fix(menu): the Save my pipeline list module includes the POSIX headers it uses #1807
No reviewers
Labels
No labels
area/accessibility
area/audio
area/input
area/install
area/menu
area/renderer
area/saves
area/tooling
area/upscaling
beanstalk: approved
beanstalk: failed
beanstalk: needs approval
beanstalk: running
beanstalk: tested
ci: approved
good first issue
gpu/amd
gpu/igpu
gpu/intel
gpu/nvidia
help wanted
kind
bug
kind
chore
kind
docs
kind
feature
kind
performance
kind
question
needs measurements
platform/linux
platform/macos
platform/steam-deck
platform/windows
preview
priority
critical
priority
high
priority
low
regression
risk
high
risk
low
risk
medium
status
blocked
status
confirmed
status
duplicate
status
invalid
status
needs-info
status
wontfix
ci: approved
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Longshot!1807
Loading…
Reference in a new issue
No description provided.
Delete branch "mooki/Longshot:fix-pipeline-list-posix"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Type: fix · Risk: low · Tested on: Linux, RTX 4090 · Needs: Windows, any GPU
TL;DR
Since the menu rows moved into modules (#1774), main does not build on Linux: one of the new files, the Save my pipeline list module, lost the block of system headers its Linux code needs. This puts that block back, ten lines in one file. With it the whole game builds again (152 files compiled, none failed) and starts to the playable game on an RTX 4090.
Breakdown
gpu/vk/menu/pipeline_list.cpptook the zip-and-open code out ofvk_menu.cppbut not the#includeblock that code needs on Linux.#elseof the_WIN32include it already has.Why
Main stopped building on Linux with #1774: the build compiles every file of
gpu/vk/menu/, and this one usesposix_spawn,waitpid,sigtimedwait,SIGPIPEandunamewithout including what declares them. Nothing on Linux, including the Test Farm's Linux machines and anyone's own build, can get past it.How to reproduce
Seen on: Arch Linux, clang, a standalone build with the project's builder over a finished work folder
Settings and switches: defaults
Steps: build main at
0de25c3e4afor LinuxExpected: the build finishes
Actual: it stops in
gpu/vk/menu/pipeline_list.cpp("compiled in 55s, 1 failed"):What changed
The module's include block gets, for everything but Windows, the headers
vk_menu.cpphad for this code before #1774:fcntl.h,spawn.h,csignal,cerrno,pthread.h,sys/utsname.h,sys/wait.h,unistd.handextern char** environ.gpu/vk/menu/pipeline_list.cpp: the#elsebranch of the#ifdef _WIN32include, inside#ifndef MENU_LAYOUT_ONLY, as the code that uses it is.Measurements
Verdict: nothing to measure for speed: ten lines of headers, no code and no data change.
Hardware: Linux 7.2, RTX 4090 (NVIDIA driver), Core i9-13900K
Scene: the game started windowless from a disposable save, in the Kokiri Forest
Method: the project's builder over a finished work folder of main (whose lift inputs equal this head's), once on main and once on this head; one game start in the side lane
How to reproduce:
python -I nx2/make_game.py --romfs ... --exefs ... --out <copy> --work <work folder>on both heads0de25c3e4a8a1fc7d5c1pipeline_list.cpp)Flags and settings
None.
Testing
Tested: on Linux (RTX 4090), this head built with the builder (
v3.0.0-rc.8-70-g8a1fc7d5c1 (source 14be9eeb)) and started once windowless:longshot v3.0.0-rc.8-70-g8a1fc7d5c1 (source 14be9eeb), Link controllable after 37.6 s, fatal=0 crash=0 devlost=0 nocore=0 kernel=0 nvrm=0. The whole suite was not run (ten lines of headers); the 108 game-free checks that read the menu's sources ran through the idle gate on main and on this head: 98 pass and the same 10 fail on both (test_bug_report_defaults,test_crash_dialog,test_crt_options,test_damage_debug_menu,test_dev_packs,test_hud_layout_row,test_mods_size_table,test_present_queue_default,test_upscaler_rows,test_walk_key). The Windows-target syntax of the file (x86_64-w64-mingw32-clang++ -std=c++20 -fsyntax-only -Wall -Wextra) passes on both heads.Not tested: a Windows build and run (the lines are in the branch Windows does not compile); the zip, clipboard and browser steps of Save my pipeline list in the game (the code is unchanged).
