Update jemalloc to v5.3 #96790

lqd · 2022-05-06T21:35:27Z

Now that jemalloc version 5.3 has been released, this PR updates tikv-jemalloc-sys to the corresponding release.

The crates.io publishing issue seems to have been resolved for the jemalloc-sys package, and version 5.3.0 is now also available under the historical name (and should become the preferred crate to be used). Therefore, this PR also switches back to using jemalloc-sys instead of tikv-jemalloc-sys.

lqd · 2022-05-06T21:37:03Z

@bors try @rust-timer queue

rust-timer · 2022-05-06T21:37:04Z

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

bors · 2022-05-06T21:37:15Z

⌛ Trying commit 4a3f0b89c6d0733a6e84288e3f9aafdf8429875e with merge 7c1ccf45f83f6c96b0fa945d58b116bea4b726ad...

bors · 2022-05-06T22:49:59Z

💔 Test failed - checks-actions

lqd · 2022-05-06T23:03:14Z

@bors try @rust-timer queue

rust-timer · 2022-05-06T23:03:16Z

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

bors · 2022-05-06T23:03:22Z

⌛ Trying commit e0907118f66e205a81a91c7e565ccd847e0ec58f with merge c236f4668dec3cbb6ae05f72d8339c85657b43ff...

bors · 2022-05-07T00:30:38Z

☀️ Try build successful - checks-actions
Build commit: c236f4668dec3cbb6ae05f72d8339c85657b43ff (c236f4668dec3cbb6ae05f72d8339c85657b43ff)

rust-timer · 2022-05-07T00:30:39Z

Queued c236f4668dec3cbb6ae05f72d8339c85657b43ff with parent 77652b9, future comparison URL.

rust-timer · 2022-05-07T01:47:52Z

Finished benchmarking commit (c236f4668dec3cbb6ae05f72d8339c85657b43ff): comparison url.

Summary:

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: mixed results

	Regressions 😿 (primary)	Regressions 😿 (secondary)	Improvements 🎉 (primary)	Improvements 🎉 (secondary)	All 😿 🎉 (primary)
count¹	0	1	219	173	219
mean²	N/A	0.3%	-1.0%	-1.1%	-1.0%
max	N/A	0.3%	-6.3%	-3.3%	-6.3%

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf -perf-regression

number of relevant changes ↩
the arithmetic mean of the percent change ↩

lqd · 2022-05-11T19:35:45Z

@bors try @rust-timer queue

rust-timer · 2022-05-11T19:35:46Z

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

bors · 2022-05-11T19:35:54Z

⌛ Trying commit e61073ef87eed391e23ae16ba5ff037031730411 with merge 41e4c1342cf9326b60d0500f5890059a3d4583fc...

bors · 2022-05-11T21:07:14Z

☀️ Try build successful - checks-actions
Build commit: 41e4c1342cf9326b60d0500f5890059a3d4583fc (41e4c1342cf9326b60d0500f5890059a3d4583fc)

rust-timer · 2022-05-11T21:07:15Z

Queued 41e4c1342cf9326b60d0500f5890059a3d4583fc with parent 08b4f1b, future comparison URL.

rust-timer · 2022-05-11T22:25:27Z

Finished benchmarking commit (41e4c1342cf9326b60d0500f5890059a3d4583fc): comparison url.

Summary:

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: 🎉 relevant improvements found

	Regressions 😿 (primary)	Regressions 😿 (secondary)	Improvements 🎉 (primary)	Improvements 🎉 (secondary)	All 😿 🎉 (primary)
count¹	0	2	218	187	218
mean²	N/A	0.2%	-0.9%	-1.1%	-0.9%
max	N/A	0.2%	-6.3%	-3.3%	-6.3%

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf -perf-regression

number of relevant changes ↩
the arithmetic mean of the percent change ↩

bors · 2022-05-25T06:49:50Z

⌛ Trying commit adab135 with merge 9ca7ce57dc5f41875f4d4bf4a239170463384666...

lqd · 2022-05-25T07:47:45Z

The perf.rlo queue has a few PRs already, so final perf results should be available in a while.

After this last sanity check, r? @Mark-Simulacrum maybe ? (feel free to re-roll if you don't have time)

nnethercote · 2022-05-25T08:10:11Z

So jemallocator is effectively an alias for tikv-jemallocator now, is that right?

nnethercote · 2022-05-25T08:11:17Z

@bors delegate=lqd

bors · 2022-05-25T08:11:20Z

✌️ @lqd can now approve this pull request

bors · 2022-05-25T08:17:51Z

☀️ Try build successful - checks-actions
Build commit: 9ca7ce57dc5f41875f4d4bf4a239170463384666 (9ca7ce57dc5f41875f4d4bf4a239170463384666)

rust-timer · 2022-05-25T08:17:52Z

Queued 9ca7ce57dc5f41875f4d4bf4a239170463384666 with parent 9fadabc, future comparison URL.

lqd · 2022-05-25T08:37:49Z

@nnethercote The situation is a bit unclear to me: the owners of the tikv-jemalloc* forked crates seem to have been granted publishing rights to some of the original packages now, but apparently not all of the jemallocator ones (I've gleaned this understanding from tikv/jemallocator#26). Maybe that's about jemalloc-ctl or jemallocator-global.

