Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Help LLVM better optimize slice::Iter(Mut)::len #61885

Merged
merged 1 commit into from
Jun 18, 2019
Merged

Help LLVM better optimize slice::Iter(Mut)::len #61885

merged 1 commit into from
Jun 18, 2019

Conversation

scottmcm
Copy link
Member

r? @RalfJung

I've included a codegen test that fails without this change as a demonstration of usefulness.

@rust-highfive rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jun 16, 2019
@rust-highfive

This comment has been minimized.

Sign in to view
@scottmcm scottmcm force-pushed the slice-iter-len-opt branch from be3ca63 to af0e35e Compare June 16, 2019 04:15
@hanna-kruppe
Copy link
Contributor

Nice! r=me unless you want more eyes on this first.

@RalfJung
Copy link
Member

Yes this looks great!

Do we have tests in src/liballoc that exhibit this? Would be nice to make sure Miri runs this code in a couple of situations, just to be sure.

@RalfJung
Copy link
Member

Oh also, should we see if this has any perf influence?

@RalfJung
Copy link
Member

Well let's get a try build.

@bors try

@bors
Copy link
Contributor

bors commented Jun 16, 2019

⌛ Trying commit af0e35e with merge 9dd34620937c9854a002172599a6217be05f6520...

@bors
Copy link
Contributor

bors commented Jun 16, 2019

☀️ Try build successful - checks-travis
Build commit: 9dd34620937c9854a002172599a6217be05f6520

@RalfJung
Copy link
Member

@rust-timer build 9dd34620937c9854a002172599a6217be05f6520

@rust-timer
Copy link
Collaborator

Success: Queued 9dd34620937c9854a002172599a6217be05f6520 with parent 6865502, comparison URL.

@rust-timer
Copy link
Collaborator

Finished benchmarking try commit 9dd34620937c9854a002172599a6217be05f6520, comparison URL.

@scottmcm
Copy link
Member Author

scottmcm commented Jun 17, 2019 •
edited
Loading

Looks like just noise in the perf run, which doesn't surprise me. (Anyone using is_empty is unaffected by this, since it's overridden separately, and it's a tiny change anyway.)

Would be nice to make sure Miri runs this code in a couple of situations, just to be sure.

Hmm, looks like I get an

error[E0080]: Miri evaluation error: unimplemented intrinsic: unchecked_sub

on the playground. I guess I should make a MIRI PR before this goes in?

Edit: I should have looked, seems like you're already on it rust-lang/miri#776 ❤️

@RalfJung
Copy link
Member

I can't find any explicit tests for the slice iterator length method, but running miri-test-libstd on this PR shows that the code path does get hit, e.g. through slice::to_vec.

@RalfJung
Copy link
Member

Ah, and this macro also gets used in nth and position and try_fold and more, so I think we are good in terms of coverage.

@bors r=rkruppe,RalfJung

@bors
Copy link
Contributor

bors commented Jun 17, 2019

📌 Commit af0e35e has been approved by rkruppe,RalfJung

@bors bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jun 17, 2019
@RalfJung RalfJung mentioned this pull request Jun 17, 2019
Merged
Centril added a commit to Centril/rust that referenced this pull request Jun 17, 2019
@Centril
Rollup merge of rust-lang#61885 - scottmcm:slice-iter-len-opt, r=rkru…
2d3d0b7 
…ppe,RalfJung

Help LLVM better optimize slice::Iter(Mut)::len

r? @RalfJung

I've included a codegen test that fails without this change as a demonstration of usefulness.
@Centril Centril mentioned this pull request Jun 17, 2019
Merged
bors added a commit that referenced this pull request Jun 17, 2019
@bors
Auto merge of #61915 - Centril:rollup-oire3i8, r=Centril
9f22ddf 
Rollup of 5 pull requests

Successful merges:

 - #61702 (test more variants of enum-int-casting)
 - #61836 (Replace some uses of NodeId with HirId)
 - #61885 (Help LLVM better optimize slice::Iter(Mut)::len)
 - #61893 (make `Weak::ptr_eq`s into methods)
 - #61908 (don't ICE on large files)

Failed merges:

r? @ghost
@bors bors merged commit af0e35e into rust-lang:master Jun 18, 2019
@scottmcm scottmcm deleted the slice-iter-len-opt branch June 18, 2019 03:47
@RalfJung RalfJung mentioned this pull request Jun 18, 2019
Merged
bors added a commit that referenced this pull request Jun 18, 2019
@bors
Auto merge of #61927 - RalfJung:miri, r=oli-obk
44fb88d 
update miri

Tests fail since #61885, this should fix that.

r? @oli-obk
@scottmcm scottmcm mentioned this pull request Aug 24, 2021
Closed
@scottmcm scottmcm mentioned this pull request Jan 21, 2022
Closed
@the8472 the8472 mentioned this pull request Jan 3, 2023
Merged
bors added a commit to rust-lang-ci/rust that referenced this pull request Jan 15, 2023
@bors
Auto merge of rust-lang#106393 - the8472:use-ptr-sub, r=scottmcm
9e75ddd 
Simplify manual ptr arithmetic in slice::Iter with ptr_sub

The old code was introduced in rust-lang#61885, which predates the ptr_sub method and underlying intrinsic. The codegen test still passes.

r? `@scottmcm`
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@RalfJung RalfJung RalfJung left review comments

@vitiral vitiral vitiral left review comments

Assignees

@RalfJung RalfJung

Labels
S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion.
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
7 participants
@scottmcm @rust-highfive @hanna-kruppe @RalfJung @bors @rust-timer @vitiral