2026/04/14

Newest at the top

2026-04-14 13:25:32 +0000karenw(~karenw@user/karenw) karenw
2026-04-14 13:25:27 +0000Digit(~user@user/digit) Digit
2026-04-14 13:23:25 +0000ystael(~ystael@user/ystael) ystael
2026-04-14 13:20:39 +0000AlexZenon(~alzenon@178.34.151.36)
2026-04-14 13:18:06 +0000uli-fem(~uli-fem@203.87.114.209) (Ping timeout: 246 seconds)
2026-04-14 13:16:49 +0000leppard(~noOne@ipservice-092-208-182-236.092.208.pools.vodafone-ip.de) Inline
2026-04-14 13:15:41 +0000 <Leary> gentauro: It's possible, but that doesn't mean it's a good idea. I rather suspect it will induce a bunch of GC overhead you don't want, which could probably be avoided by some kind of work queue.
2026-04-14 13:15:33 +0000 <mauke> (number of threads, I mean)
2026-04-14 13:15:25 +0000 <mauke> it definitely went into the 100,000s
2026-04-14 13:15:06 +0000 <mauke> I did a linear bucket chain of threads connected by MVars, maybe 15 years ago?
2026-04-14 13:14:19 +0000 <mauke> I don't remember what my record was
2026-04-14 13:13:52 +0000uli-fem(~uli-fem@203.87.114.209)
2026-04-14 13:11:53 +0000tromp(~textual@2001:1c00:340e:2700:f5bd:97ff:8f76:c38c)
2026-04-14 13:11:29 +0000 <gentauro> would it be possible to spawn 1m concurrent `https://hackage-content.haskell.org/package/async-2.2.6/docs/Control-Concurrent-Async.html` (green threads right?) as we can do in Rust/dotnet: https://hez2010.github.io/async-runtimes-benchmarks-2024/take2.html
2026-04-14 13:08:58 +0000Digit(~user@user/digit) (Ping timeout: 276 seconds)
2026-04-14 13:08:46 +0000AlexNoo(~AlexNoo@178.34.151.36)
2026-04-14 13:06:47 +0000tromp(~textual@2001:1c00:340e:2700:f5bd:97ff:8f76:c38c) (Quit: My iMac has gone to sleep. ZZZzzz…)
2026-04-14 12:58:14 +0000uli-fem(~uli-fem@203.87.114.209) (Ping timeout: 256 seconds)
2026-04-14 12:57:31 +0000AlexNoo(~AlexNoo@178.34.151.36) (Quit: Leaving)
2026-04-14 12:56:42 +0000AlexZenon(~alzenon@178.34.151.36) (Quit: ;-)
2026-04-14 12:56:17 +0000alter2000(~alter2000@user/alter2000) alter2000
2026-04-14 12:53:30 +0000uli-fem(~uli-fem@203.87.114.209)
2026-04-14 12:49:37 +0000alter2000(~alter2000@user/alter2000) (Ping timeout: 244 seconds)
2026-04-14 12:44:56 +0000puke(~puke@user/puke) (Ping timeout: 250 seconds)
2026-04-14 12:42:24 +0000uli-fem(~uli-fem@203.87.114.209) (Ping timeout: 245 seconds)
2026-04-14 12:38:48 +0000tromp(~textual@2001:1c00:340e:2700:f5bd:97ff:8f76:c38c)
2026-04-14 12:37:54 +0000uli-fem(~uli-fem@203.87.114.209)
2026-04-14 12:33:26 +0000alter2000(~alter2000@user/alter2000) alter2000
2026-04-14 12:25:39 +0000alter2000(~alter2000@user/alter2000) (Ping timeout: 255 seconds)
2026-04-14 12:15:04 +0000 <Milan_Vanca> Program executed on all cores only first 5s from 25s total. So it looks like it oveflowed sparks or something.
2026-04-14 12:13:11 +0000 <Milan_Vanca> I have found this in ghc docs Any remaining sparks are discarded at the end of execution, so “converted” plus “pruned” does not necessarily add up to the total.
2026-04-14 12:12:21 +0000 <lambdabot> 40960
2026-04-14 12:12:21 +0000oskarw(~user@user/oskarw) (ERC 5.6.1 (IRC client for GNU Emacs 30.2))
2026-04-14 12:12:20 +0000 <ski> > 49151 - (7336 + 855)
2026-04-14 12:07:52 +0000uli-fem(~uli-fem@203.87.114.209) (Ping timeout: 276 seconds)
2026-04-14 12:02:57 +0000uli-fem(~uli-fem@203.87.114.209)
2026-04-14 12:01:52 +0000 <Milan_Vanca> Maybe info in () shows last known state, and first number is total for whole program run.
2026-04-14 12:01:14 +0000arandombit(~arandombi@user/arandombit) (Remote host closed the connection)
2026-04-14 12:00:51 +0000tromp(~textual@2001:1c00:340e:2700:f5bd:97ff:8f76:c38c) (Quit: My iMac has gone to sleep. ZZZzzz…)
2026-04-14 12:00:28 +0000jreicher(~joelr@user/jreicher) jreicher
2026-04-14 11:59:40 +0000 <Milan_Vanca> I mean it should sum right? Or is overflowed counted as more than converted?
2026-04-14 11:59:00 +0000humasect(~humasect@dyn-192-249-132-90.nexicom.net) humasect
2026-04-14 11:58:40 +0000haritz(~hrtz@user/haritz) haritz
2026-04-14 11:58:40 +0000haritz(~hrtz@140.228.70.141) (Changing host)
2026-04-14 11:58:40 +0000haritz(~hrtz@140.228.70.141)
2026-04-14 11:58:27 +0000 <lambdabot> False
2026-04-14 11:58:25 +0000 <Milan_Vanca> > 49151 == 7336 + 855
2026-04-14 11:58:11 +0000 <Milan_Vanca> Hmm is output of +RTS -N4 -s in line that talks about sparks SPARKS: 49151 (7336 converted, 855 overflowed, 0 dud, 0 GC'd, 0 fizzled)
2026-04-14 11:57:18 +0000CiaoSen(~Jura@p549cbfb1.dip0.t-ipconnect.de) (Ping timeout: 255 seconds)
2026-04-14 11:33:15 +0000weary-traveler(~user@user/user363627) user363627