Logs: freenode/#haskell
| 2021-03-05 01:49:09 | × | alar[m] quits (alarmxalat@gateway/shell/matrix.org/x-bpvylpwwtlcejjwj) (Ping timeout: 272 seconds) |
| 2021-03-05 01:49:09 | × | MrMuffles[m] quits (mrmufflesm@gateway/shell/matrix.org/x-cbeqjcjrdmhenexl) (Ping timeout: 272 seconds) |
| 2021-03-05 01:49:15 | × | mly[m] quits (mlydisenco@gateway/shell/matrix.org/x-qjhvevhpltpfoqbc) (Ping timeout: 240 seconds) |
| 2021-03-05 01:49:20 | × | alexfmpe quits (alexfmpema@gateway/shell/matrix.org/x-girzcdcudxbpysje) (Ping timeout: 244 seconds) |
| 2021-03-05 01:49:20 | × | maerwald[m] quits (maerwaldma@gateway/shell/matrix.org/x-bjunztlamgbiqsmk) (Ping timeout: 244 seconds) |
| 2021-03-05 01:49:22 | × | kadoban quits (kadobanemp@gateway/shell/matrix.org/x-orwvobvmsxoskufi) (Ping timeout: 258 seconds) |
| 2021-03-05 01:49:26 | × | invent[m] quits (inventmatr@gateway/shell/matrix.org/x-vckgemxonksixvml) (Ping timeout: 240 seconds) |
| 2021-03-05 01:49:26 | × | heck-to-the-gnom quits (heck-to-th@gateway/shell/matrix.org/x-lpqeckkflomaivst) (Ping timeout: 240 seconds) |
| 2021-03-05 01:49:45 | × | fgaz quits (fgazmatrix@gateway/shell/matrix.org/x-altyhdajrdgkwiga) (Ping timeout: 258 seconds) |
| 2021-03-05 01:50:01 | → | jpds joins (~jpds@gateway/tor-sasl/jpds) |
| 2021-03-05 01:50:11 | → | alunduil joins (alunduil@gateway/web/irccloud.com/x-vzlleahqqgvhcqsg) |
| 2021-03-05 01:51:07 | → | luigy joins (~luigy@104.236.106.229) |
| 2021-03-05 01:51:38 | × | Guest74820 quits (~tristan@luna.whatbox.ca) (Ping timeout: 260 seconds) |
| 2021-03-05 01:52:12 | → | Tristan joins (~tristan@luna.whatbox.ca) |
| 2021-03-05 01:52:36 | Tristan | is now known as Guest41116 |
| 2021-03-05 01:53:29 | → | rdivyanshu joins (uid322626@gateway/web/irccloud.com/x-mlrllmnzmzixcljr) |
| 2021-03-05 01:56:38 | → | gitgood joins (~gitgood@82-132-217-85.dab.02.net) |
| 2021-03-05 01:56:57 | → | tromp joins (~tromp@dhcp-077-249-230-040.chello.nl) |
| 2021-03-05 01:57:20 | <ezzieyguywuf> | can I set ghc-options in cabal.project so it applies to all packages? |
| 2021-03-05 01:57:31 | <ezzieyguywuf> | or do I have to set it in each cabal file separately |
| 2021-03-05 01:58:31 | → | hyiltiz-M joins (hyiltizkde@gateway/shell/kde/matrix/x-ziqtfceeiaptadfr) |
| 2021-03-05 01:59:22 | → | VarikValefor[m] joins (varikvalef@gateway/shell/matrix.org/x-pwtjczbvhoktmglx) |
| 2021-03-05 01:59:23 | × | ukari quits (~ukari@unaffiliated/ukari) (Remote host closed the connection) |
| 2021-03-05 01:59:39 | × | jespada quits (~jespada@90.254.243.187) (Ping timeout: 245 seconds) |
| 2021-03-05 01:59:47 | → | ep1ctetus joins (~epictetus@ip72-194-215-136.sb.sd.cox.net) |
| 2021-03-05 01:59:47 | × | ep1ctetus quits (~epictetus@ip72-194-215-136.sb.sd.cox.net) (Client Quit) |
| 2021-03-05 01:59:51 | <infinisil> | ezzieyguywuf: You can specify `package <package>\nghc-options: ...` in cabal.project for it to apply to all *components* of that package at least |
| 2021-03-05 02:00:18 | → | conal joins (~conal@64.71.133.70) |
| 2021-03-05 02:00:22 | <monochrom> | "package *" is legal and is a wildcard. |
| 2021-03-05 02:00:31 | <infinisil> | Oh nice |
| 2021-03-05 02:00:31 | → | ukari joins (~ukari@unaffiliated/ukari) |
| 2021-03-05 02:00:33 | <ezzieyguywuf> | I could do the same with a `common` stanza in the packages cabal file though, and if I have to do it once per package anyway this is probably the more correct way |
| 2021-03-05 02:00:40 | <ezzieyguywuf> | hrm.... |
| 2021-03-05 02:01:35 | <ezzieyguywuf> | is it `package *\n` or `package *:\n`? |
