Eclipse OMR Architecture, 27 Feb 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: OMR Architecture Meeting 20200227

Description

Agenda:
* Re-architect Compiler Memory Infrastructure (#4783) [ @dsouzai ]

A

Meeting is being recorded, everyone.

B

To this week market extra meeting, so this week we have a continuation from a topic that we discussed last week so Erwin after last week's or urban went away and did a number of experiments with with the memory hierarchy in in the compiler and he's going to report back on what he's found.

B

C

So I was initially going to first only touch the raw allocator and deal with the problem that marked the saying he ran into, but then I sort of figured I'll just go all the way and just try and get something working.

C

The approach I took was in order to deal with some of the concerns that were brought up before I essentially made it so that it's the same number of delegations that happen anyways right now. So it was a relatively simple one-to-one mapping between the existing.

C

Alligators in this new world segment pool which no one's using and we should probably does remove that code anyways so for the most part, I was little surprised, it'll work, but it does I'm able to run both canine and Omar no problems. I did have and I put the links to my commits on the issue.

C

The things I ran into were, and I listed them out as tasks.

C

We should do if, if we're gonna go down this world, the smaller tasks that we were talking about in terms of what can we do or this we definitely want to change the code that passes these alligators by value, specifically the wrong royal hater and the TR alligator material allocated particularly backs I thought we were passing reference, but we're not and I, don't know if it's okay not to pass it not to make a copy of it, because the tear alligator is a cs2 alligator that when you create a new one, takes the base alligator, as the thing is going to use to allocate memory, but it also has its own segments that it keeps track of and stuff and I was like.

C

Okay, if we, if I change it to just use the one alligator that the comp object has I, don't know that affect the performance or not. So that's something that didn't think of any reason why we would have passed it by value in the first place. No, no I, don't I I was genuinely shocked, I I thought it was done and then the compilers are complaining.

B

About have we always thought you know if we've always got that way, or is it whenever, when we read our context this a couple years ago, it was changed to be.

C

By value we never needed we weren't using a scale before, but we never a high PR alligator. We were doing shared alligator directly so that we then looped wrap it up and uh the comp commentator call which returns it by the value which was extremely frustrating because it required a lot more changes that I had to do that necessary.

C

The wrong alligator same thing.

C

The the raw alligator at least makes a bit more sense because in the old way and was stateless and it's still stateless sort of, but it's now it's a hierarchy so as it be a theater, but before it was essentially just it didn't matter you one Gras alligator was the same as another raw alligator, because it was a wrapper or a lock, so that was annoying, but at least that made sense and then, in order to get the last thing we should do so one is change on Raw alligators to be passed by reference.

C

Second, one is change, even even if we don't go down this road, it's just nicer to pass by reference. Second, one is pass all dirty alligators by reference. Third, one would be a bit typed alligator to actually use references. It'll just make life easier. If we go down the road, because if we ever change the even if we don't go down this way- and we say you know what we want to make things better- garetty's it'd be nice.

C

If the alligators were all or just more references and again like the scary part, was it built it just when I ran it would crash, because all right, but can I change tonight on we cater to take a reference. It way it had a reference to an alligator, the longest access died. So creative of that is what we don't want to happen.

C

All of the complexities I ran into we're all open. J9 Omar was surprisingly to the easy and clean and they're, mostly all the issues that mark got into. They all came from the fact that j9 used a nine backing alligator as it like the PM that provides it provides. It provides the chain I'm every segment, but this whole mechanism was everything is a TR memory, segmented, so I listed out the decisions I had to take.

C

As a consequence of it I for the regenerate memory segments approach, a v8 it from how marked it is, but essentially mark land, it's the same problem when I was talking to I, just sort of kept it as dumb as possible, just extended it extended here memory segments and then and then essentially the consequences that the Java core will say. Oh you used up the entire memory segments, but really you bite ro, but it's 16 minute segments you'll be off by 16, Meg's I, don't think it's terrible.

