Re: [Texmacs-dev] Mandrake (and thought about typing and ref counting)

[Stéphane Payrard]

A post from perlmonks by Dan Sugalski, the parrot architect. 
I think it is relevant to this thread.  It compares the
respective merits of real garbage collecting and reference counting
and motivates the use of a real garbage collecting scheme in parrot.

Perl5 is using reference counting.
Parrot will be the run-time of perl6 (probably among other languages).



  Reference counting has a number of problems: 

  Circular garbage leaks 
  It's horribly error-prone 
  It requires a lot of code 
  It's slow 
  It blows your cache

  To take those in more detail: 

  1 It's not tough to get circular structures at the user (i.e. the
  application programmer) level. This means that either a program leaks,
  or each application that potentially has to generate circular
  structures has to kill them.

  2 For refcounting to work right, every increment must have a matching
  decrement. Historically (which is to say, in every release of perl
  I've seen) this doesn't happen. Sometimes its only a few variables,
  sometimes it's a lot of variables. Either way, there's leakage. It
  also means that every extension writer must get refcounting right. For
  anything other than trivial extensions, they don't.

  3 To support refcounting, each and every time a variable is inserted
  into a container or removed from a container the refcount must be
  changed. This is a lot of code. Go browse through the perl 5 source
  (And don't forget about extensions) to see how often this
  happens. This is essentially makework code--it does almost nothing
  useful, and is entirely bookkeeping.

  4 Reference counting is the slowest form of garbage collection for all
  but the most extremely degenerate code. Reference counting requires
  you to spend time touching each variable at least twice. This is all
  your variables, not just the ones that are alive. Tracing garbage
  collectors (which include the mark & sweep, compacting, and
  generational style collectors) generally only touch live data. Yes,
  it's only a quick increment or decrement, but it's a lot of
  them. Refcount schemes usually spend about twice as much time doing
  garbage collection work than tracing collectors.

  5 Refcounting makes the internal data structures and code less dense,
  and that means that your program spends more time waiting on main
  memory. (Fetching data from RAM can cost up to 200 cycles on some
  systems) Your code is less dense, since you have refcount twiddling
  code all through it. Your data is also less dense, since you need to
  have space for refcounts in it. And even worse, it means more of your
  data structure has to be touched when its dealt with. This screws your
  L1 cache. Generally data comes into your CPU's L1 cache in small
  chunks--8 or 16 bytes. You want to use as few of these chunks as you
  can since, while they're fastest to access (usually a few cycles at
  most) there's not much of it. And loading a cache line can cost 10 or
  20 cycles if the data you're looking for is in your L2 cache. (And 200
  or more if it isn't) The less memory that needs to be touched to do
  something means the more efficiently the cache is used and the faster
  your program runs.

  Tracing collectors, which is what parrot's using, generally have
  different characteristics.

  No mainline code does anything with refcounts or other 'liveness'
  bookkeeping data. That means the mainline code is denser, with more of
  it actually dedicated to executing your program.  Internal data
  structures are either smaller (if the GC bookkeeping fields are kept
  out of band) or at least use less cache (since while still in the
  structure they aren't touched and thus fetched into L1 cache).
  There's less code dedicated to GC itself and, while a tracing garbage
  collector isn't exactly trivial, it's not a big deal either. More to
  the point, the code required to do the GC is collected together in one
  single spot, rather than sprinkled througout the entire codebase. (It
  took me only a few hours to write the first cut of Parrot's GC)
  Tracing collectors take less time to execute. Since they make more
  efficient use of the cache (there's only a small chunk of code doing
  GC, so it gets pinned in L1 cache while it runs, and it generally
  restricts itself to a smallish chunk of memory when it runs so it
  gains some cache locality that way) you waste less time on the memory
  bus, and since tracing GCs only look at the live data, their runtime's
  proportional to the amount of live data you have, generally a much
  smaller number than the number of data objects you've allocated since
  the last GC run.

  So, tracing collectors are generally faster, less error-prone, have
  less code, and put less burden on the people writing the core and
  extensions in C. The one (and only) advantage that refcounting has is
  deterministic destruction. (Assuming no bugs, of course) Other than
  that it's a massive pain to deal with for the internals folks.

  All of that is why we don't like refcounts. :) 

  BTW: Your LISP book's numbers may well be really, really old. Much of
  GC's bad reputation for pauses come from benchmarks run 20 or more
  years ago, on VAX 11/780 machines with 2M of RAM and an RA60 as a swap
  disk. These days things run rather a lot faster. A quick run of one of
  parrot's benchmarks shows it did 43K GC runs in 22 seconds on my
  machine, in addition to the sample program's real work. That's not
  bad... (Though, yes, it does argue that the sample program needs more
  work to generate less garbage)


-
 stef