!djTaTBQyWEPRQxrPTb:nixos.org

Nixpkgs Architecture Team

224 Members
https://github.com/nixpkgs-architecture, weekly public meetings on Wednesday 15:00-16:00 UTC at https://meet.jit.si/nixpkgs-architecture54 Servers

Load older messages


SenderMessageTime
11 Jun 2024
@Minijackson:matrix.orgMinijackson
In reply to @aliarokapis:matrix.org
Probably only works for c strings
even then, depending on compiler optimisations and other factors, the length of the string might be assumed
12:45:33
@aliarokapis:matrix.orgAlexandros LiarokapisPadding with null strings is also possible12:45:34
@aliarokapis:matrix.orgAlexandros Liarokapis* Padding with null chars is also possible12:45:49
@Minijackson:matrix.orgMinijackson
In reply to @aliarokapis:matrix.org
Padding with null chars is also possible
not sure, if the compiler optimises printf("some string") to write(stdout, "some string", 11), replacing some string with something shorter but null padded will print null bytes to stdout
12:46:43
@nbp:mozilla.orgnbp
In reply to @aliarokapis:matrix.org
Converting longer paths to shorter relative ones should work
No, Pascal for example encode the length next to the string.
12:53:00
@nbp:mozilla.orgnbpSome compilers could bake-in strlen of constants at compile time.12:55:12
@willbush:matrix.orgwillbush infinisil: Philip Taron (UTC-8): I've been kinda wanting a place to share and discuss Rust blog posts / news with you guys. Any ideas? 13:08:56
@aloisw:kde.org@aloisw:kde.org
In reply to @nbp:mozilla.org
The replaceDependencies is the closest thing there is to replacing raw path to the nix store by another one in the nix-store, but beware that if the version changes from 0.4.0 to 0.4.0a this would cause a lot of issues.
This is why it enforces the same length already.
14:10:04
@philiptaron:matrix.orgPhilip Taron (UTC-8)I'm in most of the Matrix rooms, and on NixOS discourse. If you want something else, my email is philip.taron@gmail.com 🙂14:57:17
14 Jun 2024
@infinisil:matrix.orginfinisilWeekly office hour now :) https://meet.jit.si/nixpkgs-architecture17:57:50
@infinisil:matrix.orginfinisil Cc willbush Philip Taron (UTC-8) Dominic Mills 17:59:13
@infinisil:matrix.orginfinisil With Will and Dominic in the call I created a bunch of PRs improve some misc stuff in Nixpkgs, mostly churning this issue down :) 19:50:53
@infinisil:matrix.orginfinisilIf anybody wants to help out, feel free to fix more cases mentioned in that issue :D19:51:31
17 Jun 2024
@philiptaron:matrix.orgPhilip Taron (UTC-8)I'm thinking up ways to test https://github.com/NixOS/nixpkgs/pull/315654. There aren't any tests of nixos-option that I can see right now.23:04:27
@philiptaron:matrix.orgPhilip Taron (UTC-8) * I'm thinking up ways to test https://github.com/NixOS/nixpkgs/pull/313497. There aren't any tests of nixos-option that I can see right now.23:04:49
18 Jun 2024
@philiptaron:matrix.orgPhilip Taron (UTC-8) I currently find that if I try to ask for recursive output on my configuration, I end up getting an absolutely inscrutable error about list index 0 is out of bounds in lib/generators.nix toPretty method. 00:07:04
@philiptaron:matrix.orgPhilip Taron (UTC-8)
In reply to @philiptaron:matrix.org
I currently find that if I try to ask for recursive output on my configuration, I end up getting an absolutely inscrutable error about list index 0 is out of bounds in lib/generators.nix toPretty method.
It turns out that this was Nix deciding not to show me that the actual error was in a let block in nixos/modules/tasks/filesystems.nix. I opened https://github.com/NixOS/nixpkgs/pull/320846 to fix this.
18:39:14
19 Jun 2024
@aliarokapis:matrix.orgAlexandros Liarokapis

the rust infra has this definition:

linkerForTarget = if shouldUseLLD pkgsTargetTarget.stdenv.targetPlatform
      && !pkgsTargetTarget.stdenv.cc.bintools.isLLVM # whether stdenv's linker is lld already
      then "${pkgsBuildTarget.llvmPackages.bintools}/bin/${pkgsTargetTarget.stdenv.cc.targetPrefix}ld.lld"
      else ccForTarget;

when cross compiling to arm-none-eabihf (target = "thumbv7em-none-eabihf") this ends up picking arm-none-eabi-gcc as CARGO_TARGET_*_LINKER which then fails due to being passed flags like --gc-sections, -flavor, --as-needed, eh-frame-hdr and similar.

16:47:25
@aliarokapis:matrix.orgAlexandros Liarokapis *

the rust infra has this definition:

linkerForTarget = if shouldUseLLD pkgsTargetTarget.stdenv.targetPlatform
      && !pkgsTargetTarget.stdenv.cc.bintools.isLLVM # whether stdenv's linker is lld already
      then "${pkgsBuildTarget.llvmPackages.bintools}/bin/${pkgsTargetTarget.stdenv.cc.targetPrefix}ld.lld"
      else ccForTarget;

when cross compiling to arm-none-eabihf (target = "thumbv7em-none-eabihf") this ends up picking arm-none-eabihf-gcc as CARGO_TARGET_*_LINKER which then fails due to being passed flags like --gc-sections, -flavor, --as-needed, eh-frame-hdr and similar.

16:47:51
@aliarokapis:matrix.orgAlexandros Liarokapis changing it to arm-none-eabihf-ld works. 16:48:13
@aliarokapis:matrix.orgAlexandros LiarokapisRedacted or Malformed Event16:50:19
@aliarokapis:matrix.orgAlexandros Liarokapistaking about https://github.com/NixOS/nixpkgs/blob/master/pkgs/build-support/rust/lib/default.nix16:52:02
@aliarokapis:matrix.orgAlexandros Liarokapis maybe platform.isNone should be added as a shouldUseLLD check ? 16:52:19
20 Jun 2024
@aliarokapis:matrix.orgAlexandros Liarokapisin which cases is using ccForTarget correct? mainly for the clang environments ?13:01:09
21 Jun 2024
@jade_:matrix.org@jade_:matrix.org left the room.05:15:48
@qyliss:fairydust.spaceAlyssa RossUsing LLD on aarch64 static is a hack around a rustc bug. It wasn't the intention to force lld on rustc for any architecture long term.07:25:45
@qyliss:fairydust.spaceAlyssa RossIf Rust for some reason doesn't expect to pass flags through cc on that target, we'd probably need some other check to tell it to use ld07:26:19
@janik0:matrix.org@janik0:matrix.org left the room.09:14:30
@linus:schreibt.jetzt@linus:schreibt.jetzt left the room.14:03:50
28 Jun 2024
@infinisil:matrix.orginfinisilI didn't hold my office hour last week, but let's do it again this one, anybody is free to join and chill :) https://meet.jit.si/nixpkgs-architecture17:55:46

Show newer messages


Back to Room ListRoom Version: 9