wip: mc02 v2 shared-tile

Closes task #166 (re-measure R-bands on post-buffer-pool dispatch path).

Now that all H.264 hot-path primitives have QPU shaders and the
dispatch overhead has been hammered down (tasks #160 buffer pool,
#161 persistent command buffer), bench_h264_primitives no longer
measures one column.  Two passes — CPU NEON and QPU V3D7 compute —
with a side-by-side per-kernel comparison and ratio.

Headline result on hertz (Pi 5 V3D 7.1, 30 iters x 5 warmup):

  kernel             CPU ns/op  QPU ns/op  winner
  IDCT 4x4 luma          10.79       2.47  QPU 4.36x
  IDCT 8x8 luma          29.69       9.23  QPU 3.22x
  Deblock luma_v         17.58      10.21  QPU 1.72x
  Deblock luma_h         38.41       9.98  QPU 3.85x
  qpel mc20 (8x8)        28.24       9.66  QPU 2.92x
  qpel mc02 (8x8)        16.96      20.54  CPU 1.21x
  qpel mc22 (8x8)        71.58       9.64  QPU 7.43x

  1080p worst-case sum (IDCT4 + deblock luma + qpel mc22):
    CPU NEON only:  5.57 ms
    QPU only:       1.30 ms   (CPU/QPU sum ratio = 4.30x)

Reverses PR #10's verdict (which had CPU NEON 4x faster than QPU
for IDCT-only) — the buffer-pool + persistent-cmdbuf wins land
hard.  Only qpel mc02 still shows CPU ahead, marginally (single-
axis vertical filter, row-strided memory pattern unfriendly to the
WG layout — left as a follow-up for cycle-9-style targeted tuning).

Substrate decree (2026-05-23) stays in force as policy — these
numbers retroactively justify it.

Also tightens test_api_h264's startup recipe print: the stale
"(CPU)" / "(CPU, no QPU H shader yet)" / "(CPU, bS=4 set)" labels
next to deblock_lh, deblock_cv, deblock_ch and deblock_*_intra
are now wrong since PRs #28, #29, #35 (those kernels are on QPU).

CMakeLists.txt

@@ -284,6 +284,55 @@ if (DAEDALUS_BUILD_VULKAN)
         VERBATIM
     )
 
+    set(H264DEBLOCK_H_SPV ${CMAKE_BINARY_DIR}/v3d_h264deblock_h.spv)
+    add_custom_command(
+        OUTPUT ${H264DEBLOCK_H_SPV}
+        COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                -o ${H264DEBLOCK_H_SPV}
+                ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_h.comp
+        DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_h.comp
+        COMMENT "glslang: v3d_h264deblock_h.comp -> v3d_h264deblock_h.spv"
+        VERBATIM
+    )
+
+    set(H264DEBLOCK_CHROMA_V_SPV ${CMAKE_BINARY_DIR}/v3d_h264deblock_chroma_v.spv)
+    add_custom_command(
+        OUTPUT ${H264DEBLOCK_CHROMA_V_SPV}
+        COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                -o ${H264DEBLOCK_CHROMA_V_SPV}
+                ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_chroma_v.comp
+        DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_chroma_v.comp
+        COMMENT "glslang: v3d_h264deblock_chroma_v.comp -> .spv"
+        VERBATIM
+    )
+
+    set(H264DEBLOCK_CHROMA_H_SPV ${CMAKE_BINARY_DIR}/v3d_h264deblock_chroma_h.spv)
+    add_custom_command(
+        OUTPUT ${H264DEBLOCK_CHROMA_H_SPV}
+        COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                -o ${H264DEBLOCK_CHROMA_H_SPV}
+                ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_chroma_h.comp
+        DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_chroma_h.comp
+        COMMENT "glslang: v3d_h264deblock_chroma_h.comp -> .spv"
+        VERBATIM
+    )
+
+    # Intra (bS=4) deblock shaders — strong/weak filter selector per
+    # H.264 §8.3.2.3.  4 variants (luma_v/h + chroma_v/h).
+    foreach(_kind luma_v_intra luma_h_intra chroma_v_intra chroma_h_intra)
+        set(_spv ${CMAKE_BINARY_DIR}/v3d_h264deblock_${_kind}.spv)
+        add_custom_command(
+            OUTPUT ${_spv}
+            COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                    -o ${_spv}
+                    ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_${_kind}.comp
+            DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264deblock_${_kind}.comp
+            COMMENT "glslang: v3d_h264deblock_${_kind}.comp -> .spv"
+            VERBATIM
+        )
+        set(H264DEBLOCK_${_kind}_SPV ${_spv})
+    endforeach()
+
     set(H264_IDCT4_SPV ${CMAKE_BINARY_DIR}/v3d_h264_idct4.spv)
     add_custom_command(
         OUTPUT ${H264_IDCT4_SPV}
@@ -317,7 +366,63 @@ if (DAEDALUS_BUILD_VULKAN)
         VERBATIM
     )
 
-    add_custom_target(daedalus_shaders ALL DEPENDS ${NOOP_SPV} ${IDCT8_SPV} ${LPF_SPV} ${MC_SPV} ${LPF8_SPV} ${CDEF_SPV} ${H264DEBLOCK_SPV} ${H264_IDCT4_SPV} ${H264_IDCT8_SPV} ${H264_QPEL_MC20_SPV})
+    set(H264_QPEL_MC02_SPV ${CMAKE_BINARY_DIR}/v3d_h264_qpel_mc02.spv)
+    add_custom_command(
+        OUTPUT ${H264_QPEL_MC02_SPV}
+        COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                -o ${H264_QPEL_MC02_SPV}
+                ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_mc02.comp
+        DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_mc02.comp
+        COMMENT "glslang: v3d_h264_qpel_mc02.comp -> v3d_h264_qpel_mc02.spv"
+        VERBATIM
+    )
+
+    set(H264_QPEL_MC22_SPV ${CMAKE_BINARY_DIR}/v3d_h264_qpel_mc22.spv)
+    add_custom_command(
+        OUTPUT ${H264_QPEL_MC22_SPV}
+        COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                -o ${H264_QPEL_MC22_SPV}
+                ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_mc22.comp
+        DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_mc22.comp
+        COMMENT "glslang: v3d_h264_qpel_mc22.comp -> v3d_h264_qpel_mc22.spv"
+        VERBATIM
+    )
+
+    # Quarter-pel single-axis variants (mc10/30/01/03) + diagonal
+    # variants (mc11/12/13/21/23/31/32/33) — each composes 1-2 half-pel
+    # results with optional L2 averaging.  Same WG geometry as mc20/mc02.
+    foreach(_mc mc10 mc30 mc01 mc03 mc11 mc12 mc13 mc21 mc23 mc31 mc32 mc33)
+        set(_spv ${CMAKE_BINARY_DIR}/v3d_h264_qpel_${_mc}.spv)
+        add_custom_command(
+            OUTPUT ${_spv}
+            COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                    -o ${_spv}
+                    ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_${_mc}.comp
+            DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_${_mc}.comp
+            COMMENT "glslang: v3d_h264_qpel_${_mc}.comp -> .spv"
+            VERBATIM
+        )
+        set(H264_QPEL_${_mc}_SPV ${_spv})
+    endforeach()
+
+    # avg_ biprediction variants — same shader as put_ + extra L2 with
+    # existing dst.  All 15 useful positions.
+    foreach(_mc mc20 mc02 mc22 mc10 mc30 mc01 mc03
+                mc11 mc12 mc13 mc21 mc23 mc31 mc32 mc33)
+        set(_spv ${CMAKE_BINARY_DIR}/v3d_h264_qpel_avg_${_mc}.spv)
+        add_custom_command(
+            OUTPUT ${_spv}
+            COMMAND ${GLSLANG_VALIDATOR} -V --target-env vulkan1.3
+                    -o ${_spv}
+                    ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_avg_${_mc}.comp
+            DEPENDS ${CMAKE_SOURCE_DIR}/src/v3d_h264_qpel_avg_${_mc}.comp
+            COMMENT "glslang: v3d_h264_qpel_avg_${_mc}.comp -> .spv"
+            VERBATIM
+        )
+        set(H264_QPEL_avg_${_mc}_SPV ${_spv})
+    endforeach()
+
+    add_custom_target(daedalus_shaders ALL DEPENDS ${NOOP_SPV} ${IDCT8_SPV} ${LPF_SPV} ${MC_SPV} ${LPF8_SPV} ${CDEF_SPV} ${H264DEBLOCK_SPV} ${H264DEBLOCK_H_SPV} ${H264DEBLOCK_CHROMA_V_SPV} ${H264DEBLOCK_CHROMA_H_SPV} ${H264DEBLOCK_luma_v_intra_SPV} ${H264DEBLOCK_luma_h_intra_SPV} ${H264DEBLOCK_chroma_v_intra_SPV} ${H264DEBLOCK_chroma_h_intra_SPV} ${H264_IDCT4_SPV} ${H264_IDCT8_SPV} ${H264_QPEL_MC20_SPV} ${H264_QPEL_MC02_SPV} ${H264_QPEL_MC22_SPV} ${H264_QPEL_mc10_SPV} ${H264_QPEL_mc30_SPV} ${H264_QPEL_mc01_SPV} ${H264_QPEL_mc03_SPV} ${H264_QPEL_mc11_SPV} ${H264_QPEL_mc12_SPV} ${H264_QPEL_mc13_SPV} ${H264_QPEL_mc21_SPV} ${H264_QPEL_mc23_SPV} ${H264_QPEL_mc31_SPV} ${H264_QPEL_mc32_SPV} ${H264_QPEL_mc33_SPV} ${H264_QPEL_avg_mc20_SPV} ${H264_QPEL_avg_mc02_SPV} ${H264_QPEL_avg_mc22_SPV} ${H264_QPEL_avg_mc10_SPV} ${H264_QPEL_avg_mc30_SPV} ${H264_QPEL_avg_mc01_SPV} ${H264_QPEL_avg_mc03_SPV} ${H264_QPEL_avg_mc11_SPV} ${H264_QPEL_avg_mc12_SPV} ${H264_QPEL_avg_mc13_SPV} ${H264_QPEL_avg_mc21_SPV} ${H264_QPEL_avg_mc23_SPV} ${H264_QPEL_avg_mc31_SPV} ${H264_QPEL_avg_mc32_SPV} ${H264_QPEL_avg_mc33_SPV})
 
     # v3d_runner — reusable Vulkan plumbing.
     add_library(v3d_runner STATIC src/v3d_runner.c)
@@ -392,6 +497,10 @@ endif()
 add_library(daedalus_core STATIC
     src/daedalus_core.c
     src/h264_chroma_dc.c
+    src/h264_intra_pred_4x4.c
+    src/h264_intra_pred_16x16.c
+    src/h264_intra_pred_chroma8x8.c
+    src/h264_intra_pred_8x8_luma.c
     src/v3d_runner.c
     ${FFASM_SOURCES}
     ${FFASM_LPF_SOURCES}
@@ -446,9 +555,45 @@ if (DAEDALUS_BUILD_VULKAN)
         ${LPF8_SPV}
         ${CDEF_SPV}
         ${H264DEBLOCK_SPV}
+        ${H264DEBLOCK_H_SPV}
+        ${H264DEBLOCK_CHROMA_V_SPV}
+        ${H264DEBLOCK_CHROMA_H_SPV}
+        ${H264DEBLOCK_luma_v_intra_SPV}
+        ${H264DEBLOCK_luma_h_intra_SPV}
+        ${H264DEBLOCK_chroma_v_intra_SPV}
+        ${H264DEBLOCK_chroma_h_intra_SPV}
         ${H264_IDCT4_SPV}
         ${H264_IDCT8_SPV}
         ${H264_QPEL_MC20_SPV}
+        ${H264_QPEL_MC02_SPV}
+        ${H264_QPEL_MC22_SPV}
+        ${H264_QPEL_mc10_SPV}
+        ${H264_QPEL_mc30_SPV}
+        ${H264_QPEL_mc01_SPV}
+        ${H264_QPEL_mc03_SPV}
+        ${H264_QPEL_mc11_SPV}
+        ${H264_QPEL_mc12_SPV}
+        ${H264_QPEL_mc13_SPV}
+        ${H264_QPEL_mc21_SPV}
+        ${H264_QPEL_mc23_SPV}
+        ${H264_QPEL_mc31_SPV}
+        ${H264_QPEL_mc32_SPV}
+        ${H264_QPEL_mc33_SPV}
+        ${H264_QPEL_avg_mc20_SPV}
+        ${H264_QPEL_avg_mc02_SPV}
+        ${H264_QPEL_avg_mc22_SPV}
+        ${H264_QPEL_avg_mc10_SPV}
+        ${H264_QPEL_avg_mc30_SPV}
+        ${H264_QPEL_avg_mc01_SPV}
+        ${H264_QPEL_avg_mc03_SPV}
+        ${H264_QPEL_avg_mc11_SPV}
+        ${H264_QPEL_avg_mc12_SPV}
+        ${H264_QPEL_avg_mc13_SPV}
+        ${H264_QPEL_avg_mc21_SPV}
+        ${H264_QPEL_avg_mc23_SPV}
+        ${H264_QPEL_avg_mc31_SPV}
+        ${H264_QPEL_avg_mc32_SPV}
+        ${H264_QPEL_avg_mc33_SPV}
         DESTINATION ${CMAKE_INSTALL_DATADIR}/daedalus-fourier/shaders
     )
 endif()
@@ -538,38 +683,29 @@ add_executable(test_api_opportunistic_qpu tests/test_api_opportunistic_qpu.c)
 target_link_libraries(test_api_opportunistic_qpu PRIVATE daedalus_core)
 target_compile_options(test_api_opportunistic_qpu PRIVATE -O2)
 
