Remove polymorphization

[saethlin]

This PR removes the flag -Zpolymorphize and all the infrastructure in the compiler that exists only to support it, per rust-lang/compiler-team#810.

[saethlin]

r? compiler-errors

[rustbot]

Some changes occurred to the CTFE / Miri interpreter

cc @rust-lang/miri

Some changes occurred to the CTFE machinery

cc @rust-lang/wg-const-eval

Some changes occurred in compiler/rustc_codegen_gcc

cc @antoyo, @GuillaumeGomez

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

[compiler-errors]

r=me after it's green

@bors rollup=never

(since it likely has a net-positive perf result, but let's not block this PR on perf.)

[compiler-errors]

looks like you've got an unnecessary import for TypeVisitableExt

[rustbot]

rust-analyzer is developed in its own repository. If possible, consider making this change to rust-lang/rust-analyzer instead.

cc @rust-lang/rust-analyzer

[saethlin]

@compiler-errors I'm not 100% sure I rebased properly over the change in compiler/rustc_ty_utils/src/abi.rs.

I also expanded my search a bit and found a few more mentions of polymorphization to delete. Can you take a look?

[compiler-errors]

lgtm, r=me

[saethlin]

@bors r=compiler-errors

[bors]

📌 Commit eb4a656 has been approved by compiler-errors

It is now in the queue for this repository.

[bors]

☔ The latest upstream changes (presumably #133089) made this pull request unmergeable. Please resolve the merge conflicts.

[bors]

⌛ Testing commit 711c8cc with merge ca13e91...

[bors]

☀️ Test successful - checks-actions
Approved by: compiler-errors
Pushing ca13e91 to master...

[rust-timer]

Finished benchmarking commit (ca13e91): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Our benchmarks found a performance regression caused by this PR.
This might be an actual regression, but it can also be just noise.

Next Steps:

• If the regression was expected or you think it can be justified,
  please write a comment with sufficient written justification, and add
  @rustbot label: +perf-regression-triaged to it, to mark the regression as triaged.
• If you think that you know of a way to resolve the regression, try to create
  a new PR with a fix for the regression.
• If you do not understand the regression or you think that it is just noise,
  you can ask the @rust-lang/wg-compiler-performance working group for help (members of this group
  were already notified of this PR).

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

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)	-	-	0
Regressions ❌ (secondary)	0.5%	[0.3%, 0.7%]	6
Improvements ✅ (primary)	-0.5%	[-1.3%, -0.1%]	91
Improvements ✅ (secondary)	-0.7%	[-1.7%, -0.3%]	21
All ❌✅ (primary)	-0.5%	[-1.3%, -0.1%]	91

Max RSS (memory usage)

Results (primary -0.9%, secondary -0.5%)

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)	-	-	0
Regressions ❌ (secondary)	3.5%	[3.5%, 3.5%]	1
Improvements ✅ (primary)	-0.9%	[-1.1%, -0.6%]	6
Improvements ✅ (secondary)	-2.5%	[-2.5%, -2.5%]	2
All ❌✅ (primary)	-0.9%	[-1.1%, -0.6%]	6

Cycles

Results (primary -2.0%, secondary -3.8%)

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)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.0%	[-2.5%, -1.5%]	5
Improvements ✅ (secondary)	-3.8%	[-6.1%, -1.1%]	10
All ❌✅ (primary)	-2.0%	[-2.5%, -1.5%]	5

Binary size

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

Bootstrap: 768.721s -> 769.879s (0.15%)
Artifact size: 330.95 MiB -> 330.83 MiB (-0.04%)

[lqd]

Massively positive results, regressions are small and in stress tests only, marking as triaged.

@rustbot label: +perf-regression-triaged

[nnethercote]

Is it possible to briefly summarize the source of the performance improvements? I'm just curious.

[saethlin]

This PR deletes an often-called query and its associated table. I believe most of the improvement is caused by having less nodes in the query dep graph. Even though all those queries are effectively no-ops. (I would call this query overhead)

[nnethercote]

Was it an oversight that the query was being called despite not doing anything useful? Or was it more complicated than that?

[bjorn3]

The result of the unused_generic_params query was previously encoded in the crate metadata in case a crate that depends on the local crate uses -Zpolymorphize.

[compiler-errors]

I feel like the code perhaps couldve been written to not touch the polymorphization codepaths unless the feature was enabled, though not totally certain given polymorphization's interaction with codegen 🤔

not certain if the unused parameters query couldve been pushed onto the crates that actually needed it (ie the ones build with Zpolymorphize)