Newest at the top
| 2025-01-11 20:46:42 +0100 | tnt2 | (~Thunderbi@user/tnt1) tnt1 |
| 2025-01-11 20:44:18 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds) |
| 2025-01-11 20:43:06 +0100 | tnt2 | tnt1 |
| 2025-01-11 20:43:06 +0100 | tnt1 | (~Thunderbi@user/tnt1) (Ping timeout: 276 seconds) |
| 2025-01-11 20:42:32 +0100 | tnt2 | (~Thunderbi@user/tnt1) tnt1 |
| 2025-01-11 20:39:57 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2025-01-11 20:39:32 +0100 | tnt2 | (~Thunderbi@user/tnt1) (Ping timeout: 252 seconds) |
| 2025-01-11 20:38:23 +0100 | tnt1 | (~Thunderbi@user/tnt1) tnt1 |
| 2025-01-11 20:35:30 +0100 | tnt1 | (~Thunderbi@user/tnt1) (Ping timeout: 252 seconds) |
| 2025-01-11 20:34:58 +0100 | tnt2 | (~Thunderbi@user/tnt1) tnt1 |
| 2025-01-11 20:32:02 +0100 | machinedgod | (~machinedg@d108-173-18-100.abhsia.telus.net) machinedgod |
| 2025-01-11 20:31:05 +0100 | lxsameer | (~lxsameer@Serene/lxsameer) (Ping timeout: 252 seconds) |
| 2025-01-11 20:28:54 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 260 seconds) |
| 2025-01-11 20:27:43 +0100 | <monochrom> | Yeah that too. |
| 2025-01-11 20:27:30 +0100 | <c_wraith> | the recent optimizations doing join point analysis really help out with specific types of code, too. |
| 2025-01-11 20:26:39 +0100 | <monochrom> | And my favourite example linked above, some of you asked "why doesn't GHC take one step further and just generate 'eval = fval'?" Guess what, these days it does! :) |
| 2025-01-11 20:26:19 +0100 | swistak | (~swistak@185.21.216.141) (Ping timeout: 252 seconds) |
| 2025-01-11 20:26:18 +0100 | Sgeo | (~Sgeo@user/sgeo) Sgeo |
| 2025-01-11 20:25:43 +0100 | <c_wraith> | (when both the lens and the operation are statically known) |
| 2025-01-11 20:24:47 +0100 | <c_wraith> | lens also depends on case-of-case to be efficient without tons of rewrite rules. |
| 2025-01-11 20:23:31 +0100 | <monochrom> | The vector library's stream fusion actually relies on that fundamentally. |
| 2025-01-11 20:22:37 +0100 | <monochrom> | And there is also one where if you have "case (case y of x:_ -> Just x) of Just z -> foo" the middle Just can also be eliminated. |
| 2025-01-11 20:21:56 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2025-01-11 20:21:40 +0100 | <monochrom> | There is a fairly general scheme where if you have "case m of Just z -> ... case m of Just z again -> ..." the 2nd conditional branching can be eliminated. |
| 2025-01-11 20:20:49 +0100 | <darkling> | Needle nardle noo. |
| 2025-01-11 20:20:45 +0100 | acidjnk | (~acidjnk@p200300d6e7283f90c8dc7c78c19bd00e.dip0.t-ipconnect.de) acidjnk |
| 2025-01-11 20:19:41 +0100 | <monochrom> | My favourite example is https://mail.haskell.org/pipermail/haskell-cafe/2013-April/107775.html and it's already last decade. Imagine what it can do today. :) |
| 2025-01-11 20:17:11 +0100 | <haskellbridge> | <Bowuigi> Interesting, GHC is actually very smart |
| 2025-01-11 20:16:33 +0100 | <bailsman> | so far as I can understand - I'm not an expert in reading the simpl output - it has entirely optimized out my setfield calls to just creating a new record passing in values from the original values by direct field access and/or the values that I passed in, which seems optimal. No mention of generic anything. |
| 2025-01-11 20:16:11 +0100 | <monochrom> | I haven't seen an example, but it is said that O2 can be slower than O1 for some code, hence I analogize it to gcc's O3. |
| 2025-01-11 20:14:37 +0100 | <monochrom> | OK vector is one of the few exceptions where O2 is necessary. |
| 2025-01-11 20:13:48 +0100 | <bailsman> | I don't actually really know what the difference is. |
| 2025-01-11 20:13:35 +0100 | <bailsman> | BEcause when I did my microbenchmarks with the storable vectors -O2 was much much faster, I believe because it was specializing better |
| 2025-01-11 20:12:53 +0100 | <monochrom> | And yeah, actually ghc : 1 : 2 :: gcc : 2 : 3 |
| 2025-01-11 20:12:46 +0100 | <c_wraith> | (most code is compiled at -O1, as it's the default and hackage even warns you about uploading -O2) |
| 2025-01-11 20:12:08 +0100 | <monochrom> | Yeah the -O0 code is also suspiciously much shorter. >:) |
| 2025-01-11 20:12:07 +0100 | <c_wraith> | why not -O1? that's the most relevant level |
| 2025-01-11 20:12:02 +0100 | <haskellbridge> | <Bowuigi> Check for calls to the generic methods instead |
| 2025-01-11 20:11:32 +0100 | <haskellbridge> | <Bowuigi> They might be inlined but still there |
| 2025-01-11 20:11:02 +0100 | sprotte24 | (~sprotte24@p200300d16f109200d837c3a3f02cbb16.dip0.t-ipconnect.de) |
| 2025-01-11 20:10:56 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds) |
| 2025-01-11 20:10:43 +0100 | <bailsman> | OK I checked -ddump-simpl output, comparing -O0 and -O2. In -O0, there is "setField" all over the place. In -O2, there's no setField calls anywhere. Can I conclude these are actually getting optimized out, despite being generic? |
| 2025-01-11 20:10:35 +0100 | <EvanR> | haskell script to generate haskell |
| 2025-01-11 20:08:10 +0100 | <monochrom> | If someone else has already written the TH version, of course you can just use it and do nothing. But if no one has written either version and you have to write one yourself, the generics route is easier to write. |
| 2025-01-11 20:06:51 +0100 | <monochrom> | The GHC.Generics route has a linear-time runtime cost (to and from clones your input), whereas the TH route generates direct code that skips that. To be sure, I am lazy so I just use the generics route and not care about performance. :) |
| 2025-01-11 20:06:27 +0100 | tromp | (~textual@92-110-219-57.cable.dynamic.v4.ziggo.nl) |
| 2025-01-11 20:06:10 +0100 | acidjnk | (~acidjnk@p200300d6e7283f90c8dc7c78c19bd00e.dip0.t-ipconnect.de) (Ping timeout: 272 seconds) |
| 2025-01-11 20:05:33 +0100 | ljdarj | (~Thunderbi@user/ljdarj) ljdarj |
| 2025-01-11 20:05:22 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2025-01-11 19:59:28 +0100 | pavonia | (~user@user/siracusa) siracusa |