-# H.264 Intra_4x4 luma prediction (9 modes) — reference + tests.
+# H.264 Intra_4x4 luma prediction (9 modes) — public src primitives.
-# Pure CPU + spec-derived; no daedalus_core dependency yet (this is
+# The bodies now live in src/h264_intra_pred_4x4.c (linked into
-# the bit-exact gate for the eventual shader / dispatch wiring).
+# daedalus_core for use by libavcodec.so substitution-arc consumers).
-add_executable(test_intra_pred_4x4
+# This test exercises the public symbols.
-    tests/test_intra_pred_4x4.c
+add_executable(test_intra_pred_4x4 tests/test_intra_pred_4x4.c)
-    tests/h264_intra_pred_4x4_ref.c
+target_link_libraries(test_intra_pred_4x4 PRIVATE daedalus_core)
-)
 target_compile_options(test_intra_pred_4x4 PRIVATE -O2)
 
-# H.264 Intra_16x16 luma prediction (4 modes: V, H, DC, Plane) —
+# H.264 Intra_16x16 luma prediction (4 modes) — public src primitives,
-# reference + tests.  Same spec-gate role as the 4x4 sibling.
+# linked from daedalus_core.
-add_executable(test_intra_pred_16x16
+add_executable(test_intra_pred_16x16 tests/test_intra_pred_16x16.c)
-    tests/test_intra_pred_16x16.c
+target_link_libraries(test_intra_pred_16x16 PRIVATE daedalus_core)
-    tests/h264_intra_pred_16x16_ref.c
-)
 target_compile_options(test_intra_pred_16x16 PRIVATE -O2)
 
-# H.264 Intra_8x8 chroma prediction (4 modes: DC, H, V, Plane) —
+# H.264 Intra_8x8 chroma prediction (4 modes) — public src primitives.
-# reference + tests.  DC is per-quadrant (asymmetric); Plane uses
+add_executable(test_intra_pred_chroma8x8 tests/test_intra_pred_chroma8x8.c)
-# slope coefficient 34 instead of luma's 5.
+target_link_libraries(test_intra_pred_chroma8x8 PRIVATE daedalus_core)
-add_executable(test_intra_pred_chroma8x8
-    tests/test_intra_pred_chroma8x8.c
-    tests/h264_intra_pred_chroma8x8_ref.c
-)
 target_compile_options(test_intra_pred_chroma8x8 PRIVATE -O2)
 
 # H.264 Intra_8x8 luma prediction (High profile, 9 modes + 1-2-1
-# reference-sample pre-filter).
+# pre-filter) — public src primitives.
-add_executable(test_intra_pred_8x8_luma
+add_executable(test_intra_pred_8x8_luma tests/test_intra_pred_8x8_luma.c)
-    tests/test_intra_pred_8x8_luma.c
+target_link_libraries(test_intra_pred_8x8_luma PRIVATE daedalus_core)
-    tests/h264_intra_pred_8x8_luma_ref.c
-)
 target_compile_options(test_intra_pred_8x8_luma PRIVATE -O2)
 
 # H.264 chroma DC 2x2 Hadamard pre-pass primitive.  Pure transform,

include/daedalus.h

@@ -559,6 +559,73 @@ DECLARE_QPEL_AVG(avg_mc33)
  * ----------------------------------------------------------------- */
 void daedalus_h264_chroma_dc_hadamard_2x2(int16_t c[4]);
 
+/* -------------------------------------------------------------------
+ * H.264 Intra_4x4 luma prediction (per H.264 §8.3.1.4).  9 modes.
+ *
+ * Pure CPU primitives — each is a small straightforward fill of a
+ * 4x4 output block from neighbour pixels in the same buffer.  No
+ * substrate-dispatch wrapper (the work is too small to amortise).
+ *
+ * FFmpeg-style interface: `dst` at row 0 col 0 of the 4x4 output.
+ * Reads top-left at dst[-stride-1], top at dst[-stride..-stride+7]
+ * (top-right for DDL/VL), and left at dst[r*stride - 1] for r=0..3.
+ * Caller must ensure all 13 neighbour bytes are valid (interior-MB
+ * assumption — H.264 availability fallback handled at caller).
+ *
+ * Bit-exact validated against tests/test_intra_pred_4x4.c (10-case
+ * spec-derived test suite including the asymmetric Vertical_Right
+ * 16-cell hand-derived case; see fourier PR #12).
+ * ----------------------------------------------------------------- */
+void daedalus_h264_pred_4x4_vertical  (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_horizontal(uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_dc        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_ddl       (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_ddr       (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_vr        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_hd        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_vl        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_4x4_hu        (uint8_t *dst, ptrdiff_t stride);
+
+/* -------------------------------------------------------------------
+ * H.264 Intra_16x16 luma prediction (per §8.3.2).  4 modes:
+ * Vertical / Horizontal / DC / Plane.  Same FFmpeg-style interface
+ * as the 4x4 family at 16x16 scale.  Same neighbour availability
+ * assumption (interior-MB).
+ * ----------------------------------------------------------------- */
+void daedalus_h264_pred_16x16_vertical  (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_16x16_horizontal(uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_16x16_dc        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_16x16_plane     (uint8_t *dst, ptrdiff_t stride);
+
+/* -------------------------------------------------------------------
+ * H.264 Intra_8x8 chroma prediction (per §8.3.3, 4:2:0).  4 modes:
+ * DC / Horizontal / Vertical / Plane.  Note: DC is per-quadrant
+ * asymmetric; Plane uses slope coefficient 34 (not luma's 5).
+ * ----------------------------------------------------------------- */
+void daedalus_h264_pred_chroma8x8_dc        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_chroma8x8_horizontal(uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_chroma8x8_vertical  (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_chroma8x8_plane     (uint8_t *dst, ptrdiff_t stride);
+
+/* -------------------------------------------------------------------
+ * H.264 Intra_8x8 luma prediction (High profile, per §8.3.2.1).
+ * 9 modes with the spec-defined 1-2-1 reference-sample pre-filter
+ * applied internally to the 25 neighbours before the mode arithmetic.
+ *
+ * "_8x8l" naming follows the FFmpeg h264pred_template convention
+ * (pred8x8l_<mode>_c) to keep the substitution wrappers a 1:1 name
+ * map.
+ * ----------------------------------------------------------------- */
+void daedalus_h264_pred_8x8l_vertical  (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_horizontal(uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_dc        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_ddl       (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_ddr       (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_vr        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_hd        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_vl        (uint8_t *dst, ptrdiff_t stride);
+void daedalus_h264_pred_8x8l_hu        (uint8_t *dst, ptrdiff_t stride);
+
 /* -------------------------------------------------------------------
  * Recipe query — what does the API recommend for each kernel?
  * ----------------------------------------------------------------- */

src/h264_intra_pred_16x16.c

@@ -29,7 +29,7 @@
 static inline int clip_u8(int v) { return v < 0 ? 0 : v > 255 ? 255 : v; }
 
 /* Mode 0 — Vertical: each col = top[col]. */
-void daedalus_h264_pred_16x16_vertical_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_16x16_vertical(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     for (int r = 0; r < 16; r++)
@@ -37,7 +37,7 @@ void daedalus_h264_pred_16x16_vertical_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 1 — Horizontal: each row = left[row]. */
-void daedalus_h264_pred_16x16_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_16x16_horizontal(uint8_t *dst, ptrdiff_t stride)
 {
     for (int r = 0; r < 16; r++) {
         uint8_t l = dst[r * stride - 1];
@@ -46,7 +46,7 @@ void daedalus_h264_pred_16x16_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 2 — DC: ((sum_top16 + sum_left16 + 16) >> 5) broadcast. */
-void daedalus_h264_pred_16x16_dc_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_16x16_dc(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int sum = 16;  /* rounding for >> 5 over 32 samples */
@@ -77,7 +77,7 @@ void daedalus_h264_pred_16x16_dc_ref(uint8_t *dst, ptrdiff_t stride)
  * (it does NOT participate in the H/V sums in the 16x16 case — only
  * for the chroma 8x8 plane mode).
  */
-void daedalus_h264_pred_16x16_plane_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_16x16_plane(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     /* H accumulates differences across the right vs left half of the

src/h264_intra_pred_4x4.c

@@ -52,7 +52,7 @@ static inline uint8_t avg2(int a, int b)
 }
 
 /* Mode 0 — Vertical: each col = top[col]. */
-void daedalus_h264_pred_4x4_vertical_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_vertical(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     for (int r = 0; r < 4; r++) {
@@ -61,7 +61,7 @@ void daedalus_h264_pred_4x4_vertical_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 1 — Horizontal: each row = left[row]. */
-void daedalus_h264_pred_4x4_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_horizontal(uint8_t *dst, ptrdiff_t stride)
 {
     for (int r = 0; r < 4; r++) {
         uint8_t l = dst[r * stride - 1];
@@ -70,7 +70,7 @@ void daedalus_h264_pred_4x4_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 2 — DC: mean of top 4 + left 4, broadcast. */
-void daedalus_h264_pred_4x4_dc_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_dc(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int sum = 4;  /* rounding for ((sum + 4) >> 3) */
@@ -82,7 +82,7 @@ void daedalus_h264_pred_4x4_dc_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 3 — Diagonal_Down_Left.  Uses top[0..7] (incl. top-right). */
-void daedalus_h264_pred_4x4_ddl_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_ddl(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int t0 = top[0], t1 = top[1], t2 = top[2], t3 = top[3];
@@ -102,7 +102,7 @@ void daedalus_h264_pred_4x4_ddl_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 4 — Diagonal_Down_Right.  Uses top-left + top[0..3] + left[0..3]. */
-void daedalus_h264_pred_4x4_ddr_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_ddr(uint8_t *dst, ptrdiff_t stride)
 {
     int tl = dst[-stride - 1];
     int t0 = dst[-stride + 0], t1 = dst[-stride + 1];
@@ -123,7 +123,7 @@ void daedalus_h264_pred_4x4_ddr_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 5 — Vertical_Right. */
-void daedalus_h264_pred_4x4_vr_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_vr(uint8_t *dst, ptrdiff_t stride)
 {
     int tl = dst[-stride - 1];
     int t0 = dst[-stride + 0], t1 = dst[-stride + 1];
@@ -153,7 +153,7 @@ void daedalus_h264_pred_4x4_vr_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 6 — Horizontal_Down. */
-void daedalus_h264_pred_4x4_hd_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_hd(uint8_t *dst, ptrdiff_t stride)
 {
     int tl = dst[-stride - 1];
     int t0 = dst[-stride + 0], t1 = dst[-stride + 1], t2 = dst[-stride + 2];
@@ -182,7 +182,7 @@ void daedalus_h264_pred_4x4_hd_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 7 — Vertical_Left.  Uses top[0..7]. */
-void daedalus_h264_pred_4x4_vl_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_vl(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int t0=top[0], t1=top[1], t2=top[2], t3=top[3];
@@ -211,7 +211,7 @@ void daedalus_h264_pred_4x4_vl_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 8 — Horizontal_Up.  Uses left[0..3] only. */
-void daedalus_h264_pred_4x4_hu_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_4x4_hu(uint8_t *dst, ptrdiff_t stride)
 {
     int l0 = dst[ 0*stride - 1], l1 = dst[ 1*stride - 1];
     int l2 = dst[ 2*stride - 1], l3 = dst[ 3*stride - 1];

src/h264_intra_pred_8x8_luma.c

@@ -90,7 +90,7 @@ static void filter_refs(const uint8_t *dst, ptrdiff_t stride,
 #define FTL    filt[0]          /* filtered top-left              */
 
 /* Mode 0 Vertical (§8.3.2.1.2): pred[r,c] = filt_top[c]. */
-void daedalus_h264_pred_8x8l_vertical_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_vertical(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -99,7 +99,7 @@ void daedalus_h264_pred_8x8l_vertical_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 1 Horizontal (§8.3.2.1.3): pred[r,c] = filt_left[r]. */
-void daedalus_h264_pred_8x8l_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_horizontal(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -110,7 +110,7 @@ void daedalus_h264_pred_8x8l_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
 /* Mode 2 DC (§8.3.2.1.4): ((sum_filt_top[0..7] + sum_filt_left[0..7]
  * + 8) >> 4) broadcast.  Note the +8 (not +4 like 4x4): there are
  * 16 samples summed total, so >> 4 with half-step rounding +8. */
-void daedalus_h264_pred_8x8l_dc_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_dc(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -143,7 +143,7 @@ void daedalus_h264_pred_8x8l_dc_ref(uint8_t *dst, ptrdiff_t stride)
 #define LT    FTL
 
