[Vm-dev] Weird problem when adding a method to StackInterpreter
Eliot Miranda
eliot.miranda at gmail.com
Wed Dec 28 19:28:54 UTC 2011
On Wed, Dec 28, 2011 at 3:14 AM, Mariano Martinez Peck <
marianopeck at gmail.com> wrote:
>
>
>
> On Tue, Dec 27, 2011 at 7:02 PM, Eliot Miranda <eliot.miranda at gmail.com>wrote:
>
>>
>> Hi Mariano,
>>
>> On Tue, Dec 27, 2011 at 7:05 AM, Mariano Martinez Peck <
>> marianopeck at gmail.com> wrote:
>>
>>>
>>> Hi Eliot. Now I found another thing which took my attention. I would
>>> also like to trace when objects receives messages from the special
>>> selectors (special bytecode associated). So for example, I would like to
>>> trace an object that receives the message #new, #x, etc etc etc. With a
>>> StackVM I need to call my method #traceObjectUsage: from the bytecodePrim*
>>> methods. Usually, only when those methods answers before than the
>>> #normalSend. For example, in #bytecodePrimAdd I trace both the argument and
>>> the receiver when they are floats. If I do not add my sends to
>>> #traceObjectUsage:, then they receivers are not marked (logically).
>>>
>>> Now, what I don't understand is what happens with CogVM. In Cog, even if
>>> I don't put my calls to #traceObjectUsage: the receiver is always marked.
>>> I guess this is because I have put #traceObjectUsage: in a lot of general
>>> places of Cog. The "problem" is that with #class and #== the receiver is
>>> not marked (right now I don't want to discuss whether I should trace this
>>> or not) . Previously, with StackVM, if I have the call to
>>> #traceObjectUsage: in #bytecodePrimClass and #bytecodePrimEquivalent then
>>> the receiver is marked perfectly. But with Cog I noticed that it doesn't
>>> matter what I put in #bytecodePrim* because they seem they are never
>>> executed. Is this correct? Are these special bytecode always jitted from
>>> the very first time? or they are jitted on demand (when they are found in
>>> the cache) like the rest of the normal methods ? And the main question,
>>> what can be the cause of why I can trace with all the #bytecodePrim* but
>>> not with #class and #== ? I am obviously missing a place where I should
>>> trace....
>>>
>>
>> #class and #== are always inlined in jitted code and so if you want to
>> trace you'll have to modify the jit to add the tracing code as part of the
>> inlined code.
>>
>
> Ahhh that was is :) I didn't know that. So now I see that in
> #initializeBytecodeTableForClosureV3 or friends, you define them as
> notMapped:
> #(1 198 198 genSpecialSelectorEqualsEquals needsFrameNever:
> notMapped -1). "not mapped because it is directly inlined (for now)"
> #(1 199 199 genSpecialSelectorClass needsFrameNever: notMapped 0).
> "not mapped because it is directly inlined (for now)"
> And you have comments there and in the beginning of the method. Ok got it
> :)
>
>
>> Note that #class and #== must be inlined and not sent for the semantics
>> to be the same as the interpreter. In the interpreter these are never
>> sent, but the bytecode for them is executed, just as in jitted code, the
>> fetch of class and the comparison are executed but not sent.
>>
>>
> I understand and it makes sense. I have only one small doubt. With the
> rest of the special shortcut bytecodes such us #bytecodePrimAdd,
> #bytecodePrimNew, #bytecodePrimGraterThan, etc. there is usually the same
> behavior: check whether the receiver is of a certain type (like integers,
> floats, booleans, arrays etc) and if true then perform a C code instead of
> the regular message send. Then, if the receiver or argument are not of the
> expected type, then you follow with a #commonSend. Some other shortcut
> bytecodes just set the selector and argument count, such us
> #bytecodePrimAtEnd. And then of course you have #class and #==.
>
> Now, in the jit, you seem to use the same method for all of them (all but
> #class and #==) and it is #genSpecialSelectorSend. Such method seems to
> only set the selector and argument count. That is the style of the
> #bytecodePrimAtEnd that I mentioned. So..... my question is... is it ok to
> assume that when you JIT those special method they "stop making much sense"
> (in fact, they have less sense) since the only thing you do is to just set
> the selector and argumentCount? What I mean is that the jitted version of
> #+ for example will be generated as a regular jit (using genSend: selector
> numArgs: numArgs) rather than checking that the receivers are integers and
> if true answer directly (as #bytecodePrimAdd does). Am I correct?
