--- Log opened Tue Apr 22 00:00:22 2025
00:16 -!- vasild [~vd@user/vasild] has joined #bitcoin-core-builds
04:00 -!- gribble [~gribble@bitcoin/bot/gribble] has quit [Remote host closed the connection]
04:23 -!- gribble [~gribble@bitcoin/bot/gribble] has joined #bitcoin-core-builds
04:29 -!- gribble [~gribble@bitcoin/bot/gribble] has quit [Remote host closed the connection]
04:55 -!- gribble [~gribble@bitcoin/bot/gribble] has joined #bitcoin-core-builds
06:28 -!- pablomartin [~pablomart@89.35.30.158] has joined #bitcoin-core-builds
06:30 -!- gribble [~gribble@bitcoin/bot/gribble] has quit [Remote host closed the connection]
06:50 -!- gribble [~gribble@bitcoin/bot/gribble] has joined #bitcoin-core-builds
07:13 < willcl-ark> hebasto: I'm curious if a) there is any reason we dont currently set CMAKE_PREFIX_PATH in the depends toolchain, and b) whether there would be any opposition to doing so? Currently on NixOS, because of how it mangles certain CMAKE vars (sigh), it's a bit tricky to build depends and successfully use the generated toolchain. However, appending the depends/triplet path to
07:13 < willcl-ark> CMAKE_PREFIX_PATH in the toolchain file fixes it up pretty nicely. i.e. this patch: https://paste.256k1.dev/rommenjurr.patch
07:15 -!- pablomartin [~pablomart@89.35.30.158] has quit [Ping timeout: 248 seconds]
07:16 < hebasto> while working on the staging branch, I did consider using `CMAKE_PREFIX_PATH`. I don't remember the exact reason why I ended with  using only `CMAKE_FIND_ROOT_PATH`
07:18 < hebasto> willcl-ark: also please notice that the toolchain file is executed multiple times, so `list(APPEND ...)` is not great
07:19 < hebasto> what exactly is not working on the master branch?
07:19 < willcl-ark> Oh, thanks. That is mildly irritating :P
07:19 < willcl-ark> I think CMAKE_FIND_ROOT_PATH might be preferred in cross-compiling environments
07:25 < willcl-ark> Well the NixOS cmake package has to patch cmake with it's non-standard paths (see e.g. patches here: https://github.com/NixOS/nixpkgs/tree/nixos-24.11/pkgs/by-name/cm/cmake) and some of this appears to override the toolchain! So if, in a "pure" nix shell I build depends, then try and use the toolchain, I will see that my just-built packages are not found. I don't consider this a
07:25 < willcl-ark> "problem with master", but an OS quirk, and I'm not quite sure if I should try and fix in my environment or in the toolchain. I was hop
07:26 < willcl-ark> ...ing I might be able to fix cleanly using the toolchain file, but perhaps it's not as easy as I thought
08:16 -!- pablomartin [~pablomart@89.35.30.159] has joined #bitcoin-core-builds
09:04 -!- pablomartin [~pablomart@89.35.30.159] has quit [Ping timeout: 276 seconds]
09:09 -!- pablomartin [~pablomart@89.35.30.158] has joined #bitcoin-core-builds
09:45 -!- pablomartin [~pablomart@89.35.30.158] has quit [Ping timeout: 245 seconds]
12:53 -!- pablomartin [~pablomart@89.35.30.155] has joined #bitcoin-core-builds
12:58 -!- pablomartin [~pablomart@89.35.30.155] has quit [Quit: Leaving]
13:17 -!- gribble [~gribble@bitcoin/bot/gribble] has quit [Read error: Connection reset by peer]
13:36 -!- gribble [~gribble@bitcoin/bot/gribble] has joined #bitcoin-core-builds
18:34 -!- gribble [~gribble@bitcoin/bot/gribble] has quit [Remote host closed the connection]
18:42 -!- gribble [~gribble@bitcoin/bot/gribble] has joined #bitcoin-core-builds
21:47 -!- darosior [~darosior@109.205.214.46] has quit [Quit: Ping timeout (120 seconds)]
21:48 -!- darosior [~darosior@109.205.214.46] has joined #bitcoin-core-builds
22:00 -!- darosior [~darosior@109.205.214.46] has quit [Quit: Ping timeout (120 seconds)]
22:01 -!- darosior [~darosior@109.205.214.46] has joined #bitcoin-core-builds
--- Log closed Wed Apr 23 00:00:24 2025