 /* Mode 3 DDL (Diagonal_Down_Left) — uses TOP + TOP_RIGHT, no LEFT. */
-void daedalus_h264_pred_8x8l_ddl_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_ddl(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -165,7 +165,7 @@ void daedalus_h264_pred_8x8l_ddl_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 4 DDR (Diagonal_Down_Right). */
-void daedalus_h264_pred_8x8l_ddr_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_ddr(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -187,7 +187,7 @@ void daedalus_h264_pred_8x8l_ddr_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 5 VR (Vertical_Right). */
-void daedalus_h264_pred_8x8l_vr_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_vr(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -216,7 +216,7 @@ void daedalus_h264_pred_8x8l_vr_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 6 HD (Horizontal_Down). */
-void daedalus_h264_pred_8x8l_hd_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_hd(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -245,7 +245,7 @@ void daedalus_h264_pred_8x8l_hd_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 7 VL (Vertical_Left) — uses TOP + TOP_RIGHT only. */
-void daedalus_h264_pred_8x8l_vl_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_vl(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);
@@ -274,7 +274,7 @@ void daedalus_h264_pred_8x8l_vl_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 8 HU (Horizontal_Up) — uses LEFT only. */
-void daedalus_h264_pred_8x8l_hu_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_8x8l_hu(uint8_t *dst, ptrdiff_t stride)
 {
     uint8_t filt[25];
     filter_refs(dst, stride, filt);

src/h264_intra_pred_chroma8x8.c

@@ -43,7 +43,7 @@ static inline int clip_u8(int v) { return v < 0 ? 0 : v > 255 ? 255 : v; }
  * quadrant ignores the top-left-half because that half is "vertically
  * above" the top-left quadrant; the spec uses top[4..7] only.
  */
-void daedalus_h264_pred_chroma8x8_dc_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_chroma8x8_dc(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int top_lo = 0, top_hi = 0, left_lo = 0, left_hi = 0;
@@ -68,7 +68,7 @@ void daedalus_h264_pred_chroma8x8_dc_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 1 — Horizontal: each row = left[row]. */
-void daedalus_h264_pred_chroma8x8_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_chroma8x8_horizontal(uint8_t *dst, ptrdiff_t stride)
 {
     for (int r = 0; r < 8; r++) {
         uint8_t l = dst[r * stride - 1];
@@ -77,7 +77,7 @@ void daedalus_h264_pred_chroma8x8_horizontal_ref(uint8_t *dst, ptrdiff_t stride)
 }
 
 /* Mode 2 — Vertical: each col = top[col]. */
-void daedalus_h264_pred_chroma8x8_vertical_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_chroma8x8_vertical(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     for (int r = 0; r < 8; r++)
@@ -97,7 +97,7 @@ void daedalus_h264_pred_chroma8x8_vertical_ref(uint8_t *dst, ptrdiff_t stride)
  *   - Centre is (x-3, y-3) (not x-7, y-7).
  *   - Spans 4 differences per sum (not 8).
  */
-void daedalus_h264_pred_chroma8x8_plane_ref(uint8_t *dst, ptrdiff_t stride)
+void daedalus_h264_pred_chroma8x8_plane(uint8_t *dst, ptrdiff_t stride)
 {
     const uint8_t *top = dst - stride;
     int H = 0, V = 0;

src/v3d_h264_qpel_avg_mc01.comp

@@ -0,0 +1,52 @@
+// daedalus-fourier — H.264 luma qpel avg_mc01 (biprediction) (8x8, ¼-pel vertical),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "d" position:
+//
+//   dst[r,c] = ((clip255(mc02(s)[r,c]) + s[r,c] + 1) >> 1)
+//
+// Sibling of v3d_h264_qpel_mc02.comp with L2 step against src[r, c].
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc01_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint col_base = src_off + c;
+
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int vp = clamp(v >> 5, 0, 255);
+
+    int avg = (vp + s_0 + 1) >> 1;    // L2 with src[r, c]
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc02.comp

@@ -0,0 +1,77 @@
+// daedalus-fourier — H.264 luma qpel avg_mc02 (biprediction) (8x8, vertical half-pel), V3D 7.1.
+//
+// Sibling of cycle 9's v3d_h264_qpel_mc20.comp.  Same 6-tap filter,
+// transposed to vertical direction:
+//
+//   dst[r,c] = clip255(
+//       ( s[r-2,c]
+//         - 5 * s[r-1,c]
+//         + 20 * s[r,  c]
+//         + 20 * s[r+1,c]
+//         -  5 * s[r+2,c]
+//         +      s[r+3,c]
+//         + 16
+//       ) >> 5)
+//
+// src+src_off points at row 0 col 0 of the OUTPUT block; the filter
+// reads rows -2..+3 (2 rows of top context, 3 rows of bottom).
+//
+// Same WG layout as mc20: 64 lanes / 1 block-per-WG / 1 lane-per-pixel.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc02_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+
+layout(push_constant) uniform PC {
+    uint n_blocks;
+    uint stride_u8;
+    uint _pad0, _pad1;
+} pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3;
+    uint c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    // Read the 6 rows of vertical context at col (c) of THIS output row.
+    // src_off+r*stride+c is at the OUTPUT pixel position; the kernel
+    // samples r-2..r+3 along the column.  Unsigned-safe because the
+    // public API contract guarantees src_off >= 2*stride.
+    uint col_base = src_off + c;
+
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int p = clamp(v >> 5, 0, 255);
+
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + p + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc03.comp

@@ -0,0 +1,52 @@
+// daedalus-fourier — H.264 luma qpel avg_mc03 (biprediction) (8x8, ¾-pel vertical),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "n" position:
+//
+//   dst[r,c] = ((clip255(mc02(s)[r,c]) + s[r+1, c] + 1) >> 1)
+//
+// Same as mc01 but L2-averages with src[r+1, c] instead of src[r, c].
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc03_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint col_base = src_off + c;
+
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int vp = clamp(v >> 5, 0, 255);
+
+    int avg = (vp + s_p1 + 1) >> 1;   // L2 with src[r+1, c]
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc10.comp

@@ -0,0 +1,55 @@
+// daedalus-fourier — H.264 luma qpel avg_mc10 (biprediction) (8x8, ¼-pel horizontal),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "a" position:
+//
+//   dst[r,c] = ((clip255(mc20(s)[r,c]) + s[r,c] + 1) >> 1)
+//
+// = horizontal half-pel filter, clipped to u8, then L2 rounded-averaged
+// with the integer source pixel at the SAME position.  Sibling of
+// v3d_h264_qpel_mc20.comp with the L2 step added at the tail.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc10_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint row_base = src_off + r * stride + c;
+
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int hp = clamp(v >> 5, 0, 255);
+
+    // L2 average with the integer source at the SAME (r, c) position.
+    int avg = (hp + s_0 + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc11.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc11 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc11[r,c] = avg(mc20(r, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc11_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc12.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc12 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc12[r,c] = avg(mc22(r, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc12_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc13.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc13 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc13[r,c] = avg(mc20(r+1, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc13_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r+1u, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc20.comp

@@ -0,0 +1,91 @@
+// daedalus-fourier — H.264 luma qpel avg_mc20 (biprediction) (8x8, horizontal half-pel), V3D 7.1.
+//
+// H.264 spec §8.4.2.2.1 horizontal 6-tap luma interpolation:
+//
+//   dst[r,c] = clip255(
+//       ( s[r,c-2]
+//         - 5 * s[r,c-1]
+//         + 20 * s[r,c]
+//         + 20 * s[r,c+1]
+//         -  5 * s[r,c+2]
+//         +      s[r,c+3]
+//         + 16
+//       ) >> 5)
+//
+// Single-stride: dst and src share `stride` (H264QpelContext
+// convention).  src+src_off already points at the leftmost output
+// column (col 0); the filter reads cols -2..+3.  Caller guarantees
+// edge-padding context per the public API docstring.
+//
+// Workgroup layout: 64 invocations = 1 lane per output pixel.
+// 1 block per WG; n_blocks WGs total.  This is the simplest layout
+// that avoids any inter-lane communication — each lane independently
+// reads its 6 src samples and writes its 1 dst sample.  V3D's L2
+// cache handles the redundant reads from adjacent lanes.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc20_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Src {
+    uint8_t src[];
+} u_src;
+
+layout(binding = 1) buffer Dst {
+    uint8_t dst[];
+} u_dst;
+
+layout(binding = 2) readonly buffer Meta {
+    uvec4 meta[];       // .x = dst_off, .y = src_off
+} u_meta;
+
+layout(push_constant) uniform PC {
+    uint n_blocks;
+    uint stride_u8;
+    uint _pad0, _pad1;
+} pc;
+
+void main()
+{
+    // 1 block per WG, 64 lanes covering the 8x8 output block.
+    uint wg_id      = gl_WorkGroupID.x;
+    uint block_idx  = wg_id;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3;    // 0..7 (row)
+    uint c = lane & 7u;    // 0..7 (column)
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    // src points at output col 0 of the block; filter reads cols -2..+3
+    // of the current row.  Negative col arithmetic is unsigned-safe
+    // because src_off >= 2 (caller-guaranteed left context).
+    uint row_base = src_off + r * stride + c;
+
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base + 0u]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int p = clamp(v >> 5, 0, 255);
+
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + p + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc21.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc21 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc21[r,c] = avg(mc22(r, c),
+//                     mc20(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc21_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_h(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc22.comp

@@ -0,0 +1,94 @@
+// daedalus-fourier — H.264 luma qpel avg_mc22 (biprediction) (8x8, 2D half-pel "j" position).
+// V3D 7.1.
+//
+// Cascaded H+V 6-tap per H.264 §8.4.2.2.1 / FFmpeg ff_put_h264_qpel8_mc22_neon:
+//
+//   tmp[r,c] = src[r,c-2] - 5*src[r,c-1] + 20*src[r,c] + 20*src[r,c+1]
+//              - 5*src[r,c+2] + src[r,c+3]                    (int16)
+//
+//   dst[r,c] = clip255((tmp[r-2,c] - 5*tmp[r-1,c] + 20*tmp[r,c]
+//                       + 20*tmp[r+1,c] - 5*tmp[r+2,c] + tmp[r+3,c]
+//                       + 512) >> 10)
+//
+// The +512 >> 10 final scale compensates for both 6-tap scalings.
+// CANNOT just cascade mc20→mc02 because intermediate must be int16
+// (no per-stage clip), so this is a dedicated kernel.
+//
+// Per-lane structure: each lane computes its own (r, c) output by
+// running the FULL cascade — 6 horizontal lowpass int16 values for
+// rows r-2..r+3, then a vertical lowpass on those.  ~50 ALU ops per
+// lane.  No shared memory / barriers needed; V3D L2 absorbs the
+// redundant src reads across lanes.
+//
+// WG layout: 64 lanes / 1 block-per-WG / 1 lane-per-output-pixel
+// (same as mc20 / mc02).
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc22_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+
+layout(push_constant) uniform PC {
+    uint n_blocks;
+    uint stride_u8;
+    uint _pad0, _pad1;
+} pc;
+
+// Horizontal 6-tap filter at (row_off, c) — reads src at cols c-2..c+3
+// of the row identified by row_off, returns int16 intermediate (NOT
+// scaled — the v-pass does the +512 >> 10 for both stages).
+int hpel_h(uint row_off, uint c)
+{
+    int s_m2 = int(u_src.src[row_off + c - 2u]);
+    int s_m1 = int(u_src.src[row_off + c - 1u]);
+    int s_0  = int(u_src.src[row_off + c       ]);
+    int s_p1 = int(u_src.src[row_off + c + 1u]);
+    int s_p2 = int(u_src.src[row_off + c + 2u]);
+    int s_p3 = int(u_src.src[row_off + c + 3u]);
+    return s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3;
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3;
+    uint c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    // Compute 6 horizontal lowpass values at rows r-2..r+3 (relative
+    // to the output row r) of column c.  src_off+r*stride+c is the
+    // output pixel position; we sample rows r-2..r+3.
+    // Unsigned-safe because src_off >= 2*stride per the caller contract.
+    int t0 = hpel_h(src_off + (r - 2u) * stride, c);
+    int t1 = hpel_h(src_off + (r - 1u) * stride, c);
+    int t2 = hpel_h(src_off +  r       * stride, c);
+    int t3 = hpel_h(src_off + (r + 1u) * stride, c);
+    int t4 = hpel_h(src_off + (r + 2u) * stride, c);
+    int t5 = hpel_h(src_off + (r + 3u) * stride, c);
+
+    int v = t0 - 5 * t1 + 20 * t2 + 20 * t3 - 5 * t4 + t5 + 512;
+    int p = clamp(v >> 10, 0, 255);
+
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + p + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc23.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc23 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc23[r,c] = avg(mc22(r, c),
+//                     mc20(r+1, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc23_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_h(src_off, stride, r+1u, c);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc30.comp

