Program 203 reuses prefixSumCache for a second subgroup reduction before every workgroup invocation has consumed the first result. Add a fingerprint-gated SPIR-V pass that inserts the missing Workgroup acquire-release barrier while preserving native subgroup operations.
Keep the repair opt-in behind MOBILEGL_ITERATIONRP_FIX_BARRIER, cover insertion, pass-through, and idempotence, and enable it together with the existing iterationRP subgroup repairs for the matching Linux and Android CI retraces.
iterationRP's Program 203 declares shared vec2 prefixSumCache[32] for a
512-invocation workgroup indexed by gl_SubgroupID; any device narrower
than 16 lanes partitions into more than 32 subgroups and the pack writes
shared memory out of bounds (heap corruption on lavapipe's CPU
rasterizer, ssim 0.028 on the CI retrace). Fix it where the fault lies -
in the fixture - and keep the GL contract sound everywhere else:
- FixIterationRPSubgroupScratchPass: fingerprint-gated SPIR-V pass that
grows exactly that array to ceil(invocations/width) entries on sub-16-lane devices; every other module passes through byte-identical.
- DeriveNumSubgroupsPass stays default-on for the Adreno topology bug
and is made spec-sound: pipelines request REQUIRE_FULL_SUBGROUPS
whenever the workgroup shape makes the flag legal (computeFullSubgroups
enabled, local_size_x a multiple of the native width, subgroup count
within maxComputeWorkgroupSubgroups).
- EmulateSubgroupsPass: 32-lane virtual-subgroup lowering kept in-tree
as a last resort, enabled only by MOBILEGL_MAGMA_EMULATE_SUBGROUP=1 on
devices with no native subgroup support; fails closed on extended
subgroup instructions and on modules whose added scratch would exceed
maxComputeSharedMemorySize.
- IterationRPFirstReductionScenario skips gracefully outside the pack's
16..256-lane source domain; the new IterationRPScratchFixScenario runs
the fixture-shaped reduction on any width and asserts the exact
width-independent total. DriverPost keeps reporting FAIL on
out-of-domain devices.
- Program203 -> IterationRP rename throughout; the per-trace
num_subgroups_quirk plumbing is removed from the trace replayer, JNI
chain, and CI workflows.
The Minecraft-shaped driver benchmark could only be run from a desktop shell
against a desktop driver, which is the wrong machine: MobileGL exists to run on
mobile GPUs, and nothing said what its translation costs there. This puts the
same cases on an Android device, both in the plugin's POST screen and from a
shell, and adds the native-driver baseline they have to be read against.
The cases move into DriverBenchCases.inc so both harnesses run byte-identical
bodies - the desktop program resolving entry points from one EGL provider, and
DriverBenchJni.cpp calling MobileGL's frontend in-process. The JNI file binds
every gl*/egl* name to MG_Impl by macro rather than by linkage: this library
legitimately has the platform libEGL and libGLESv3 in its own lookup scope, and
a benchmark that quietly measured the device driver instead of the translation
layer would have looked like very good news.
Frames are now closed with a fence wait instead of glFinish. MobileGL implements
glFinish and glFlush as no-ops, so the old loop timed submit-plus-GPU on a native
driver and submit-only on a MobileGL backend, and the two numbers did not
describe the same work.
To measure a device's own driver the cases needed to be expressible in GLES:
ESSL 3.20 twins of the four shaders (chosen at runtime from GL_VERSION, since
MobileGL is deliberately still fed desktop GLSL - translating it is the thing
under test), a multi-draw hook that loops DrawElementsBaseVertex where the
multi-draw entry point does not exist, and an EGL bootstrap that falls back from
desktop GL to GLES 3. The binary cross-compiles for arm64 unchanged.
BenchService hosts each run in its own process and exits afterwards. That is not
caution: the backend is latched from MOBILEGL_BACKEND_TYPE at initialization, so
Espryt and Magma can never share a process, and Espryt's teardown terminates the
process-default EGL display, which would take the POST activity's own EGL
objects with it.
Running it found that Magma could not create a windowless context on Mali at
all - CreateInstance required VK_EXT_headless_surface, which no mobile driver
here exposes, and aborted the process. The Xlib path already probes and falls
back to a hidden window for the same reason on NVIDIA; Android now probes too
and hands the WSI an AImageReader's ANativeWindow, a real producer surface
attached to no display whose images are never acquired. DriverPost reports the
extension's absence as a WARN so the fallback is visible rather than silent.
Measured on a Mali-G77 MC9 (native / Espryt / Magma, ns per operation):
5495 chunk draws 14397 / 36934 / 33763, the 26.2 per-draw uniform-range pattern
13710 / 31205 / 21252, sodium-style multi-draw 256956 / 238389 / 209527. The
translation costs about 2.4x per draw here against 5-9x on the desktop, because
the mobile driver's own per-call cost dwarfs it - and both backends beat the
native driver on multi-draw, which it has to emulate.
Desktop unit tests 421/421; the POST screen and both Run Bench buttons verified
on the device.
Moves snapshot capture entirely into the apitrace retrace layer (glReadPixels + PNG encode). Drops the MOBILEGL_PRESENT_DUMP_PATH / MOBILEGL_PRESENT_STATS / MOBILEGL_PRESENT_DUMP_CALL / MOBILEGL_PRESENT_CURRENT_CALL / MOBILEGL_TRACE_CURRENT_CALL_OVERRIDE plumbing from Config, ConfigLoader, VulkanRenderer (GetPresentedDumpPixel/WritePresentedDumpPpm + present-stats readback), the EGL/GLX/Android ws shims, and the Android trace_replay_core PPM reader.
DirectVulkan ReadPixels on the default framebuffer now remaps raw swapchain pixels (top-left origin, preTransform-rotated) to GL orientation (bottom-left origin) so the retrace snapshot matches the golden; SwapchainObject also resizes the default-FBO stencil attachment to the swapchain extent to fix GL_INVALID_FRAMEBUFFER_OPERATION under the glReadPixels completeness check.
Checks render as a two-column table (name | colored status chip) with
alternating row stripes; per-check detail text is hidden until the row
is tapped, and the raw JSON report collapses behind a bottom toggle.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Opening a MobileGL plugin APK now shows a POST screen that probes the
device's GLES and Vulkan drivers independently against MobileGL's
expectations - a device may satisfy only one backend - and reports a
per-backend verdict (OK / DEGRADED / UNSUPPORTED) with per-check rows.
The GLES probe builds its own ES3 pbuffer context on the system driver
and reuses FillInGLESCapabilities, including the indirect-draw
gl_InstanceID semantics probe; the Vulkan probe checks instance/device
requirements and the optional features each DirectVulkan path degrades
without. Results serialize as ASCII-safe JSON through a JNI entry in
libMobileGL.so; PostActivity renders them and caches the run per
process (single-flight, rotation-safe). PluginActivity keeps its
NoDisplay stub but the launcher entry moves to the POST screen; FCL
plugin discovery reads application meta-data and is unaffected.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- compare actual output against primary and alternate golden images
- record the matched golden path in trace replay results
- allow APK and Linux retrace fixtures to pass alternate golden paths
- keep nostalgia validation accepting both Mesa and PC goldens
swung0x48
