Zephyr support for ESP32

madscientist_42
Posts: 95
Joined: Tue Feb 21, 2017 10:17 pm

Re: Zephyr support for ESP32

Postby madscientist_42 » Mon Jun 19, 2017 6:40 pm

BuddyCasino wrote:Is there anything out there that can compare to Jorge Aparicio's RTFM? Its IMHO the natural design for memory-constrained devices. Shared memory, multitasking, events, hardware assisted scheduling, verifiably safe by virtue of Rust.
Can't be "verifiably" safe in the sense of logic errors. Verifiably safe means you're not touching it for programming purposes. Is it better than some? Perhaps. But they said the **SAME** things about Java, D, and a few others. It's harder to make mistakes- but you can **STILL** make them. You just can't make the dumber f-ups.

BuddyCasino
Posts: 263
Joined: Sun Jun 19, 2016 12:00 am

Re: Zephyr support for ESP32

Postby BuddyCasino » Mon Jun 19, 2017 7:03 pm

madscientist_42 wrote:Can't be "verifiably" safe in the sense of logic errors.
No one in their right mind would claim that. But, you know - I'll take
  • memory safety
  • freedom of data-races
  • guaranteed deadlock free execution
over what C and C++ guarantee, which is pretty much nothing.

madscientist_42
Posts: 95
Joined: Tue Feb 21, 2017 10:17 pm

Re: Zephyr support for ESP32

Postby madscientist_42 » Mon Jun 19, 2017 8:47 pm

BuddyCasino wrote:
madscientist_42 wrote:Can't be "verifiably" safe in the sense of logic errors.
No one in their right mind would claim that. But, you know - I'll take
  • memory safety
  • freedom of data-races
  • guaranteed deadlock free execution
over what C and C++ guarantee, which is pretty much nothing.
Memory safety is something of a luxury as a Systems Programming language.
Same with the Freedom of data-races.

And guaranteed deadlock free execution? THIS claim, which is technically IMPOSSIBLE, is why I don't give Rust or it's adherents much credit.

Therein lies the rub. It's a decent enough language, but the things you claim? Not viable or usable (or even doable) for an OS or much of anything else. At some point you're down in the guts where you don't get those things. Rust won't save you from that layer and you're kidding yourself if you believe it so. 3 decades of people promising the **EXACT SAME THINGS** you're claiming. 3 decades of failure. There's a bit of a hint there, if you contemplate this for a long moment.

You can keep believing and be disappointed...or you can gain the skill needed to do what needs to be done at those lowest levels. I'm there with doing what needs to be done to not have mishaps WITH those languages you deride.

madscientist_42
Posts: 95
Joined: Tue Feb 21, 2017 10:17 pm

Re: Zephyr support for ESP32

Postby madscientist_42 » Mon Jun 19, 2017 8:52 pm

If it's so great, why don't you implement it?

If so, great. It'll showcase that you're right and I'm wrong.

If not, perhaps you should gain a bit of chops programming at that level before you comment on such things. Someone that HAS that is trying to tell you several somethings...and you blew it off.

Who is online

Users browsing this forum: No registered users and 121 guests