mirror of
https://github.com/Droid-Deck/DroidDeck.git
synced 2026-11-18 20:36:04 +00:00
198 lines
12 KiB
Markdown
198 lines
12 KiB
Markdown
# American Nightmare on Pixel: rendering workaround and remaining limits
|
|
|
|
Local engineering results from 2026-10-11/12. Steam game 202750 was tested on a
|
|
Pixel 11 Pro Fold using the Pixel Lab runtime. These observations do not establish
|
|
compatibility across the Pixel family or a fix for the system resets.
|
|
|
|
## Rendering result
|
|
|
|
The bright dotted triangle edges over textured rocks, boards and the player are
|
|
not an application wireframe setting: traced input and driver pipelines all used
|
|
`VK_POLYGON_MODE_FILL`, with no dynamic polygon mode. The original affected path
|
|
used 2x MSAA, RGBA16F color and D24S8 depth. The game's minimum UI AA setting and
|
|
`d3d11.disableMsaa` did not disable its D3D9 multisampling.
|
|
|
|
A real single-sample D3D9 comparison removed the visible bright triangle edges.
|
|
Trial 16 verified 114 input and 114 driver pipelines, all single-sample. Trial 17
|
|
retained correct meshes/textures through movement and turning, with 119 pipelines
|
|
in each phase, all single-sample. This isolates a useful workaround to the MSAA
|
|
path; it does not identify the precise faulty operation or prove a vendor bug.
|
|
The override also changes effective surface descriptors, so the application may
|
|
select different AA-aware shader/rendering paths.
|
|
|
|
`tools/dxvk/patches/0004-d3d9-force-single-sample.patch` adds an explicit per-game
|
|
`d3d9.forceSingleSample` option, default **False**. After validating the original
|
|
sample request, D3D9 surface normalization selects `D3DMULTISAMPLE_NONE` and quality
|
|
0. GetDesc, image creation, resolves and sample-dependent state use this effective
|
|
description. This intentionally disables MSAA, per-sample masks and
|
|
alpha-to-coverage. It is a compatibility override, not transparent MSAA emulation
|
|
or a universal PowerVR default. Existing low32, BC-format, present synchronization
|
|
and tiler workarounds must remain intact.
|
|
|
|
Swapchain backbuffers have a separate creation path. With the override enabled,
|
|
that path now validates a copy of the original requested descriptor and transfers
|
|
the effective sample type/quality before allocation, so automatic depth surfaces
|
|
and backbuffers agree. Validation uses the application's lockable-backbuffer flag;
|
|
the internal lockability required for GDI fallback is retained on the allocated
|
|
backbuffer. With the override disabled the original backbuffer path is unchanged.
|
|
This correction addresses official PR602 review 5486295088. Its CPU harness
|
|
compiles the actual DXVK sample decoder, surface normalizer and backbuffer method,
|
|
plus the automatic-depth descriptor and framebuffer depth-binding excerpts.
|
|
Fifteen cases cover on/off, NONMASKABLE quality, matching depth, no-depth operation,
|
|
unsupported/invalid/lockable requests, allocation cleanup and recreation. Backend
|
|
images and adapter capabilities are mocked; these are not GPU or gameplay tests.
|
|
|
|
## Independent HUD correction
|
|
|
|
The diagnostic frametimes HUD bound an R8_UINT storage texel buffer to a shader
|
|
whose SPIR-V image format was Unknown. This required formatless buffer writes
|
|
which the device did not report for that format, producing VUID 07029.
|
|
`0003-hud-explicit-r8-texel-format.patch` declares `r8ui` to match the actual host
|
|
view. No capabilities are fabricated. The build now compiles and validates the
|
|
actual shader, and checks binding 3 for DimBuffer/storage/R8ui. The original shader
|
|
fails that contract; the patched shader passes. Real-game trial 15 used the new
|
|
DLLs, removed 07029, and retained the original triangle edges. This HUD bug is
|
|
separate from the rendering workaround, and is not a proven reset cause.
|
|
|
|
VUID 06479 for color-format comparison descriptors remains under investigation.
|
|
The D3D9 shader uses a conditional comparison path; static validation output alone
|
|
does not prove that the offending branch executes. No warning suppression or
|
|
false format-feature claims are proposed. Trial 14's LOG+FAIL with a duplicate
|
|
limit blocked only the first message of each kind, not every invalid API call;
|
|
subsequent trials used logging only. Do not describe these runs as fully blocked
|
|
validation or GPU-assisted validation.
