Re: CCL on Big Sur and Apple M1 chips

Hi Dave,

Yes. Myself and a bright summer intern did significant “advanced scouting” (analysis and planning) of the CCL runtime system and compiler output for the ARMv8-A. We wrote straw man examples of compiled code and ported a small portion of the runtime, enough to get these examples working both in simulation and on Raspberry Pi 400 hardware.

We have a status report to share upon request.

I was able to raise $3,000 to pay my intern. I didn’t draw a salary for the 80+ hours I’ve put into the project so far.

The estimate I heard was that it would take Gary Byers six months full-time to do the port, but he isn’t available. Theoretically, it would take me 20-30% longer (if I didn’t have a job that I would be foolish to quit). Matt Emerson is similarly not available. Gary Palter is available, but he would need to get paid for his time on an ongoing basis. He won’t do this “on spec.” Also I don’t know if he could devote full time to the project.

My plan is predicated on assistance from someone like Gary Palter who is a Mac OS expert. The project would need to get further along before we can open it up to a larger team. If I can, I would gladly rehire my intern next summer.

I estimate the cost of the project at $50-70k. That may be optimistic, as it doesn’t include a QA budget or allow for contingencies. There is a test suite, but it wasn’t designed for computer development. The compiler code is not designed for easy porting, but our proposed runtime architecture is similar to the x86-64. I would strongly recommend not rewriting the GC. That would add significantly to the cost.

The big question is: Where is the money going to come from? We have had funding offers, but no one was willing to take on the risk that the project might not get completed. Many of us love CCL, but the project does not have huge commercial value.

What is missing is, in movie industry terms, a “completion bond.”

I am happy to help interview potential developers. Let’s take further.

--Tim

On Oct 12, 2022, at 21:12, Dave Cooper <[email protected]> wrote:



Hi,

Has anyone been discussing or trying to coordinate a CCL apple silicon port lately?

I'm supposed to assemble names of any willing devs for the project, and maybe some ballpark cost estimates.

Please let me know if you know any. Also let me know if any other efforts are afoot.

Thanks,

Dave Cooper

Board Member, Common Lisp Foundation

On Wed, Dec 16, 2020 at 4:25 AM Michał "phoe" Herda <[email protected] > wrote:

SBCL went with https://fundrazr.com/61kPgb as their platform of choice

when creating the SBCL-porting fundraiser. It seems that they pass 100%

of the money to the fundraiser owner and make money from tips that are

suggested but optional.

~phoe

On 16.12.2020 10:17, Chris Hanson wrote:

> On Nov 18, 2020, at 11:47 PM, Ron Garret <[email protected] > wrote:

>> PSA: I am willing to fund a significant portion of the work to port CCL to run natively on Apple silicon. But not 100%. The las time I made this offer, that constraint turned out to be a show-stopper so I thought I’d cast a wider net before giving up.

> I concur with those interested in crowdfunding the Apple Silicon port. I’d be happy to kick in some amount, ideally via a mechanism that lets most or all of the contribution go to Clozure.

>

> — Chris

>

> _______________________________________________

> Openmcl-devel mailing list

> [email protected]

> https://lists.clozure.com/mailman/listinfo/openmcl-devel

_______________________________________________

Openmcl-devel mailing list

[email protected]

https://lists.clozure.com/mailman/listinfo/openmcl-devel

--

My Best,

Dave Cooper, [email protected]
genworks.com , gendl.org
+1 248-330-2979