@@ -0,0 +1,52 @@
+// daedalus-fourier — H.264 luma qpel avg_mc30 (biprediction) (8x8, ¾-pel horizontal),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "c" position:
+//
+//   dst[r,c] = ((clip255(mc20(s)[r,c]) + s[r,c+1] + 1) >> 1)
+//
+// Same as mc10 but L2-averages with src[r, c+1] instead of src[r, c].
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc30_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint row_base = src_off + r * stride + c;
+
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int hp = clamp(v >> 5, 0, 255);
+
+    int avg = (hp + s_p1 + 1) >> 1;   // L2 with src[r, c+1]
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc31.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc31 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc31[r,c] = avg(mc20(r, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc31_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc32.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc32 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc32[r,c] = avg(mc22(r, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc32_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_avg_mc33.comp

@@ -0,0 +1,96 @@
+// daedalus-fourier — H.264 luma qpel avg_mc33 (biprediction) (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc33[r,c] = avg(mc20(r+1, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+//
+// avg_ variant for B-slice biprediction per H.264 §8.4.2.3.1:
+//   dst[r,c] = avg(dst[r,c], mc33_value)
+// Caller pre-loads dst with the list0 prediction; this shader
+// folds in the list1 contribution.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r+1u, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    uint final_off = dst_off + r * stride + c;
+    int prev = int(u_dst.dst[final_off]);
+    u_dst.dst[final_off] = uint8_t((prev + avg + 1) >> 1);
+}

src/v3d_h264_qpel_mc01.comp

@@ -0,0 +1,44 @@
+// daedalus-fourier — H.264 luma qpel mc01 (8x8, ¼-pel vertical),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "d" position:
+//
+//   dst[r,c] = ((clip255(mc02(s)[r,c]) + s[r,c] + 1) >> 1)
+//
+// Sibling of v3d_h264_qpel_mc02.comp with L2 step against src[r, c].
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint col_base = src_off + c;
+
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int vp = clamp(v >> 5, 0, 255);
+
+    int avg = (vp + s_0 + 1) >> 1;    // L2 with src[r, c]
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc02.comp

@@ -0,0 +1,110 @@
+// daedalus-fourier — H.264 luma qpel mc02 (8x8, vertical half-pel), V3D 7.1.
+//
+// v2: cooperative-load shared-memory tile.
+//
+//   dst[r,c] = clip255(
+//       ( s[r-2,c]
+//         - 5 * s[r-1,c]
+//         + 20 * s[r,  c]
+//         + 20 * s[r+1,c]
+//         -  5 * s[r+2,c]
+//         +      s[r+3,c]
+//         + 16
+//       ) >> 5)
+//
+// src+src_off points at row 0 col 0 of the OUTPUT block; the filter
+// reads rows -2..+3 (2 rows of top context, 3 rows of bottom), total
+// 13 distinct source rows × 8 cols = 104 bytes per 8x8 output.
+//
+// v1 had each of the 64 lanes do 6 SSBO loads → 384 loads/WG to cover
+// 104 unique bytes (3.7x redundant), and each lane's loads were stride-
+// spaced (one cache line per byte under V3D's TMU).  PR #36 bench
+// showed mc02 was the only qpel position where CPU NEON still beat
+// QPU (16.96 ns/op CPU vs 20.54 ns/op QPU; 1.21x CPU favoring).
+//
+// v2 splits the work into a coalesced load phase + a shared-memory
+// compute phase:
+//
+//   Phase 1: each of the 64 lanes cooperatively loads the 104-byte
+//   source tile into shared memory.  Lanes 0..63 load bytes at indices
+//   0..63 (covers source rows 0..7 of the 13-row tile); lanes 0..39
+//   second-load bytes 64..103 (rows 8..12).  Reads within a row are
+//   contiguous so the SIMD groups coalesce; total SSBO loads = 104,
+//   matching the unique-byte count.
+//
+//   Phase 2: all 64 lanes compute one output pixel each, reading 6
+//   bytes from shared.  Shared-memory access on V3D is local-store
+//   backed (no TMU round-trip).
+//
+// Same WG layout as v1: 64 lanes / 1 block-per-WG / 1 lane-per-pixel.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+
+layout(push_constant) uniform PC {
+    uint n_blocks;
+    uint stride_u8;
+    uint _pad0, _pad1;
+} pc;
+
+// 13 source rows × 8 cols.  int storage (4 bytes each) — wasteful vs
+// uint8_t but avoids 8-bit-shared interop concerns on glslang+v3dv;
+// 416 bytes shared/WG is well within any reasonable local-store budget.
+shared int s_tile[13 * 8];
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    // Source-tile base: src_off points at output-row-0 col-0, the tile
+    // starts 2 rows above.  Unsigned-safe because the public API
+    // contract guarantees src_off >= 2*stride.
+    uint tile_base = src_off - 2u * stride;
+
+    // Phase 1: cooperative load — 64 lanes load 104 bytes.
+    {
+        uint sr = lane >> 3;        // 0..7
+        uint sc = lane & 7u;
+        s_tile[lane] = int(u_src.src[tile_base + sr * stride + sc]);
+    }
+    if (lane < 40u) {
+        uint idx = lane + 64u;      // 64..103
+        uint sr = idx >> 3;         // 8..12
+        uint sc = idx & 7u;
+        s_tile[idx] = int(u_src.src[tile_base + sr * stride + sc]);
+    }
+
+    barrier();
+
+    // Phase 2: each lane computes one output pixel from the shared tile.
+    uint r = lane >> 3;
+    uint c = lane & 7u;
+
+    int s_m2 = s_tile[(r + 0u) * 8u + c];
+    int s_m1 = s_tile[(r + 1u) * 8u + c];
+    int s_0  = s_tile[(r + 2u) * 8u + c];
+    int s_p1 = s_tile[(r + 3u) * 8u + c];
+    int s_p2 = s_tile[(r + 4u) * 8u + c];
+    int s_p3 = s_tile[(r + 5u) * 8u + c];
+
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int p = clamp(v >> 5, 0, 255);
+
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(p);
+}

src/v3d_h264_qpel_mc03.comp

@@ -0,0 +1,44 @@
+// daedalus-fourier — H.264 luma qpel mc03 (8x8, ¾-pel vertical),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "n" position:
+//
+//   dst[r,c] = ((clip255(mc02(s)[r,c]) + s[r+1, c] + 1) >> 1)
+//
+// Same as mc01 but L2-averages with src[r+1, c] instead of src[r, c].
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint col_base = src_off + c;
+
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int vp = clamp(v >> 5, 0, 255);
+
+    int avg = (vp + s_p1 + 1) >> 1;   // L2 with src[r+1, c]
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc10.comp

@@ -0,0 +1,47 @@
+// daedalus-fourier — H.264 luma qpel mc10 (8x8, ¼-pel horizontal),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "a" position:
+//
+//   dst[r,c] = ((clip255(mc20(s)[r,c]) + s[r,c] + 1) >> 1)
+//
+// = horizontal half-pel filter, clipped to u8, then L2 rounded-averaged
+// with the integer source pixel at the SAME position.  Sibling of
+// v3d_h264_qpel_mc20.comp with the L2 step added at the tail.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint row_base = src_off + r * stride + c;
+
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int hp = clamp(v >> 5, 0, 255);
+
+    // L2 average with the integer source at the SAME (r, c) position.
+    int avg = (hp + s_0 + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc11.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc11 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc11[r,c] = avg(mc20(r, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc12.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc12 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc12[r,c] = avg(mc22(r, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc13.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc13 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc13[r,c] = avg(mc20(r+1, c),
+//                     mc02(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r+1u, c);
+    int b = hpel_v(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc21.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc21 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc21[r,c] = avg(mc22(r, c),
+//                     mc20(r, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_h(src_off, stride, r, c);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc22.comp

@@ -0,0 +1,86 @@
+// daedalus-fourier — H.264 luma qpel mc22 (8x8, 2D half-pel "j" position).
+// V3D 7.1.
+//
+// Cascaded H+V 6-tap per H.264 §8.4.2.2.1 / FFmpeg ff_put_h264_qpel8_mc22_neon:
+//
+//   tmp[r,c] = src[r,c-2] - 5*src[r,c-1] + 20*src[r,c] + 20*src[r,c+1]
+//              - 5*src[r,c+2] + src[r,c+3]                    (int16)
+//
+//   dst[r,c] = clip255((tmp[r-2,c] - 5*tmp[r-1,c] + 20*tmp[r,c]
+//                       + 20*tmp[r+1,c] - 5*tmp[r+2,c] + tmp[r+3,c]
+//                       + 512) >> 10)
+//
+// The +512 >> 10 final scale compensates for both 6-tap scalings.
+// CANNOT just cascade mc20→mc02 because intermediate must be int16
+// (no per-stage clip), so this is a dedicated kernel.
+//
+// Per-lane structure: each lane computes its own (r, c) output by
+// running the FULL cascade — 6 horizontal lowpass int16 values for
+// rows r-2..r+3, then a vertical lowpass on those.  ~50 ALU ops per
+// lane.  No shared memory / barriers needed; V3D L2 absorbs the
+// redundant src reads across lanes.
+//
+// WG layout: 64 lanes / 1 block-per-WG / 1 lane-per-output-pixel
+// (same as mc20 / mc02).
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+
+layout(push_constant) uniform PC {
+    uint n_blocks;
+    uint stride_u8;
+    uint _pad0, _pad1;
+} pc;
+
+// Horizontal 6-tap filter at (row_off, c) — reads src at cols c-2..c+3
+// of the row identified by row_off, returns int16 intermediate (NOT
+// scaled — the v-pass does the +512 >> 10 for both stages).
+int hpel_h(uint row_off, uint c)
+{
+    int s_m2 = int(u_src.src[row_off + c - 2u]);
+    int s_m1 = int(u_src.src[row_off + c - 1u]);
+    int s_0  = int(u_src.src[row_off + c       ]);
+    int s_p1 = int(u_src.src[row_off + c + 1u]);
+    int s_p2 = int(u_src.src[row_off + c + 2u]);
+    int s_p3 = int(u_src.src[row_off + c + 3u]);
+    return s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3;
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3;
+    uint c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    // Compute 6 horizontal lowpass values at rows r-2..r+3 (relative
+    // to the output row r) of column c.  src_off+r*stride+c is the
+    // output pixel position; we sample rows r-2..r+3.
+    // Unsigned-safe because src_off >= 2*stride per the caller contract.
+    int t0 = hpel_h(src_off + (r - 2u) * stride, c);
+    int t1 = hpel_h(src_off + (r - 1u) * stride, c);
+    int t2 = hpel_h(src_off +  r       * stride, c);
+    int t3 = hpel_h(src_off + (r + 1u) * stride, c);
+    int t4 = hpel_h(src_off + (r + 2u) * stride, c);
+    int t5 = hpel_h(src_off + (r + 3u) * stride, c);
+
+    int v = t0 - 5 * t1 + 20 * t2 + 20 * t3 - 5 * t4 + t5 + 512;
+    int p = clamp(v >> 10, 0, 255);
+
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(p);
+}

src/v3d_h264_qpel_mc23.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc23 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc23[r,c] = avg(mc22(r, c),
+//                     mc20(r+1, c))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_h(src_off, stride, r+1u, c);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc30.comp

@@ -0,0 +1,44 @@
+// daedalus-fourier — H.264 luma qpel mc30 (8x8, ¾-pel horizontal),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 "c" position:
+//
+//   dst[r,c] = ((clip255(mc20(s)[r,c]) + s[r,c+1] + 1) >> 1)
+//
+// Same as mc10 but L2-averages with src[r, c+1] instead of src[r, c].
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+    uint row_base = src_off + r * stride + c;
+
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5 * s_m1 + 20 * s_0 + 20 * s_p1 - 5 * s_p2 + s_p3 + 16;
+    int hp = clamp(v >> 5, 0, 255);
+
+    int avg = (hp + s_p1 + 1) >> 1;   // L2 with src[r, c+1]
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc31.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc31 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc31[r,c] = avg(mc20(r, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc32.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc32 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc32[r,c] = avg(mc22(r, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_hv(src_off, stride, r, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264_qpel_mc33.comp

