RFC: Preventing Retain Cycles (Memory Ownership Model)

ownership

(Joe Groff) #1

Rust attempted this sort of type-directed cycle prevention early in its development, when it was attempting to provide language-supported shared references without GC, and it didn't work out well for them in practice.

-Joe

···

On Jul 29, 2016, at 6:42 PM, Andrew Bennett via swift-evolution <swift-evolution@swift.org> wrote:

I'd like an opt-in way to verify and prevent unintentional strong references in Swift.

This can be used to verify ownership structures, and ultimately avoid retain cycles.

Read a draft proposal here:
https://github.com/therealbnut/swift-evolution/blob/therealbnut-explicit-ownership/proposals/NNNN-explicit-ownership-type-attribute.md

TL;DR:

If you have any questions please read the proposal before asking here.

It's an opt-in attribute that defines a whitelist of types something can own. For example:

@owns(TypeA, TypeB) struct TypeC { ... }

I wrote this a few months ago, but we weren't accepting additive proposals. Now we're explicitly looking for something like this:

Memory ownership model: Adding an (opt-in) Cyclone/Rust inspired memory ownership model to Swift is highly desired by systems programmers and folks who want predictable and deterministic performance (for example, in real time audio processing code). More pertinent to the goals of Swift 4, this feature is important because it fundamentally shapes the ABI. It informs code generation for “inout", how low-level “addressors” work in the ABI, impacts the Swift runtime, and will have a significant impact on the type system and name mangling.

- Chris

----

Here's a link to the version of the proposal when I sent this email.
_______________________________________________
swift-evolution mailing list
swift-evolution@swift.org
https://lists.swift.org/mailman/listinfo/swift-evolution


(Andrew Bennett) #2

Interesting, I'm not familiar with the history of the feature in Rust. Are
you able to provide more details on what went wrong in practice?

This proposal may be different, this is not attempting to do it without GC.
What I'm suggesting still uses ARC (assuming you were counting ARC as GC).
It does provide the compiler many opportunities to optimise ARC though.

I really like what Rust ended up with, but I'm concerned that it's very
complicated, and I want to explore alternatives. From
https://doc.rust-lang.org/book/ownership.html:

However, this system does have a certain cost: learning curve. Many new

···

users to Rust experience something we like to call ‘fighting with the
borrow checker’.

On Sat, Jul 30, 2016 at 1:22 PM, Joe Groff via swift-evolution < swift-evolution@swift.org> wrote:

Rust attempted this sort of type-directed cycle prevention early in its
development, when it was attempting to provide language-supported shared
references without GC, and it didn't work out well for them in practice.

-Joe

> On Jul 29, 2016, at 6:42 PM, Andrew Bennett via swift-evolution < > swift-evolution@swift.org> wrote:
>
> I'd like an opt-in way to verify and prevent unintentional strong
references in Swift.
>
> This can be used to verify ownership structures, and ultimately avoid
retain cycles.
>
> Read a draft proposal here:
>
https://github.com/therealbnut/swift-evolution/blob/therealbnut-explicit-ownership/proposals/NNNN-explicit-ownership-type-attribute.md
>
> TL;DR:
>
> If you have any questions please read the proposal before asking here.
>
> It's an opt-in attribute that defines a whitelist of types something can
own. For example:
>
> @owns(TypeA, TypeB) struct TypeC { ... }
>
> I wrote this a few months ago, but we weren't accepting additive
proposals. Now we're explicitly looking for something like this:
>
> Memory ownership model: Adding an (opt-in) Cyclone/Rust inspired memory
ownership model to Swift is highly desired by systems programmers and folks
who want predictable and deterministic performance (for example, in real
time audio processing code). More pertinent to the goals of Swift 4, this
feature is important because it fundamentally shapes the ABI. It informs
code generation for “inout", how low-level “addressors” work in the ABI,
impacts the Swift runtime, and will have a significant impact on the type
system and name mangling.
>
> - Chris
>
> ----
>
> Here's a link to the version of the proposal when I sent this email.
> _______________________________________________
> swift-evolution mailing list
> swift-evolution@swift.org
> https://lists.swift.org/mailman/listinfo/swift-evolution

_______________________________________________
swift-evolution mailing list
swift-evolution@swift.org
https://lists.swift.org/mailman/listinfo/swift-evolution