| 2021-03-05 02:01:36 | × | tromp quits (~tromp@dhcp-077-249-230-040.chello.nl) (Ping timeout: 246 seconds) |
| 2021-03-05 02:01:57 | → | dbmikus joins (~dbmikus@cpe-76-167-86-219.natsow.res.rr.com) |
| 2021-03-05 02:02:07 | <infinisil> | You'll get an error if you use the wrong one :) |
| 2021-03-05 02:02:16 | → | jespada joins (~jespada@90.254.243.187) |
| 2021-03-05 02:02:40 | <ezzieyguywuf> | lol |
| 2021-03-05 02:02:52 | <ezzieyguywuf> | indeed I did |
| 2021-03-05 02:04:10 | <ezzieyguywuf> | hrm, why do only my test-suit stanzas have "ghc-options: -threaded -rtsopts -with-rtsopts=-N", I have a feeling this came from `cabal init` but I don't knw what they do |
| 2021-03-05 02:05:08 | → | apache8080 joins (~rishi@wsip-70-168-153-252.oc.oc.cox.net) |
| 2021-03-05 02:05:26 | <Clint> | so that they'll break on all platforms without -N |
| 2021-03-05 02:05:55 | <ezzieyguywuf> | so that's a good thing? |
| 2021-03-05 02:06:15 | × | ericsagnes quits (~ericsagne@2405:6580:0:5100:42a8:8650:90d4:54bc) (Ping timeout: 240 seconds) |
| 2021-03-05 02:06:25 | <monochrom> | No colon. Just newline. |
| 2021-03-05 02:06:32 | <ezzieyguywuf> | is there a reason it's only needed on the test-suite though? |
| 2021-03-05 02:06:42 | <monochrom> | I mean for "package foo" and "package *" |
| 2021-03-05 02:06:42 | <Clint> | it's not needed at all, it's largely an annoyance |
| 2021-03-05 02:06:59 | <MarcelineVQ> | Which 'it'? |
| 2021-03-05 02:07:00 | × | dbmikus quits (~dbmikus@cpe-76-167-86-219.natsow.res.rr.com) (Ping timeout: 258 seconds) |
| 2021-03-05 02:07:20 | → | SupaYoshii joins (~supayoshi@213-10-140-13.fixed.kpn.net) |
| 2021-03-05 02:07:26 | × | lambdabot quits (~lambdabot@haskell/bot/lambdabot) (Ping timeout: 265 seconds) |
| 2021-03-05 02:07:36 | <monochrom> | I don't think "cabal init" auto-gens any ghc-options at all. |
| 2021-03-05 02:07:42 | <ezzieyguywuf> | Clint: great, I'll get rid of it. |
| 2021-03-05 02:07:51 | <ezzieyguywuf> | monochrom: I must have gotten it from the interwebs then |
| 2021-03-05 02:07:55 | <MarcelineVQ> | those particular options are standard with new stack projects however |
| 2021-03-05 02:08:02 | <ezzieyguywuf> | I STARTED WITH STACK! |
| 2021-03-05 02:08:04 | × | SupaYoshi quits (~supayoshi@213.10.140.13) (Ping timeout: 260 seconds) |
| 2021-03-05 02:08:04 | × | gentauro quits (~gentauro@unaffiliated/gentauro) (Ping timeout: 260 seconds) |
| 2021-03-05 02:08:05 | <ezzieyguywuf> | so that's probably why |
| 2021-03-05 02:08:42 | → | gentauro joins (~gentauro@unaffiliated/gentauro) |
| 2021-03-05 02:08:45 | → | waleee-cl joins (uid373333@gateway/web/irccloud.com/x-jksqnlfrjzjlkhlm) |
| 2021-03-05 02:08:54 | → | lambdabot joins (~lambdabot@haskell/bot/lambdabot) |
| 2021-03-05 02:10:11 | → | nbloomf joins (~nbloomf@2600:1700:ad14:3020:ad70:b790:2132:9c51) |
| 2021-03-05 02:11:22 | <MarcelineVQ> | -threaded uses the threaded (as opposed to non-threaded) rts, -rtsopts allows extra rts options to be specifies at runtime you wouldn't be able to otherwise, -with-rtsopts=-N passes the rts the -N flag which says to create as many capabilities as the pc has cores available, one per core. capabilities are little virtual machines that manage ghc's threads |
| 2021-03-05 02:11:31 | → | themsay[m] joins (themsaymat@gateway/shell/matrix.org/x-hbatkioaxwxoyeob) |
| 2021-03-05 02:12:05 | <MarcelineVQ> | I'm unsure what Clint is referring to about platform breaking or which particular annoyances |
| 2021-03-05 02:12:09 | <ezzieyguywuf> | hrm, that all sounds actually useful |
