Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 635 636 637 638 639 640 641 642 643 644 645 .. 5022
— 502,152 events total
2020-10-14 17:39:21 <gentauro> maerwald: I will not talk bad about people working with other `tech stacks`, but it's very clear who I would work with ;)
2020-10-14 17:39:22 <hseg> merijn: it ended up not being that much more efficient than a straight map
2020-10-14 17:39:46 <hseg> *shrugs* happens
2020-10-14 17:40:00 <maerwald> gentauro: I've worked with javascript devs who were pretty high on the engineer skill scale
2020-10-14 17:40:14 <maerwald> I don't think the language has anything to do with it, it's just community bias
2020-10-14 17:40:38 <maerwald> it might be true that low-slikk is less annoying in haskell
2020-10-14 17:40:44 <maerwald> but that's a different thing
2020-10-14 17:40:54 <maerwald> s/low-slikk/low-skill/
2020-10-14 17:40:56 <gentauro> maerwald: lucky you. I worked with JS "peeps" (I wouldn't even call them devs) that didn't know what they were doing
2020-10-14 17:41:18 <maerwald> you have the same in haskell, except that jobs are harder to get
2020-10-14 17:41:23 <gentauro> maerwald: my point is that it's very difficult to hear that sentence with people working with Scala ;)
2020-10-14 17:41:29 × kritzefitz quits (~kritzefit@212.86.56.80) (Remote host closed the connection)
2020-10-14 17:41:35 <merijn> maerwald: i think that's not entirely true
2020-10-14 17:41:43 <gentauro> s/Scala/FP/
2020-10-14 17:42:13 <merijn> maerwald: I think there is a historical (and decreasing) higher average skill in the Haskell community, but as it grows and expands it regresses back to the mean
2020-10-14 17:42:55 <maerwald> which is a good thing
2020-10-14 17:43:16 <merijn> Simply from sheer selection bias of "who feels attracted to small niche languages"
2020-10-14 17:43:31 <gentauro> btw, I can see that I just jumped into the disccusion :) but did anybody see my question? 19:35 < gentauro> is there a Haskell project that reminds of `.NET LINQ`?
2020-10-14 17:43:34 <gentauro> xD
2020-10-14 17:44:25 <dolio> I answered it.
2020-10-14 17:44:30 <dolio> And you didn't read it.
2020-10-14 17:44:51 <gentauro> dolio: you didn't tagged me :P
2020-10-14 17:44:58 × xerox_ quits (~xerox@unaffiliated/xerox) (Ping timeout: 260 seconds)
2020-10-14 17:45:16 <phadej> gentauro: https://downloads.haskell.org/~ghc/latest/docs/html/users_guide/glasgow_exts.html#extension-MonadComprehensions
2020-10-14 17:45:20 × kori quits (~kori@arrowheads/kori) (Read error: Connection reset by peer)
2020-10-14 17:45:24 <[exa]> Hm so sometimes when starting a scotty app on a mac, I'm getting: "file descriptor 202794296 out of range for select (0--1024). Recompile with -threaded to work around this. Is there any good explanation of what's happening?
2020-10-14 17:45:29 <phadej> here go your tag.
2020-10-14 17:45:53 → Amras joins (~Amras@unaffiliated/amras0000)
2020-10-14 17:46:05 <monochrom> There is also this angle: If you can learn several different languages --- and very different, having learned n-1 of them doesn't help learning the nth one --- it is a good show of what your mind can do.
2020-10-14 17:46:20 × bahamas quits (~lucian@unaffiliated/bahamas) (Ping timeout: 272 seconds)
2020-10-14 17:46:34 <gentauro> phadej: and dolio: it's just what I was looking for. Thx
2020-10-14 17:46:51 → justanotheruser joins (~justanoth@unaffiliated/justanotheruser)
2020-10-14 17:47:41 <merijn> [exa]: "you need to recompile with -threaded"? :p
2020-10-14 17:47:52 <merijn> [exa]: Looks like you hit a limitation of the non-threaded RTS
2020-10-14 17:47:56 → xerox_ joins (~xerox@unaffiliated/xerox)
2020-10-14 17:48:06 <ghoulguy> I bet 202794296 isn't a real file descriptor
2020-10-14 17:48:20 <merijn> Also that
2020-10-14 17:48:57 <[exa]> well yeah, the question is where this came from (can't find anything reasonable). I'm not aware of anything in the app that could break such stuff, it just starts scotty in a perfectly normal way, no magic, no workarounds, no nothing
2020-10-14 17:49:09 <monochrom> threaded RTS is not going to get a higher quota for file descriptors.
