Cycle 3 (MC interpolation) closure: M1'''=100%, R'''=0.067 RED, M4=-19.5%

Third daedalus-fourier kernel — VP9 8-tap regular subpel filter,
horizontal direction, 8-wide output. Multiply-heavy by design to
stress V3D's no-DP4A deficit. Full cycle Phase 1-7 + M4'''.

Phase 5''' second-model review delivered cleanly — caught 1 RED
bug pre-implementation (src_off off-by-3 indexing convention) and
2 YELLOW gaps (assert MUST language, shaderdb filter-LUT gate).
Without the review, M1''' would have failed silently on first run
with cryptic "high-index source pixels wrong" symptoms.

Phase 6 v1 first-light: M1''' 100.0000% bit-exact (65536/65536
blocks across all 16 mx phases). Phase 5''' filter-LUT prediction
materialised exactly: 197 uniforms (gate was 144), 2 threads (down
from cycle-2's 4 due to register pressure).

Performance:

  M2''' = 1.413 Mblock/s     (707.9 ns/block)
  M3''' = 20.997 Mblock/s    (NEON baseline phase3)
  R'''  = 0.067              (RED band — structural mismatch)
  shaderdb: 488 inst, 2 threads, 197 uniforms, 25 max-temps, 0 spills

M4''' concurrent matrix (8s windows):

  NEON 1-core           14.479 Mblock/s
  NEON 4-core           15.248 Mblock/s   <- baseline (compute-bound,
                                              not bandwidth-saturated
                                              like cycles 1+2!)
  QPU only               1.380 Mblock/s
  MIXED NEON-3 + QPU    12.277 Mblock/s   <- -19.5% (FAIL gate)
  MIXED NEON-4 + QPU    12.158 Mblock/s   <- -20.3%

NEW cross-cycle finding (Phase 9 lesson 2): compute-bound CPU
workloads make the QPU-offload story collapse. Cycles 1+2 were
bandwidth-saturated (4-core scaling 0.56-0.82x of 1-core), so
freeing a CPU core via QPU offload added throughput. Cycle 3 MC
is compute-bound (4-core scaling 1.05x of 1-core — near-linear),
no free cycles to free. QPU contribution (0.45 Mblock/s in
contention) doesn't compensate for losing 1 NEON core delivering
~3.8 Mblock/s.

But 30fps@1080p floor: PASS in every config (1.4x to 15.7x
isolation margin). Per project_30fps_floor_is_fine.md, user-facing
test never fails — daily YouTube playback works fine on any CPU/QPU
split.

DEPLOYMENT RECIPE for higgs (cycle 3 confirmed split):

  IDCT (k1)  -> QPU   (R=0.92, +7% mixed, frees CPU core)
  LPF  (k2)  -> QPU   (R=0.41, +7% mixed, frees CPU core)
  MC   (k3)  -> CPU   (R=0.067, -19.5% mixed — stays on CPU)
  Entropy    -> CPU   (structurally serial)

Mixed-substrate deployment, not "QPU does everything". Realistic for
higgs: entropy + MC on 2-3 ARM cores; IDCT + LPF dispatched to QPU
concurrently; 1-2 ARM cores left for vscode etc.

New artifacts:
- src/v3d_mc_8h.comp               — GLSL kernel
- tests/vp9_mc_ref.c               — standalone C ref (REGULAR filter
                                     embedded; clean transcription)
- tests/bench_neon_mc.c            — M1'''_c + M3''' bench
- tests/bench_v3d_mc.c             — M1''' + M2''' bench with contract
                                     asserts + 30fps margin display
- tests/bench_concurrent_mc.c      — M4''' pthread bench
- external/ffmpeg-snapshot/libavcodec/aarch64/vp9mc_neon.S    (vendored)
- external/ffmpeg-snapshot/libavcodec/vp9_subpel_filters_table.c
                                     (hand-extracted; provides
                                      ff_vp9_subpel_filters symbol
                                      without dragging in full vp9dsp.c)
- docs/k3_mc_phase{1,2,3,4,5,7}.md — full cycle documentation