| 2021-03-05 02:12:18 | → | lambdaclan joins (lambdaclan@gateway/shell/matrix.org/x-sallzvkvkldcgkut) |
| 2021-03-05 02:12:46 | <waleee-cl> | I'm trying to apply this https://github.com/llvm-hs/llvm-hs/issues/315#issuecomment-755229982 in a stack.yaml , but during the build it seems to be parsed as xx=-std=c++11 instead of -optcxx=-std=c++11 . Is it malformed in some way ? |
| 2021-03-05 02:13:02 | <MarcelineVQ> | it's useful if your progam uses libs that take advantage of threads, common libs such as async for example |
| 2021-03-05 02:13:14 | → | the_1_[m] joins (the1matrix@gateway/shell/matrix.org/x-ydnkooibksztllyy) |
| 2021-03-05 02:13:14 | → | srid joins (sridmatrix@gateway/shell/matrix.org/x-naqkszpohzlsbubr) |
| 2021-03-05 02:13:53 | <Clint> | MarcelineVQ: if you're on a platform where ghc doesn't "Support SMP" it will fail to build and most testsuites don't actually seem to take advantage of threading |
| 2021-03-05 02:14:13 | × | Sheilong quits (uid293653@gateway/web/irccloud.com/x-uuniqtfysevxnvsh) () |
| 2021-03-05 02:14:41 | → | boistordu joins (boistordum@gateway/shell/matrix.org/x-epqhlcgltbmupbxi) |
| 2021-03-05 02:14:57 | → | JaakkoLuttinen[m joins (jluttinema@gateway/shell/matrix.org/x-sivibcbidnslmvhv) |
| 2021-03-05 02:15:20 | → | speakerspivakeem joins (speakerdea@gateway/shell/matrix.org/x-grrchxbekhuqojog) |
| 2021-03-05 02:15:21 | <infinisil> | Hmm, why is -N not the default? |
| 2021-03-05 02:15:22 | → | oreoking[m] joins (oreokingma@gateway/shell/matrix.org/x-sfbsvckrwskkmqtb) |
| 2021-03-05 02:15:36 | → | domenkozar[m] joins (domenkozar@NixOS/user/domenkozar) |
| 2021-03-05 02:15:46 | → | Lurkki[m] joins (lurkkipriv@gateway/shell/matrix.org/x-xzyglswxynebzkit) |
| 2021-03-05 02:15:50 | → | alecs[m] joins (malumorema@gateway/shell/matrix.org/x-dlpnoymfcpnetnsl) |
| 2021-03-05 02:16:05 | <shapr> | infinisil: I think there's a ghc proposal to make it the default |
| 2021-03-05 02:16:07 | → | Jd007 joins (~Jd007@162.156.11.151) |
| 2021-03-05 02:16:14 | → | Hanma[m] joins (hanmamatri@gateway/shell/matrix.org/x-jgikxfrvozhpbsoi) |
| 2021-03-05 02:16:15 | → | ManofLetters[m] joins (manoflette@gateway/shell/matrix.org/x-ntjjesgcogujguzg) |
| 2021-03-05 02:16:30 | → | toasty_avocado[m joins (toastyavoc@gateway/shell/matrix.org/x-wkhjyjmxecrptsuj) |
| 2021-03-05 02:17:17 | → | mjlbach joins (atriusmatr@gateway/shell/matrix.org/x-ietoxhbonvcmpxmf) |
| 2021-03-05 02:17:50 | <infinisil> | Oh there's https://github.com/ghc-proposals/ghc-proposals/pull/240 |
| 2021-03-05 02:18:08 | <infinisil> | Though it's only -threaded by default, not -N (and gives a reason why, because of bad performance) |
| 2021-03-05 02:18:15 | → | ericsagnes joins (~ericsagne@2405:6580:0:5100:ce0d:4010:b248:c5be) |
| 2021-03-05 02:18:37 | <monochrom> | For most of the years since -N was first invented, people found that it performs worse than, say, -N($ncpus - constant). |
| 2021-03-05 02:19:56 | <infinisil> | Is there an outlook for improving multithreading? |
| 2021-03-05 02:20:21 | <monochrom> | Yes, IIRC the culprit is GC, and it's worked on. Maybe even already solved. |
| 2021-03-05 02:20:22 | <infinisil> | Or is it a matter of carefully optimizing the code and GC |
| 2021-03-05 02:20:50 | <infinisil> | Nice |
| 2021-03-05 02:21:35 | <monochrom> | I think either 8.10 or 9.0 is better, I forgot which, I saw something along that line in a release note. |
| 2021-03-05 02:22:18 | → | chenshen joins (~chenshen@2620:10d:c090:400::5:dc3c) |
All times are in UTC.