2023-10-26 Triage Log

A largely positive week with really only 1 true regression and 1 large improvement both by cjgillot proving that the compiler giveth and taketh away. The improvement, an improvement in the borrow checker, lead to a 0.7% improvement across ~80 benchmarks while the regression had the same magnitude but only impacted 18 benchmarks.

Triage done by @rylev. Revision range: b9832e7..698db85

Summary:

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.5%	[0.2%, 3.4%]	17
Regressions ❌ (secondary)	1.2%	[0.9%, 1.7%]	11
Improvements ✅ (primary)	-0.8%	[-3.0%, -0.1%]	85
Improvements ✅ (secondary)	-1.0%	[-1.8%, -0.1%]	31
All ❌✅ (primary)	-0.6%	[-3.0%, 3.4%]	102

7 Regressions, 6 Improvements, 4 Mixed; 5 of them in rollups 85 artifact comparisons made in total

Regressions

Bump COINDUCTIVE_OVERLAP_IN_COHERENCE to deny + warn in deps #116493 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.7%	[0.7%, 0.8%]	2
Regressions ❌ (secondary)	0.4%	[0.3%, 0.6%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.7%	[0.7%, 0.8%]	2

Noise

Rollup of 7 pull requests #116923 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.3%	[0.2%, 0.4%]	9
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

While this very much does not look like noise, the regressions are small enough that I don't think this is really worth any investigation.

Rollup of 4 pull requests #116940 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.7%	[0.2%, 1.2%]	18
Regressions ❌ (secondary)	1.0%	[0.7%, 1.2%]	9
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.8%	[-1.8%, -1.8%]	1
All ❌✅ (primary)	0.7%	[0.2%, 1.2%]	18

From lqd “I think this is all in mir transforms, likely the additional pass name calls and string finds in turn -- and since is all static data, maybe should be cached, and stored in a better-suited data structure.”
#116896 is the culprit. Following up with the author, cjgillot.

Lint non_exhaustive_omitted_patterns by columns #116734 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	1.5%	[1.5%, 1.6%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Given that the impacted benchmark is the stress test for the code in question, I think the small regression is fine.

Suggest removing redundant arguments in format!() #115324 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	2.1%	[0.3%, 3.9%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.1%	[0.3%, 3.9%]	2

The perf regression is likely noise.

Rollup of 5 pull requests #117087 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	4.1%	[4.1%, 4.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	4.1%	[4.1%, 4.1%]	1

Noise

Improve android-ndk property interface #116998 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	3.9%	[3.9%, 3.9%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.9%	[3.9%, 3.9%]	1

Noise

Improvements

Panic when the global allocator tries to register a TLS destructor #116402 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.7%	[-0.7%, -0.7%]	2
Improvements ✅ (secondary)	-0.4%	[-0.7%, -0.2%]	14
All ❌✅ (primary)	-0.7%	[-0.7%, -0.7%]	2

On Windows make read_dir error on the empty path #116606 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.8%	[-2.8%, -2.8%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.8%	[-2.8%, -2.8%]	1

This is noise.

Separate move path tracking between borrowck and drop elaboration. #116300 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.2%	[0.1%, 0.2%]	5
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	2
Improvements ✅ (primary)	-0.7%	[-3.6%, -0.2%]	77
Improvements ✅ (secondary)	-1.1%	[-1.8%, -0.4%]	30
All ❌✅ (primary)	-0.7%	[-3.6%, 0.2%]	82

time: use clock_gettime on macos #116238 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-3.6%	[-3.6%, -3.6%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-3.6%	[-3.6%, -3.6%]	1

This is noise.

Uplift ClauseKind and PredicateKind into rustc_type_ir #116993 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.5%	[-0.8%, -0.2%]	14
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.5%	[-0.8%, -0.2%]	14

Refactor type visitor walking #117076 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.2%	[-0.3%, -0.1%]	11
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.2%	[-0.3%, -0.1%]	11

Mixed

Rollup of 5 pull requests #116885 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.5%	[0.5%, 0.5%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.3%	[-0.5%, -0.2%]	9
Improvements ✅ (secondary)	-0.6%	[-0.9%, -0.3%]	13
All ❌✅ (primary)	-0.2%	[-0.5%, 0.5%]	10

Improvements outweigh regressions.

Update hashbrown to 0.14.2 #116956 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.4%	[0.2%, 0.5%]	7
Regressions ❌ (secondary)	0.2%	[0.2%, 0.3%]	3
Improvements ✅ (primary)	-0.6%	[-1.2%, -0.2%]	5
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.0%	[-1.2%, 0.5%]	12

Mostly neutral - max-rss regression seems to have corrected itself in the meantime

report unused_import for empty reexports even it is pub #116033 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	0.4%	[0.3%, 0.5%]	13
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-3.7%	[-3.7%, -3.7%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.1%	[-3.7%, 0.5%]	14

Small regression - asked the author and reviewer if they feel like they had a quick idea how to address it.

Rollup of 7 pull requests #117180 (Comparison Link)

(instructions:u)	mean	range	count
Regressions ❌ (primary)	3.7%	[3.7%, 3.7%]	1
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.3%	[-1.3%, -1.3%]	1
All ❌✅ (primary)	3.7%	[3.7%, 3.7%]	1

The one primary 3.7% regression looks real. The tiny secondary one looks either noise or has basically returned to its prior value. The small improvement also seems to have sustained at the new levels.