[perf experiment] Hash interned values, not pointers

[saethlin]

r? ghost

Assessing how bad this is and/or which Hash impls are hot and when

[saethlin]

@bors try @rust-timer queue

[bors]

⌛ Trying commit 4dee948 with merge 6c5885c...

[bors]

☀️ Try build successful - checks-actions
Build commit: 6c5885c (6c5885c827ab41d5deff7ccfa9009cbba83e5c14)

[rust-timer]

Finished benchmarking commit (6c5885c): comparison URL.

Overall result: ❌ regressions - please read the text below

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.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

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

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	1.6%	[0.2%, 14.9%]	245
Regressions ❌ (secondary)	1.8%	[0.2%, 10.5%]	198
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.6%	[0.2%, 14.9%]	245

Max RSS (memory usage)

Results (primary 1.4%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.4%	[1.4%, 1.4%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.4%	[1.4%, 1.4%]	1

Cycles

Results (primary 3.2%, secondary 4.0%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	3.4%	[1.1%, 14.7%]	125
Regressions ❌ (secondary)	4.0%	[1.8%, 11.9%]	57
Improvements ✅ (primary)	-4.4%	[-4.9%, -4.0%]	4
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.2%	[-4.9%, 14.7%]	129

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 773.063s -> 790.285s (2.23%)
Artifact size: 362.34 MiB -> 362.79 MiB (0.12%)

[compiler-errors]

I assume a large portion of the regression (especially nalgebra, since it does a lot of trait solving) is b/c we're interning a ton of type system stuff (like clauses, etc) which are probably somewhat large/expensive to hash.

[saethlin]

I did some slightly more deep inspection in the compiler by making the Hash impl for raw pointers panic, and it turns out that of course the compiler hashes pointers all over. I looked into the regressions above and I think there are better ways to stabilize our artifacts than undertaking such widespread changes.