|
|
|
|
## Responsiveness, performance and stability
|
|
|
|
Paired screenshots prove movement and camera changes, rather than relying on an
|
|
input receipt. Trial 17 completed eight samples without a black viewport. Its
|
|
cold sample was 3.2 FPS; later recognized samples were 19.1-29.9 FPS, with roughly
|
|
2.54 CPU-core equivalents. Camera/workload, capture overhead and thermals were
|
|
not controlled; this is not a reliable A/B benchmark.
|
|
|
|
The planned eight-minute trial 17 stopped at a focus/session guard after the last
|
|
confirmed sample at 315.6 seconds. Steam owned focus; the target game was still
|
|
available for a single client quit. Its temporary settings were successfully
|
|
restored, and the phone boot counter remained 12. The focus change's cause is
|
|
unknown. The run is **incomplete**, not an eight-minute pass or a demonstrated
|
|
game crash. Controls, focus ownership and sustained stability still need broader
|
|
coverage beyond these short tests.
|
|
|
|
### Final production configuration and user confirmation
|
|
|
|
The clean production configuration was persisted only for game 202750 through
|
|
the application's per-game environment UI. It selects the tested DXVK component
|
|
and sets `d3d9.forceSingleSample = True`, retaining the existing low32, BC,
|
|
presentation-sync and `dxvk.tilerMode = False` workarounds. Shared configuration
|
|
and other games were unchanged. Shader audit layers, validation and the
|
|
frametimes HUD are not enabled in this production configuration.
|
|
|
|
A final automated production trial loaded the existing saved game, verified
|
|
turning/movement with five paired non-black screenshots, and observed 115 seconds
|
|
without the bright triangle edges or another phone reset. It then quit normally;
|
|
the saved per-game environment remained unchanged. This is bounded observation,
|
|
not a long-duration stability or controlled performance benchmark.
|
|
|
|
On 2026-10-11 (America/Toronto), the device owner separately confirmed that this
|
|
stage worked and authorized publication. No duration, frame rate or full-game
|
|
completion was supplied with that confirmation. It is additional user validation
|
|
of the stage, not a claim that all Pixel devices or the GNOC resets are fixed.
|
|
|
|
## Native guards, resets and diagnostic controls
|
|
|
|
The Pixel base is e589e3ae21959c9c74c490d917cebf256acc8ac8. PR561's 9e6fe314 and
|
|
3c363910 guards were integrated in an isolated worktree; 26 selected production-C
|
|
regressions and a full APK/runtime verification passed. Pixel AHB support remains
|
|
present. The existing Lab app was updated without uninstalling or resetting its
|
|
Steam account. This APK preceded a second system reset, so the native guards are
|
|
not a verified reset fix.
|
|
|
|
Two resets changed the boot counter 10 -> 11 -> 12. After personal unlock, narrow
|
|
kernel/firmware projections confirmed GNOC/access-control details for the first,
|
|
and `reset,warm,gnoc_error` with a fresh GNOC numeric record for the second. The
|
|
specific faulty component/operation and any relation to rendering are unknown.
|
|
`tools/pixel_reset_diagnostics.py` exports only fixed fault categories/numeric
|
|
fields, lock state, boot counter and entry times; never arbitrary log text or
|
|
account data. Boot-property and kernel last-boot reasons use the same exact
|
|
allowlist of canonical categories and the observed `reset,warm,gnoc_error` value;
|
|
unknown reasons and custom suffixes produce `unknown`, never their original text.
|
|
Primary-user states are likewise limited to recognized lifecycle categories.
|
|
Output files are created exclusively with mode `0600` before any bytes are
|
|
written (POSIX), and existing files/symlinks are refused. Host regressions cover
|
|
these boundaries as well as freshness, unlock gating and narrow output.
|
|
Dropbox's offset-free local timestamp is not assigned a guessed instant during
|
|
a DST transition: repeated hours report `ambiguous_device_local_time`, skipped
|
|
hours report `nonexistent_device_local_time`. Neither is attributed to the trial
|
|
or used to project kernel-body categories until its time can be established.
|
|
|
|
Private opt-in diagnostics captured maintenance5 inline modules, which the older
|
|
shim path missed. Materialization passed nine actual-C ASan/UBSan cases; benign
|
|
pipeline-chain planning passed twelve. Real-game NoContraction, scaled vertex
|
|
conversion and force-clip rewrites were actually applied and did not remove the
|
|
edges. None is a production default. Sixty-two valid bounded, owned offscreen GPU
|
|
cases passed exact pixel readback, fences and validation, covering interpolation,
|
|
seams, perspective meshes, interface widening, scaled fetch, HDR/depth prepasses,
|
|
real clipping and nonconstant depth. Invalid initial fixtures are excluded. These
|
|
passing controls do not exonerate every driver/rendering path.
