356e446a49
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>
238 lines
8.4 KiB
CMake
238 lines
8.4 KiB
CMake
# daedalus-fourier — Phase 3 baseline + (later) Phase 6 implementation.
|
|
#
|
|
# Builds:
|
|
# bench_neon_idct — NEON throughput baseline (Phase 3 M3) +
|
|
# bit-exact correctness gate (Phase 1 M1).
|
|
# bench_vulkan_dispatch — Vulkan compute dispatch-overhead baseline (M5).
|
|
#
|
|
# Linkage note: bench_neon_idct statically links the vendored
|
|
# FFmpeg n7.1.3 NEON snapshot (LGPL-2.1+); see
|
|
# external/ffmpeg-snapshot/PROVENANCE.md.
|
|
|
|
cmake_minimum_required(VERSION 3.20)
|
|
project(daedalus-fourier C ASM)
|
|
|
|
set(CMAKE_C_STANDARD 11)
|
|
set(CMAKE_C_STANDARD_REQUIRED ON)
|
|
|
|
if (NOT CMAKE_BUILD_TYPE)
|
|
set(CMAKE_BUILD_TYPE Release)
|
|
endif()
|
|
|
|
if (NOT CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64")
|
|
message(FATAL_ERROR
|
|
"daedalus-fourier targets aarch64 (Pi 5 / BCM2712). "
|
|
"Cross-compile not yet wired.")
|
|
endif()
|
|
|
|
add_compile_options(-Wall -Wextra -Wno-unused-parameter)
|
|
|
|
# ---- Vendored FFmpeg snapshot (LGPL-2.1+) -----------------------------------
|
|
|
|
set(FFSNAP ${CMAKE_SOURCE_DIR}/external/ffmpeg-snapshot)
|
|
|
|
# Assembly preamble (config.h shim + FFmpeg's asm helpers) used by the
|
|
# vendored .S file. -I flags expose:
|
|
# - FFSNAP/ so `#include "config.h"` finds our shim
|
|
# - FFSNAP/libavcodec/aarch64/ so `#include "neon.S"` finds the helper
|
|
# - FFSNAP/ so `#include "libavutil/aarch64/asm.S"`
|
|
# resolves against the vendored copy
|
|
set(FFASM_FLAGS
|
|
-I${FFSNAP}
|
|
-I${FFSNAP}/libavcodec/aarch64
|
|
-I${FFSNAP}
|
|
)
|
|
|
|
set(FFASM_SOURCES
|
|
${FFSNAP}/libavcodec/aarch64/vp9itxfm_neon.S
|
|
)
|
|
|
|
# Cycle 2 — VP9 loop filter NEON source (vendored 2026-05-18).
|
|
set(FFASM_LPF_SOURCES
|
|
${FFSNAP}/libavcodec/aarch64/vp9lpf_neon.S
|
|
)
|
|
set_source_files_properties(${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}"
|
|
LANGUAGE ASM)
|
|
|
|
# ---- NEON baseline microbenches --------------------------------------------
|
|
|
|
add_executable(bench_neon_idct
|
|
tests/bench_neon_idct.c
|
|
tests/vp9_idct8_ref.c
|
|
${FFASM_SOURCES}
|
|
)
|
|
target_compile_options(bench_neon_idct PRIVATE -O3 -march=armv8-a+simd)
|
|
|
|
# Cycle 2 — VP9 loop filter NEON baseline.
|
|
add_executable(bench_neon_lpf
|
|
tests/bench_neon_lpf.c
|
|
tests/vp9_lpf_ref.c
|
|
${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) ----------------------
|
|
# Stub: written in a follow-up step. Toggle ON with -DDAEDALUS_BUILD_VULKAN=ON
|
|
# once tests/bench_vulkan_dispatch.c exists.
|
|
|
|
option(DAEDALUS_BUILD_VULKAN "Build Vulkan compute-dispatch microbench" ON)
|
|
|
|
if (DAEDALUS_BUILD_VULKAN)
|
|
find_package(Vulkan REQUIRED)
