Florian Angeletti
Florian Angeletti
Thanks for opening the issue, I agree that this is a regression for `base`. The general solution is probably to enable `-short-path` as mentioned by @gasche. But it may make...
@tmcgilchrist , @Julow has started reviewing this PRs but more eyes are always welcome.
For future reference, the issue is still open after https://github.com/ocaml/ocaml/pull/12505, because it requires a longer-term fix in 5.2 .
Note that I am planning to cherry-pick the 5.1 fix to trunk before branching 5.2 to be sure that the initial bug is fixed in all future versions.
This is a complex performance regression for which as far as I know neither @kayceesrk nor @sadiqj could identify the source after few weeks. Fixing such regression is in scope...
Potentially related, I have been reported another example of regression for a [numerical code using `owl`](https://gitlab.inria.fr/fangelet/ocaml-gc-regression-examples), where the performance regression appears even on a single domain code. Measuring time, the...
I am reopening this issue because it still trips up homebrew users (I am seen one more report this week for instance).
Are you aware that dune is already using its own list of default warnings? Your proposed change will be invisible to all dune users. In other words, using warnings like...
I believe that using `the method 'm'` or `the field 'f'` doesn't work because both don't have a type on their own. Consider for instance: ```ocaml type ('a,'b,'c) t =...
Overall, I like the change; and if @Julow has the time I would be happy to revisit it.