recent parser performance regression #15262
Labels
parser
Language parsing and surface syntax
performance
Must go faster
regression
Regression in behavior compared to a previous version
With the recent work on modifying julia-syntax, I assumed it should be faster now. However, I was surprised to find that f5033ad seemed to be about twice as fast on one of my torture benchmarks (#9729 (comment)) than both v0.4 and master today (636916e).
(there's nothing special about that first commit, i just usually try to have one build on my machine that lags behind my current work)
The text was updated successfully, but these errors were encountered: