On the Steam Frame, VR_Init failed with "Installation path could not be
located": the SDK's lib/linuxarm64/libopenvr_api.so (2.15.6) reads the VR
path registry from the fixed path /data/work/openvrpaths.vrpath instead
of ~/.config/openvr. The SDK's bin/linuxarm64 library of the same version
uses ~/.config/openvr; the vrdashboard template ships it and was verified
on the Frame. It is added as a new file (third-party/openvr had no
bin/linuxarm64, and "bin" is gitignored, hence force-added), and aarch64
builds link against and ship it. x86_64 and Windows are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The first deploy to the Steam Frame aborted in install.sh: OVRAS creates
its QApplication before it handles --force-install-manifest, and Qt's
default platform plugin (xcb) aborts when there is no display, as over
SSH. Registering and unregistering draw nothing, so the generated
install.sh and uninstall.sh now run them with QT_QPA_PLATFORM=offscreen.
The normal start keeps DISPLAY=:0, as rendering needs a real context.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- docs/steamframe/README.md: entry point with status, branches/remotes, setup,
commands, Steam Frame facts, open points.
- CLAUDE.md: working rules and code conventions (fork-only).
- docs/steamframe/OVRAS-FEATURES.md: OVRAS's features, how they are
implemented, issues found while reading.
- PROGRESS.md: status and wrap-up.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The CMake build had the version hard-coded as 5.7.4, while the qmake build
read it from compile_version_string.txt (5.8.17-release). The stale number
also made the version check report an update. As in the qmake build, the
full string becomes APPLICATION_VERSION and its numeric part the project
version.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- build_scripts/steamframe/: Debian trixie (Qt 6.8) containers for the
aarch64 cross build, native build and tests; toolchain; scripts to build,
test (Vulkan validation layers), package (system Qt, ~3 MB) and deploy over
SSH. Adapted from the vrdashboard template.
- test/render/: render test for the OpenGL and Vulkan backends, driven like
OverlayController (option OVRAS_BUILD_TESTS).
- RPATH only $ORIGIN, without the build machine's link directories.
- Ignore dist/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- manifest.vrmanifest: binary_path_linux and binary_path_linux_arm. The
Steam Frame's SteamVR reads only binary_path_linux_arm.
- handleManifests(): connect as Utility app first (works without a running
SteamVR, e.g. from the Windows installer); if SteamVR refuses Utility apps,
as on the Steam Frame, connect as Background app. Fail right away on other
errors instead of going through initializeOpenVR()'s retry loop.
- main(): check --graphics-api before connecting to OpenVR, so a wrong value
fails right away.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Replace the QRhi rendering in OverlayController (private Qt API) with the
render backends of the vrdashboard template (src/render/), which use public
Qt API only (QQuickRenderTarget::fromOpenGLTexture / fromVulkanImage). The
binary then runs on another Qt 6 patch release, e.g. the Steam Frame's system
Qt.
The Vulkan backend also does what SteamVR requires and the QRhi path lacked:
the required instance/device extensions and the HMD's GPU (via
openvr_init::renderInterop()), VK_FORMAT_R8G8B8A8_UNORM, and the transition to
TRANSFER_SRC_OPTIMAL after each frame.
- New option --graphics-api opengl|vulkan (default opengl, no fallback).
- Shutdown() calls VR_Shutdown before the render target and Qt's device go
(crash on the Steam Frame with Vulkan otherwise); it runs only once.
- Mouse y is flipped on every platform: SteamVR reports it in GL convention.
- Render delay 5 -> 16 ms (about 60 fps).
- GuiPrivate is no longer needed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- CMake option OVRAS_STEAM_FRAME. It is evaluated only in
src/utils/platform_capabilities.h, which turns it into named capability
constants and platform defaults.
- Where the chaperone can't be committed (Steam Frame: it shifts the play
area), MoveCenterTabController no longer commits the working copy at start-up
or when applying reset data, and doesn't create the init autosave profile.
Skips are logged once.
- In "vsync disabled" mode the pump timer now runs directly at the custom tick
rate instead of firing every 1 ms and counting. The Frame defaults to this
mode at 11 ms (~90 Hz).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Remove Core5Compat (unused, not part of the Steam Frame's system Qt).
- Request GuiPrivate explicitly (required since Qt 6.10 for QRhi); temporary
until rendering moves to public API.
- Compile QML to bytecode only: C++ generated from QML uses private Qt API
that only matches the exact Qt version it was built with.
- Pick the OpenVR library per architecture (linux64, linuxarm64, win64) for
linking and for the copy next to the binary; RPATH $ORIGIN.
- Link libXtst, needed by input_sender_X11.cpp.
- Include d3d11.h and handle QRhi::D3D11 only on Windows.
- Replace leftover easylogging LOG(INFO) calls with qInfo().
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Internal fork based on upstream's qt6 branch (2d4db7e). Goal: OVRAS runs
natively on the Steam Frame (SteamOS, aarch64) with a minimal change set
against qt6 and only public Qt API.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* add delay if interface not found
* adjust initialization behavior
* pushed for debugging reasons for concerned parties
* comment out simple recenter
* remove old style motion, some minor optimizations
* optimize default case check
* optimized default condition
* change manifest install behavior to trigger on desktop mode as well
* add desktop mode binding launch
* add tracker overlay option to utils
* commit to check status
* attempt to fix native LH issues with chaperone when exiting
* add desktop mode toggle and some cleanup
* remove adjust chaperone
* remove clang format from ci, increment version, cleanup
* update circle ci
* fix conversion warning
* add F13-19 to keyboard parser
* add additional logging around failure to initialize
* add unsaved changes
* update version 5.8.1
* fix some collisions in enum declaration
* add controller powers on steamvr, and no HMD to steamvr tab, increment version
* slightly alter drag movement, fix qml issue
* code cleanup
* fix logspam/attempt to delete non existant overlays
* increment version 5.8.3
* add zero-offset to true as potential edge case fix
* remove dead code in qml
* fix both brightness sliders to be uniform .05 minimum via set, and .2 minimum via slider
* attempt to fix recenter, and fix floor
* further space drag/recenter improvements
* get rid of old code
* increment version 5.8.6
* remove old commented code settings page
* update documentation
* re-add mistakenly removed line
* add nlhomann json header
* add get scene app fn
* add get current binding and related logic
* add save binding file to settings folder
* apply binding fn added fixed uid bug
* add button logic to QML, add enable Bind Setting
* add finding logic for all custom bind
* fix minor qml issues
* fix get controllername bug, fix bug with save bind
* fix regex issue
* change webrequest for selectconfig
* change to properly urlify path to send
* fix race condition web requests
* change steamvr tab qml formatting
* change color scheme
* change colors of icons to text color
* change bind qml better for better ux
* add possible workaround for space drag issues
* add deact color to pushbutton
* increment version
* adjust button size in movecenter
* move nlhomann json
* fix PKGBUILD to include missing dependency (#606) (#613)
It turns out qt5-websockets needs to be included for a successful build.
