!UNVBThoJtlIiVwiDjU:nixos.org

Staging

315 Members
Staging merges | Running staging cycles: https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+is%3Aopen+head%3Astaging-next+head%3Astaging-next-25.05 | Review Reports: https://malob.github.io/nix-review-tools-reports/108 Servers

Load older messages

SenderMessageTime
19 Sep 2025
@k900:0upti.meK900OK three browsers left and we should be good12:39:31
@emilazy:matrix.orgemilywe probably have one -next after this one before freeze right?13:25:31
@grimmauld:grapevine.grimmauld.deGrimmauld (any/all)i sure hope, i'd want systemd 258 and gcc 15 in 25.1113:26:32
@k900:0upti.meK900At least one yes definitely13:27:50
@emilazy:matrix.orgemilyyes my fear is just that doing that stuff in the last pre-freeze cycle is miserable13:28:35
@emilazy:matrix.orgemilybut we can probably make it work13:28:44
@k900:0upti.meK900Also down to two browsers now13:28:55
@k900:0upti.meK900Progress13:28:59
@emilazy:matrix.orgemily have you done KDE packages too @k900:0upti.me? or just ISO 13:29:16
@k900:0upti.meK900KDE packages is all done13:29:23
@k900:0upti.meK900ISO is waiting on webkitgtk, GNOME stuff that depends on webkitgtk, and Firefox13:29:37
@k900:0upti.meK900webkitgtk should be ~30 more minutes and then Firefox should be another hour or so13:30:37
@k900:0upti.meK900But I am building just your cmake PR on top of staging13:30:51
@k900:0upti.meK900So there's no other weird things left13:31:07
@grimmauld:grapevine.grimmauld.deGrimmauld (any/all)systemd 258 should be there in ~a week, i'd hope. No idea how bad gcc 15 would be13:31:12
@k900:0upti.meK900I'd expect this to be a long cycle13:33:55
@emilazy:matrix.orgemilyit's more about how long it takes to start another one13:34:16
@grimmauld:grapevine.grimmauld.deGrimmauld (any/all)it'll be fiiiine13:35:05
@k900:0upti.meK900 Well it'll take at least as long as this one takes to complete 13:36:41
@emilazy:matrix.orgemilyif we neglect 25.05, sure13:49:52
@emilazy:matrix.orgemilyyou weren't there when we did GCC 14 for IIRC the last pre-freeze cycle in 24.11 and had to revert it and scrap the whole cycle because it was too buggy 🙃13:50:40
@grimmauld:grapevine.grimmauld.deGrimmauld (any/all) i am hoping gcc 15 would not be that 13:51:48
@k900:0upti.meK900webkitgtk (the other one) is done14:20:00
@k900:0upti.meK900Firefox is profiling14:20:04
@emilazy:matrix.orgemilydoes anyone feel like rubber-stamping https://github.com/NixOS/nixpkgs/pull/442473 btw14:20:10
@emilazy:matrix.orgemilyat this rate they'll probably have another release out by the time we get to it…14:20:19
@emilazy:matrix.orgemilyI should probably just self-merge these because nobody else wants to touch them ever14:20:47
@emilazy:matrix.orgemily 
    # Fix a GCC‐incompatible use of a private trait.
    #
    # Per Folly’s own documentation:
    #
    #     /// Under gcc, the builtin is available but does not mangle. Therefore, this
    #     /// trait must not be used anywhere it might be subject to mangling, such as in
    #     /// a return-type expression.

    # In file included from /build/source/folly/lang/test/BitsTest.cpp:17:
    # In member function 'constexpr bool folly::get_bit_at_fn::operator()(const Uint*, std::size_t) const [with Uint = short unsigned int]',
    #     inlined from 'void folly::BitsAllUintsTest_GetBitAtLE_Test<gtest_TypeParam_>::TestBody() [with gtest_TypeParam_ = short unsigned int]' at /build/source/folly/lang/test/BitsTest.cpp:640:5:
    # /build/source/folly/lang/Bits.h:494:10: warning: 'in' is used uninitialized [-Wuninitialized]

    # Facebook Thrift requires C++20 now but Watchman hasn’t been
    # updated yet… (Aren’t these things meant to be integrated together
    # in a monorepo?)
    substituteInPlace CMakeLists.txt \
      --replace-fail 'set(CMAKE_CXX_STANDARD 17)' 'set(CMAKE_CXX_STANDARD 20)'
14:21:39
@emilazy:matrix.orgemilyGood Software14:21:41
@k900:0upti.meK900haha wangle14:22:16

Show newer messages

Back to Room ListRoom Version: 6