C

I mean 16 Meg's yeah I mean it might. It might matter.

B

Have my comments.

C

Like that, but.

D

We can't ever want to lose my opponents.

C

Precisely for that, because of the proof of concept, I was taking a bunch of little just decisions and I was like I know this won't fly and any real design, but just to show that it works. You know, I, don't want to get bogged down in all these tiny details that come up, say right mistake. I made there were a couple of other enforcement decisions like.

C

They construct the destructors had to know like if the API says it has to be a TR memory segment, but in j9 it kind of had to know that it was actually the child type and again it was encapsulated in one file. But it's still sort of unfortunate same with the raw alligator. It had to know that the alligator you're passing in to copy construct from even though the type of the API type will actually be a very specific type, and you better get that right. When you're compiling.

A

C

Code, so it was that and then the last one was just the persistent alligator was mostly fine, except for the fact that now that we in this prototype since I switched it to all be here memory segments references.

C

What the old approach did was that it would say, do I have an existing segment, that I can reuse or to fulfill. Sorry, do I have an existing segment that I can use to fulfill the request of being made right now and if there wasn't it, which is returned knowledge that it was null, would allocate a new segment that doesn't work with references. So I've talked to Leo. It returns an iterator right now, which is not something I like because the.

A

D

C

Fine in this world, because in this particular case, because it's only being called there and pushed it locked, but if anyone else inside there will see use it without knowing the iterator could be invalidated because the persistent alligator is global and it's not I, guess something that you have to be careful of. So that's those are the decisions. I have to go through to make this work, but it does actually work shockingly expecting to see no crashes but I.

C

B

Is the thing that you have put together in a sane enough state that you could actually do performance runs with it like with it? Would they be meaningful performance rooms to get an idea as to whether or not this is.

C

D

B

Are problems of this approach? I could do it um if it would they make sense if.

C

B

C

I think so I mean they're essentially the same. It's all exactly the same. Actually like I left all heuristics the same. Nothing is actually different.

C

There are in expect remember how, when I first described the design we're talking about having multiple delegation levels, this one we have the same number of delegation levels. The segment sizes are the same. How you choose to free back to the VM is the same. Nothing has actually changed in any way, so I expect performance to be the same.

C

That said, some of the decisions are less than ideal, and so, if even if we did performance runs it wouldn't matter if like when we change the design up, it's a fundamental way. They affect performance, so I think I'd say it only makes us different from when we're actually I do.

B

C

B

I guess the question is whether or not if we have to do a lot of work to get to that point, and then we realize oh, the performance. Isn't this isn't a great solution? We shouldn't have done it. You can answer that question now already listened indication as to whether or not I.

C

Can try it and hopefully, there's no wonder like bugs that I've missed in my groups, concept I mean I did one.

A

Benchmark where there was a whole bunch.

C

Of hot compile- and there was no bugs so thinking it's fine, but never done so I can try that out assistant experiment to see what happens. The thing I didn't do in this proof. Concept was I, didn't do what mark needed at what trill needs, and the reason is because I, like I, don't know which approach will be take.

C

It would be that you want a standard class that recaps weights all these layers or would be that the the type Dictionary and the method filter will take in an alligator I, don't know what the velocity was so I was, like. You know what I just sort of said. Okay, if we switch this model, it'll work, how I mean to fix that problem can be done in multiple ways just the way it stands right now like. If you passed in an alligator, it would also work right now. It would be the same thing.

C

So that's why I figured that's a different discussion.

C

But it is the case right now that you only talk about a segment alligator, a presenter or a rawhide advertiser. That's the only two classes you are ever talking about in the entire code base, except for very specific places, and so it's consistent on that point. I.

C

Believe then, for the code- and he had some comment- yeah, mostly just to do with field being of reference types, but I.

D

Think that's something that could be fixed enough, revised design, it wouldn't change any performance, characteristics or.

C

Anything it was math, it would mess up, move assignment and obvious sign of operators.