It seems that depending on which of these jemalloc crates one was using, you could now switch back to jemallocator (much like you can switch back to jemalloc-sys if you're only using that).

I'm under the impression that in the most common cases, the end result indeed behaves as an alias. And otherwise in the rarer cases where you depend on more than these main crates, you would see that it's the same code published under 2 different names, with the satellite crates likely depending on the tikv-jemalloc* ones only.

rust-timer · 2022-05-25T10:11:54Z

Finished benchmarking commit (9ca7ce57dc5f41875f4d4bf4a239170463384666): comparison url.

Instruction count

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: mixed results

	mean¹	max	count²
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	0.2%	0.2%	1
Improvements 🎉 (primary)	-0.9%	-6.1%	214
Improvements 🎉 (secondary)	-1.1%	-3.4%	188
All 😿🎉 (primary)	-0.9%	-6.1%	214

Max RSS (memory usage)

Results

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: mixed results

	mean¹	max	count²
Regressions 😿 (primary)	1.0%	1.2%	4
Regressions 😿 (secondary)	1.9%	3.9%	8
Improvements 🎉 (primary)	-2.7%	-5.6%	48
Improvements 🎉 (secondary)	-7.7%	-21.6%	60
All 😿🎉 (primary)	-2.4%	-5.6%	52

Cycles

Results

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: 🎉 relevant improvements found

	mean¹	max	count²
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	N/A	N/A	0
Improvements 🎉 (primary)	-2.4%	-3.4%	4
Improvements 🎉 (secondary)	-2.7%	-3.7%	8
All 😿🎉 (primary)	-2.4%	-3.4%	4

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf -perf-regression

the arithmetic mean of the percent change ↩ ↩² ↩³
number of relevant changes ↩ ↩² ↩³

nnethercote · 2022-05-25T10:31:10Z

Very nice improvements on max-rss.

lqd · 2022-05-25T12:31:34Z

I'll let mark weigh in on whether we need to do some additional checks when updating such important dependencies, that's why I'm not going to r=you immediately, to give him some time to do so.

Mark-Simulacrum · 2022-05-27T17:48:28Z

We're pretty early in the release cycle and I think this looks like a good win without major regressions, so I think we can move forward.

@bors r+ rollup=never

bors · 2022-05-27T17:48:30Z

📌 Commit adab135 has been approved by Mark-Simulacrum

bors · 2022-05-27T18:28:20Z

⌛ Testing commit adab135 with merge ebbcbfc...

bors · 2022-05-27T20:57:49Z

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing ebbcbfc to master...

rust-timer · 2022-05-27T23:14:47Z

Finished benchmarking commit (ebbcbfc): comparison url.

Instruction count

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: mixed results

	mean¹	max	count²
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	0.3%	0.3%	1
Improvements 🎉 (primary)	-0.9%	-6.2%	212
Improvements 🎉 (secondary)	-1.1%	-3.2%	191
All 😿🎉 (primary)	-0.9%	-6.2%	212

Max RSS (memory usage)

Results

Primary benchmarks: 🎉 relevant improvements found
Secondary benchmarks: mixed results

	mean¹	max	count²
Regressions 😿 (primary)	1.2%	1.2%	2
Regressions 😿 (secondary)	2.4%	4.3%	11
Improvements 🎉 (primary)	-2.6%	-5.5%	48
Improvements 🎉 (secondary)	-7.7%	-20.7%	53
All 😿🎉 (primary)	-2.5%	-5.5%	50

Cycles

Results

Primary benchmarks: no relevant changes found
Secondary benchmarks: mixed results

	mean¹	max	count²
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	4.5%	4.5%	1
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	-2.6%	-3.0%	3
All 😿🎉 (primary)	N/A	N/A	0

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

@rustbot label: -perf-regression

the arithmetic mean of the percent change ↩ ↩² ↩³
number of relevant changes ↩ ↩² ↩³

switch `jemalloc-sys` back to `tikv-jemalloc-sys`, and update to 0.6.0 Some context: - we used to use jemalloc bindings from https://github.com/gnzlbg/jemallocator, since rust-lang#55238 - that crate was abandoned, picked up as a fork in https://github.com/tikv/jemallocator, so we switched to that in rust-lang#83152. - then they were able to publish to the original `jemalloc-sys` bindings crate, and `jemalloc-sys` and `tikv-jemalloc-sys` became the same thing -- so I switched back to the OG crate in rust-lang#96790 - they're now having publishing problems again: I've been waiting for tikv/jemallocator#96 for the `jemalloc-sys` 0.6.0 update for a few months, but `tikv-jemalloc-sys` is already updated to 0.6.0. A perf run showed some improvements, so this PR switches back to `tikv-jemalloc-sys` to update to 0.6.0. r? ghost

