Stop caching typeck on disk

[Noratrieb]

I would naively assume that if the inputs to typeck are unchanged, the inputs of the dependendents of typeck would also stay unchanged, making the caching redundant.

I have no data to back this up, so let's try getting some.

#111026 got some nice wins from caching less typeck, let's see what happens here. Perf results are not enough data to determine whether this is good, but it's a starting point.

r? @ghost

[Noratrieb]

@bors try @rust-timer queue

[bors]

⌛ Trying commit eebe0fe with merge 800b4489e24a641f03354624f76693d1190e4d1d...

[rust-log-analyzer]

The job x86_64-gnu-llvm-14 failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

failures:

---- [incremental] tests/incremental/change_private_fn/struct_point.rs stdout ----

error in revision `cfail2`: test compilation failed although it shouldn't!
status: exit status: 1
command: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2/bin/rustc" "/checkout/tests/incremental/change_private_fn/struct_point.rs" "-Zthreads=1" "--sysroot" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2" "--target=x86_64-unknown-linux-gnu" "--cfg" "cfail2" "-C" "incremental=/checkout/obj/build/x86_64-unknown-linux-gnu/test/incremental/change_private_fn/struct_point/struct_point.inc" "-Z" "incremental-verify-ich" "-O" "--error-format" "json" "--json" "future-incompat" "-Zui-testing" "-Zdeduplicate-diagnostics=no" "-C" "prefer-dynamic" "--out-dir" "/checkout/obj/build/x86_64-unknown-linux-gnu/test/incremental/change_private_fn/struct_point" "-Crpath" "-Cdebuginfo=0" "-Lnative=/checkout/obj/build/x86_64-unknown-linux-gnu/native/rust-test-helpers" "-L" "/checkout/obj/build/x86_64-unknown-linux-gnu/test/incremental/change_private_fn/struct_point/auxiliary" "-Z" "query-dep-graph"
stdout: none
--- stderr -------------------------------
--- stderr -------------------------------
error: `typeck(fn_calls_methods_in_same_impl::check)` should have been loaded from disk but it was not
   |
LL |     pub fn check() {
   |     ^^^^^^^^^^^^^^

[bors]

☀️ Try build successful - checks-actions
Build commit: 800b4489e24a641f03354624f76693d1190e4d1d (800b4489e24a641f03354624f76693d1190e4d1d)

[rust-timer]

Finished benchmarking commit (800b4489e24a641f03354624f76693d1190e4d1d): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

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 a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	32.1%	[1.2%, 166.1%]	98
Regressions ❌ (secondary)	61.3%	[0.2%, 454.7%]	30
Improvements ✅ (primary)	-2.4%	[-5.3%, -0.4%]	20
Improvements ✅ (secondary)	-3.0%	[-7.9%, -0.2%]	37
All ❌✅ (primary)	26.2%	[-5.3%, 166.1%]	118

Max RSS (memory usage)

Results

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)	10.5%	[1.8%, 28.6%]	96
Regressions ❌ (secondary)	11.6%	[2.1%, 35.0%]	22
Improvements ✅ (primary)	-2.6%	[-4.8%, -1.4%]	6
Improvements ✅ (secondary)	-4.4%	[-5.7%, -1.8%]	13
All ❌✅ (primary)	9.8%	[-4.8%, 28.6%]	102

Cycles

Results

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)	38.8%	[1.0%, 140.7%]	99
Regressions ❌ (secondary)	71.6%	[2.6%, 478.3%]	28
Improvements ✅ (primary)	-2.9%	[-5.8%, -1.2%]	10
Improvements ✅ (secondary)	-4.0%	[-6.3%, -2.7%]	16
All ❌✅ (primary)	35.0%	[-5.8%, 140.7%]	109

Binary size

Results

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)	0.1%	[0.0%, 0.2%]	19
Regressions ❌ (secondary)	0.1%	[0.1%, 0.2%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.1%	[0.0%, 0.2%]	19

Bootstrap: 656.873s -> 653.777s (-0.47%)

[Noratrieb]

The test failure shows why this is bad. We have non incremental passes (😢) that use typeck.