D

So generally, it's not recommended.

C

The other thing is that this and maybe I'll, maybe I, should put up a diagram of how this prototype looks because it doesn't match the original PDF put in there.

C

So when Leo me and David were first talking, we sort of figured that we would implement the debug segment provider as instead of using the malloc raw alligator, would use the debug raw alligator to allocate because after time, APIs actually because that doesn't work, because when you deallocate, you don't want to m1 map, you want to M protect and you want to deallocate sub earlier, so that wouldn't work with the existing set up. So I sort of said well, just the debug statement.

D

Provider would be another segment alligators. It.

C

It didn't it didn't, have a backing provider, it just did the work directly, so I can put up an updated diagram of how this looks, which holds map one to one exactly with the existing infrastructure. But it's easier to reason about in terms of names, and you can rename a bunch of the stuff to keep the name. I get the name the same because again, I sort of just said: well, simplify all the field names and all that stuff.

C

But realistically all the field names could just be changed to segments alligator everywhere and it doesn't matter what the type is. The knife is, always all the fields are all tier ii, alligator reference so and in fact it was actually very easy for me to construct some of these things because they all took a reference of saving alligators. So if, like literally build things around I thought they broke.

B

um One of the things he talked about last time was if we were to go down this path and having the ability to have both the new scheme coexist, perhaps with a built flag or something like that, is that that's.

C

B

Right down there.

C

Everything is guarded by new memory. Okay, I can build both and.

B

Then some changes would potentially be common, like making the passing the alligator.

C

By reference that has to be done for us, but right now it I thought it would be common to both, so it would be coming the boat there, I mean I, think I, don't know how you guys feel about this I, don't know if it's worth the effort to change all those things, references if you're not going to use it in any way. I, don't know how you can feel about it.

A

Turning you mean changing things to references if we're not ready who's been your memory, yeah does.

C

It still work you think going through this because there's a lot of places and it will be a dependent change like you change, if you change, for example, the return type of comparing a locator that breaks a whole bunch of code, but you have to change at the same time. The majority I believe is in or more but there's like one place in open j9 and there might be others in extending projects.

C

So it's one of those things where, when you change the return type of compound allocator, it affects by the lock same with the raw alligator you actually have to, and if you look at the way, I have to do it it's a little silly.

C

But in order to be the raw alligator I personally instantiate a raw alligator on the other local then use that raw alligator to that one hit another wild Gator, which is a pointer and then have that pointer, be the reference, because now it lives that ethically otherwise it would be on the stag to die well as before it didn't matter, because you did just pass them around so the liking. The one thing I said that we have to do here is what better understand life divers.

C

All of these objects didn't care about before, but sure and the TR alligator is a question that just I don't really know. Why that's the case? Why why? Why do we even have one in the compound date if we're gonna pass it by value I've annoyed it? It's.

B

Weird it looks, it looks like to your alligator right now points to what's effectively trying to be a reference wrapper, very string, not.

C

Sure why? But if it's not because it'll create a new one right, like I thought it didn't matter at first, but it does because if you look at the ps/2 alligator, it has like it resets the punching I think it does we bucket bucket emily and every time you create a new rock, I'm paula catering, you bucket, and that might have changed the performance characteristics like if we switch to only using one, the various buckets at home versus the one. I don't know what the consequences that is, so that's why to do it.

C

I left it as each one gets their own version.

C

Kind of expensive- and we actually like it's kind of ridiculous, because we're passing these two like nope, like multiple functions, won't like ask them along, so we're constructing and destructing. This thing like crazy it just just to get one of these is so no.

B

I need I need to look at this.

D

Further because it looks to me like it's just a reference wrapper, so it's not actually like when you're copying and it doesn't look like it's constructing anything new. Okay. This is a very bad reference. Rather, it itself has a deal to de-stress I. Don't know why.

C

That was okay. If you can take a look- and let me know, yeah I'll take a look into that, because.

