Skip to content

Improve clone_from impls for collections - #28602

Merged
bors merged 2 commits into
rust-lang:masterfrom
apasel422:clone_from
Sep 24, 2015
Merged

Improve clone_from impls for collections#28602
bors merged 2 commits into
rust-lang:masterfrom
apasel422:clone_from

Conversation

@apasel422

@apasel422apasel422 commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

@dotdash

dotdash commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

@bors r+

Nice!

@bors

bors commented Sep 23, 2015

Copy link
Copy Markdown
Collaborator

📌 Commit 85be55b has been approved by dotdash

@bors

bors commented Sep 23, 2015

Copy link
Copy Markdown
Collaborator

⌛ Testing commit 85be55b with merge 8de4886...

@bluss

bluss commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

Did you push a new commit after bors tested the old one. What was the change?

bluss
bluss reviewed Sep 23, 2015
View reviewed changes
Comment threadsrc/libcollections/vec.rs Outdated

@blussblussSep 23, 2015

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This trick might warrant a comment.

@apasel422

apasel422 commented Sep 23, 2015

Copy link
Copy Markdown
ContributorAuthor

@bluss Yeah, I pushed a new commit, so bors hasn't tested the new one yet. I removed an unnecessary call to cmp::min, since truncate guarantees that the destination's length is less than or equal to the source's.

I'll add a comment that explains the trick and re-push.

@bluss

bluss commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

Great

@apasel422

apasel422 commented Sep 23, 2015

Copy link
Copy Markdown
ContributorAuthor

Hmm, instead of using the trick, should I just use a more explicit (and less finicky) get_unchecked loop like @Veedrac did in the related issue?

@bluss

bluss commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

It's almost about time we package up this "zip slices" pattern now in a function or iterator actually, since we have to use it in a lot of places.

I think your question is interesting:

  1. Write direct, do what I say, unsafe code (get_unchecked)
  2. Write a particular pattern that we know the optimizer understands

(2) can absolutely be called fragile. I still like (2) better since it's safe, it degrades well (still correct even if the optimizer's behavior changes).

@apasel422

apasel422 commented Sep 23, 2015

Copy link
Copy Markdown
ContributorAuthor

It's actually already present in clone_from_slice, from what I can tell.

@bluss

bluss commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

yes, for clone in particular it seems it is. We can reuse that here, then.

@apasel422

apasel422 commented Sep 23, 2015

Copy link
Copy Markdown
ContributorAuthor

Updated.

@bluss

bluss commented Sep 23, 2015

Copy link
Copy Markdown
Contributor

Now it's beautiful

@bors r+

@bors

bors commented Sep 23, 2015

Copy link
Copy Markdown
Collaborator

📌 Commit 2f1084a has been approved by bluss

@bors

bors commented Sep 24, 2015

Copy link
Copy Markdown
Collaborator

⌛ Testing commit 2f1084a with merge 450f87f...

@bors

bors commented Sep 24, 2015

Copy link
Copy Markdown
Collaborator

💔 Test failed - auto-mac-64-opt

@apasel422
Optimize Vec::clone_from
97f2a32
Before:
test dst_bigger::src_100_dst_1000::clone ... bench: 34 ns/iter (+/- 1)
test dst_bigger::src_100_dst_1000::clone_from ... bench: 75 ns/iter (+/- 3)
test dst_bigger::src_10_dst_100::clone ... bench: 25 ns/iter (+/- 0)
test dst_bigger::src_10_dst_100::clone_from ... bench: 9 ns/iter (+/- 1)
test eq::src_1000_dst_1000::clone ... bench: 105 ns/iter (+/- 2)
test eq::src_1000_dst_1000::clone_from ... bench: 593 ns/iter (+/- 21)
test eq::src_100_dst_100::clone ... bench: 34 ns/iter (+/- 1)
test eq::src_100_dst_100::clone_from ... bench: 75 ns/iter (+/- 1)
test src_bigger::src_1000_dst_100::clone ... bench: 103 ns/iter (+/- 5)
test src_bigger::src_1000_dst_100::clone_from ... bench: 148 ns/iter (+/- 5)
test src_bigger::src_100_dst_10::clone ... bench: 34 ns/iter (+/- 1)
test src_bigger::src_100_dst_10::clone_from ... bench: 20 ns/iter (+/- 0)
After:
test dst_bigger::src_100_dst_1000::clone ... bench: 34 ns/iter (+/- 2)
test dst_bigger::src_100_dst_1000::clone_from ... bench: 15 ns/iter (+/- 1)
test dst_bigger::src_10_dst_100::clone ... bench: 26 ns/iter (+/- 1)
test dst_bigger::src_10_dst_100::clone_from ... bench: 7 ns/iter (+/- 0)
test eq::src_1000_dst_1000::clone ... bench: 103 ns/iter (+/- 1)
test eq::src_1000_dst_1000::clone_from ... bench: 85 ns/iter (+/- 4)
test eq::src_100_dst_100::clone ... bench: 34 ns/iter (+/- 2)
test eq::src_100_dst_100::clone_from ... bench: 15 ns/iter (+/- 1)
test src_bigger::src_1000_dst_100::clone ... bench: 103 ns/iter (+/- 4)
test src_bigger::src_1000_dst_100::clone_from ... bench: 90 ns/iter (+/- 2)
test src_bigger::src_100_dst_10::clone ... bench: 34 ns/iter (+/- 2)
test src_bigger::src_100_dst_10::clone_from ... bench: 20 ns/iter (+/- 0)
Closes#28601.
@apasel422

apasel422 commented Sep 24, 2015

Copy link
Copy Markdown
ContributorAuthor

Fixed the test failure.

@bluss

bluss commented Sep 24, 2015

Copy link
Copy Markdown
Contributor

@bors r+

thanks!

@bors

bors commented Sep 24, 2015

Copy link
Copy Markdown
Collaborator

📌 Commit 97f2a32 has been approved by bluss

bors added a commit that referenced this pull request Sep 24, 2015
@bors
Auto merge of#28602- apasel422:clone_from, r=bluss
355bbfb
r? @bluss
@bors

bors commented Sep 24, 2015

Copy link
Copy Markdown
Collaborator

⌛ Testing commit 97f2a32 with merge 355bbfb...

@bors

bors commented Sep 24, 2015

Copy link
Copy Markdown
Collaborator

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt, auto-win-msvc-32-opt, auto-win-msvc-64-opt

@bors
bors merged commit 97f2a32 into rust-lang:masterSep 24, 2015
@borsbors mentioned this pull request Sep 24, 2015
Closed
@apasel422
apasel422 deleted the clone_from branch September 24, 2015 17:08
@blussbluss added the relnotes Marks issues that should be documented in the release notes of the next release. label Sep 24, 2015
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

@blussbluss

Labels

relnotesMarks issues that should be documented in the release notes of the next release.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@apasel422@dotdash@bors@bluss