@@ -0,0 +1,88 @@
+// daedalus-fourier — H.264 luma qpel mc33 (8x8, diagonal quarter-pel),
+// V3D 7.1.  Per H.264 §8.4.2.2.1 (table 8-4) — composes two half-pel
+// anchors via L2 rounded-average:
+//
+//   mc33[r,c] = avg(mc20(r+1, c),
+//                     mc02(r, c+1))
+//
+// Per-lane structure: each lane computes BOTH anchor outputs at its
+// own (r, c) target offset, then L2 averages.  No shared memory.
+// Same WG geometry as the other qpel shaders.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage             : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Src  { uint8_t src[]; } u_src;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(binding = 2) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(push_constant) uniform PC { uint n_blocks, stride_u8, _p0, _p1; } pc;
+
+int hpel_h(uint src_off, uint stride, uint r, uint c) {
+    uint row_base = src_off + r * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_v(uint src_off, uint stride, uint r, uint c) {
+    uint col_base = src_off + c;
+    int s_m2 = int(u_src.src[col_base + (r - 2u) * stride]);
+    int s_m1 = int(u_src.src[col_base + (r - 1u) * stride]);
+    int s_0  = int(u_src.src[col_base +  r       * stride]);
+    int s_p1 = int(u_src.src[col_base + (r + 1u) * stride]);
+    int s_p2 = int(u_src.src[col_base + (r + 2u) * stride]);
+    int s_p3 = int(u_src.src[col_base + (r + 3u) * stride]);
+    int v = s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3 + 16;
+    return clamp(v >> 5, 0, 255);
+}
+
+int hpel_hv_row(uint src_off, uint stride, uint rr, uint c) {
+    // Single row's int16 horizontal lowpass (NOT clipped — used as
+    // intermediate for the vertical pass of hpel_hv).
+    uint row_base = src_off + rr * stride + c;
+    int s_m2 = int(u_src.src[row_base - 2u]);
+    int s_m1 = int(u_src.src[row_base - 1u]);
+    int s_0  = int(u_src.src[row_base       ]);
+    int s_p1 = int(u_src.src[row_base + 1u]);
+    int s_p2 = int(u_src.src[row_base + 2u]);
+    int s_p3 = int(u_src.src[row_base + 3u]);
+    return s_m2 - 5*s_m1 + 20*s_0 + 20*s_p1 - 5*s_p2 + s_p3;
+}
+
+int hpel_hv(uint src_off, uint stride, uint r, uint c) {
+    int t0 = hpel_hv_row(src_off, stride, r - 2u, c);
+    int t1 = hpel_hv_row(src_off, stride, r - 1u, c);
+    int t2 = hpel_hv_row(src_off, stride, r,       c);
+    int t3 = hpel_hv_row(src_off, stride, r + 1u, c);
+    int t4 = hpel_hv_row(src_off, stride, r + 2u, c);
+    int t5 = hpel_hv_row(src_off, stride, r + 3u, c);
+    int v = t0 - 5*t1 + 20*t2 + 20*t3 - 5*t4 + t5 + 512;
+    return clamp(v >> 10, 0, 255);
+}
+
+void main()
+{
+    uint block_idx = gl_WorkGroupID.x;
+    if (block_idx >= pc.n_blocks) return;
+
+    uint lane = gl_LocalInvocationID.x;
+    uint r = lane >> 3, c = lane & 7u;
+
+    uint dst_off = u_meta.meta[block_idx].x;
+    uint src_off = u_meta.meta[block_idx].y;
+    uint stride  = pc.stride_u8;
+
+    int a = hpel_h(src_off, stride, r+1u, c);
+    int b = hpel_v(src_off, stride, r, c+1u);
+    int avg = (a + b + 1) >> 1;
+    u_dst.dst[dst_off + r * stride + c] = uint8_t(avg);
+}

src/v3d_h264deblock_chroma_h.comp

@@ -0,0 +1,69 @@
+// daedalus-fourier — H.264 chroma 4:2:0 H loop filter (horizontal
+// filter across a vertical edge), non-intra bS<4 variant.
+//
+// Sibling of v3d_h264deblock_chroma_v.comp; same kernel transposed
+// to read pix[-2..+1] (cols) instead of pix[-2*stride..+1*stride]
+// (rows).  Same 8-cell × 4-segment geometry, same WG layout (lanes
+// 8..15 of each edge early-return — only 8 active per edge).
+//
+// 4:2:0-only: 4:2:2 chroma_h has a 16-row edge that this shader
+// doesn't address.  daedalus_dispatch_h264_deblock_chroma_h is
+// 4:2:0-only by design; caller (libavcodec init) gates accordingly.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+
+layout(push_constant) uniform PC {
+    uint n_edges;
+    uint dst_stride_u8;
+    uint _pad0;
+    uint _pad1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;        // 0..15
+    uint row_in_edge = lane_in_wg & 15u;       // 0..15 — only 0..7 active
+
+    uint edge_idx = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+    if (row_in_edge >= 8u) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint stride  = pc.dst_stride_u8;
+    uint dst_off = m.x + row_in_edge * stride;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+
+    uint seg = row_in_edge >> 1;
+    uint tc0_byte = (m.z >> (seg * 8u)) & 0xffu;
+    int tc0_s = int(tc0_byte);
+    if (tc0_s >= 128) tc0_s -= 256;
+
+    if (alpha == 0 || beta == 0) return;
+    if (tc0_s < 0) return;
+
+    int p1 = int(u_dst.dst[dst_off - 2u]);
+    int p0 = int(u_dst.dst[dst_off - 1u]);
+    int q0 = int(u_dst.dst[dst_off       ]);
+    int q1 = int(u_dst.dst[dst_off + 1u]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    int tc = tc0_s + 1;
+    int delta = clamp(((q0 - p0) * 4 + (p1 - q1) + 4) >> 3, -tc, tc);
+
+    u_dst.dst[dst_off - 1u] = uint8_t(clamp(p0 + delta, 0, 255));
+    u_dst.dst[dst_off       ] = uint8_t(clamp(q0 - delta, 0, 255));
+}

src/v3d_h264deblock_chroma_h_intra.comp

@@ -0,0 +1,44 @@
+// daedalus-fourier — H.264 chroma 4:2:0 intra (bS=4) H deblock —
+// V3D 7.1.  Transpose of v3d_h264deblock_chroma_v_intra.comp.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(push_constant) uniform PC {
+    uint n_edges, dst_stride_u8, _p0, _p1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;
+    uint row_in_edge = lane_in_wg & 15u;
+    uint edge_idx    = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+    if (row_in_edge >= 8u) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint stride  = pc.dst_stride_u8;
+    uint dst_off = m.x + row_in_edge * stride;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+    if ((alpha | beta) == 0) return;
+
+    int p1 = int(u_dst.dst[dst_off - 2u]);
+    int p0 = int(u_dst.dst[dst_off - 1u]);
+    int q0 = int(u_dst.dst[dst_off       ]);
+    int q1 = int(u_dst.dst[dst_off + 1u]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    u_dst.dst[dst_off - 1u] = uint8_t(clamp((2*p1 + p0 + q1 + 2) >> 2, 0, 255));
+    u_dst.dst[dst_off       ] = uint8_t(clamp((2*q1 + q0 + p1 + 2) >> 2, 0, 255));
+}

src/v3d_h264deblock_chroma_v.comp

@@ -0,0 +1,76 @@
+// daedalus-fourier — H.264 chroma 4:2:0 V loop filter (vertical
+// filter across a horizontal edge), non-intra bS<4 variant.
+//
+// Per H.264 §8.7.2.4: chroma kernel is simpler than luma's bS<4 —
+// only p0 / q0 are updated (chroma never modifies p1, p2, q1, q2),
+// tC = tc0_seg + 1 (no luma-style ap/aq side bonus), and the edge
+// spans 8 cells (4 segments × 2 cells/seg).
+//
+// V3D 7.1 via Mesa v3dv compute.  WG geometry kept identical to the
+// luma shader (16 edges × 16 lanes/WG) for uniform dispatch math
+// across the deblock family; lanes 8..15 of each edge early-return.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Meta {
+    uvec4 meta[];   // per edge: (dst_off, alpha|beta<<8, packed_tc0, _pad)
+} u_meta;
+
+layout(binding = 1) buffer Dst {
+    uint8_t dst[];
+} u_dst;
+
+layout(push_constant) uniform PC {
+    uint n_edges;
+    uint dst_stride_u8;
+    uint _pad0;
+    uint _pad1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;        // 0..15
+    uint col_in_edge = lane_in_wg & 15u;       // 0..15 — only 0..7 active
+
+    uint edge_idx = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+    if (col_in_edge >= 8u) return;             // 8 cells per chroma edge
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint dst_off = m.x + col_in_edge;
+    uint stride  = pc.dst_stride_u8;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+
+    // 8 cells / 4 segments = 2 cells per segment.
+    uint seg = col_in_edge >> 1;
+    uint tc0_byte = (m.z >> (seg * 8u)) & 0xffu;
+    int tc0_s = int(tc0_byte);
+    if (tc0_s >= 128) tc0_s -= 256;
+
+    if (alpha == 0 || beta == 0) return;
+    if (tc0_s < 0) return;
+
+    int p1 = int(u_dst.dst[dst_off - 2u * stride]);
+    int p0 = int(u_dst.dst[dst_off - 1u * stride]);
+    int q0 = int(u_dst.dst[dst_off]);
+    int q1 = int(u_dst.dst[dst_off + 1u * stride]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    int tc = tc0_s + 1;
+    int delta = clamp(((q0 - p0) * 4 + (p1 - q1) + 4) >> 3, -tc, tc);
+
+    u_dst.dst[dst_off - 1u * stride] = uint8_t(clamp(p0 + delta, 0, 255));
+    u_dst.dst[dst_off            ]   = uint8_t(clamp(q0 - delta, 0, 255));
+    // p1, q1 untouched — chroma kernel only updates p0/q0.
+}

src/v3d_h264deblock_chroma_v_intra.comp

@@ -0,0 +1,54 @@
+// daedalus-fourier — H.264 chroma 4:2:0 intra (bS=4) V deblock —
+// V3D 7.1.  Per H.264 §8.3.2.3 chroma intra path: simpler than luma
+// — always weak filter, only p0/q0 updated, 8 cells per edge.
+//
+// p0' = (2*p1 + p0 + q1 + 2) >> 2
+// q0' = (2*q1 + q0 + p1 + 2) >> 2
+//
+// Same 16-edges × 16-lanes/edge WG shape as luma; lanes 8..15 of each
+// edge early-return (chroma edges are only 8 cells wide).
+//
+// 4:2:0-only — caller-side gating handles 4:2:2 (chroma_format_idc>1)
+// at the libavcodec init layer.
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(push_constant) uniform PC {
+    uint n_edges, dst_stride_u8, _p0, _p1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;
+    uint col_in_edge = lane_in_wg & 15u;
+    uint edge_idx    = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+    if (col_in_edge >= 8u) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint dst_off = m.x + col_in_edge;
+    uint stride  = pc.dst_stride_u8;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+    if ((alpha | beta) == 0) return;
+
+    int p1 = int(u_dst.dst[dst_off - 2u * stride]);
+    int p0 = int(u_dst.dst[dst_off - 1u * stride]);
+    int q0 = int(u_dst.dst[dst_off]);
+    int q1 = int(u_dst.dst[dst_off + 1u * stride]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    u_dst.dst[dst_off - 1u * stride] = uint8_t(clamp((2*p1 + p0 + q1 + 2) >> 2, 0, 255));
+    u_dst.dst[dst_off            ]   = uint8_t(clamp((2*q1 + q0 + p1 + 2) >> 2, 0, 255));
+}

src/v3d_h264deblock_h.comp

@@ -0,0 +1,111 @@
+// daedalus-fourier — H.264 luma "h_loop_filter" (horizontal filtering
+// across a vertical edge), non-intra bS<4 variant.  Sibling of cycle 8's
+// v3d_h264deblock.comp; same algorithm with row/col access transposed.
+//
+// V3D 7.1 via Mesa v3dv compute.  Same WG geometry as the V shader:
+//   - 256 invocations / WG, 16 edges/WG (16 lanes/edge = 1 sg/edge)
+//   - uint8_t dst SSBO via storageBuffer8BitAccess
+//   - No barrier (each lane independent)
+//   - lane_in_edge = ROW index (0..15) along the vertical edge
+//   - meta.dst_off points to (row 0, col 0) of the RIGHT block;
+//     the kernel reads cols [-4..+3] of each row and writes [-2..+1].
+//
+// Filter contract (per H.264 §8.7.2.4):
+//   1. (m.x % pc.dst_stride_u8) ≥ 4   (kernel reads p3 at pix[-4])
+//   2. pc.dst_stride_u8 = byte stride between rows
+//   3. tc0_s pre-stored as signed int8 in m.z packed 4 bytes (one per
+//      4-row segment along the 16-row edge)
+//
+// License: BSD-2-Clause.  Algorithm transcribed from
+// tests/h264_h_loop_filter_luma_ref.c which mirrors FFmpeg
+// ff_h264_h_loop_filter_luma_neon (LGPL-2.1+).
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+
+layout(binding = 0) readonly buffer Meta {
+    uvec4 meta[];   // per edge: (dst_off, alpha|beta<<8, packed_tc0, _pad)
+} u_meta;
+
+layout(binding = 1) buffer Dst {
+    uint8_t dst[];
+} u_dst;
+
+layout(push_constant) uniform PC {
+    uint n_edges;
+    uint dst_stride_u8;
+    uint _pad0;
+    uint _pad1;
+} pc;
+
+void main()
+{
+    uint gid          = gl_GlobalInvocationID.x;
+    uint wg_id        = gl_WorkGroupID.x;
+    uint lane_in_wg   = gid & 255u;
+    uint edge_in_wg   = lane_in_wg >> 4;       // 0..15 (16 edges/WG)
+    uint row_in_edge  = lane_in_wg & 15u;      // 0..15 — ROW along the V edge
+
+    uint edge_idx = wg_id * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint stride = pc.dst_stride_u8;
+    // dst_off addresses row 0 col 0 of the right block; advance by row * stride
+    // to land at this lane's row.  The kernel reads pix[-4..+3] AT THIS ROW.
+    uint dst_off = m.x + row_in_edge * stride;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+
+    // tc0 segment = 0..3 indexed by (row_in_edge / 4).
+    uint seg = row_in_edge >> 2;
+    uint tc0_byte = (m.z >> (seg * 8u)) & 0xffu;
+    int tc0_s = int(tc0_byte);
+    if (tc0_s >= 128) tc0_s -= 256;
+
+    if (alpha == 0 || beta == 0) return;
+    if (tc0_s < 0) return;                     // segment skip
+
+    // Horizontal access pattern — read cols at offsets [-3..+2] of this row.
+    // p3 (col -4) unused in bS<4; same DCE comment as the V shader.
+    int p2 = int(u_dst.dst[dst_off - 3u]);
+    int p1 = int(u_dst.dst[dst_off - 2u]);
+    int p0 = int(u_dst.dst[dst_off - 1u]);
+    int q0 = int(u_dst.dst[dst_off       ]);
+    int q1 = int(u_dst.dst[dst_off + 1u]);
+    int q2 = int(u_dst.dst[dst_off + 2u]);
+
+    // Edge preconditions (same as V).
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    int ap = abs(p2 - p0);
+    int aq = abs(q2 - q0);
+    bool ap_lt = ap < beta;
+    bool aq_lt = aq < beta;
+    int tc = tc0_s + int(ap_lt) + int(aq_lt);
+
+    int delta = clamp(((q0 - p0) * 4 + (p1 - q1) + 4) >> 3, -tc, tc);
+    int p0p = clamp(p0 + delta, 0, 255);
+    int q0p = clamp(q0 - delta, 0, 255);
+
+    int p1p = p1;
+    if (ap_lt) {
+        int d_p1 = clamp((p2 + ((p0 + q0 + 1) >> 1) - 2*p1) >> 1, -tc0_s, tc0_s);
+        p1p = clamp(p1 + d_p1, 0, 255);
+    }
+    int q1p = q1;
+    if (aq_lt) {
+        int d_q1 = clamp((q2 + ((p0 + q0 + 1) >> 1) - 2*q1) >> 1, -tc0_s, tc0_s);
+        q1p = clamp(q1 + d_q1, 0, 255);
+    }
+
+    u_dst.dst[dst_off - 2u] = uint8_t(p1p);
+    u_dst.dst[dst_off - 1u] = uint8_t(p0p);
+    u_dst.dst[dst_off       ] = uint8_t(q0p);
+    u_dst.dst[dst_off + 1u] = uint8_t(q1p);
+}

