News for hackers. Stuff that matters.
Re:Ruby's Proposed STM by rurban on Friday October 30, 2020 @08:35AM (#24941567)
Not only bets are off in non-functional language, it's the wrong solution.

STM puts the cost of locking into everything, ie slower default operations, whilst still being concurrency unsafe. Deadlocks can occur.

There are proven good solutions for fast and safe concurrency with ownership tracking and a good compiler/scheduler. No, not Rust. And not STM. We fought with this for years.

Re:Ruby's Proposed STM by rrss on Friday October 30, 2020 @08:45AM (#24941636)
> There are proven good solutions for fast and safe concurrency with ownership tracking and a good compiler/scheduler.

Do you have any pointers to these solutions for the uninitiated?

Re:Ruby's Proposed STM by rurban on Friday October 30, 2020 @04:35PM (#24946630)
All concurrency-safe languages. Eg pony, concurrent pascal, the parrot vm, ...
Re:Ruby's Proposed STM by amelius on Friday October 30, 2020 @11:19AM (#24943154)
Eh, STM is often used as a solution to avoid locking.

https://en.wikipedia.org/wiki/Software_transactional_memory

Re:Ruby's Proposed STM by brandmeyer on Friday October 30, 2020 @12:22PM (#24943843)
If you read into the implementation papers a little more deeply, you'll find that they are typically implemented with locks under the covers.