D

I, it is the the one I'm thinking is pretty bad at me, so yeah. Maybe it's overridden somewhere and I'm, not being the override here, but looks like it's just a reference wrapper which.

C

Actually has at least a little bit better, there's a I mean one of them. That was.

D

In that case, the new TR elevator could just be the new technology.

B

So I mean so assuming, let's assume that this is something that's it's going to work and it's going to it's going to be not meet our needs.

B

Is it? Could you put together sort of a transition plan from where we are right now to the call that I? Don't then the steps that we need to accomplish along the way together there talk to these guys in terms of the idea.

C

I, don't miss. Are you with.

B

B

Yeah I'm just sort of thing as a we can, we get potentially staged like if we can do a lot of the stuff completely independently of actually somebody doing the entire thing. We can give ya some of this workouts, like the changing of this values, to references for example. So, but just something we have an idea in our mind and throw it when would be a great time to do performance, analysis, sure, I,.

C

One of the things I wanted to get out of this was that this right now is the one-to-one mapping exactly the way things are, as I said, I wasn't sure it. There was a desire to change that a bit or not like in terms of right. Now, every single alligator will do exactly what if they're one-to-one mapping does, whereas before we were saying you know, we could let this so that every alligator has its own policy. Maybe doesn't make sense to do that because you're delegating too much, maybe some policy should be merged.

C

Essentially what I wanted people to look at it and see? This is now that you have it cuz it's an API. Does it make sense, break it pick up things or not, but I don't.

B

What any easier to do that, sir? Is it easier and less error-prone to do the one to one mapping first and then, once you get there, that your changes because worried about changing too many things at once, yeah actually just kind of goes back to the previous thing about the transition plan, because the one thing that I would really not like is if we started on this path and that we gave up on it. So we would have yet another set of memory allocation.

B

This other apparatus in place that is partly used. Maybe we've got this other stuff, so we were left with this in an incomplete state, again, I.

C

Think this one is one of those things where you've never.

B

Partly use it would.

C

Have to be one or the other different.

C

But did you have that like if we give up it'll, be any given up without user, but I.

D

C

Come up with a plan: okay,.

A

For a question I'd, like you, we re architecting so that there's more delegation, whatever going on I, might feel like that's almost like a per project decision right, because the part of the point of this design is that om are just deals with everything through the interfaces and it's up to the users to decide how they compose various alligators together.

A

So I think that's all mostly just up to downstream projects to decide how they want to use those interfaces and how they want to split things up.

C

Do we turn that, like some of us are also old engine activators here, we need to have an equivalent discussion with the open chain and compiler side now like if I just say, Walmart decides to go for this I. Don't think what is the process.

B

um At the very least, you should have an issue open, I, guess an open j9. So you can have some kind of a discussion there.

C

B

The most part is odds.

C

Because meeting forth on a compiler and very few people who work on the memory so.

B

It's going to be the same people in this room and on this.

C

B

Just for transparency purposes, you should probably do that. You.

B

D

Work out what what doesn't look it out open, J, no I think you should discuss what the transition plan would be. What the effect would be. The open j9 not moving to the new reason line. They could make that choice. They can't make that choice, then open j9 compiler. Our community needs to make that choice. I mean.

C

D

C

They can take like one or two things from Walmart and move it to App Engine in right, as it isn't in fact we're pretty small I think it did to adopt. It now hope I won't throw that noise in terms of what you want to do with trilogy builder will be pretty massive because you wouldn't be able to. You would have to then implement the alligator that trill in the new builder 1 yep.

D

Look I said: I hope they won't meet that once.

D

D

Boys back up what what testing have you done on this so far, testing.

C

Or no, not even that far I I sort of got very basic Java does running and then I ran another batch more that did get too hot. So I did see that I can reduce. I didn't go super crazy because I was like there's decisions made.

A

In here that I want.