src/v3d_h264deblock_luma_h_intra.comp

@@ -0,0 +1,70 @@
+// daedalus-fourier — H.264 luma intra (bS=4) H deblock — V3D 7.1.
+//
+// Sibling of v3d_h264deblock_luma_v_intra.comp transposed to the
+// horizontal axis: lane → row, reads pix[-4..+3] (cols) instead of
+// pix[-4*stride..+3*stride] (rows).  Same strong/weak filter
+// selector + same write-back algebra.
+//
+// dst_off contract: (m.x % stride) ≥ 4 (kernel reads p3 at pix[-4]).
+//
+// License: BSD-2-Clause.
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(push_constant) uniform PC {
+    uint n_edges, dst_stride_u8, _p0, _p1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;
+    uint row_in_edge = lane_in_wg & 15u;
+    uint edge_idx    = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint stride = pc.dst_stride_u8;
+    uint dst_off = m.x + row_in_edge * stride;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+    if ((alpha | beta) == 0) return;
+
+    int p3 = int(u_dst.dst[dst_off - 4u]);
+    int p2 = int(u_dst.dst[dst_off - 3u]);
+    int p1 = int(u_dst.dst[dst_off - 2u]);
+    int p0 = int(u_dst.dst[dst_off - 1u]);
+    int q0 = int(u_dst.dst[dst_off       ]);
+    int q1 = int(u_dst.dst[dst_off + 1u]);
+    int q2 = int(u_dst.dst[dst_off + 2u]);
+    int q3 = int(u_dst.dst[dst_off + 3u]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    bool strong_common = abs(p0 - q0) < (alpha >> 2) + 2;
+    bool strong_p = strong_common && abs(p2 - p0) < beta;
+    bool strong_q = strong_common && abs(q2 - q0) < beta;
+
+    if (strong_p) {
+        u_dst.dst[dst_off - 1u] = uint8_t(clamp((p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4) >> 3, 0, 255));
+        u_dst.dst[dst_off - 2u] = uint8_t(clamp((p2 + p1 + p0 + q0 + 2) >> 2, 0, 255));
+        u_dst.dst[dst_off - 3u] = uint8_t(clamp((2*p3 + 3*p2 + p1 + p0 + q0 + 4) >> 3, 0, 255));
+    } else {
+        u_dst.dst[dst_off - 1u] = uint8_t(clamp((2*p1 + p0 + q1 + 2) >> 2, 0, 255));
+    }
+
+    if (strong_q) {
+        u_dst.dst[dst_off       ] = uint8_t(clamp((q2 + 2*q1 + 2*q0 + 2*p0 + p1 + 4) >> 3, 0, 255));
+        u_dst.dst[dst_off + 1u] = uint8_t(clamp((q2 + q1 + q0 + p0 + 2) >> 2, 0, 255));
+        u_dst.dst[dst_off + 2u] = uint8_t(clamp((2*q3 + 3*q2 + q1 + q0 + p0 + 4) >> 3, 0, 255));
+    } else {
+        u_dst.dst[dst_off       ] = uint8_t(clamp((2*q1 + q0 + p1 + 2) >> 2, 0, 255));
+    }
+}

src/v3d_h264deblock_luma_v_intra.comp

@@ -0,0 +1,81 @@
+// daedalus-fourier — H.264 luma intra (bS=4) V deblock — V3D 7.1.
+//
+// Per H.264 §8.3.2.3: at I-MB edges and certain inter-MB edges that
+// force boundary strength to 4, the deblock kernel is structurally
+// different from bS<4 — it has a per-side strong/weak filter
+// selector that decides whether to update 3 cells (strong) or 1
+// (weak), reads p3/q3, and ignores tc0.
+//
+// strong_common = |p0-q0| < (α>>2) + 2
+// strong_p      = strong_common AND |p2-p0| < β
+// strong_q      = strong_common AND |q2-q0| < β
+//
+// Strong-p updates p0/p1/p2 with specific 5-/4-/3-tap blends.
+// Weak-p updates p0 only with (2*p1 + p0 + q1 + 2) >> 2.
+// Mirror for q-side.
+//
+// WG geometry identical to v3d_h264deblock.comp (16 edges × 16 lanes/WG).
+// dst_off contract: m.x ≥ 4*stride (kernel reads p3 at -4*stride).
+//
+// License: BSD-2-Clause.  Algorithm transcribed from
+// tests/h264_intra_loop_filter_ref.c (PR #11).
+
+#version 450
+#extension GL_EXT_shader_8bit_storage              : require
+#extension GL_EXT_shader_explicit_arithmetic_types : require
+
+layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
+layout(binding = 0) readonly buffer Meta { uvec4 meta[]; } u_meta;
+layout(binding = 1) buffer Dst { uint8_t dst[]; } u_dst;
+layout(push_constant) uniform PC {
+    uint n_edges, dst_stride_u8, _p0, _p1;
+} pc;
+
+void main()
+{
+    uint lane_in_wg  = gl_GlobalInvocationID.x & 255u;
+    uint edge_in_wg  = lane_in_wg >> 4;
+    uint col_in_edge = lane_in_wg & 15u;
+    uint edge_idx    = gl_WorkGroupID.x * 16u + edge_in_wg;
+    if (edge_idx >= pc.n_edges) return;
+
+    uvec4 m = u_meta.meta[edge_idx];
+    uint dst_off = m.x + col_in_edge;
+    uint stride  = pc.dst_stride_u8;
+    int alpha = int(m.y & 0xffu);
+    int beta  = int((m.y >> 8) & 0xffu);
+    if ((alpha | beta) == 0) return;
+
+    int p3 = int(u_dst.dst[dst_off - 4u * stride]);
+    int p2 = int(u_dst.dst[dst_off - 3u * stride]);
+    int p1 = int(u_dst.dst[dst_off - 2u * stride]);
+    int p0 = int(u_dst.dst[dst_off - 1u * stride]);
+    int q0 = int(u_dst.dst[dst_off]);
+    int q1 = int(u_dst.dst[dst_off + 1u * stride]);
+    int q2 = int(u_dst.dst[dst_off + 2u * stride]);
+    int q3 = int(u_dst.dst[dst_off + 3u * stride]);
+
+    if (abs(p0 - q0) >= alpha) return;
+    if (abs(p1 - p0) >= beta)  return;
+    if (abs(q1 - q0) >= beta)  return;
+
+    bool strong_common = abs(p0 - q0) < (alpha >> 2) + 2;
+    bool strong_p = strong_common && abs(p2 - p0) < beta;
+    bool strong_q = strong_common && abs(q2 - q0) < beta;
+
+    if (strong_p) {
+        u_dst.dst[dst_off - 1u * stride] = uint8_t(clamp((p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4) >> 3, 0, 255));
+        u_dst.dst[dst_off - 2u * stride] = uint8_t(clamp((p2 + p1 + p0 + q0 + 2) >> 2, 0, 255));
+        u_dst.dst[dst_off - 3u * stride] = uint8_t(clamp((2*p3 + 3*p2 + p1 + p0 + q0 + 4) >> 3, 0, 255));
+    } else {
+        u_dst.dst[dst_off - 1u * stride] = uint8_t(clamp((2*p1 + p0 + q1 + 2) >> 2, 0, 255));
+    }
+
+    if (strong_q) {
+        u_dst.dst[dst_off              ] = uint8_t(clamp((q2 + 2*q1 + 2*q0 + 2*p0 + p1 + 4) >> 3, 0, 255));
+        u_dst.dst[dst_off + 1u * stride] = uint8_t(clamp((q2 + q1 + q0 + p0 + 2) >> 2, 0, 255));
+        u_dst.dst[dst_off + 2u * stride] = uint8_t(clamp((2*q3 + 3*q2 + q1 + q0 + p0 + 4) >> 3, 0, 255));
+    } else {
+        u_dst.dst[dst_off              ] = uint8_t(clamp((2*q1 + q0 + p1 + 2) >> 2, 0, 255));
+    }
+}

tests/bench_h264_primitives.c

@@ -2,25 +2,22 @@
 /* CLOCK_MONOTONIC under -std=c11 -CMAKE_C_EXTENSIONS=OFF. */
 #define _POSIX_C_SOURCE 200809L
 /*
- * bench_h264_primitives — NEON-path latency baseline for the H.264
+ * bench_h264_primitives — latency baseline for the H.264 primitive
- * primitive library landed across PRs #9–#23.
+ * library landed across PRs #9–#35.
  *
  * Each kernel is exercised at a representative per-frame N for 1080p
- * (8160 MBs); the per-kernel total + ns/op + ms/frame are reported.
+ * (8160 MBs); the per-kernel total + ns/op + ms/frame are reported,
- * Lets us answer "what's the total NEON-only budget for the H.264
+ * once per substrate (CPU NEON, QPU V3D7 compute).  The QPU column
- * decode at 1080p" — useful for sizing intercept-patch decisions
+ * appears only when the host has a usable Vulkan device.  When both
- * (which kernels NEED QPU shaders vs which are budget-fine on NEON).
+ * columns exist a CPU/QPU ratio is printed; that's the per-kernel
+ * data the QPU-substrate decree (2026-05-23) deliberately overrides
+ * but which is still useful to track over time as dispatch overhead
+ * shrinks (buffer pool, persistent cmdbuf, dmabuf import — tasks 160-162).
  *
  * NOT a ctest — produces wall-time numbers, doesn't pass/fail.
  *
- * Invoke:   ./build/bench_h264_primitives [iters]
+ * Invoke:   ./build/bench_h264_primitives [iters [warmup]]
- *           (default iters = 50, post-warmup = 5)
+ *           (default iters = 50, warmup = 5)
- *
- * NB: results are inherently approximate — single-core, includes
- * loop overhead + memory access patterns that may not match what
- * a real decode would hit (we touch a small set of pages repeatedly).
- * The numbers are useful for relative comparison and order-of-
- * magnitude sizing, not absolute perf claims.
  */
 
 #include "daedalus.h"
@@ -46,11 +43,6 @@ static double now_ms(void) {
 
 /* Per-1080p-frame counts (8160 MBs at 1920x1088). */
 #define MBS_1080P  8160
-#define LUMA_4x4_PER_MB   16   /* if transform_8x8=0 */
-#define LUMA_8x8_PER_MB   4    /* if transform_8x8=1 */
-#define CHROMA_4x4_PER_MB 8    /* 4 Cb + 4 Cr */
-#define DEBLOCK_LUMA_EDGES_PER_MB 4    /* 4 horiz + 4 vert internal+MB-edge — ~4 each */
-#define DEBLOCK_CHROMA_EDGES_PER_MB 2  /* 2 each direction */
 