Co-authored-by: James R <71466513+ColdIce1605@users.noreply.github.com>
* fix warning
* improved logging messages around perapp bindings
* improve getsceneappID logging
* add HMD based controller Finding
* improve logging controller name
* change logging around initial chaperone calibration
* increment version
* improve logging/error handling around sceneappid failures
* added fallback for getting scene app by pid
* update version string
* update documentation
Co-authored-by: James R <71466513+ColdIce1605@users.noreply.github.com>
* add Select Dongle Pairing, fix clipping on long SN names (quest)
* remove test log, add ws log, add pair protection non steamvr
* add missing install commands for new qt modules
* remove bad package
* add auto apply chaperone option, remove old seated/uncal/openxr fixes in favor of always on
* adjust battery overlay for tundras
* fix tundra tracker bat indicator positioning, remove offset page restrictions from non-standing universe, make pairing more responsive and display brief instructions
* add Drag Multiplier
* update documentation
* change version to release
* update readme
* fix format issues readme
* update readme with setting description
* change uid to prevent collision
* changed rounding for grav fling to 2 decimals to be consistand with other values on drag page
* Crash Recovery Defaults to Off, add option to use space drag w/ uncalibrated playspace
* increment version 5.3.3
* Feat/apply offsets binding (#576)
* add apply offsets steamvr input
* add new binding to Readme
Co-authored-by: TheMrGong <gongora656@gmail.com>
* increment version 5.4.0
* added LH Console Interface class, WIP TX/RX list via lh console
* switch over to property based dongle information
* add ovr system wrapper
* add use of parameter to avoid warning as error
* openXR fix, and Change Display State of Chaperone Disabled
* update text
* format files
* clang 9.0.0 format
* add documentation update
* remove windows header include
* remove filesystem header
* fix uncal motion switching issue
Co-authored-by: TheMrGong <gongora656@gmail.com>
Compiling C++ on various different platforms with as many out of tree
dependencies as we have is volatile at best.
In order to prevent a third party warning from ruining the build for
people who don't care about third party warnings, we remove
warnings-as-errors by default.
They are still enabled by default on CI.
* fixed input sending bug not properly releasing keys
* updated to ovr 1.12.5
* added disable oculus api button experimental
* small refactor and fix float and bool bugs
* added nokeyup events to modifier keys when read in, and adjusted token send to have proper logic
* fixed literal held issue
* added backslash
* fixed sender token parser
* updated version string to dev whoops
* added backslash to documentation for keyboard input
* Implement (most of) pop-up notification when autoturn is enabled
* Add neato fading: doesn't appear to impact performance
* formatting
* added macro and UI for notification of auto-turn
* changed notification default to true for auto-turn
* added common issue for trackers
* added warning for custom keys
* minor documentation markdown format fixes
* fix qml syntax error in atgb
* gasd
Co-authored-by: ykeara <ykeara1@gmail.com>
The old OpenVR overlays did not function for unknown reasons, and it was
not possible to get them to work.
With this a white overlay with ALARM ACTIVATED will be shown for 3 seconds.
It's not pretty, but it effectively communicates that an alarm has been
activated.
This fixes
https://github.com/OpenVR-Advanced-Settings/OpenVR-AdvancedSettings/issues/417
* updated documentation/screenshots, let seated-recenter show in all modes, fixed default value of brigthness
* fixed formatting
* changed PTT to System
* added comma to documentation
Source outputs must all be moved to new default source.
Reference topic
https://askubuntu.com/a/72076
* Adds microphone controls on Linux via PulseAudio (PA) for several
points in
https://github.com/OpenVR-Advanced-Settings/OpenVR-AdvancedSettings/issues/43.
It can not be overstated how unsuited the public AudioManager API is for
using the PA API.
It can also not be overstated how undocumented and horrible the PA API
is. The PA way of doing audio is very unlike the Windows way of doing
audio, which the AudioManager is heavily influenced by.
Since I don't have a Linux machine I have not been able to personally
test this. https://github.com/feilen/ has graciously helped with
testing.
PulseAudio is very undocumented and not very widely used for what we
need outside of official utilities. As such there's most likely a better
way of doing things than I've done.
Additional information can be found at
https://github.com/username223/pulseaudio-device-changing.
The "complex" PA API is asynchronous. Since our current AudioManager is
running synchronously and would not work asynchronously (functions are
expected to return valid data on function return) it is necessary to
massage the PA API until it runs synchronously.
This is done with a general workflow of
1. Register a callback function (`pa_context_get_source_info_list()`,
for example).
2. Run the PA loop through `customPulseLoop()`. This will trigger
any registered callbacks. Sometimes a `void*` is passed through the
callback that can be used for custom data. `successCallback()` uses
this to set the success variable.
3. Run the callback function and do required work.
4. Return to original function, do work or return values.
It is not possible to run a `customPulseLoop()` inside of an already
running loop (inside a callback), as this will lead to a PA assert
error on runtime.