|
|
|
|
# Compile GLSL compute shaders to SPIR-V via glslangValidator.
|
|
# The binary loads them at runtime from the build dir (cwd-relative).
|
|
find_program(GLSLANG_VALIDATOR
|
|
NAMES glslangValidator glslang
|
|
REQUIRED)
|
|
|
|
set(NOOP_SPV ${CMAKE_BINARY_DIR}/noop.spv)
|
|
add_custom_command(
|
|
OUTPUT ${NOOP_SPV}
|
|
COMMAND ${GLSLANG_VALIDATOR} -V -o ${NOOP_SPV}
|
|
${CMAKE_SOURCE_DIR}/tests/shaders/noop.comp
|
|
DEPENDS ${CMAKE_SOURCE_DIR}/tests/shaders/noop.comp
|
|
COMMENT "glslang: noop.comp -> noop.spv"
|
|
VERBATIM
|
|
)
|
|
|
|
set(IDCT8_SPV ${CMAKE_BINARY_DIR}/v3d_idct8.spv)
|
|
add_custom_command(
|
|
OUTPUT ${IDCT8_SPV}
|
|
COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
|
|
-o ${IDCT8_SPV}
|
|
${CMAKE_SOURCE_DIR}/src/v3d_idct8.comp
|
|
DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_idct8.comp
|
|
COMMENT "glslang: v3d_idct8.comp -> v3d_idct8.spv"
|
|
VERBATIM
|
|
)
|
|
|
|
set(LPF_SPV ${CMAKE_BINARY_DIR}/v3d_lpf_h_4_8.spv)
|
|
add_custom_command(
|
|
OUTPUT ${LPF_SPV}
|
|
COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
|
|
-o ${LPF_SPV}
|
|
${CMAKE_SOURCE_DIR}/src/v3d_lpf_h_4_8.comp
|
|
DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_lpf_h_4_8.comp
|
|
COMMENT "glslang: v3d_lpf_h_4_8.comp -> v3d_lpf_h_4_8.spv"
|
|
VERBATIM
|
|
)
|
|
|
|
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)
|
|
target_include_directories(v3d_runner PUBLIC src)
|
|
target_link_libraries(v3d_runner PUBLIC Vulkan::Vulkan)
|
|
target_compile_options(v3d_runner PRIVATE -O2)
|
|
|
|
add_executable(bench_vulkan_dispatch tests/bench_vulkan_dispatch.c)
|
|
add_dependencies(bench_vulkan_dispatch daedalus_shaders)
|
|
target_link_libraries(bench_vulkan_dispatch PRIVATE Vulkan::Vulkan)
|
|
target_compile_options(bench_vulkan_dispatch PRIVATE -O2)
|
|
|
|
add_executable(bench_v3d_idct
|
|
tests/bench_v3d_idct.c
|
|
tests/vp9_idct8_ref.c
|
|
)
|
|
add_dependencies(bench_v3d_idct daedalus_shaders)
|
|
target_link_libraries(bench_v3d_idct PRIVATE v3d_runner Vulkan::Vulkan)
|
|
target_compile_options(bench_v3d_idct PRIVATE -O2)
|
|
|
|
# Cycle 2 — QPU LPF bench.
|
|
add_executable(bench_v3d_lpf
|
|
tests/bench_v3d_lpf.c
|
|
tests/vp9_lpf_ref.c
|
|
)
|
|
add_dependencies(bench_v3d_lpf daedalus_shaders)
|
|
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.
|
|
add_executable(bench_concurrent
|
|
tests/bench_concurrent.c
|
|
${FFASM_SOURCES}
|
|
)
|
|
add_dependencies(bench_concurrent daedalus_shaders)
|
|
target_link_libraries(bench_concurrent PRIVATE v3d_runner Vulkan::Vulkan pthread)
|
|
target_compile_options(bench_concurrent PRIVATE -O3 -march=armv8-a+simd)
|
|
|
|
# Cycle 2 M4'' — concurrent LPF.
|
|
add_executable(bench_concurrent_lpf
|
|
tests/bench_concurrent_lpf.c
|
|
${FFASM_LPF_SOURCES}
|
|
)
|
|
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 ----------------------------------------------------------------
|
|
|
|
message(STATUS "daedalus-fourier build configured for ${CMAKE_SYSTEM_PROCESSOR}")
|
|
message(STATUS " FFmpeg snapshot: ${FFSNAP}")
|
|
message(STATUS " Build type: ${CMAKE_BUILD_TYPE}")
|
|
message(STATUS " Targets: bench_neon_idct"
|
|
"$<$<BOOL:${DAEDALUS_BUILD_VULKAN}>:; bench_vulkan_dispatch>")
|