et al did, add the following: readDataFrom:size: storeDataOn:

I'm not sure how to generally handle #yourself. I have this in my decorators as well because I don't want to undecorate when the decorator receives it.

typically use of the mirror protocol, at its most basic, looks like

thisContext object: foo instVarAt: index

so #yourself is handled by ending the cascade early, and answering the object (foo in the above).  Yes that's a tedious refactor, but straightforward, no?   --

_,,,^..^,,,_

best, Eliot

Hi Marcel,

yes, I already discovered both. I thought I introduced this into the initial problem statement :-)

Best,

Christoph

________________________________ Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Taeumel, Marcel Gesendet: Sonntag, 16. Februar 2020 11:57:33 An: Javier Diaz-Reinoso via Squeak-dev Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Hi Christoph.

At the byte-code level, one can see the inlining. The #class send is C7, while any other unary message send seems to be D0:

[cid:544f287c-1cf6-4a0c-9c09-a62ac75eefd7]

Take look at EncoderV3 class >> #scanBlockOrNilForLiteral: and also explore "Smalltalk specialSelectors".

Best, Marcel

Am 14.02.2020 18:51:42 schrieb Thiede, Christoph christoph.thiede@student.hpi.uni-potsdam.de:

Hi Eliot,

Or provide a vm flag which controls whether special selector #class (the bytecode for sending #class) is inlined or not. This flag can also control #== & #~~.

Could you give me some (code) pointers about this inlining? :-)

There is a good solution to this. Low-level Inspectors *should not* use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

I like and do agree to your vision for tidying up ProtoObject! Maybe we could also consider some "MetaDecorator" that inherits from ProtoObject and implements #class, #instVarAt: etc. for arbitrary targets? This might simplify the access to the mirror protocol, and it would be a bit more object-oriented:

(MetaDecorator on: aMorph) class "--> Morph" (MetaDecorator on: aProxy) class "--> Proxy" (MetaDecorator on: aProxy) instVarsInclude: 'target' "--> true"

Hopefully I’ve answered your questions and given you pointers. I’m interested in collaborating to make this a reality in the next major release.

You did. I'm too.

Best, Christoph

________________________________ Von: Squeak-dev squeak-dev-bounces@lists.squeakfoundation.org im Auftrag von Eliot Miranda eliot.miranda@gmail.com Gesendet: Freitag, 14. Februar 2020 18:27:14 An: The general-purpose Squeak developers list Betreff: Re: [squeak-dev] [Bug & design issue] Messages understood or not understood by ProtoObject

Hi Christoph,

On Feb 14, 2020, at 5:18 AM, Thiede, Christoph Christoph.Thiede@student.hpi.uni-potsdam.de wrote:



Hi all! :-)

When we were fiddling around with some proxy implementations today, Marcel and I discovered an interesting set of issues regarding ProtoObjects.

Let's start with a concrete bug:

Code to reproduce <do it>:

ProtoObject new class.

Expected behavior:

DNU exception, because #class is implemented on Object only.

Actual behavior:

Outputs ProtoObject.

Even stranger: If you debug the method instead and simulate the #class send, you get a "simulated message class not understood" error.

If you subclass ProtoObject and override/forward #class, the results will deviate based on whether you are debugging or executing the code.

Some thoughts:

There are two options: Either to implement #class on ProtoObject (by moving the primitive definition up from Object), or to remove it from the specialObjectsArray so that a send to #class is not compiled differently than a send to any other regular message.

Or provide a vm flag which controls whether special selector #class (the bytecode for sending #class) is inlined or not. This flag can also control #== & #~~.

This leads us to the rather general question: What special messages should an instance of ProtoObject understand?

IMO *only* #doesNotUnderstand:

See below on the mirror protocol

We don't answer this question completely consistent at the moment: For example, #instVarAt: is implemented on Object, but #instVarsInclude: is implemented on ProtoObject. Again, it is a bit weird that the implementation of #instVarsInclude: calls #class which is implemented on Object only. And so it goes on:

* #someObject is implemented on Object, but #nextObject is available in ProtoObject. * #withArgs:executeMethod: is implemented on ProtoObject, whereas #perform:withArguments: is on Object. * #ifNotNil: is implemented on ProtoObject; however, #ifNotNilDo: is implemented on Object. (btw: does the latter still have any raison d'être?) * #flag: exists on ProtoObject, and #yourself exists on Object only. Isn't #yourself rather a syntactical element we would like to behave exactly the same way for every possible message receiver? (btw: Is there a good reason not to speed up #yourself via specialObjectsArray)? * (I don't claim for completeness)

And just some other problems regarding to ProtoObject (just collecting them here instead of forgetting them forever):

* ObjectTracer is broken due to several reasons. Will fix this soon. * Inspectors cannot inspect ProtoObjects correctly (for example: Inspector openOn: (ObjectTracer on: Display)). This is because #instVarAt: is implemented on Object only so this message is forwarded. Maybe we should replace these critical calls from Inspector by something like #tryPrimitive:173withArgs:? But should we do so in Inspector or rather in a new subclass named ProtoInspector or similar? Hm, then we would need to implement #inspectorClass on ProtoObject, which is bad either ... We will keep investigating this issue.

There is a good solution to this. Low-level Inspectors *should not* use instVarAt: et al. Instead they should access object intervals via “the mirror protocol”, see the methods in Context, #object:instVarAt:[put:] #object:at:[put:] #objectClass:

You see hacks everywhere (the xxxFoo messages in serialization) because we don’t have a well-defined notion of what is in a proxy. Having one that is as minimal as possible is IMNERHO the clearest definition possible, and the most minimal is just having a #doesNotUnderstand: that prevents recursive does not understand vm exit and displays a stack trace without sending further messages to the proxy (because it uses the mirror primitives).

Note that the mirror primitives are what the execution machinery in Context use to ensure that bytecodes do what they are expected to do. An I et car access should *not* send instVarAt: to a proxy; it should access the inst var of the proxy, etc.

However, the crucial question is: What special messages should an instance of ProtoObject understand? On the one hand, the idea of ProtoObjects is to work as total proxies with a maximum amount of forwarding potential, which implies a minimum number of methods. On the other hand, certain aspects such as accessing instvars or executing messages are really identity-related. Looking forward to your comments :-)

Hopefully I’ve answered your questions and given you pointers. I’m interested in collaborating to make this a reality in the next major release.

Best, Christoph