>
Nearly correct :) There are two JITs, SimpleStackBasedCogit that does no
inlining except for #class and #== (because Squeak assumes these are
executed without lookup) and StackToRegisterMappingCogit that inlines
SmallInteger arithmetic and comparison, #class and #==, and short-cuts
SmallInteger comparison followed by conditional jumps.
SimpleStackBasedCogit compiles the special selector bytecodes for #+, #-,
#<, #> et al merely by generating normal sends.
StackToRegisterMappingCogit compiles (currently) #+ #- #bitAnd: #bitOr: as
a test for SmallInteger arguments, inlined code, possibly an overflow test,
and a fall-back conventional send if not SmallIntegers or overflow (see
genSpecialSelectorArithmetic). It compiles #< #> #<= #>= #= #~= as tests
for SmallIntegers followed by inline comparison, and possibly, if followed
by a conditional branch, the inlined conditional branch, with a fall-back
conventional send if not SmallIntegers (see genSpecialSelectorComparison).
It will also constant fold #+, #-, #bitAnd: & #bitOr: so that e.g. (1 + 2
bitAnd: 5) bitOr: 8 is compiled to a load of 9. And I reserve the right to
add additional optimizations as time passes ;)
So in summary, the old simple JIT did nothing special, compiling the
special selectors to normal sends, the new JIT does some simple inlining,
just for SmallIntegers.
I think this has no implications for your tracing code. You're unlikely to
unload the SmallIntegers and so you don't need to trace them. Instead, I
would try and define the abstract semantic model for Smalltalk and come up
with the minimal set of trace points. For example, for any non-immediate
object not created as a side-effect of execution (by which I mean contexts,
blocks and indirection vectors for closed-over variables) it can only be
accessed via a send. So it seems to me that the only six places in the VM
you need to trace objects are sends in the interpreter, sends in jitted
code, and the inlining of #class & #== in the interpreter and jitted code.
For performance, you could inline the bit test on the receiver in jitted
code either into each method's prolog or into the ceTraceLinkedSend
trampoline, avoiding going to C on every send, which kills performance.
>
>> But given that the stack vm and the cog vm are semantically equivalent do
>> you even need to add tracing code to the jit? If you're tracing e.g. to
>> discover how much of the object graph a given computation uses and you;re
>> going to use this information for something later on, like creating a
>> kernel image or something, why not just use the stack vm for tracing?
>>
>
> Indeed :)
> Thanks for going beyond my questions. For this thing I am doing (I call it
> ObjectUsageTracer) we have so far 2 users:
> - Luc is trying to do boostrap/kernel. In such scenario he can PERFECTLY
> use the StackVM since the computation of used/unused objects is mostly done
> once and then such information is used.
> - In Marea (what I am doing for my PhD), I want to dynamically detect
> unused objects, swap them out and replace them by proxies. It means that
> the system needs detecting these unused objects all the time. It is not
> something I just do once. Anyway, I could use the StackVM, no problem.
> But....with Cog I can improve the performance of my solution hehehhe. So I
> wanted to give it a try and see if I could make the ObjectUsageTracer work
> in Cog. So far it is working more or less good and I only found the problem
> of #class and #==. And I am not even sure if that's a problem in my case (I
> need to think a little bit about it).
>
Think about the abstract semantics. How can an object be used?
Encapsulation is your friend.
> Best regards,
>
>
>
>
>>
>>
>>>
>>> Thanks a lot in advance,
>>>
>>>
>>>
>>> On Mon, Dec 26, 2011 at 10:00 AM, Mariano Martinez Peck <
>>> marianopeck at gmail.com> wrote:
>>>
>>>>
>>>>
>>>>>> Then the test fails in self assert: (tracer isMarked: obj2). I
>>>>>> imagine it is because it is executing the machine code of #foo: . So my
>>>>>> question is if there is a way where I could intercept and trace the
>>>>>> receiver also there? I tried to do it but I failed.