C

The agree with anyways, so I figured for this particular meeting. I didn't bother in terms of and the reason. Why is because there's it's very much a one-to-one mapping is I think maybe two places where it's different. Almost everything else is exactly the same. The two places are.

A

D

We allocate from.

C

The j9 memory, which is itself a classic new.

D

Bill those are not composable classes right now, right right.

C

D

They're ready to test a look yeah.

C

Here's base, if you look at it I have explicitly you know, include all more tests whatever at j9 test, whatever only because I I don't fully understand how to build an accessible class from scratch. I'll say you know what I don't want to copy an existing template and in return all I try to minimize Union doesn't know.

D

This, presumably, that would be the way to do this, for real and Walmart would provide a like major segment provider in clouds. A segment.

C

Alligator interface, I, don't think it might be sore. So I was thinking about this and I. Don't know how like, if we're gonna make anything sensible. The only thing that can be accessible is the api card. One, that's the after I've done I. Think else ask that should be the only reason for that is in j9. We have extra queries of you know how much distance technical assistant member, maybe I, hated how much region memory and refer segment size. So I just put all that in Walmart. One.

A

C

It and they put like a tear of certain there shouldn't be calling this marriage.

D

C

Thing but those are all those are all API having top-level and.

D

C

Two approaches there: one is you make that an abstract class accessible the other.

A

C

Does think where it creates a second abstract class below it, sort of what length look did with Sigma providers and alligators, where Sigma provider only had requests and release, and it's like an alligator at the remaining. No thank him again. The reason I didn't go into all this is because I was like these are things to discuss with actual design of the real way.

A

I'm, sorry are those extra queries. You said that are only used in open j9, so that the reason they right now they're defined in om are is so that they're defined on the interface class. The.

C

A

Did that was because.

C

In Jane I and we passed that we passed a reference to a few. Oh, yes, and if I could change it to be the one that I know it's going to be yeah, but the whole idea was if we change that future.

A

C

Maybe it's fine I just.

A

C

A

C

A

Because I was trying to understand why they had to be in omr if they were.

C

Yet he was literally just for the reason that I think you said, there's different. We look together, yeah, that's up for debate, I'm, not let it do any one of them. So.

D

At this point, we wouldn't actually be able to test.

D

What you've done here is for something like Dictionary. We.

C

Could the way you would do it is either the way I put in the comment which is create another classroom, same name, j9 and Walmart or more. You can hold it one way and generate a proposal other way, or we extend us that we pass in and allocated to it and that little, but that could work in the current approach. Also, the same thing works both ways. This is the current one we could create.

C

Another class I have the same name that extends from the sister segment provider, and you know, use a compiler magic will just perhaps the right class do all the right things. I.

D

Remind myself how.

C

This works so the time dictionary and.

A

The metal filter taken.

C

A system segment provider there is a whole Martha, Sigma provider. There's a Jane I think the providers they have different instructors because they're not accessible we're. Sorry. Where are they getting? They are just created on like in in there like in the pencil there itself is about getting it from anywhere else yeah.

C

So these the two approaches, encapsulated class or faceted.

A

Well, yeah yeah.

C

D

Else to create that default, yeah.

A

So wisdom can gauge it builder drill or it can be the user and the user constructs for capacity. Okay, so.

D

Burchett builder can't be the user.

D

Because they don't want them to yeah I, imagine we don't want to be engaging with all these. They could builder people independent of the Omar compiler I. Don't want the dependence on the Omar compiler for gentlemen, so.

C

For something like say a night where we already include on the memory stuff, wouldn't it make them there to be able to construct as a pallet or in the first part of what you said.

D

C

Of the big reasons you didn't want to facet and allocated because you don't want any use it to did builder to include all normal compiler.

A

Headers in the yeah.

C

But when you were making it work with j9, as we include compiler memory, headers and a whole bunch of stuff anyways, that particular scenario: wouldn't it be okay to faster than alligator. The Jana in case is different. Yes, you're right, so I think it's only that we we have the default one, and then he I guess the only problem.

