On Tue, Jul 30, 2013 at 5:36 AM, Ian P. Cooke <[email protected]> wrote:

> I don't have time to read through the MPS docs now but this seemed
relevant to some of the discussion here.
>
> http://www.ravenbrook.com/project/mps/

This page has some details about their collection..

  http://www.ravenbrook.com/project/mps/master/design/collection

At a quick skim, it looks like they use a full-cost barrier to support
no-world-stop. (they claim to only stop for the root-set snapshot) They
support Appel-Ellis-Li hardware accelerated write barrier.

http://www.hpl.hp.com/techreports/Compaq-DEC/SRC-RR-25.pdf

While digging up references I ran across this paper I hadn't seen before..
It's promising as he worked at Azul and now works at Google.

http://www.cs.man.ac.uk/~irogers/Reducing_and_Eliding_Read_Barriers.pdf

I don't have time to read through the MPS docs now but this seemed relevant
to some of the discussion here.

http://www.ravenbrook.com/project/mps/

[apologies to those others that subscribe to both lists for the cross-post]

Begin forwarded message:

*From: *"I.T. Daniher" <[email protected]>
*Subject: **Re: [rust-dev] MPS investigation for Rust GC*
*Date: *July 29, 2013 10:33:37 PM CDT
*To: *Kevin Cantu <[email protected]>
*Cc: *"[email protected]" <[email protected]>

If the licensing terms aren't suitable for you (for example, you're
developing a closed-source commercial product or a *compiler run-time*
*system*) you can easily license the MPS under different terms from
Ravenbrook.  Please write to us at `<*[email protected]*>`_
for more information.

from https://github.com/Ravenbrook/mps-temporary/blob/master/license.txt

If there's serious interest in this, we could reach out. I don't know the
state of the gc library project.

On Mon, Jul 29, 2013 at 10:59 PM, Kevin Cantu <[email protected]> wrote:

> I'm personally not very familiar with the Sleepycat license (as MPS seems
> to use: https://github.com/Ravenbrook/mps-temporary).
>  Would proprietary code have to be shipped in separate binaries than the
> Rust runtime, or would commercial licensing be required?
>
>
> Kevin
>
>
>
>
> --
> Kevin Cantu
>
>
> On Mon, Jul 29, 2013 at 10:41 AM, Thad Guidry <[email protected]>wrote:
>
>> REGARDING:
>> - Graydon wants to investigate using the Memory Pool System as the Rust
>> GC,
>>   rather than a bespoke one. The
>> [MPS](http://www.ravenbrook.com/project/mps/) is
>>   a very mature and robust memory management library.
>>
>> Reading through and seeing one of the MPS creators initial design goals :
>>
>> "The MPS was not designed for C++, but as a memory manager for dynamic
>> language run-time systems. In fact, it was specifically designed not to
>> require C++."
>>
>> ...instantly gained my respect for this system and its potential.
>>
>> --
>> -Thad
>> Thad on Freebase.com <http://www.freebase.com/view/en/thad_guidry>
>> Thad on LinkedIn <http://www.linkedin.com/in/thadguidry/>
>>
>> _______________________________________________
>> Rust-dev mailing list
>> [email protected]
>> https://mail.mozilla.org/listinfo/rust-dev
>>
>>
>
> _______________________________________________
> Rust-dev mailing list
> [email protected]
> https://mail.mozilla.org/listinfo/rust-dev
>
>
_______________________________________________
Rust-dev mailing list
[email protected]
https://mail.mozilla.org/listinfo/rust-dev



_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to