Uh oh!
There was an error while loading. Please reload this page.
incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. - #46368
Conversation
…ans for type definitions.
eddyb
commented
Nov 30, 2017
@bors r+ |
bors
commented
Nov 30, 2017
📌 Commit 89dc8ae has been approved by |
…uginfo-spans, r=eddyb incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. r? @nikomatsakis
…uginfo-spans, r=eddyb incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. r? @nikomatsakis
…uginfo-spans, r=eddyb incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. r? @nikomatsakis
…uginfo-spans, r=eddyb incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. r? @nikomatsakis
bors
commented
Dec 2, 2017
…s, r=eddyb incr.comp.: Remove an unnecessary HIR access which enables hashing spans for type definitions. r? @nikomatsakis
bors
commented
Dec 2, 2017
☀️ Test successful - status-appveyor, status-travis |
Mark-Simulacrum
commented
Dec 3, 2017
This appears to have regressed performance on perf.rlo (http://perf.rust-lang.org/compare.html?start=9053fdd411e81910a86b68f0a1c3b3ca084bff62&end=377decc352e58aa2bb0e2deb2eb66c3e7241a622&stat=wall-time): was that expected? |
michaelwoerister
commented
Dec 4, 2017
Thanks for the heads up, @Mark-Simulacrum. It's not entirely surprising, since we are hashing more and we detect more changes. Most of these newly detected changes are probably false positives of some kind. |
r? @nikomatsakis