[Vm-dev] [COG] primitiveClosureValue assert is failing with
StackVM (possible BUG in bytecodes!!)
stephane ducasse
stephane.ducasse at gmail.com
Wed Dec 22 21:56:36 UTC 2010
Mariano
I really want to thank you for your stubborness (probably not positive in english), I wish I could spend the time you spend on gdb and other and learn as much as you do. Keep pushing.
Stef
On Dec 22, 2010, at 10:34 PM, Eliot Miranda wrote:
> Hi Mariano,
>
> indeed it's a bug. Your fix (objectMemory is: rcvr instanceOf: (self splObj: ClassBlockClosure) compactClassIndex: ClassBlockClosureCompactIndex) is the correct one. There are a few other places where this needs to be done also. I'll roll out new code soon.
>
> best
> Eliot
>
> On Wed, Dec 22, 2010 at 12:03 PM, Mariano Martinez Peck <marianopeck at gmail.com> wrote:
>
> Hi Eliot. If I compile a StackVM, and I just press the left arrow in a code pane of a OB windows, in a pharo image, I get an asser that fails:
> (oop & 1) 19202
>
> Sorry I bother with these asserts, but I am trying to solve the crashes I have and I think starting for the asserts is a good idea.
>
> The assert that fails is in primitiveClosureValue() the part:
>
> blockClosure = longAt(GIV(stackPointer) + (GIV(argumentCount) * BytesPerWord));
> /* begin quickFetchInteger:ofObject: */
> oop = longAt((blockClosure + BaseHeaderSize) + (ClosureNumArgsIndex << ShiftForWord));
> assert((oop & 1));
> numArgs = (oop >> 1);
> if (!(GIV(argumentCount) == numArgs)) {
> /* begin primitiveFail */
> if (GIV(primFailCode) == 0) {
> GIV(primFailCode) = 1;
> }
> return;
> }
>
>
> numArgs results to be a large number (using #printNum: was 718613205), and if I try to do a #printOop: I got a " 0x55aa55aa0x55aa55aa is not on the heap"
>
> Now, of course, the "if (!(GIV(argumentCount) == numArgs)) " fails.
>
> What it is funny is that I did a #printOop: of blockClosure instVar, and this is what I get: 0x1f59b578: a(n) Association
>
> Association? WTF ? so...I guess something weird is happening. I continue debugging it..and I noticed that:
>
> - Several classes were called in addition to Association,like Character. These seems to be compact classes and implement #value and #value: ;)
> - Both, #bytecodePrimValue and #bytecodePrimValueWithArg were not calling #primitiveClosureValue. So, "isBlock" was false...
>
> I tried modifying #bytecodePrimValueWithArg and #bytecodePrimValue this line:
>
> isBlock := objectMemory is: rcvr instanceOf: ClassBlockClosure compactClassIndex: ClassBlockClosureCompactIndex.
>
> to this one:
>
> isBlock := objectMemory is: rcvr instanceOf: (self splObj: ClassBlockClosure) compactClassIndex: ClassBlockClosureCompactIndex.
>
> WIth this change, at least, the assert doesn't fail anymore. I made progress (maybe). But still, for normal closures, "isBlock" seems to be false. So.....all block values are being managed as a normal send ????
>
> Now, I don't understand why ClassBlockClosureCompactIndex is 0. I think this may be the problem. Why it is not 12? (in Pharo Smalltalk compactClassesArray at: 12 ->>> BlockClosure.)
> I notice that ObjectMemory >> initializeCompactClassIndices does:
>
> ClassBlockClosureCompactIndex := 0 "12". "Prospective. Currently TranslatedMethod class"
>
>
> Notice that this ZERO changes the sematic of #is: oop instanceOf: classOop compactClassIndex: compactClassIndex
> called from the #bytecodePrimValue and bytecodePrimValueWithArg
>
> Finally, I changed the mentioned line to this:
>
> isBlock := objectMemory is: rcvr instanceOf: (self splObj: ClassBlockClosure) compactClassIndex: 12.
>
> and finally, with block closures, it is calling the primitive, and for Association it doesn't fail the assert. ahh and the image doesn't crash ;)
>
> I don't know if this is a bug, if my solution was ok, nor it impact. Please let me know.
>
> Cheers
>
> Mariano
>
>
>
>
>
>
More information about the Vm-dev
mailing list