Must be tested:
Risks
What could break: a Linux build that already worked cannot be made worse by headers; a name from these headers colliding with the file's own would stop the build, and the build above says it does not.
Turning it off: revert the commit.
Compatibility: Nothing changes.
Issues
See #1774
Checklist
Everyone
type(scope): description) of the status line's type.can't be run by hand is in Testing's Not testable by hand, with why.
in Risks.
main.python -I -B tests/run_all.pypasses (any failure that also fails onmainis named).third_party/untouched (a new folder with aPROVENANCE.mdinstead).When it applies
Assisted-by:/Co-Authored-By:trailers.New on this forge; not opened on GitHub.
Main has not built on Linux since the menu model (#1774):
gpu/vk/menu/pipeline_list.cpplost the block of POSIX headers thatvk_menu.cpphad for Save my pipeline list, so the one file fails withposix_spawn_file_actions_t,waitpid,SIGPIPE,sigtimedwaitandutsnameundeclared.This PR puts that block back in the module, under the
_WIN32split the code already has; nothing else changes (10 lines, one file).On Linux (RTX 4090) the build of this head is clean (152 files compiled, 0 failed) and it started once, windowless, to the playable game with no fatal, crash or device-loss line; main alone stops at that file. The Windows branch is unchanged and the file still passes a Windows-target syntax check.
Every open menu PR that is built on Linux needs this first.
WIP: fix(menu): the Save my pipeline list module includes the POSIX headers it usesto fix(menu): the Save my pipeline list module includes the POSIX headers it uses@Beanstalk
Beanstalk test farm
Commit
8a1fc7d· tier reviewed · approved by mooki · dashboardWhat will be tested · AI plan (
gpt-6.1-terra)The farm builds main and the PR on Linux and Windows, then loads the Kokiri Forest checkpoint and waits for play. The smoke test takes about 10 minutes per target, excluding builds; a failed PR build or start fails the test. Main is reported not to build on Linux, which may block the Linux comparison rather than indicate a PR regression.
kokiri-start; 1 run on each side.Done (started 2026-11-19 22:41 UTC, last report 2026-11-19 23:06 UTC). Comment
@Beanstalkto run it again.NVIDIA GeForce RTX 5070, 620.70.6; 2: linux+amd+radv+igpuAMD Radeon 760M Graphics (RADV PHOENIX), 26.0.8AMD Radeon RX 9060 XT, 2.0.416; 2: windows+nvidia+nvidia-proprietary+dgpuNVIDIA GeForce RTX 4070 Ti SUPER, 621.14.0Measurements · linux
Verdict: could not be judged on linux (2/1 report).
Hardware: 1: linux+nvidia+nvidia-proprietary+dgpu
NVIDIA GeForce RTX 5070, 620.70.6AMD Ryzen 5 9600X 6-Core Processor, 32 GB RAM; 2: linux+amd+radv+igpuAMD Radeon 760M Graphics (RADV PHOENIX), 26.0.8AMD Ryzen 5 230 w/ Radeon 760M Graphics, 16 GB RAMScene: kokiri-start, from the repository's save kokiri-start; each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); kokiri-start: 1 game start a side; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (
@Beanstalkon this PR runs it again)Nothing was measured: no metric came back.
Flags and settings
None: the plan varied no setting, switch or build option.
Markdown for the PR description
Measurements · windows
Verdict: pass on windows (2/1 report): no rule failed and no PR run crashed.
Hardware: 1: windows+amd+amd-proprietary+dgpu
AMD Radeon RX 9060 XT, 2.0.416AMD Ryzen 5 5600 6-Core Processor, 32 GB RAM; 2: windows+nvidia+nvidia-proprietary+dgpuNVIDIA GeForce RTX 4070 Ti SUPER, 621.14.0AMD Ryzen 7 7800X3D 8-Core Processor, 32 GB RAMScene: kokiri-start, from the repository's save kokiri-start; each cell's settings and switches as the table names them
Method: the Beanstalk test farm: main and this PR built and run on the same PC in one session, game starts alternating (ABBA); kokiri-start: 1 game start a side; each number the median of the runs, then of 2 reports
How to reproduce: the farm's plan above (
@Beanstalkon this PR runs it again)Nothing was measured: no metric came back.
Flags and settings
None: the plan varied no setting, switch or build option.
Markdown for the PR description
gab9281 referenced this pull request2026-11-20 05:12:20 +00:00