C

There is that anyone who was not in one of those two worlds wants to have a different alligator, but doesn't wander into the compiler that have the problem, where maybe they they're forced to use the default class. Unless we say everyone has to implement, have to create a class of this.

A

C

Then that'll be rigged after.

D

C

D

I mean there's a there is a use case or good filter that doesn't have the Omar compiler on the in the library. We don't currently support that, but but that is a direction that I'm trying to talk about. This isn't one of the next I guess the next one for the next particular yeah.

D

All right, I'll for shot up a little bit.

D

The general strategy there's I'm, trying to turn JIT builder into a compositor constructor of its own, which can which operates above the owner compilers. The olmeyer compiler becomes kind of a back-end or JIT builder, but you can actually work with Jin builder itself and manipulate the JIT builder services huemul as an intermediate representation.

D

In that scenario, the output of that process could be to stay persistent form of the jiggler il and not actually even invoke the Omar compiler editor, so so that that might not intersect with what you're talking about, because the place where the place where you need the compilers memory. Allocator is in the Omar builder implementation files, which wouldn't have in that scenario, but I.

C

Guess the question becomes: do you envision and he uses of that wanna.

D

Share of own memory: okay, alligators, so.

A

D

A

Okay, I think I have an answer to that about being able to read marks, mine was a word of the quote.

A

Yes, mark is right.

A

I can put pieces together of what he's thinking. So, if the OM, our compiler right now, is just as a back-end right the so when JIT builder is in a mode where it uses me one more compiler, it would have to provide the allocator the OM our compiler will use now that allocator does not need to be the same as the one that JIT builder is going to use. So JIT builder can have its own separate way of providing custom allocation mechanisms to the user, or it cannot that doesn't really matter.

A

It's going to be whatever magic and the implementation. When it's using do a more compiler, it is going to wrap whatever it has in a segment, a locator and a royai locator, and pass that to the UM our compiler. So.

C

As to what you're saying is that if the user.

B

C

Have an API where you told user, give me your alligator deallocate function, then it will. Then it will use. Then it'll construct the segment provider such that it uses ya users once the user doesn't provide the to allocate.

A

The alligator pointers it will just it'll use its own default one and when it hooks into the OMR compiler, it will pass those along through the interface we've set up for you, one more compiler. That's.

C

Better I mean it.

A

Suppose you're trying to think further down the.

D

Line from where I am.

C

To your custom allocated.

D

Less exists, yeah I mean in in general, like not using. You want more compiler as a back-end means not using the Omar compiler memory. Raishin infrastructure here, yeah right, so at that point you could, it could be, doesn't have to be in the prototype that I've written is you're just allocating stuff with C++, new and silly and wherever you want right, it's not even using special allocators to do things right. It's more.

D

The use of the Omar compiler allocator framework introduced itself when the Omar compiler becomes a central part of the story and not it should not impose itself on the rest of the story. It's fine, my basic position there but I think I, don't think that's very hard, like a having thought a little bit more about it. While I was talking, I, don't think that's actually a big issue. So nice thing is design.

C

Is that, in terms of that, the persistent minor in the oil market Pilar right now takes in a Malick raw alligator, it would be trivial to just say thinking the custom alligator everyone all the rest of the code would be exactly the same.

A

C

A

That's one of the benefits of using this interface approaches, your more composers, don't care where stuff comes from. If tooth builder, that's important to use it, it doesn't use it.

B

So we haven't had a discussion on the compiler API in quite some time for version 0.2 or 1.0, but I would imagine that where this is going should be part of the compiler API.

B

Yes, so where this is the folio and you live every allocation interface, so I think.

D

B

D

Even be a good place to start, oh yeah, how do I get memory all right figure.

C

That out with the compiler API, is that something where you have to expose the compiler headers like? If you want to use another API you're, building a pouter headers there, or is that the thing where you're still.

