John Högberg

Thanks for your report! Just as a heads-up, we're a bit short-handed over the summer vacation period so we likely won't get around to fixing this until most of us...

`on_load/1` complicates code loading in general (it still needs to appear atomic) so it's a fairly bad idea to do anything complicated in it, but it nevertheless shouldn't fail silently...

Thanks for your report, and sorry for the late reply. I got back from long-term leave late last week and didn't manage to fix this in time for the upcoming...

I've merged a fix into `maint` now, which will be released as `25.3` early next year. Thanks again for your report!

We had another technical board meeting today to try to reconcile the differences between Unix and Windows, and have realized that the functionality differs too much between operating systems to...

Thanks for your report, https://github.com/jhogberg/otp/tree/john/erts/fix-win32-sz/GH-8903/OTP-19283 ought to fix it but we haven't got a Windows machine to test on at the moment. Could you try it out? :-)

I've merged a fix to `maint` now, to be released as OTP 27.2 later this autumn. Thanks again for your report! :)

Hi! We haven't forgotten about this. The problem is that we had a bug in how virtual heap sizes were calculated, leading to full-sweep garbage collections being less frequent than...

I've made some tweaks in #8347 that brings most performance back on small sizes, and makes larger ones faster than before. Can you try it out?

Thanks, I've pushed another tweak of the virtual heap growth curve that brings performance back to where it was (or better) on my machine (edit: it varies a lot though,...