Newest at the top
| 2025-02-03 18:00:44 +0100 | jespada | (~jespada@2800:a4:2315:e400:dc3a:2af1:31c:f0bc) (Quit: My Mac has gone to sleep. ZZZzzz…) |
| 2025-02-03 17:57:12 +0100 | lxsameer | (~lxsameer@Serene/lxsameer) lxsameer |
| 2025-02-03 17:55:29 +0100 | tzh | (~tzh@c-76-115-131-146.hsd1.or.comcast.net) tzh |
| 2025-02-03 17:43:23 +0100 | dolio | (~dolio@130.44.140.168) dolio |
| 2025-02-03 17:42:11 +0100 | dolio | (~dolio@130.44.140.168) (Client Quit) |
| 2025-02-03 17:41:56 +0100 | urdh_ | urdh |
| 2025-02-03 17:41:56 +0100 | urdh | (~urdh@user/urdh) (Ping timeout: 272 seconds) |
| 2025-02-03 17:40:29 +0100 | urdh_ | (~urdh@user/urdh) urdh |
| 2025-02-03 17:40:17 +0100 | dolio | (~dolio@130.44.140.168) dolio |
| 2025-02-03 17:34:32 +0100 | dolio | (~dolio@130.44.140.168) (Quit: ZNC 1.9.1 - https://znc.in) |
| 2025-02-03 17:26:51 +0100 | alfiee | (~alfiee@user/alfiee) (Ping timeout: 252 seconds) |
| 2025-02-03 17:26:14 +0100 | agent314 | (~quassel@146.70.172.177) (Ping timeout: 244 seconds) |
| 2025-02-03 17:24:03 +0100 | <Athas> | (I did not look very diligently, however.) |
| 2025-02-03 17:23:45 +0100 | <Athas> | The only reason I didn't disable it is because I can't figure out how to override the GHC configuration in Nixpkgs. |
| 2025-02-03 17:23:23 +0100 | <Athas> | I assume it's the reason GHC produces better stack traces now, but I don't need that for my release executables anyway. |
| 2025-02-03 17:22:44 +0100 | alfiee | (~alfiee@user/alfiee) alfiee |
| 2025-02-03 17:22:40 +0100 | <tomsmeding> | ah |
| 2025-02-03 17:22:32 +0100 | <Athas> | tomsmeding: it is a dependency of the GHC RTS. |
| 2025-02-03 17:21:23 +0100 | <tomsmeding> | Athas: I think dminuoso's point was that while elfutils is a dependency of GHC, statically linking your program should not need a statically-linked GHC, and hence should not need statically linking elfutils in the first place |
| 2025-02-03 17:18:33 +0100 | euleritian | (~euleritia@ip4d17fae8.dynamic.kabel-deutschland.de) |
| 2025-02-03 17:18:15 +0100 | euleritian | (~euleritia@dynamic-176-006-146-039.176.6.pool.telefonica.de) (Read error: Connection reset by peer) |
| 2025-02-03 17:14:40 +0100 | <tomsmeding> | doesn't appear in any of the 9.6.* release notes |
| 2025-02-03 17:12:23 +0100 | <tomsmeding> | O.o |
| 2025-02-03 17:12:16 +0100 | <Athas> | Maybe not anywhere? My understanding is that GHC-with-DWARF is the default since 9.6. |
| 2025-02-03 17:12:00 +0100 | <tomsmeding> | ah. |
| 2025-02-03 17:11:57 +0100 | <Athas> | Not in Nixpkgs. |
| 2025-02-03 17:11:50 +0100 | <tomsmeding> | (isn't the GHC-without-DWARF the _default_?) |
| 2025-02-03 17:11:45 +0100 | <Athas> | I don't care about the linking of GHC. I want to statically link my Haskell program. |
| 2025-02-03 17:11:34 +0100 | <Athas> | Tom Sydney mentions that you can compile GHC without DWARF support, and I guess that's what I'll eventually have to figure out how to do. |
| 2025-02-03 17:11:24 +0100 | <dminuoso> | Which one is it? |
| 2025-02-03 17:11:20 +0100 | <dminuoso> | But what you just said suggests you want a statically linked GHC binary |
| 2025-02-03 17:11:19 +0100 | <tomsmeding> | heh |
| 2025-02-03 17:11:08 +0100 | <dminuoso> | 15:59:23 Athas │ tomsmeding: the latter, sorry. |
| 2025-02-03 17:11:01 +0100 | <dminuoso> | Athas: Hold on, Im confused. |
| 2025-02-03 17:10:56 +0100 | <Athas> | tomsmeding: yes. |
| 2025-02-03 17:10:49 +0100 | <tomsmeding> | Athas: then what's nix necessary for? Getting static versions of a bunch of C libraries you need? |
| 2025-02-03 17:10:43 +0100 | <Athas> | dminuoso: I have to, don't I? GHC depends on libdw and libdw depends on elfutils. |
| 2025-02-03 17:10:29 +0100 | <dminuoso> | Sorry for the double question mark. |
| 2025-02-03 17:10:23 +0100 | <dminuoso> | Athas: So you're trying to statically link elfutils into your haskell binary?? |
| 2025-02-03 17:10:03 +0100 | <carbolymer> | tomsmeding: thanks, that lambda trick works. I can never wrap my brain around mixing viewpatterns and regular haskell syntax |
| 2025-02-03 17:10:01 +0100 | <Athas> | I even link statically with glibc! It works fine if you understand the caveats. |
| 2025-02-03 17:09:43 +0100 | <Athas> | tomsmeding: I am linking all-statically. I think it is just that static linking is not fashionable, so it is not a well tested path. That is probably why something broke in elfutils. |
| 2025-02-03 17:09:25 +0100 | <tomsmeding> | (hence needing to statically link libc if you want to statically link anything, and you can't statically link glibc, hence you need to use alpine with musl or something?) |
| 2025-02-03 17:08:59 +0100 | <tomsmeding> | what is the source of all this madness again? That system linkers are unable to statically link some libraries in, but keep some others as dynamic instead? |
| 2025-02-03 17:08:29 +0100 | <dminuoso> | Haskell and nix are not fun together, in my experience. :( |
| 2025-02-03 17:08:11 +0100 | <Athas> | It's a land of contrasts and hypocrisy. |
| 2025-02-03 17:08:00 +0100 | <Athas> | (But not so much that I bump nixpkgs all that often.) |
| 2025-02-03 17:07:59 +0100 | <dminuoso> | Yeah fair enough. |
| 2025-02-03 17:07:49 +0100 | <Athas> | dminuoso: yeah, but I actually care very much about my code working with precisely those versions. |
| 2025-02-03 17:07:29 +0100 | <dminuoso> | Athas: Also regarding interference, this is not about niv but rather nixpkgs pinning hackage versions to some arbitrary stuff that mostly works together, but sometimes does not. |