Re: gc question

[Daniel Berlin <dberlin at dberlin dot org>]

> > I have a collector that is based on the zone collector that can do
> > this, because it marks each allocation with the typecode (by having a
> > ton of different ggc_alloc macros).
> >
> > Other than that, you probably can't do this easily.
> I don't mind doing some post-processing, or hacking into the allocator 
> to
> do something like build a hash table of things I care about, 
> invalidating
> them if they're released, etc etc.

Well, okay, in that case, the easiest way to do this is as follows:

1. Change the ggc_alloc_typed stub in ggc-page to record the memory it 
assigns for the things you want to track.
2. Make sure you remove these things when they get freed.
3. Change the varray/phi_node allocations to use ggc_alloc_typed with 
an argument (You can find the enum list in gtype-desc.h in the build 
dir).
4. Do what you want with the results.

It might be easier to do this with ggc-zone (configure --with=gc-zone), 
since it already stores the typecodes with each object allocated using 
ggc_alloc_typed in the object.
So you can tell, when an object is freed, what it's typecode was, 
without having to mess around with hashtables and whatnot.


> Basically I want to accomplish two things.  First, nodes that I know 
> should
> have been collected were actually collected.  Second, I'm seeing some 
> odd
> stuff with varrays which makes me wonder if somewhere, somehow we're
> holding on to some pointers to them and not collecting the varrays.
>
> jeff