Enabling the switch on release builds will optimize the binaries as
normal, but will still include debug symbols. This is advantageous for
bugs that do not show up in debug builds without optimizations.
After looking at the assembly output it became apparent that the Windows
compiler did not compile out the `getSettingsController()` function
call.
This led to an unnecessary function call in order to get the
SettingsController. Turning it into a static object in the compilation
unit scope removes this function call and turns getSettings call into a
single instruction (not counting zeroing the register for comparison and
the jump instruction afterwards).
----
Adding the following code to the `main` function:
```cpp
f ( settings::getSetting( settings::BoolSetting::UTILITY_vrcDebug ) )
{
LOG( INFO ) << "Enabled";
}
else
{
LOG( INFO ) << "Disabled";
}
```
Leads to this with the `getSettingsController` function:
```assembly
[...]
47 if ( settings::getSetting( settings::BoolSetting::UTILITY_vrcDebug ) )
0x7ff6c96819a0 <+ 1024> call AdvancedSettings!settings::getSettingsController (00007ff6`c96cd280)
0x7ff6c96819a5 <+ 1029> xorps xmm0,xmm0
0x7ff6c96819a8 <+ 1032> cmp byte ptr [rax+5FCh],0
0x7ff6c96819af <+ 1039> je AdvancedSettings!main+0x4fd (00007ff6`c9681a9d)
49 LOG( INFO ) << "Enabled";
0x7ff6c96819b5 <+ 1045> mov qword ptr
[rsp+288h],r15
0x7ff6c96819bd <+ 1053> mov qword ptr
[rsp+290h],rsi
[...]
```
And this without the function:
```assembly
[...]
47 if ( settings::getSetting( settings::BoolSetting::UTILITY_vrcDebug ) )
0x7ff7d97a19bb <+ 1019> xorps xmm0,xmm0
0x7ff7d97a19be <+ 1022> cmp byte ptr [AdvancedSettings!settings::settingController+0x5fc (00007ff7`d9854d4c)],0
0x7ff7d97a19c5 <+ 1029> je AdvancedSettings!main+0x4f3 (00007ff7`d97a1ab3)
49 LOG( INFO ) << "Enabled";
0x7ff7d97a19cb <+ 1035> mov qword ptr
[rsp+288h],r15
0x7ff7d97a19d3 <+ 1043> mov qword ptr
[rsp+290h],rsi
[...]
```
Notice the missing `call AdvancedSettings![...]` without the function
and that the settings call literally cannot be made any faster.
The details of the API are in the `/docs/specsSettings-API.md` file.
The enums do not list all available settings, more are to be added as
the settings API replaces inline QSettings calls.
The page was not made with Arch in mind.
The currently required were tested on Manjaro, more packages may be
explicitly required due to the bare-bones nature of Arch.
Adds simple logging calls makes debugging easier.
The initial build script was basically ported over directly from
Windows. It was written because I at the time didn't understand
compiling on Linux, and the existing standards, very well.
Moving as many build components into a program specifically made for
building C++ makes everything much easier.
The old build script is still used on CI where control flow can only be
changed using environment variables.
`bear` and `clang-format` are only used on the CI, there's no reason to
include them here.
This enables a `sudo make install` that copies the build directory to
the PREFIX directory (/opt/OpenVR-AdvancedSettings/AdvancedSettings by
default).
The entire directory is being copied because Qt is incredibly weird
about how the INSTALLS variable works. From testing it appears that only
files that exist before the build process has started (even the
file_copy process) will be eligible for installation.
If the /OpenVR-AdvancedSettings/ directory is excluded, Qt will attempt
to remove the /opt/ folder on `make uninstall`, this is obviously not
desirable.
The logs are stored in `~/.local/share/AdvancedSettings-Team/OpenVRAdvancedSettings/`, but the config file is stored in `~/.config/AdvancedSettings-Team/`.
matzmanandClaude Opus 5.5
Matthias JanetschekandClaude Opus 5.5
ykeara
Vasily Utkin
Chelsea Jaggiandykeara
Kung-vr