RValue DIP Draft.

Oct 19, 2021

12345swordy

Oct 19, 2021

Mike Parker

Oct 19, 2021

Oct 19, 2021

Oct 19, 2021

Oct 19, 2021

Oct 19, 2021

Oct 19, 2021

On Tuesday, 19 October 2021 at 00:28:47 UTC, 12345swordy wrote:

Any issues that you have regarding the DIP can be posted here for the other 3 to look at and discuss with them.

Until then stop spamming the PR with commits please.

Link: https://github.com/dlang/DIPs/pull/218

I've left a comment in that thread. But I'll repeat the gist of it here.

As described in the review documentation, the Draft Review began the moment you submitted the DIP. Feedback on its current state is not spam. That is where comments on the DIP belong, not in this thread. The intent of Draft Review is to fix any obvious flaws that jump out at the people who care to participate.

You may not have been the original author of the DIP, but having submitted it, you are responsible for any revisions going forward. You're the Point of Contact, as described in [the authoring documentation][2]. That's not to say that you need to make actually make any changes yourself. If you don't intend to, then it's on you to coordinate with the original author(s) to address feedback, either by making changes or responding to review comments.

[2] https://github.com/dlang/DIPs/blob/master/docs/process-authoring.md#step-1---determine-the-point-of-contact

I support this DIP. However: > For objects larger than 1-2 registers in size, passing by pointer may be more efficent than passing by value I don't buy this Really, there are two issues being brought up: 1. Copying large objects is slow 2. Copy constructors may be slow 1 is exclusively an ABI issue, and should be solved at the ABI level (cf https://elronnd.net/writ/boring/2021-05-07_abi-wrong.html). Regarding 2, consider that this DIP applies to rvalues. That is, it applies to expressions like: some_func(SomeStructWithAnExpensiveCopy()) The proper fix for this is to turn the last use of a struct into a move instead of a copy. (I believe there was a DIP proposing this.)

October 19, 2021

Re: RValue DIP Draft.

Posted by Tejas
in reply to Elronnd

Permalink

Tejas

Posted in reply to Elronnd

Permalink

On Tuesday, 19 October 2021 at 03:53:45 UTC, Elronnd wrote:

I support this DIP. However:

For objects larger than 1-2 registers in size, passing by pointer may be more efficent than passing by value

I don't buy this

Really, there are two issues being brought up:

Copying large objects is slow
Copy constructors may be slow

1 is exclusively an ABI issue, and should be solved at the ABI level (cf https://elronnd.net/writ/boring/2021-05-07_abi-wrong.html).

Even preview=in subscribes to this philosophy and implements the same semantics, so idk if anyone will ever implement changes at ABI level.

Regarding 2, consider that this DIP applies to rvalues. That is, it applies to expressions like:

some_func(SomeStructWithAnExpensiveCopy())

The proper fix for this is to turn the last use of a struct into a move instead of a copy. (I believe there was a DIP proposing this.)

It was DIP 1040 that was introducing move constructors, moveAssign (and possiby rvalue references), but it's currently not being worked on :(

Forums