Newest at the top
| 2025-01-24 18:27:51 +0100 | Googulator7 | (~Googulato@20014C4D1DC76E0020B6DF4FD38204F1.dsl.pool.telekom.hu) (Client Quit) |
| 2025-01-24 18:26:59 +0100 | alfiee | (~alfiee@user/alfiee) alfiee |
| 2025-01-24 18:26:47 +0100 | <hellwolf> | but I think I am letting it go; I digressed too much. |
| 2025-01-24 18:26:40 +0100 | <hellwolf> | *no |
| 2025-01-24 18:26:37 +0100 | <hellwolf> | now idea how to lift >>> show an unlifted type |
| 2025-01-24 18:26:34 +0100 | Pixi__ | Pixi |
| 2025-01-24 18:26:23 +0100 | <hellwolf> | print ((# (wordToWord8# 8##), (wordToWord8# 9##) #) :: F#) ... |
| 2025-01-24 18:25:43 +0100 | <hellwolf> | losing arms and legs using ghc |
| 2025-01-24 18:25:25 +0100 | Googulator7 | (~Googulato@20014C4D1DC76E0020B6DF4FD38204F1.dsl.pool.telekom.hu) |
| 2025-01-24 18:25:21 +0100 | <mauke> | what happened to the left arm? (does your irc client speak markdown?) |
| 2025-01-24 18:22:33 +0100 | Everything | (~Everythin@195.138.86.118) (Ping timeout: 252 seconds) |
| 2025-01-24 18:22:24 +0100 | <hellwolf> | ¯_(ツ)_/¯ |
| 2025-01-24 18:22:11 +0100 | <hellwolf> | well, the solution to all your problems -XUnboxedTuples |
| 2025-01-24 18:21:17 +0100 | <hellwolf> | let me read some docs |
| 2025-01-24 18:21:06 +0100 | <hellwolf> | yes |
| 2025-01-24 18:20:54 +0100 | <geekosaur> | did you turn on MagicHash? |
| 2025-01-24 18:20:03 +0100 | <hellwolf> | " Operator applied to too few arguments: #" |
| 2025-01-24 18:18:46 +0100 | <geekosaur> | (with commas; note that with | it's unboxed sum) |
| 2025-01-24 18:18:42 +0100 | tzh | (~tzh@c-76-115-131-146.hsd1.or.comcast.net) |
| 2025-01-24 18:18:27 +0100 | <geekosaur> | yes |
| 2025-01-24 18:18:12 +0100 | <hellwolf> | is it `(# ... #)` |
| 2025-01-24 18:18:03 +0100 | <hellwolf> | what's the syntax for unboxed tuple anyways |
| 2025-01-24 18:16:40 +0100 | <geekosaur> | although maybe 9.10.2-to-be already has it fixed |
| 2025-01-24 18:16:24 +0100 | <geekosaur> | interesting. I'd report that |
| 2025-01-24 18:15:36 +0100 | <hellwolf> | wrong code, I guess. |
| 2025-01-24 18:15:32 +0100 | <hellwolf> | crash GHC 9.10, but meaningful error in 9.8 |
| 2025-01-24 18:15:23 +0100 | <hellwolf> | https://play.haskell.org/saved/YjifSESh |
| 2025-01-24 18:14:22 +0100 | kuribas | (~user@ip-188-118-57-242.reverse.destiny.be) (Remote host closed the connection) |
| 2025-01-24 18:12:18 +0100 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) (Ping timeout: 252 seconds) |
| 2025-01-24 18:11:13 +0100 | <geekosaur> | (the bytecode interpreter can't tell the difference between boxed and unboxed tuples, specifically) |
| 2025-01-24 18:10:50 +0100 | <geekosaur> | requires 9.8 or later; older versions crash with unboxed tuples |
| 2025-01-24 18:10:18 +0100 | <geekosaur> | it's supposed to in recent versions but you need to turn on a flag (-fno-unoptimized-core-for-interpreter, I think) |
| 2025-01-24 18:10:10 +0100 | Googulator7 | (~Googulato@81.183.235.203) (Ping timeout: 240 seconds) |
| 2025-01-24 18:09:45 +0100 | <merijn> | hellwolf: You can try to start ghci with -fobject-code |
| 2025-01-24 18:09:16 +0100 | <hellwolf> | I proceed to crash https://play.haskell.org/ then |
| 2025-01-24 18:09:10 +0100 | <lambdabot> | [86,55,164,116,47] |
| 2025-01-24 18:09:08 +0100 | <homo> | > let f i = if i < 256 then [i] else (mod i 256) : f (div i 256) in f 203820382038 |
| 2025-01-24 18:08:57 +0100 | <lambdabot> | [86,55,164,116,47] |
| 2025-01-24 18:08:55 +0100 | <homo> | > let f i = if i < 256 then [i] else uncurry (flip (:)) . (\(a,b) -> (f a, b)) $ quotRem i 256 in f 203820382038 |
| 2025-01-24 18:08:08 +0100 | vanishingideal | (~vanishing@user/vanishingideal) vanishingideal |
| 2025-01-24 18:08:08 +0100 | <haskellbridge> | <sm> Don't Panic, GHC |
| 2025-01-24 18:08:04 +0100 | <merijn> | hellwolf: I'm not sure bytecode can support unboxed tuples |
| 2025-01-24 18:07:51 +0100 | <merijn> | hellwolf: ghci doesn't compile to machine code, it compiles to bytecode that's interpreted |
| 2025-01-24 18:07:39 +0100 | <haskellbridge> | <sm> six mentions of Call Stack, making clear there was panic! |
| 2025-01-24 18:07:28 +0100 | <merijn> | hellwolf: ah, you can't do that in ghci I expect |
| 2025-01-24 18:07:13 +0100 | <homo> | merijn the reason to not use quotRem and uncurry is because resulting code becomes longer than it is with div and mod/rem |
| 2025-01-24 18:07:01 +0100 | <merijn> | Congrats! |
| 2025-01-24 18:06:43 +0100 | vanishingideal | (~vanishing@user/vanishingideal) (Ping timeout: 264 seconds) |
| 2025-01-24 18:05:46 +0100 | <hellwolf> | I don't know what I am doing, but I crashed GHC |
| 2025-01-24 18:05:39 +0100 | <hellwolf> | https://paste.tomsmeding.com/0xsGQoyr |