08:10 <average> https://prog21.dadgum.com/38.html

08:10 <average> tl;dr allegedly Haskell, Erlang, Forth, J are puzzle languages whereas Python, Ruby, Lua, C are not puzzle languages

08:30 <octachron_> That reads like another instance of confusing familiarity with ease of use when subconsiously restraining oneself to a well-treaded set of problems.

08:33 <reynir> I get DNS timeouts when I try to install an opam package. Is anyone else observing the same?

08:36 <reynir> hm no must be a local issue - I don't observe it on another machine

12:57 <bencef> Hi, quick question I can't seem to find the answer to: How do I refer to an earlier result in utop?

13:02 <octachron> utop -implicit-bindings

13:05 <bencef> ahhh I should have consulted `utop --help'

13:13 <bencef> `_0' doesn't seem to work still

13:28 <bencef> maybe it is not compatible with `-emacs'?

13:30 <bencef> for the LambdaTerm interface it works as expected

13:31 <bencef> for the emacs interface I still see "- : <expr>" results and can't refer to them

14:09 <discocaml> <mbacarella> How do I prevent OCaml from OOMing on an admittedly low memory VM? The true working set of this app is only about 70MB but it seems to collect trash past the 3-400MB of space on the box and then Linux kills it

14:09 <discocaml> <mbacarella> Do I just run Gc.compact every minute or is there a heap size setting

14:15 <discocaml> <undu> you probably want to change the max_overhead https://v2.ocaml.org/api/Gc.html

14:15 <discocaml> <undu> > Heap compaction is triggered when the estimated amount of "wasted" memory is more than max_overhead percent of the amount of live data. If max_overhead is set to 0, heap compaction is triggered at the end of each major GC cycle (this setting is intended for testing purposes only). If max_overhead >= 1000000, compaction is never triggered. If compaction is permanently disabled, it is strongly suggested to set allocation_policy to 2. Default: 500

14:15 <discocaml> <undu> you probably want to change the max_overhead https://v2.ocaml.org/api/Gc.html

14:15 <discocaml> <undu> > Heap compaction is triggered when the estimated amount of "wasted" memory is more than max_overhead percent of the amount of live data. If max_overhead is set to 0, heap compaction is triggered at the end of each major GC cycle (this setting is intended for testing purposes only). If max_overhead >= 1000000, compaction is never triggered. If compaction is permanently disabled, it is strongly suggested to set allocation_policy to 2. **Default: 5

14:57 <discocaml> <__jco__> Hi everyone, happy to release bindings (and wrapper) to an algebra library https://discuss.ocaml.org/t/ann-bindings-to-the-pari-algebra-library/13157

15:37 <discocaml> <mbacarella> I think no. I think I want GC to run more often *and* to do compaction each time.

15:38 <discocaml> <mbacarella> I think not quite. I think I want GC to run more often *and* to do compaction each time.

15:48 <discocaml> <undu> setting the parameter to zero forces compaction on each cycle 🙂

15:49 <discocaml> <undu> I have to say that compaction doesn't work in ocaml 5.x

15:50 <discocaml> <undu> so for long-running processing allocating the new versions are not viable yet

16:25 <discocaml> <mbacarella> the problem, IMO, is GC isn't running before the box OOMs

16:25 <discocaml> <mbacarella> though compaction might matter too

17:17 <hannes> mbacarella: do you use OCaml 4 or OCaml 5? AFAIU OCaml 5 still does not compact / free memory -- see https://github.com/ocaml/ocaml/pull/12193

19:09 <h0rror> hello, I can't get Big_int to work on utop, it can auto-complete its values, but then it complains about "reference to undefined global"

19:16 <companion_cube> You probably need to `#require "bignum" ` or something like that.

19:16 <companion_cube> Note that it's deprecated for zarith though

19:20 <h0rror> oh, are there no arbitrary-precision integers in the standard lib?

19:26 <companion_cube> Nope

19:28 <h0rror> ok, zarith's interface is nicer, thanks companion_cube :)

19:28 <companion_cube> Nope

19:32 <companion_cube> It's also a lot faster