D

Compiler, it's a hallmark compiler in this price conversation right so I mean all the other, all the other, most of the other, certainly most of the other compiler. Most of the other components. Omar have a notion of a public API that they exposed and that they expect people to program against.

D

Maybe some issues with those public API is and not all of them are stable at cetera, but they have a notion of a public API, a compiler right now we don't have. We haven't documented any real notion of a public API for people to build compilers using our english program principle. Everything is it's open season on api's and classes inside the compiler, which is nice from a flexibility standpoint, but not great from a what am I allowed to do. What should I do? How can I do this? The right.

C

To do, oh, you mean they've, got point in terms of things like init, the compiler. It.

D

D

If I'm going to write a compiler using the Omar compiler, and why is that? What are the things I have to change? What are the things? I did what's one of the public integration points that I need to hit, and if that story is well, you go inside a system provider and you go wrong allocator, but don't do this that thing and torque with that thing, if you want like it, if it's a complicated story like that, then no one's gonna come yeah.

D

If we can start to build up some documentation and and and real actual like raw allocator, has this API implement this and poof your compiler will be able to allocate memory, you know, etc, etc. For the set of classes that you need to implement, then all of a sudden you've got an API that people can actually look at me to vote and have some hope of being able to in like interact with.

D

We need to need to do better on this as a general rule, where we've done a lot of work to make things accessible and so that you can actually implement compilers on top of the Omar compiler framework, but I think it's still kind of very difficult for someone to learn how to do that without sitting in on the floor of this building and soaking in it for a year or two in our team, which is not great to do that. Then.

C

Maybe, as part of our roadmap plan, we can also formalize the api's.

D

Course is about to get apartment. That was, that would be.

C

B

Any other comments on on this.

D

The next steps are somewhat clear for putting together the putting data as an extension hardening.

D

They one of the other things that having a defined public API, allows us to do, is document changes that are happening in the API that are going to affect Deborah idream implementation, which is something that we have no, okay, basically anything can change underneath you not everything. Does.

A

I have just one question when, in your third point, you said that you that constructor, that in j9 are a that takes a Const or allocate or reference, and it that just because you haven't gone through and changed all of the B, all of the aspires to be as by reference or was there another reason for that.

C

It's because I copied raw alligator as it was, and they had the same copy. Constructor controls like.

D

I, don't know what are all.

C

The places that might implicitly have this thing, yeah I, believe now, except maybe now that I fix everything to return by reference. Maybe this isn't a problem anymore: okay, Oh.

A

Cuz when we were trying I thought that was originally the reason you needed this I think.

C

Maybe there maybe it wasn't that time, yeah well, maybe I just really think at the time we were talking, I, don't think I'd even got around to building it right. Okay, but I saw the copy, constructor and I was thinking about it going and for only ever dealing with a API raw alligator. There are friends. How is the compiler supposed to know which copycats are yeah, but maybe we don't need that copy of instructor anymore yeah.

B

D

C

To the problem that I ran I try.

D

C

D

It my virtual hierarchy, so.

C

I have an ugly like reinterpret rat's run like.

A

C

You know what you are.

A

My understanding was, if we're passing everything by reference, it doesn't matter what never copy constructing anything. Maybe.

C

I can just remind.

A

To conscious strokes to see what happens? Okay, if it needs to be there, that's fine, just not.

C

Only why in now that I think about it, but again that's one of the things. I am there's a lot of like things glossed over that need more information, yep, that's right and.

C

That's where the whole point of this was just to get something.

D

Minor nit use of j9 re makes me read: register allocator everywhere.

C

Yeah really hurting.

B

That's I've done reprogram.

D

D

It's a minor thing yesterday, but.

C

The reason it was there was because, in that design dog we put Janet say Jen are a you know what I'm lazy, oh yeah,.

A

D

B

A little using reading, okay, any more questions or discussion on this topic for adjourn, no okay, all right thanks, Erwin and a little adjourn and convene again in two weeks thanks. Everyone hi.