perf report for 2023-09-13.

Author: pnkfelix

triage/2023-09-13.md

@@ -1,8 +1,14 @@
 # 2023-09-13 Triage Log
 
-TODO: Summary
-
-Triage done by **@???**.
+An interesting week. We saw a massive improvement to instruction-counts across
+over a hundred benchmarks, thanks to #110050 an improved encoding scheme for the
+dependency graphs that underlie incremental-compilation. However, these
+instruction-count improvements did not translate to direct cycle time
+improvements. We also saw an improvement to our artifact sizes due to #115306.
+Beyond that, we had a scattering of small regressions to instruction-counts that
+were justified because they were associated with bug fixes.
+
+Triage done by **@pnkfelix**.
 Revision range: [15e52b05..7e0261e7](https://perf.rust-lang.org/?start=15e52b05ca8f63e0da27c808680388717e5b997e&end=7e0261e7ea2085bdc0bc3d0fd6776bf343473858&absolute=false&stat=instructions%3Au)
 
 **Summary**:
@@ -16,7 +22,7 @@ Revision range: [15e52b05..7e0261e7](https://perf.rust-lang.org/?start=15e52b05c
 | All ❌✅ (primary)                 | -1.3% | [-5.9%, 10.2%] | 123   |
 
 
-3 Regressions, 2 Improvements, 5 Mixed; ??? of them in rollups
+3 Regressions, 2 Improvements, 5 Mixed; 2 of them in rollups
 84 artifact comparisons made in total
 
 #### Regressions
@@ -31,6 +37,8 @@ Add `FreezeLock` type and use it to store `Definitions` [#115401](https://github
 | Improvements ✅ <br /> (secondary) | -    | -            | 0     |
 | All ❌✅ (primary)                 | 0.3% | [0.2%, 0.4%] | 11    |
 
+* The impact here is [hypothesized to be](https://github.com/rust-lang/rust/pull/115401#issuecomment-1709461275) due to serial/parallel trade-off; we benchmark the serial case and observe a small regression, while the parallel case is observing an improvement of roughly the same caliber.
+* Marked as triaged
 
 Rollup of 6 pull requests [#115672](https://github.com/rust-lang/rust/pull/115672) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=cd71a37f320c379df47ff64abd934f3a2da94c26&end=309af3442a1808888e3ceb2eacccbf4140eba1e0&stat=instructions:u)
 
@@ -42,6 +50,8 @@ Rollup of 6 pull requests [#115672](https://github.com/rust-lang/rust/pull/11567
 | Improvements ✅ <br /> (secondary) | -    | -            | 0     |
 | All ❌✅ (primary)                 | 4.2% | [0.8%, 9.8%] | 5     |
 
+* already marked as triaged
+* all five regressions are to doc benchmarks, due to new feature added in https://github.com/rust-lang/rust/pull/115201
 
 Use the same DISubprogram for each instance of the same inlined function within a caller [#115417](https://github.com/rust-lang/rust/pull/115417) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=ffc48e3eda36e288f76b4022d72d94321887ebf5&end=62ebe3a2b177d50ec664798d731b8a8d1a9120d1&stat=instructions:u)
 
@@ -53,6 +63,8 @@ Use the same DISubprogram for each instance of the same inlined function within
 | Improvements ✅ <br /> (secondary) | -    | -            | 0     |
 | All ❌✅ (primary)                 | 1.0% | [0.6%, 1.3%] | 3     |
 
+* already marked as triaged
+* regression was expected, though we may be able to claw back performance after resolving rust#115455
 
 #### Improvements
 
@@ -90,6 +102,10 @@ Represent MIR composite debuginfo as projections instead of aggregates [#115252]
 | Improvements ✅ <br /> (secondary) | -0.4% | [-0.4%, -0.3%] | 4     |
 | All ❌✅ (primary)                 | 1.1%  | [-0.3%, 3.9%]  | 3     |
 
+* The single regression is to exa-0.10.1-opt-full
+* However, nnethercote noted that this PR introduced broad (if small)
+  regressions to linked artifact (aka binary) sizes (in both opt and debug settings)
+* not marking as triaged
 
 Use a specialized varint + bitpacking scheme for DepGraph encoding [#110050](https://github.com/rust-lang/rust/pull/110050) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=4e5b31c2b0023dba53a1b2827f4b7ac42aaaa18f&end=f00c1399987c60b4e884afc42f4aa6226855e9ae&stat=instructions:u)
 
@@ -101,6 +117,10 @@ Use a specialized varint + bitpacking scheme for DepGraph encoding [#110050](htt
 | Improvements ✅ <br /> (secondary) | -1.4% | [-2.9%, -0.5%] | 32    |
 | All ❌✅ (primary)                 | -1.7% | [-5.8%, -0.3%] | 104   |
 
+* on its surface, the improvements to instruction counts here clearly outweigh the regressions
+* it is worth noting that the cycle counts did not see the same trends;
+  there were zero improvements and 7 primary regressions to cycle counts.
+* still, marking as triaged; this PR has gone through enough performance evaluation already.
 
 Rollup of 7 pull requests [#115665](https://github.com/rust-lang/rust/pull/115665) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=3d249706aa8b0167dd49efa1b3ce7cc0e9cbba08&end=de4cba3a98a15a891ad708a049c7fb5682083d97&stat=instructions:u)
 
@@ -112,6 +132,8 @@ Rollup of 7 pull requests [#115665](https://github.com/rust-lang/rust/pull/11566
 | Improvements ✅ <br /> (secondary) | -0.3% | [-0.3%, -0.3%] | 1     |
 | All ❌✅ (primary)                 | 0.7%  | [0.6%, 0.7%]   | 2     |
 
+* primary regressions were helloworld-check (incr-unchanged and incr-patched:println)
+* marking as triaged; not worth investigating a rollup for that benchmark.
 
