I don't remember who I got it from (probably @enhancedscurry?) but I got to use "recursive locks are for people who don't actually understand their concurrency model" again this week.

[numist@xoxo.zone]

I don't remember who I got it from (probably @enhancedscurry?) but I got to use "recursive locks are for people who don't actually understand their concurrency model" again this week.

(unfortunately "…or they do understand it, but made the mistake of publishing a reentrant API" really takes the punch out of it)

[catfish_man@mastodon.social]

@numist @enhancedscurry it’s one I’m fond of telling people about, but really I’m just repeating what the guy who added them to posix said: https://groups.google.com/g/comp.programming.threads/c/tcrTKnfP8HI/m/me2K7_byNdgJ

[enhancedscurry@mastodon.social]

@numist I probably said that, but if I did, all credit to dispatch_async(3):

[numist@xoxo.zone]

@enhancedscurry "We found that the vast majority of recursive locks are deployed retroactively when ill-defined lock hierarchies are discovered." Damn, it's like @madcoder saw my parachute open over the project I just "fixed".

[madcoder@infosec.exchange]

@numist @enhancedscurry I did t write that but I still agree.