switch `jemalloc-sys` back to `tikv-jemalloc-sys`, and update to 0.6.0 Some context: - we used to use jemalloc bindings from https://github.com/gnzlbg/jemallocator, since rust-lang#55238 - that crate was abandoned, picked up as a fork in https://github.com/tikv/jemallocator, so we switched to that in rust-lang#83152. - then they were able to publish to the original `jemalloc-sys` bindings crate, and `jemalloc-sys` and `tikv-jemalloc-sys` became the same thing -- so I switched back to the OG crate in rust-lang#96790 - they're now having publishing problems again: I've been waiting for tikv/jemallocator#96 for the `jemalloc-sys` 0.6.0 update for a few months, but `tikv-jemalloc-sys` is already updated to 0.6.0. A perf run showed some improvements, so this PR switches back to `tikv-jemalloc-sys` to update to 0.6.0.

rustbot added the T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. label May 6, 2022

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label May 6, 2022

bors added the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label May 6, 2022

This comment has been minimized.

Sign in to view

lqd force-pushed the update_jemalloc branch from 4a3f0b8 to e090711 Compare May 6, 2022 23:02

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels May 7, 2022

lqd mentioned this pull request May 7, 2022

Remove useless #[global_allocator] from rustc and rustdoc. #92222

Merged

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label May 11, 2022

lqd changed the title ~~Test jemalloc update to 5.3RC~~ Test jemalloc update to 5.3 May 11, 2022

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label May 11, 2022

lqd marked this pull request as ready for review May 25, 2022 07:45

rust-highfive assigned Mark-Simulacrum May 25, 2022

bors added the merged-by-bors This PR was explicitly merged by bors. label May 27, 2022

bors merged commit ebbcbfc into rust-lang:master May 27, 2022

rustbot added this to the 1.63.0 milestone May 27, 2022

bors mentioned this pull request May 27, 2022

Add WIP stable MIR crate #97385

Merged

lqd deleted the update_jemalloc branch May 28, 2022 06:14

lqd mentioned this pull request Dec 3, 2024

switch jemalloc-sys back to tikv-jemalloc-sys, and update to 0.6.0 #133792

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update jemalloc to v5.3 #96790

Update jemalloc to v5.3 #96790

lqd commented May 6, 2022 •

edited

Loading

lqd commented May 6, 2022

rust-timer commented May 6, 2022

bors commented May 6, 2022

bors commented May 6, 2022

This comment has been minimized.

lqd commented May 6, 2022

rust-timer commented May 6, 2022

bors commented May 6, 2022

bors commented May 7, 2022

rust-timer commented May 7, 2022

rust-timer commented May 7, 2022

lqd commented May 11, 2022

rust-timer commented May 11, 2022

bors commented May 11, 2022

bors commented May 11, 2022

rust-timer commented May 11, 2022

rust-timer commented May 11, 2022

bors commented May 25, 2022

lqd commented May 25, 2022 •

edited

Loading

nnethercote commented May 25, 2022

nnethercote commented May 25, 2022

bors commented May 25, 2022

bors commented May 25, 2022

rust-timer commented May 25, 2022

lqd commented May 25, 2022

rust-timer commented May 25, 2022

nnethercote commented May 25, 2022

lqd commented May 25, 2022 •

edited

Loading

Mark-Simulacrum commented May 27, 2022

bors commented May 27, 2022

bors commented May 27, 2022

bors commented May 27, 2022

rust-timer commented May 27, 2022

Update jemalloc to v5.3 #96790

Update jemalloc to v5.3 #96790

Conversation

lqd commented May 6, 2022 • edited Loading

lqd commented May 6, 2022

rust-timer commented May 6, 2022

bors commented May 6, 2022

bors commented May 6, 2022

This comment has been minimized.

lqd commented May 6, 2022

rust-timer commented May 6, 2022

bors commented May 6, 2022

bors commented May 7, 2022

rust-timer commented May 7, 2022

rust-timer commented May 7, 2022

Footnotes

lqd commented May 11, 2022

rust-timer commented May 11, 2022

bors commented May 11, 2022

bors commented May 11, 2022

rust-timer commented May 11, 2022

rust-timer commented May 11, 2022

Footnotes

bors commented May 25, 2022

lqd commented May 25, 2022 • edited Loading

nnethercote commented May 25, 2022

nnethercote commented May 25, 2022

bors commented May 25, 2022

bors commented May 25, 2022

rust-timer commented May 25, 2022

lqd commented May 25, 2022

rust-timer commented May 25, 2022

Instruction count

Max RSS (memory usage)

Cycles

Footnotes

nnethercote commented May 25, 2022

lqd commented May 25, 2022 • edited Loading

Mark-Simulacrum commented May 27, 2022

bors commented May 27, 2022

bors commented May 27, 2022

bors commented May 27, 2022

rust-timer commented May 27, 2022

Instruction count

Max RSS (memory usage)

Cycles

Footnotes

lqd commented May 6, 2022 •

edited

Loading

lqd commented May 25, 2022 •

edited

Loading

lqd commented May 25, 2022 •

edited

Loading