2020-10-14 17:49:18 × wroathe quits (~wroathe@c-73-24-27-54.hsd1.mn.comcast.net) (Ping timeout: 260 seconds)
2020-10-14 17:49:42 <merijn> Well, it looks like he hit a *select* limit
2020-10-14 17:49:47 <geekosaur> but it may use poll instead of select() which by inherent design has a low file descriptor limit
2020-10-14 17:49:48 <merijn> Not a file descriptor limit
2020-10-14 17:49:48 <ghoulguy> The -threaded bit is a distraction. Something is generating or corrupting an invalid file descriptor
2020-10-14 17:50:07 <Ariakenom> threaded rts behaves nicer in a bunch of ways, right?
2020-10-14 17:50:14 <merijn> Ariakenom: Yes
2020-10-14 17:50:16 <geekosaur> that said, I agree that does not look to me like it was a valid fd in the first place
2020-10-14 17:50:36 <merijn> Ariakenom: You probably always want the threaded RTS (but with parallel GC disabled! which should become the default sometime soon-ish)
2020-10-14 17:50:37 <monochrom> However, non-threaded RTS asks for periodic SIGVTALRM, this can interfere with syscalls, and after a long story may lead to weird return values.
2020-10-14 17:51:20 <merijn> The parallel GC is bad in most cases, unless you know what you're doing
2020-10-14 17:51:22 <Ariakenom> merijn: soonTM :p. It's been the less buggy ines for years I think
2020-10-14 17:51:32 <Ariakenom> less buggy one
2020-10-14 17:51:40 <ghoulguy> OK, it could be the case that the non-threaded runtime is completely broken and shouldn't be used and in this case error message recommendation coincidentally applies
2020-10-14 17:51:44 <[exa]> like, I shall just compile with threaded and hope that this stops happening (the random nature of this occurring is what scares me), but a better explanation that "please just workaround" would be really cool
2020-10-14 17:52:12 <[exa]> certainly the number is not a valid FD
2020-10-14 17:52:15 → dirediresalt joins (DirefulSal@gateway/vpn/privateinternetaccess/direfulsalt)
2020-10-14 17:53:12 <geekosaur> especially not on a Mac which has rather low fd limits out of the box, unlike linux
2020-10-14 17:53:20 <[exa]> yeah
2020-10-14 17:53:37 <[exa]> "non-threaded runtime is completely broken" oh noes
2020-10-14 17:53:50 <monochrom> Hrm, this can be very evil.
2020-10-14 17:53:53 × GyroW_ quits (~GyroW@unaffiliated/gyrow) (Quit: Someone ate my pie)
2020-10-14 17:53:56 <geekosaur> well, I wouldnt say completely broken, but it's not very stable
2020-10-14 17:54:03 <[exa]> I can collect versions of everything just for reference
2020-10-14 17:54:03 → GyroW joins (~GyroW@d54c03e98.access.telenet.be)
2020-10-14 17:54:03 × GyroW quits (~GyroW@d54c03e98.access.telenet.be) (Changing host)
2020-10-14 17:54:03 → GyroW joins (~GyroW@unaffiliated/gyrow)
2020-10-14 17:54:15 <[exa]> (it's not my box in particular)
2020-10-14 17:54:18 × DirefulSalt quits (DirefulSal@gateway/vpn/privateinternetaccess/direfulsalt) (Ping timeout: 260 seconds)
2020-10-14 17:54:56 <merijn> [exa]: It's not "completely broken", but there's a bunch of libraries that cannot work correctly with it (for example, process)
2020-10-14 17:55:56 <geekosaur> process can't work correctly anyway :/
2020-10-14 17:55:56 <monochrom> Here is the evil possibility. The suggestion "use -threaded" may be a blanket suggestion that applies to other scenerios but not yours.
