Using concurrency in GC usually implies one of
- Parallel GC. Stop the mutator(s) and use multiple processors to do GC; when
they are done, start the mutators. This isn't real-time, because there's an
unbounded pause for GC. This is what Simon and Roshan's parallel
Hi
> I am curious, because I have a project in mind that would benefit
> greatly from real-time, parallel garbage collection :)
Is Haskell real-time? Doesn't lazy evaluation rather destroy lots of
the real-time properties that you might like in a language. If you
want a real-time functional langu
Bryan O'Sullivan wrote:
Ravi Nanavati wrote:
It isn't an immediately pressing need a Bluespec, but we do get asked
every once in a while about parallelizing the Bluespec compiler.
Roshan James was working on parallel GC at MSR Cambridge last year.
Unfortunately, it's no
Hello,
Is real-time, parallel garbage collection at all feasible?
My thinking is, real-time garbage collection requires the garbage
collector to be able to work on the problem in small, predictable,
pieces. That seems like something which would also be useful for
scaling up GC to multiple cores?
Ravi Nanavati wrote:
It isn't an immediately pressing need a Bluespec, but we do get asked
every once in a while about parallelizing the Bluespec compiler.
Roshan James was working on parallel GC at MSR Cambridge last year.
Unfortunately, it's not an internship-sized project.
I seem to recall reading in a GHC status report that there was some
work going on with parallel garbage-collection. Looking at 6.8.1, it
seems pretty clear that it didn't make it. Is there any update about
where things are on that front?
It isn't an immediately pressing need a Bluespec, but we do
Sounds great, thanks!
-Chad
On 10/16/06, Simon Marlow <[EMAIL PROTECTED]> wrote:
GHC 6.6 does run code in parallel on SMP hardware, but the GC is still single
threaded. We have a prototype parallel GC, and I'm planning to work on it some
more in the next few months... you'll
have come up
with parallelization, or whether something is still in the works.
Excitedly awaiting GHC's SMP GC,
GHC 6.6 does run code in parallel on SMP hardware, but the GC is still single
threaded. We have a prototype parallel GC, and I'm planning to work on it some
more in
IIRC, there was some work being done this past summer on parallelizing
GC for SMP GHC (I think that's my current record for most acronyms in
one sentence). GHC 6.6 is reported as having single-threaded GC, and
I'm wondering if an unforeseen technical problem might have come up
with parallelization