|
|
|
|
## Reproducibility and recovery
|
|
|
|
The runtime artifact supplies both architectures, exact DLL hashes, public patched
|
|
DXVK sources, all four patches, wrapper and build recipe. It contains no game
|
|
assets, bytecode, saves, account files, screenshots or raw device logs. Source
|
|
patches apply cleanly to pinned DXVK revision 25ca63f17f34bdc05a39873ee63907d3cbbfa030.
|
|
All five runtime DLL targets were built for both architectures.
|
|
|
|
Stop further load tests after a full reset. Wait for personal unlock and inspect a
|
|
fresh narrow projection before another launch. Temporary profile restoration checks
|
|
baseline/current hashes and unchanged shared/other-game settings; it never rolls
|
|
back saved game progress. Persistent selection must affect only 202750 and leave
|
|
private shader audit layers, validation settings and frametimes HUD disabled.
|
|
Routine Heartbeat must not launch games or automatically resume this engineering.
|
|
|
|
## Submission scope
|
|
|
|
These opt-in runtime tools are submitted directly against official DroidDeck
|
|
`main`. The standalone DXVK build recipe, source pin, BC-format prerequisite and
|
|
presentation-semaphore patch/check are imported verbatim from the PowerVR proposal
|
|
<https://github.com/Droid-Deck/DroidDeck/pull/570> at
|
|
`d0f7264ea8e402d30c868779198dc2fd74e7431c`. That proposal's attribution and existing
|
|
source/license provenance are retained. The semaphore correction originated in
|
|
<https://github.com/wallentx/DroidDeck/pull/3>. No Android app changes, managed pack
|
|
or the experimental fork's development history are imported with those recipes.
|
|
|
|
The source tools can be reviewed and built independently. Automatic PowerVR
|
|
integration and delivery in DroidDeck require the separately reviewed #570
|
|
foundation and a new managed-runtime release. This submission does not change
|
|
managed-pack manifests/checksums or globally enable any override. The earlier
|
|
fork-only stage <https://github.com/wallentx/DroidDeck/pull/4> is superseded by the
|
|
official submission; its review history is preserved.
|
|
|
|
For standalone verification, run
|
|
`python3 -m unittest discover -s tools/tests -p test_pixel_reset_diagnostics.py`
|
|
on the host. On native 64-bit Linux/glibc,
|
|
`tools/hybris/low32/build.sh /absolute/output-dir` builds and runs the CPU integrity
|
|
and independent mmap/mmap64 preload-chain regressions (both library orders in
|
|
off, trace and low32 modes). Optional Vulkan headers produce `gpu-copy-native`
|
|
for the selected native host architecture, not a hardcoded ARM64 filename.
|
|
`python3 -m unittest discover -s tools/tests -p test_dxvk_build.py` verifies output
|
|
isolation with a synthetic toolchain, without fetching or compiling DXVK.
|
|
The DXVK component recipe is `tools/dxvk/build.sh /absolute/output-dir`;
|
|
it requires the pinned DXVK build prerequisites, including Meson/Ninja, MinGW
|
|
cross compilers, glslangValidator and SPIR-V Tools. It retains both architectures,
|
|
source/license provenance and downstream patches, without editing an installed
|
|
runtime or APK. Its output directory must be new or empty (including hidden
|
|
entries). Prior or partially built outputs are refused before fetching sources,
|
|
left untouched and never included in a new DLL manifest. Use a fresh directory
|
|
after a failed build.
|
|
The recipe also requires a native C++17 compiler for
|
|
`python3 tools/dxvk/check_d3d9_single_sample.py /path/to/patched-dxvk` before DLL
|
|
compilation. Add `--sanitize` on supported hosts to run its ASan/UBSan controls.
|
|
|
|
The compositor ownership safeguards are independently proposed in
|
|
<https://github.com/Droid-Deck/DroidDeck/pull/561>. Those native changes were present
|
|
in the integrated Pixel Lab APK used for the game experiments, but are not part
|
|
of this runtime PR's diff. The original BC emulation is a prerequisite supplied by
|
|
the separately proposed PowerVR runtime. This submission supplies source and reproducibility tools,
|
|
not an APK, proprietary driver, game data or private diagnostic archive.
|