From https://news.ycombinator.com/item?id=25894233

Could make a great blog post. Anyone know the author?

https://github.com/jnxx/ > Johannes Nix > jnxx > This account is deactivated. Please look at http://gitlab.com\|gitlab.com or http://notabug.org\|notabug.org for updates to FOSS projects.

Thank you very much! This helped quite a bit!

@mflatt @notjack The recent change to Racket CS srcloc in errors broke this test: http://drdr.racket-lang.org/56431/cs/racket/share/pkgs/rackunit-test/tests/rackunit/pr10950.rkt

Maybe https://ieeexplore.ieee.org/author/37299616600

Someone recently reminded me that Racket CS has some limited support for running Racket code in separate OS threads. I’ve added an option to sqlite3 connections to use a separate OS thread for running queries, and the test suite times suggest it has less time overhead than places. I would expect it to be less expensive in memory use too, but I haven’t measured that difference yet.

Looks like (error-print-context-length 0) disables location...: in BC.

Thanks — repair pushed