2021-05-18 Triage Log
A lot of noise in the benchmark results this week. We are discussing (zulip archive, live zulip) how best to update the benchmark set to eliminate the noisy cases that are bouncing around. Beyond that, some large improvements to a few individual benchmarks.
The memory usage (max-rss) seemed largely flat. Except for an upward trend on tuple-stess
that indicates 4% more memory from a week ago.
Triage done by @pnkfelix. Revision range: 382f748f23979e37e3e012b090e5a0313463f182..25a277f03df7e44643ddfcc240d034409cb2f505
5 Regressions, 7 Improvements, 2 Mixed 1 of them in rollups
Reachable statics have reachable initializers #84549
incr-unchanged
builds of ctfe-stress-4-opt
)ctfe-stress-4-debug
also saw a 1.2% regression in its incr-unchanged
builds.have on_completion record subcycles #85186
incr-unchanged
builds of syn-opt
)Store VariantIdx to distinguish enum variants #85195
full
builds of match-stress-enum-check
)match-stress-enum
variants observed a similar improvement. This makes sense since that is a microbenchmark stressing the compilation of match
expressions.Do not allocate or unwind after fork #81858
incr-patched: b9b3e592dd cherry picked
builds of style-servo-debug
)BTree: no longer copy keys and values before dropping them #84904
incr-unchanged
builds of ctfe-stress-4-check
and ctfe-stress-4-opt
and ctfe-stress-4-debug
)incr-patched: println
builds of cargo-opt
)rustc_codegen_ssa: only create backend BasicBlock
s as-needed. #84993
full
builds of syn-opt
)full
builds of deeply-nested-async-check
)syn-opt
was anticipated from perf runs on the PR itself.deeply-nested-async-check
was not predicted by that run.rustc_driver cleanup #83610
full
builds of coercions-debug
)Add auto traits and clone trait migrations for RFC2229 #84730
full
builds of coercions-debug
)Provide ExitStatusError #82973
full
builds of coercions-debug
)Fix --remap-path-prefix
not correctly remapping rust-src
component paths and unify handling of path mapping with virtualized paths #83813
full
builds of coercions-debug
)Rollup of 5 pull requests #85231
full
builds of coercions-debug
)Rollup of 8 pull requests #85414
incr-full
builds of coercions-debug
)coercions-check
and coercions-opt
had smaller improvements, -0.7% and -0.5%, respectively.Add support for const operands and options to global_asm! #84107
full
builds of html5ever-opt
)Update cc crate #85190
full
builds of html5ever-opt
)coercions-debug
also saw a -1.0% improvement on its full
builds.Double-check whether regression to deeply-nested-async-check
from #84993 is noise or something real.
Look into the 4% tuple-stress
regression. The first seems like gradual creep, plus a 3% jump that seems to be associated with #84571, “Parse unnamed fields of struct and union type.” (Perhaps its inherent given the nature of that microbenchmark, but some other benchmarks also had their memory usage impacted.