2021-07-27 Triage Log

A very quiet week with only improvements. There was one possible regression, but it was removed from consideration due to only barely impacting a somewhat noisy stress-test benchmark. Untriaged pull requests continue to pile up, but there is still not a good process for dealing with them.

Triage done by @rylev. Revision range: 5c0ca08c662399c1c864310d1a20867d3ab68027..998cfe5aad7c21eb19a4bca50f05a13354706970

0 Regressions, 3 Improvements, 0 Mixed; 0 of them in rollups

Regressions

Support HIR wf checking for function signatures #87265

• Moderate regression in instruction counts (up to 1.1% on full builds of externs-opt)
• Given that this only impacts the somewhat noisy externs stress-test benchmark, it's not clear that we should treat this as an actual regression.

Improvements

Improve caching during trait evaluation #86946

• Moderate improvement in instruction counts (up to -3.7% on full builds of hyper-2-check)

Make mir borrowck‘s use of opaque types independent of the typeck query’s result #87287

• Moderate improvement in instruction counts (up to -1.6% on full builds of deeply-nested-async-check)

Store all HIR owners in the same container #83723

• Moderate improvement in instruction counts (up to -4.4% on incr-unchanged builds of unicode_normalization-check)

Mixed

Untriaged Pull Requests

• #87244 Better diagnostics with mismatched types due to implicit static lifetime
• #86777 Include terminators in instance size estimate
• #86698 Move OnDiskCache to rustc_query_impl.
• #86588 Rollup of 8 pull requests
• #86034 Change entry point to 🛡️ against 💥 💥-payloads
• #84560 Inline Iterator as IntoIterator.

Nags requiring follow up