Memory updates: project_30fps_floor_is_fine.md (user's 30fps target
recalibration), MEMORY.md index updated.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-05-18 12:51:43 +00:00
parent 36eca40ff2
commit 356e446a49
15 changed files with 2545 additions and 1 deletions
+53 -1
View File
@@ -55,6 +55,19 @@ set_source_files_properties(${FFASM_LPF_SOURCES} PROPERTIES
COMPILE_OPTIONS "${FFASM_FLAGS}"
LANGUAGE ASM)
# Cycle 3 — VP9 MC interpolation NEON source + filter coefficient table
# (vendored 2026-05-18). The .c table provides ff_vp9_subpel_filters
# symbol which vp9mc_neon.S references via movrel.
set(FFASM_MC_SOURCES
${FFSNAP}/libavcodec/aarch64/vp9mc_neon.S
)
set(FFC_MC_SOURCES
${FFSNAP}/libavcodec/vp9_subpel_filters_table.c
)
set_source_files_properties(${FFASM_MC_SOURCES} PROPERTIES
COMPILE_OPTIONS "${FFASM_FLAGS}"
LANGUAGE ASM)
# Tell CMake/gas to preprocess .S sources.
set_source_files_properties(${FFASM_SOURCES} PROPERTIES
COMPILE_OPTIONS "${FFASM_FLAGS}"
@@ -76,6 +89,15 @@ add_executable(bench_neon_lpf
${FFASM_LPF_SOURCES}
)
target_compile_options(bench_neon_lpf PRIVATE -O3 -march=armv8-a+simd)
# Cycle 3 — VP9 MC interpolation NEON baseline.
add_executable(bench_neon_mc
tests/bench_neon_mc.c
tests/vp9_mc_ref.c
${FFASM_MC_SOURCES}
${FFC_MC_SOURCES}
)
target_compile_options(bench_neon_mc PRIVATE -O3 -march=armv8-a+simd)
# bench_neon_idct doesn't need vulkan/drm — pure CPU baseline.
# ---- Vulkan dispatch-overhead microbench (next chunk) ----------------------
@@ -125,7 +147,18 @@ if (DAEDALUS_BUILD_VULKAN)
VERBATIM
)
add_custom_target(daedalus_shaders ALL DEPENDS ${NOOP_SPV} ${IDCT8_SPV} ${LPF_SPV})
set(MC_SPV ${CMAKE_BINARY_DIR}/v3d_mc_8h.spv)
add_custom_command(
OUTPUT ${MC_SPV}
COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
-o ${MC_SPV}
${CMAKE_SOURCE_DIR}/src/v3d_mc_8h.comp
DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_mc_8h.comp
COMMENT "glslang: v3d_mc_8h.comp -> v3d_mc_8h.spv"
VERBATIM
)
add_custom_target(daedalus_shaders ALL DEPENDS ${NOOP_SPV} ${IDCT8_SPV} ${LPF_SPV} ${MC_SPV})
# v3d_runner — reusable Vulkan plumbing.
add_library(v3d_runner STATIC src/v3d_runner.c)
@@ -155,6 +188,15 @@ if (DAEDALUS_BUILD_VULKAN)
target_link_libraries(bench_v3d_lpf PRIVATE v3d_runner Vulkan::Vulkan)
target_compile_options(bench_v3d_lpf PRIVATE -O2)
# Cycle 3 — QPU MC bench.
add_executable(bench_v3d_mc
tests/bench_v3d_mc.c
tests/vp9_mc_ref.c
)
add_dependencies(bench_v3d_mc daedalus_shaders)
target_link_libraries(bench_v3d_mc PRIVATE v3d_runner Vulkan::Vulkan)
target_compile_options(bench_v3d_mc PRIVATE -O2)
# M4 — concurrent CPU(NEON) + QPU bench. Links the FFmpeg NEON
# snapshot so we can run real NEON kernels on pinned CPU cores
# while the QPU runs its dispatch loop concurrently.
@@ -174,6 +216,16 @@ if (DAEDALUS_BUILD_VULKAN)
add_dependencies(bench_concurrent_lpf daedalus_shaders)
target_link_libraries(bench_concurrent_lpf PRIVATE v3d_runner Vulkan::Vulkan pthread)
target_compile_options(bench_concurrent_lpf PRIVATE -O3 -march=armv8-a+simd)
# Cycle 3 M4''' — concurrent MC.
add_executable(bench_concurrent_mc
tests/bench_concurrent_mc.c
${FFASM_MC_SOURCES}
${FFC_MC_SOURCES}
)
add_dependencies(bench_concurrent_mc daedalus_shaders)
target_link_libraries(bench_concurrent_mc PRIVATE v3d_runner Vulkan::Vulkan pthread)
target_compile_options(bench_concurrent_mc PRIVATE -O3 -march=armv8-a+simd)
endif()
# ---- Summary ----------------------------------------------------------------