>>>>>>
>>>>>
>>>>> See the flag word traceLinkedSends in cogit.c. A bit in the flags
>>>>> causes the JIT to generate a call at the start of a method for tracing:
>>>>>
>>>>> #define recordSendTrace() (traceLinkedSends & 2)
>>>>>
>>>>> The result is that ceTraceLinkedSend is called on every send.
>>>>>
>>>>>
>>>> Wow. I cannot believe how easy it was :) Thanks Eliot. So what I did
>>>> is to change Cogit class >> declareCVarsIn:
>>>> to set 2 rather than 8:
>>>>
>>>> var: #traceLinkedSends
>>>> declareC: 'int traceLinkedSends = 2';
>>>>
>>>>
>>>> And then just add my tracing stuff in #ceTraceLinkedSend
>>>>
>>>> Thank you very much Eliot and Happy Christmas to all VM hackers
>>>>
>>>>
>>>>
>>>>> HTH
>>>>> Eliot
>>>>>
>>>>>
>>>>>> Thanks a lot in advance,
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Fri, Dec 23, 2011 at 11:23 AM, Mariano Martinez Peck <
>>>>>> marianopeck at gmail.com> wrote:
>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>> Weird ehh, because you use #internalStackValue: along
>>>>>>>>> StackInterpreter in a lot of other places and you don't have problems
>>>>>>>>> there.
>>>>>>>>>
>>>>>>>>
>>>>>>>> Turns out it's not weird at all. Since
>>>>>>>> lookupInMethodCacheSel:class: is used outside of interpret in
>>>>>>>> findNewMethodInClass: and in callback lookup it can't be inlined and hence
>>>>>>>> can't access localSP.
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Hi Eliot. Thanks for you answer. It also turns out that I don't know
>>>>>>> enough about SLANG ;) so it was not weird at all but expected. Ok, I am
>>>>>>> learning in the way. So I understand that sentence. But (down)
>>>>>>>
>>>>>>>
>>>>>>>> If you want to get the receiver you'll need to use stackValue:
>>>>>>>> *and* you'll need to ensure that stackPointer is updated when
>>>>>>>> calling lookupInMethodCacheSel:class: from internalFindNewMethod (see
>>>>>>>> externalizeFPandSP), which may slow down the interpreter slightly.
>>>>>>>>
>>>>>>>>
>>>>>>> I DO understand what #externalizeFPandSP does, but what I don't
>>>>>>> understand is why I should only do it in #internalFindNewMethod. I mean,
>>>>>>> what happens with all the rest of the senders of
>>>>>>> #lookupInMethodCacheSel:class: ? maybe if one of those senders do not
>>>>>>> update stackPointer (externalizeFPandSP), then in
>>>>>>> #lookupInMethodCacheSel:class: I will be accessing something wrong ?
>>>>>>>
>>>>>>> Anyway, I wanted to trace the receiver in
>>>>>>> #lookupInMethodCacheSel:class: to avoid doing it in all its senders. But
>>>>>>> with the problem found, I workarrounded by tracing the receiver in its
>>>>>>> senders (only those inlined) and that seems to work :)
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> You're going to have to delve into the inliner in Slang. This
>>>>>>>>>> is, um, not fun. I liken it to getting hit on the head with a stick by
>>>>>>>>>> your guru, except that no enlightenment results. Good luck.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> :( thanks.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> Thanks in advance,
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> Mariano
>>>>>>>>>>> http://marianopeck.wordpress.com
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> best,
>>>>>>>>>> Eliot
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Mariano
>>>>>>>>> http://marianopeck.wordpress.com
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> best,
>>>>>>>> Eliot
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Mariano
>>>>>>> http://marianopeck.wordpress.com
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Mariano
>>>>>> http://marianopeck.wordpress.com
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> best,
>>>>> Eliot
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Mariano
>>>> http://marianopeck.wordpress.com
>>>>
>>>>
>>>
>>>
>>> --
>>> Mariano
>>> http://marianopeck.wordpress.com
>>>
>>>
>>>
>>
>>
>> --
>> best,
>> Eliot
>>
>>
>>
>
>
> --
> Mariano
> http://marianopeck.wordpress.com
>
>
>
--
best,
Eliot
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/vm-dev/attachments/20111228/6c24c86a/attachment-0001.htm
More information about the Vm-dev
mailing list