 Avoid a `source_span` query when encoding Spans into query results [#115657](https://github.com/rust-lang/rust/pull/115657) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=5ede9408945b46ab183dd228253297bdf62304f7&end=38bbc2ce03a2369d96898d58cc0aa06f1a4b5dcf&stat=instructions:u)
 
@@ -123,6 +145,9 @@ Avoid a `source_span` query when encoding Spans into query results [#115657](htt
 | Improvements ✅ <br /> (secondary) | -0.5% | [-0.6%, -0.4%] | 4     |
 | All ❌✅ (primary)                 | -0.0% | [-0.4%, 0.4%]  | 4     |
 
+* primary regressions are to diesel-check (full and incr-full).
+* This is fixing a soundness issue with the dep-graph maintenance; therefore, these regressions seem tolerable.
+* Marking as triaged
 
 Encode only MIR reachable from other crates [#115306](https://github.com/rust-lang/rust/pull/115306) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8e37c509fda1f7387895e33783cba94ea3960a29&end=7418413a7fad1c4e8b82f970bd78af030e5f813e&stat=instructions:u)
 
@@ -134,41 +159,6 @@ Encode only MIR reachable from other crates [#115306](https://github.com/rust-la
 | Improvements ✅ <br /> (secondary) | -0.9% | [-1.2%, -0.7%] | 5     |
 | All ❌✅ (primary)                 | -0.1% | [-2.7%, 2.4%]  | 27    |
 
-
-#### Untriaged Pull Requests
-
-- [#115665 Rollup of 7 pull requests](https://github.com/rust-lang/rust/pull/115665)
-- [#115657 Avoid a `source_span` query when encoding Spans into query results](https://github.com/rust-lang/rust/pull/115657)
-- [#115401 Add `FreezeLock` type and use it to store `Definitions`](https://github.com/rust-lang/rust/pull/115401)
-- [#115306 Encode only MIR reachable from other crates](https://github.com/rust-lang/rust/pull/115306)
-- [#115252 Represent MIR composite debuginfo as projections instead of aggregates](https://github.com/rust-lang/rust/pull/115252)
-- [#115082 Fix races conditions with `SyntaxContext` decoding](https://github.com/rust-lang/rust/pull/115082)
-- [#114892 Remove conditional use of `Sharded` from query caches](https://github.com/rust-lang/rust/pull/114892)
-- [#114481 Rollup of 9 pull requests](https://github.com/rust-lang/rust/pull/114481)
-- [#114459 Do not run ConstProp on mir_for_ctfe.](https://github.com/rust-lang/rust/pull/114459)
-- [#114321 get auto traits for parallel rustc](https://github.com/rust-lang/rust/pull/114321)
-- [#114023 Warn on inductive cycle in coherence leading to impls being considered not overlapping](https://github.com/rust-lang/rust/pull/114023)
-- [#114004 Add `riscv64gc-unknown-hermit` target](https://github.com/rust-lang/rust/pull/114004)
-- [#113858 Always const-prop scalars and scalar pairs](https://github.com/rust-lang/rust/pull/113858)
-- [#113758 Turn copy into moves during DSE.](https://github.com/rust-lang/rust/pull/113758)
-- [#113485 Bump version to 1.73](https://github.com/rust-lang/rust/pull/113485)
-- [#113370 Rollup of 8 pull requests](https://github.com/rust-lang/rust/pull/113370)
-- [#113320 Add some extra information to opaque type cycle errors](https://github.com/rust-lang/rust/pull/113320)
-- [#113306 Update debuginfo test runner to provide more useful output](https://github.com/rust-lang/rust/pull/113306)
-- [#113304 Upgrade to indexmap 2.0.0](https://github.com/rust-lang/rust/pull/113304)
-- [#113270 perform TokenStream replacement in-place when possible in expand_macro](https://github.com/rust-lang/rust/pull/113270)
-- [#113057 Rollup of 2 pull requests](https://github.com/rust-lang/rust/pull/113057)
-- [#112963 Stop bubbling out hidden types from the eval obligation queries](https://github.com/rust-lang/rust/pull/112963)
-- [#112882 Rewrite `UnDerefer`](https://github.com/rust-lang/rust/pull/112882)
-- [#112420 Rollup of 4 pull requests](https://github.com/rust-lang/rust/pull/112420)
-- [#112157 Resurrect: rustc_target: Add alignment to indirectly-passed by-value types, correcting the alignment of byval on x86 in the process.](https://github.com/rust-lang/rust/pull/112157)
-- [#112083 Make `TrustedStep` require `Copy`](https://github.com/rust-lang/rust/pull/112083)
-- [#112016 Rollup of 6 pull requests](https://github.com/rust-lang/rust/pull/112016)
-- [#112001 Enable MatchBranchSimplification](https://github.com/rust-lang/rust/pull/112001)
-- [#111869 Rollup of 6 pull requests](https://github.com/rust-lang/rust/pull/111869)
-- [#111850 Specialize `StepBy<Range<{integer}>>`](https://github.com/rust-lang/rust/pull/111850)
-
-#### Nags requiring follow up
-
-TODO: Nags
-
+* the big (>1%) primary regressions were to three check-incr-unchanged cases: cranelift-codegen-0.82.1, html5ever-0.26.0, and hyper-0.14.18
+* the regressions seem unfortunate, but tolerable given the improvement to linked artifact sizes
+* marking as triaged