 /* Standard benchmark loop.  fn() is called n times per iteration. */
 typedef void (*bench_fn)(void);
@@ -64,16 +56,18 @@ static double bench_ns(const char *name, int iters, int warmup,
     double t1 = now_ms();
     double total_ms = (t1 - t0);
     double ns_per_op = (total_ms * 1e6) / ((double) iters * ops_per_iter);
-    printf("  %-32s %8.2f ns/op  (%d iters x %d ops)\n",
+    printf("  %-32s %10.2f ns/op  (%d iters x %d ops)\n",
            name, ns_per_op, iters, ops_per_iter);
     return ns_per_op;
 }
 
 /* ---- Per-kernel scaffolding.  Each section sets up the buffers +
  * meta, then defines a static fn() that calls the corresponding
- * dispatch with a representative N. */
+ * dispatch with a representative N.  The substrate is read from the
+ * global g_sub so the same fn() can be re-driven with CPU then QPU. */
 
-static daedalus_ctx *ctx;
+static daedalus_ctx          *ctx;
+static daedalus_substrate     g_sub = DAEDALUS_SUBSTRATE_CPU;
 
 /* --- IDCT 4x4 luma: N = 16 blocks per MB.  Bench with 1024 blocks
  *     per call (64 MBs worth).  Per-MB the dispatch overhead is the
@@ -83,7 +77,7 @@ static daedalus_h264_block_meta idct4_meta[1024];
 static uint8_t              idct_dst[64 * 4 * 16 * 16];   /* 64 MB-rows × ... */
 
 static void bench_idct4(void) {
-    daedalus_dispatch_h264_idct4(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_idct4(ctx, g_sub,
                                   idct_dst, 64*16, idct4_coeffs, 1024, idct4_meta);
 }
 
@@ -92,7 +86,7 @@ static int16_t              idct8_coeffs[256 * 64];
 static daedalus_h264_block_meta idct8_meta[256];
 
 static void bench_idct8(void) {
-    daedalus_dispatch_h264_idct8(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_idct8(ctx, g_sub,
                                   idct_dst, 64*16, idct8_coeffs, 256, idct8_meta);
 }
 
@@ -101,12 +95,12 @@ static daedalus_h264_deblock_meta deblock_meta[256];
 static uint8_t deblock_dst[256 * 16 * 16];
 
 static void bench_deblock_v(void) {
-    daedalus_dispatch_h264_deblock_luma_v(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_deblock_luma_v(ctx, g_sub,
                                            deblock_dst, 16, 256, deblock_meta);
 }
 
 static void bench_deblock_h(void) {
-    daedalus_dispatch_h264_deblock_luma_h(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_deblock_luma_h(ctx, g_sub,
                                            deblock_dst, 16, 256, deblock_meta);
 }
 
@@ -116,18 +110,43 @@ static uint8_t qpel_dst[256 * 16 * 16];
 static daedalus_h264_qpel_meta qpel_meta[256];
 
 static void bench_qpel_mc20(void) {
-    daedalus_dispatch_h264_qpel_mc20(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_qpel_mc20(ctx, g_sub,
                                       qpel_dst, qpel_src, 16, 256, qpel_meta);
 }
 static void bench_qpel_mc02(void) {
-    daedalus_dispatch_h264_qpel_mc02(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_qpel_mc02(ctx, g_sub,
                                       qpel_dst, qpel_src, 16, 256, qpel_meta);
 }
 static void bench_qpel_mc22(void) {
-    daedalus_dispatch_h264_qpel_mc22(ctx, DAEDALUS_SUBSTRATE_CPU,
+    daedalus_dispatch_h264_qpel_mc22(ctx, g_sub,
                                       qpel_dst, qpel_src, 16, 256, qpel_meta);
 }
 
+/* ---- One row of bench output:
+ *   - kernel name + N
+ *   - CPU ns/op
+ *   - QPU ns/op (or "n/a" if Vulkan absent)
+ *   - CPU/QPU ratio (>1 means QPU wins; <1 means CPU wins) */
+struct row {
+    const char *name;
+    int         n_per_call;
+    bench_fn    fn;
+    double      cpu_ns;
+    double      qpu_ns;   /* -1 if not measured */
+    int         frame_n;  /* count per 1080p frame */
+};
+
+static struct row rows[] = {
+    {"IDCT 4x4 luma",       1024, bench_idct4,     0, -1, MBS_1080P * 16},
+    {"IDCT 8x8 luma",        256, bench_idct8,     0, -1, MBS_1080P *  4},
+    {"Deblock luma_v",       256, bench_deblock_v, 0, -1, MBS_1080P *  4},
+    {"Deblock luma_h",       256, bench_deblock_h, 0, -1, MBS_1080P *  4},
+    {"qpel mc20 (8x8)",      256, bench_qpel_mc20, 0, -1, MBS_1080P *  4},
+    {"qpel mc02 (8x8)",      256, bench_qpel_mc02, 0, -1, MBS_1080P *  4},
+    {"qpel mc22 (8x8)",      256, bench_qpel_mc22, 0, -1, MBS_1080P *  4},
+};
+#define N_ROWS ((int)(sizeof(rows)/sizeof(rows[0])))
+
 int main(int argc, char **argv)
 {
     int iters  = argc > 1 ? atoi(argv[1]) : 50;
@@ -138,6 +157,7 @@ int main(int argc, char **argv)
         fprintf(stderr, "ctx create failed (Vulkan?)\n");
         return 1;
     }
+    int has_qpu = daedalus_ctx_has_qpu(ctx);
 
     /* Pre-fill all input buffers with random data so the NEON inner
      * loops see realistic memory access patterns. */
@@ -147,8 +167,7 @@ int main(int argc, char **argv)
         idct8_coeffs[i] = (int16_t)((int)(xs64() % 1024) - 512);
     for (size_t i = 0; i < sizeof(qpel_src); i++) qpel_src[i] = (uint8_t)(xs64() & 0xff);
 
-    /* IDCT meta: each block at offset i*16 (row layout matters less
+    /* IDCT meta. */
-     * here since we're just measuring per-block latency). */
     for (size_t i = 0; i < 1024; i++)
         idct4_meta[i].dst_off = (uint32_t)((i / 16) * 64 + (i % 16) * 4);
     for (size_t i = 0; i < 256; i++)
@@ -162,58 +181,88 @@ int main(int argc, char **argv)
         for (int s = 0; s < 4; s++) deblock_meta[i].tc0[s] = (int8_t)(s + 1);
     }
 
-    /* qpel meta: src and dst at row 3 col 3 of each 16x16 tile. */
+    /* qpel meta. */
     for (size_t i = 0; i < 256; i++) {
         qpel_meta[i].src_off = (uint32_t)(i * 256 + 3 * 16 + 3);
         qpel_meta[i].dst_off = (uint32_t)(i * 256 + 3 * 16 + 3);
     }
 
-    printf("bench_h264_primitives: %d iters (%d warmup), substrate=CPU NEON\n",
+    printf("bench_h264_primitives: %d iters (%d warmup)\n", iters, warmup);
-           iters, warmup);
+    printf("  ctx has_qpu=%d  (CPU pass always runs; QPU pass skipped without Vulkan)\n\n", has_qpu);
-    printf("Per-call N is set per kernel; ns/op is per BLOCK or EDGE.\n\n");
 
-    double idct4_ns   = bench_ns("IDCT 4x4 luma",     iters, warmup, 1024, bench_idct4);
+    /* Pass 1: CPU NEON. */
-    double idct8_ns   = bench_ns("IDCT 8x8 luma",     iters, warmup,  256, bench_idct8);
+    g_sub = DAEDALUS_SUBSTRATE_CPU;
-    double debl_v_ns  = bench_ns("Deblock luma_v",    iters, warmup,  256, bench_deblock_v);
+    printf("== CPU NEON ==\n");
-    double debl_h_ns  = bench_ns("Deblock luma_h",    iters, warmup,  256, bench_deblock_h);
+    for (int i = 0; i < N_ROWS; i++)
-    double qmc20_ns   = bench_ns("qpel mc20 (8x8)",   iters, warmup,  256, bench_qpel_mc20);
+        rows[i].cpu_ns = bench_ns(rows[i].name, iters, warmup, rows[i].n_per_call, rows[i].fn);
-    double qmc02_ns   = bench_ns("qpel mc02 (8x8)",   iters, warmup,  256, bench_qpel_mc02);
-    double qmc22_ns   = bench_ns("qpel mc22 (8x8)",   iters, warmup,  256, bench_qpel_mc22);
 
-    /* Per-frame budget summary at 1080p (8160 MBs).  Worst-case
+    /* Pass 2: QPU compute (if available). */
-     * assumptions:
+    if (has_qpu) {
-     *   - All MBs are transform_4x4 (16 4x4 IDCTs each) — so 130,560
+        g_sub = DAEDALUS_SUBSTRATE_QPU;
-     *     IDCT 4x4 blocks per frame.  If High profile transform_8x8,
+        printf("\n== QPU V3D7 compute ==\n");
-     *     it'd be 32,640 IDCT 8x8 blocks instead.
+        for (int i = 0; i < N_ROWS; i++)
-     *   - All MBs are intra (no MC — qpel zero) OR all inter (no
+            rows[i].qpu_ns = bench_ns(rows[i].name, iters, warmup, rows[i].n_per_call, rows[i].fn);
-     *     intra prediction).  We report MC at "all inter, all qpel
+    }
-     *     mc22" worst case.
+
-     *   - Deblock: ~4 luma_v + 4 luma_h edges per MB; assume all 8
+    /* Summary table — both substrates side by side. */
-     *     edges trigger filtering. */
+    printf("\n== Per-kernel comparison ==\n");
-    printf("\nProjected 1080p frame budgets (worst-case, CPU NEON only):\n");
+    printf("  %-24s %12s %12s %8s   %7s\n",
-    printf("  IDCT 4x4 (all-4x4 MBs):       %7.2f ms   (%d blocks)\n",
+           "kernel", "CPU ns/op", "QPU ns/op", "winner", "ms/frame");
-           idct4_ns * MBS_1080P * 16 / 1e6, MBS_1080P * 16);
+    for (int i = 0; i < N_ROWS; i++) {
-    printf("  IDCT 8x8 (all-8x8 MBs):       %7.2f ms   (%d blocks)\n",
+        double cpu_ms = rows[i].cpu_ns * rows[i].frame_n / 1e6;
-           idct8_ns * MBS_1080P * 4 / 1e6, MBS_1080P * 4);
+        double qpu_ms = rows[i].qpu_ns > 0 ? rows[i].qpu_ns * rows[i].frame_n / 1e6 : -1;
-    printf("  Deblock luma_v (all MBs):     %7.2f ms   (%d edges)\n",
+        const char *winner;
-           debl_v_ns * MBS_1080P * 4 / 1e6, MBS_1080P * 4);
+        char ratio[16];
-    printf("  Deblock luma_h (all MBs):     %7.2f ms   (%d edges)\n",
+        if (rows[i].qpu_ns <= 0) {
-           debl_h_ns * MBS_1080P * 4 / 1e6, MBS_1080P * 4);
+            winner = "CPU";  /* QPU n/a */
-    printf("  qpel mc22 (all 8x8 blocks):   %7.2f ms   (%d blocks)\n",
+            snprintf(ratio, sizeof(ratio), "n/a");
-           qmc22_ns * MBS_1080P * 4 / 1e6, MBS_1080P * 4);
+        } else if (rows[i].cpu_ns < rows[i].qpu_ns) {
+            winner = "CPU";
+            snprintf(ratio, sizeof(ratio), "%.2fx", rows[i].qpu_ns / rows[i].cpu_ns);
+        } else {
+            winner = "QPU";
+            snprintf(ratio, sizeof(ratio), "%.2fx", rows[i].cpu_ns / rows[i].qpu_ns);
+        }
+        char qpu_field[16];
+        if (rows[i].qpu_ns > 0) snprintf(qpu_field, sizeof(qpu_field), "%.2f", rows[i].qpu_ns);
+        else                    snprintf(qpu_field, sizeof(qpu_field), "n/a");
+        char ms_field[24];
+        if (qpu_ms > 0)
+            snprintf(ms_field, sizeof(ms_field), "%.2f/%.2f", cpu_ms, qpu_ms);
+        else
+            snprintf(ms_field, sizeof(ms_field), "%.2f/n/a", cpu_ms);
+        printf("  %-24s %12.2f %12s   %3s %s   %s\n",
+               rows[i].name, rows[i].cpu_ns, qpu_field, winner, ratio, ms_field);
+    }
+
+    /* Per-frame budget summary at 1080p (8160 MBs). */
+    double cpu_idct4 = rows[0].cpu_ns * MBS_1080P * 16 / 1e6;
+    double cpu_debl  = (rows[2].cpu_ns + rows[3].cpu_ns) * MBS_1080P * 4 / 1e6;
+    double cpu_mc    = rows[6].cpu_ns * MBS_1080P * 4 / 1e6;   /* mc22 worst-case */
+    double cpu_sum   = cpu_idct4 + cpu_debl + cpu_mc;
+
+    printf("\n== Projected 1080p worst-case (CPU NEON only) ==\n");
+    printf("  IDCT 4x4 + deblock luma + qpel mc22:  %.2f ms (30fps deadline 33.33)\n", cpu_sum);
+    printf("  Margin:                               %+.2f ms\n", 33.33 - cpu_sum);
+
+    if (has_qpu) {
+        double qpu_idct4 = rows[0].qpu_ns * MBS_1080P * 16 / 1e6;
+        double qpu_debl  = (rows[2].qpu_ns + rows[3].qpu_ns) * MBS_1080P * 4 / 1e6;
+        double qpu_mc    = rows[6].qpu_ns * MBS_1080P * 4 / 1e6;
+        double qpu_sum   = qpu_idct4 + qpu_debl + qpu_mc;
+        printf("\n== Projected 1080p worst-case (QPU V3D7 compute only) ==\n");
+        printf("  IDCT 4x4 + deblock luma + qpel mc22:  %.2f ms (30fps deadline 33.33)\n", qpu_sum);
+        printf("  Margin:                               %+.2f ms\n", 33.33 - qpu_sum);
+        printf("\n  CPU vs QPU sum ratio: %.2fx  (>1 means QPU wins)\n",
+               qpu_sum > 0 ? cpu_sum / qpu_sum : 0.0);
+    }
 