2020-10-14 17:56:00 <merijn> Just remember "+RTS -qg" :p (or use -with-rtsopts)
2020-10-14 17:57:12 × cheater quits (~user@unaffiliated/cheater) (Quit: (BitchX) Time wasted: all of it)
2020-10-14 17:57:13 × raehik quits (~raehik@cpc96984-rdng25-2-0-cust109.15-3.cable.virginm.net) (Ping timeout: 264 seconds)
2020-10-14 17:57:14 <ghoulguy> Why -qg?
2020-10-14 17:57:33 <geekosaur> disable parallel gc, I think
2020-10-14 17:57:38 <merijn> Disables parallel GC
2020-10-14 17:57:45 <ghoulguy> Yeah, I gathered that much
2020-10-14 17:58:05 → raehik joins (~raehik@cpc96984-rdng25-2-0-cust109.15-3.cable.virginm.net)
2020-10-14 17:58:17 <geekosaur> because that's only useful with specific workloads currently, as already discussed (and comes up in here every couple weeks)
2020-10-14 17:58:25 <merijn> ghoulguy: Because parallel GC often causes *huge* slowdown, an order of magnitude in my own code
2020-10-14 17:59:38 <ghoulguy> Do you know what kinds of workloads it is good for?
2020-10-14 18:00:02 × funkatron quits (~funkatron@185.104.184.43) ()
2020-10-14 18:00:11 → cheater joins (~user@unaffiliated/cheater)
2020-10-14 18:00:52 <merijn> No clue, haven't seen one yet :p
2020-10-14 18:01:15 <merijn> And even when it's good you'll want to explicitly configure how many parallel threads, instead of 1 per capability
2020-10-14 18:01:50 <monochrom> Yeah, I now feel that perhaps the non-threaded RTS was not the cause, rather the messenger. It notices bad FDs more often because of its other limitations. If you switch to the threaded RTS, you are not eliminating the cause, only eliminating a messenger.
2020-10-14 18:03:04 → ensyde joins (~ensyde@2600:1702:2e30:1a40:f09c:a012:516e:b253)
2020-10-14 18:03:21 → conal joins (~conal@64.71.133.70)
2020-10-14 18:03:34 × dyeplexer quits (~lol@unaffiliated/terpin) (Remote host closed the connection)
2020-10-14 18:04:55 × conal quits (~conal@64.71.133.70) (Client Quit)
2020-10-14 18:05:17 × alp quits (~alp@2a01:e0a:58b:4920:35bf:57b0:490a:24ef) (Remote host closed the connection)
2020-10-14 18:05:44 → alp joins (~alp@2a01:e0a:58b:4920:a521:3fa3:56aa:9ff2)
2020-10-14 18:06:39 → conal joins (~conal@64.71.133.70)
2020-10-14 18:06:45 × conal quits (~conal@64.71.133.70) (Client Quit)
2020-10-14 18:07:09 × geekosaur quits (82659a0e@host154-014.vpn.uakron.edu) (Ping timeout: 245 seconds)
2020-10-14 18:07:42 × ensyde quits (~ensyde@2600:1702:2e30:1a40:f09c:a012:516e:b253) (Ping timeout: 260 seconds)
2020-10-14 18:08:01 × raichoo quits (~raichoo@dslb-092-073-214-254.092.073.pools.vodafone-ip.de) (Quit: Lost terminal)
2020-10-14 18:09:20 <nihilazo> is it true that ghc can only cross-compile itself and not other programs? Why?
2020-10-14 18:09:24 <nihilazo> (I'm struggling with compiling stuff on my slow tablet that keeps freezing up)
←Prev  Next→
Page 1 .. 635 636 637 638 639 640 641 642 643 644 645 .. 5022

All times are in UTC.