This week‘s report started with 6 regressions; after eliminating truly obvious noise, we are left with just 2 minor regressions. Of the cases that regressed, I think the only interesting one is keccak (regressed by 1.73% in PR #91549). But don’t be too depressed: keccak was also improved up to 23% by PR #85013 (!); thanks to @simulacrum for that PR.
Triage done by @pnkfelix. Revision range: e2116acae59654bfab2a9729a024f3e2fd6d4b02..404c8471aba60c2d837fa728e7c729a0f52d5830
2 Regressions, 5 Improvements, 3 Mixed; 1 of them in rollups 48 comparisons made in total
“Support AVR for inline asm!” #91224
incr-unchanged
builds of externs
)externs
benchmark.“Eliminate ConstnessAnd again” #91549
full
builds of inflate
)inflate
and keccak
, and I am suspicious of inflate
based on its sensitivity in other PR's.“Avoid string validation in rustc_serialize, check a marker byte instead” #91407
full
builds of helloworld
)“Sync portable-simd to remove autosplats” #91484
full
builds of helloworld
)“replace vec::Drain drop loops with drop_in_place” #85157
full
builds of match-stress-enum
)“replace dynamic library module with libloading” #90716
incr-unchanged
builds of externs
)externs
.“manually implement Hash for DefId” #91660
full
builds of match-stress-enum
)match-stress-enum
and externs
are both pathological cases, so don't expect your compiles to get 2% faster.“Replace dominators algorithm with simple Lengauer-Tarjan” #85013
full
builds of keccak
)incr-unchanged
builds of helloworld
)“Rollup of 6 pull requests” #91799
incr-unchanged
builds of inflate
)incr-unchanged
builds of issue-88862
)“Deduplicate projection sub-obligations” #90423
full
builds of deeply-nested
)full
builds of regression-31157
)memcmp
for more array comparisonsLayout::array
TypeFolder::fold_*
return Result
param_env
field in Obligation's TypeFoldable
implrustc_index::newtype_index
macroreverse
is UB, so replace it with something LLVM can vectorizedist
command to match the components they generate