Extra PMC allocations with new ['Integer']

[Peter Lobsinger]

On Sun, Feb 6, 2011 at 9:44 AM, Patrick R. Michaud <pmichaud at pobox.com> wrote:
> Bacek++ fixed TT #1994 so that .INTERPINFO_ACTIVE_PMCS works again
> (thanks!); while playing with it I noticed that creating the first
> instance of a PMC type such as Integer or Float results in the
> creation of 38 extra PMCs somewhere.  (Creating the second and
> subsequent instances doesn't seem to result in this extra allocation.)
>
>    pmichaud at orange:~/parrot$ cat activepmc-2.pir
>    .sub 'main'
>        .include 'interpinfo.pasm'
>
>        sweep 1
>        collect
>
>        $I0 = interpinfo .INTERPINFO_ACTIVE_PMCS
>        say $I0
>
>        .local pmc a, b
>        a = new ['Integer']
>        $I0 = interpinfo .INTERPINFO_ACTIVE_PMCS
>        say $I0
>
>        b = new ['Integer']
>        $I0 = interpinfo .INTERPINFO_ACTIVE_PMCS
>        say $I0
>    .end
>
>    pmichaud at orange:~/parrot$ ./parrot activepmc-2.pir
>    1173
>    1212
>    1213
>    pmichaud at orange:~/parrot$
>
> I'm sure that some of the PMCs are being created to support the PMCProxy
> instances.... but *38*?  That number seems surprisingly high, so I
> thought I'd bring it up to the list and see if anyone has any ideas
> about why and where so many extra PMCs are being created.

Confirmed, PMCProxy objects are being created.

It is interesting to note that of those 38, only 12 persist after a GC
run (add the sweep/collect code before the interpinfo query). Of those
12, 8 can be directly accounted for in Class.init() (parent of
PMCProxy). That leaves only 4 persisting PMCs unaccounted for.

So which of these are we concerned about the most? The 26 contributing
to GC churn? The 8 caused by Class being fat? Or the 4 I haven't
bothered to track down?

> The 38 extra PMCs seems to be per-type; for example, if after
> creating the two Integers above I then create a new Float object,
> another 38 PMCs get created somewhere.
>
> Thanks,
>
> Pm
> _______________________________________________
> http://lists.parrot.org/mailman/listinfo/parrot-dev
>