-    double sum_idct_4x4 = idct4_ns * MBS_1080P * 16 / 1e6;
-    double sum_deblock  = (debl_v_ns + debl_h_ns) * MBS_1080P * 4 / 1e6;
-    double sum_mc       = qmc22_ns * MBS_1080P * 4 / 1e6;  /* worst-case all-mc22 */
-    printf("\n  Sum (IDCT 4x4 + deblock luma + MC all-mc22): %7.2f ms\n",
-           sum_idct_4x4 + sum_deblock + sum_mc);
-    printf("  30 fps deadline:                              33.33 ms\n");
-    printf("  Margin:                                       %+.2f ms\n",
-           33.33 - (sum_idct_4x4 + sum_deblock + sum_mc));
     printf("\n(NOT included: chroma deblock, chroma IDCT, intra prediction,\n");
     printf(" CABAC/CAVLC entropy.  These bench numbers are a budget LOWER\n");
-    printf(" bound; the real decode stack adds 20-40%% on top.)\n");
+    printf(" bound; the real decode stack adds 20-40%% on top.\n");
-    (void) qmc20_ns; (void) qmc02_ns;
+    printf(" Per-kernel substrate decisions belong in daedalus_core.c recipe\n");
+    printf(" table; the QPU substrate decree (2026-05-23) keeps everything\n");
+    printf(" on QPU regardless of these numbers as a policy choice.)\n");
 
     daedalus_ctx_destroy(ctx);
     return 0;

tests/test_api_h264.c

@@ -683,13 +683,13 @@ int main(void)
     printf("  H264_QPEL_MC20 recipe substrate:  %d\n",
            (int) daedalus_recipe_substrate_for(DAEDALUS_KERNEL_H264_QPEL_MC20));
 
-    printf("  H264_DEBLOCK_LH recipe substrate: %d (CPU, no QPU H shader yet)\n",
+    printf("  H264_DEBLOCK_LH recipe substrate: %d\n",
            (int) daedalus_recipe_substrate_for(DAEDALUS_KERNEL_H264_DEBLOCK_LH));
-    printf("  H264_DEBLOCK_CV recipe substrate: %d (CPU)\n",
+    printf("  H264_DEBLOCK_CV recipe substrate: %d\n",
            (int) daedalus_recipe_substrate_for(DAEDALUS_KERNEL_H264_DEBLOCK_CV));
-    printf("  H264_DEBLOCK_CH recipe substrate: %d (CPU)\n",
+    printf("  H264_DEBLOCK_CH recipe substrate: %d\n",
            (int) daedalus_recipe_substrate_for(DAEDALUS_KERNEL_H264_DEBLOCK_CH));
-    printf("  H264_DEBLOCK_*_INTRA recipe substrate: %d (CPU, bS=4 set)\n",
+    printf("  H264_DEBLOCK_*_INTRA recipe substrate: %d (bS=4 family, all on QPU)\n",
            (int) daedalus_recipe_substrate_for(DAEDALUS_KERNEL_H264_DEBLOCK_LV_INTRA));
 
     int fail = 0;

tests/test_intra_pred_16x16.c

@@ -18,10 +18,10 @@
 #include <stdio.h>
 #include <string.h>
 
-extern void daedalus_h264_pred_16x16_vertical_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_16x16_vertical(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_16x16_horizontal_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_16x16_horizontal(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_16x16_dc_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_16x16_dc(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_16x16_plane_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_16x16_plane(uint8_t *dst, ptrdiff_t stride);
 
 #define STRIDE 17
 #define ROWS   17
@@ -84,7 +84,7 @@ int main(void)
         int t[16], l[16];
         for (int i = 0; i < 16; i++) { t[i] = 10 + i; l[i] = 0; }
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_16x16_vertical_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_16x16_vertical(&buf[1][1], STRIDE);
         struct vertical_ctx vc = { t };
         fail |= check(buf, "Vertical (mode 0)", expect_vertical, &vc);
     }
@@ -95,7 +95,7 @@ int main(void)
         int t[16] = {0}, l[16];
         for (int i = 0; i < 16; i++) l[i] = 50 + i;
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_16x16_horizontal_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_16x16_horizontal(&buf[1][1], STRIDE);
         struct horizontal_ctx hc = { l };
         fail |= check(buf, "Horizontal (mode 1)", expect_horizontal, &hc);
     }
@@ -108,7 +108,7 @@ int main(void)
         int t[16], l[16];
         for (int i = 0; i < 16; i++) { t[i] = 2; l[i] = 6; }
         set_ctx(buf, 99, t, l);
-        daedalus_h264_pred_16x16_dc_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_16x16_dc(&buf[1][1], STRIDE);
         uint8_t exp_val = 4;
         fail |= check(buf, "DC (mode 2)", expect_uniform, &exp_val);
     }
@@ -123,7 +123,7 @@ int main(void)
         int t[16], l[16];
         for (int i = 0; i < 16; i++) { t[i] = 100; l[i] = 100; }
         set_ctx(buf, 100, t, l);   /* uniform tl too — H/V sums actually zero */
-        daedalus_h264_pred_16x16_plane_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_16x16_plane(&buf[1][1], STRIDE);
         uint8_t exp_val = 100;
         fail |= check(buf, "Plane (mode 3, uniform)", expect_uniform, &exp_val);
     }
@@ -150,7 +150,7 @@ int main(void)
         int t[16], l[16];
         for (int i = 0; i < 16; i++) { t[i] = i; l[i] = i; }
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_16x16_plane_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_16x16_plane(&buf[1][1], STRIDE);
         uint8_t tl_actual = buf[1 + 0][1 + 0];
         uint8_t br_actual = buf[1 + 15][1 + 15];
         int spot_fail = 0;

tests/test_intra_pred_4x4.c

@@ -22,15 +22,15 @@
 #include <stdio.h>
 #include <string.h>
 
-extern void daedalus_h264_pred_4x4_vertical_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_vertical(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_horizontal_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_horizontal(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_dc_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_dc(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_ddl_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_ddl(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_ddr_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_ddr(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_vr_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_vr(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_hd_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_hd(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_vl_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_vl(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_4x4_hu_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_4x4_hu(uint8_t *dst, ptrdiff_t stride);
 
 #define STRIDE 9
 typedef void (*pred_fn)(uint8_t *dst, ptrdiff_t stride);
@@ -82,7 +82,7 @@ int main(void)
         int t[8] = { 10, 20, 30, 40,  0, 0, 0, 0 };
         int l[4] = {  0,  0,  0,  0 };
         set_ctx(buf, tl, t, l);
-        daedalus_h264_pred_4x4_vertical_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_vertical(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {10,20,30,40}, {10,20,30,40}, {10,20,30,40}, {10,20,30,40}
         };
@@ -95,7 +95,7 @@ int main(void)
         int t[8] = { 0,0,0,0, 0,0,0,0 };
         int l[4] = { 50, 60, 70, 80 };
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_4x4_horizontal_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_horizontal(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {50,50,50,50}, {60,60,60,60}, {70,70,70,70}, {80,80,80,80}
         };
@@ -110,7 +110,7 @@ int main(void)
         int t[8] = { 1,1,1,1, 0,0,0,0 };
         int l[4] = { 3,3,3,3 };
         set_ctx(buf, 99, t, l);   /* tl unused for DC */
-        daedalus_h264_pred_4x4_dc_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_dc(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {2,2,2,2}, {2,2,2,2}, {2,2,2,2}, {2,2,2,2}
         };
@@ -125,7 +125,7 @@ int main(void)
         int t[8] = { 100,100,100,100, 100,100,100,100 };
         int l[4] = { 0,0,0,0 };
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_4x4_ddl_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_ddl(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {100,100,100,100}, {100,100,100,100},
             {100,100,100,100}, {100,100,100,100}
@@ -140,7 +140,7 @@ int main(void)
         int t[8] = { 200,200,200,200, 0,0,0,0 };
         int l[4] = { 200,200,200,200 };
         set_ctx(buf, 200, t, l);
-        daedalus_h264_pred_4x4_ddr_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_ddr(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {200,200,200,200}, {200,200,200,200},
             {200,200,200,200}, {200,200,200,200}
@@ -155,7 +155,7 @@ int main(void)
         int t[8] = { 80,80,80,80, 0,0,0,0 };
         int l[4] = { 80,80,80,80 };
         set_ctx(buf, 80, t, l);
-        daedalus_h264_pred_4x4_vr_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_vr(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {80,80,80,80}, {80,80,80,80}, {80,80,80,80}, {80,80,80,80}
         };
@@ -168,7 +168,7 @@ int main(void)
         int t[8] = { 120,120,120,120, 0,0,0,0 };
         int l[4] = { 120,120,120,120 };
         set_ctx(buf, 120, t, l);
-        daedalus_h264_pred_4x4_hd_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_hd(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {120,120,120,120}, {120,120,120,120},
             {120,120,120,120}, {120,120,120,120}
@@ -182,7 +182,7 @@ int main(void)
         int t[8] = { 64,64,64,64, 64,64,64,64 };
         int l[4] = { 0,0,0,0 };
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_4x4_vl_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_vl(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {64,64,64,64}, {64,64,64,64}, {64,64,64,64}, {64,64,64,64}
         };
@@ -195,7 +195,7 @@ int main(void)
         int t[8] = { 0,0,0,0, 0,0,0,0 };
         int l[4] = { 200,200,200,200 };
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_4x4_hu_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_hu(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             {200,200,200,200}, {200,200,200,200},
             {200,200,200,200}, {200,200,200,200}
@@ -230,7 +230,7 @@ int main(void)
         int t[8] = { 10,20,30,40, 0,0,0,0 };
         int l[4] = { 50,60,70,0 };
         set_ctx(buf, 5, t, l);
-        daedalus_h264_pred_4x4_vr_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_4x4_vr(&buf[1][1], STRIDE);
         uint8_t exp[4][4] = {
             { 8,15,25,35},
             {18,11,20,30},

tests/test_intra_pred_8x8_luma.c

@@ -14,15 +14,15 @@
 #include <stdio.h>
 #include <string.h>
 
-extern void daedalus_h264_pred_8x8l_vertical_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_vertical(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_horizontal_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_horizontal(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_dc_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_dc(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_ddl_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_ddl(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_ddr_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_ddr(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_vr_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_vr(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_hd_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_hd(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_vl_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_vl(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_8x8l_hu_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_8x8l_hu(uint8_t *dst, ptrdiff_t stride);
 
 #define STRIDE 17
 #define ROWS   9
@@ -61,7 +61,7 @@ int main(void)
         for (int i = 0; i < 16; i++) t[i] = 50;
         for (int j = 0; j < 8; j++)  l[j] = 0;
         set_ctx(buf, 50, t, l);
-        daedalus_h264_pred_8x8l_vertical_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_8x8l_vertical(&buf[1][1], STRIDE);
         fail |= check_uniform(buf, "Vertical (mode 0, uniform top)", 50);
     }
 
@@ -71,7 +71,7 @@ int main(void)
         int t[16] = {0}, l[8];
         for (int j = 0; j < 8; j++) l[j] = 70;
         set_ctx(buf, 70, t, l);
-        daedalus_h264_pred_8x8l_horizontal_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_8x8l_horizontal(&buf[1][1], STRIDE);
         fail |= check_uniform(buf, "Horizontal (mode 1, uniform left)", 70);
     }
 
@@ -84,7 +84,7 @@ int main(void)
         for (int i = 0; i < 16; i++) t[i] = 33;
         for (int j = 0; j < 8; j++)  l[j] = 33;
         set_ctx(buf, 33, t, l);
-        daedalus_h264_pred_8x8l_dc_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_8x8l_dc(&buf[1][1], STRIDE);
         fail |= check_uniform(buf, "DC (mode 2, uniform)", 33);
     }
 
@@ -108,7 +108,7 @@ int main(void)
         int t[16], l[8] = {0};
         for (int i = 0; i < 16; i++) t[i] = i;
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_8x8l_vertical_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_8x8l_vertical(&buf[1][1], STRIDE);
         int diff = 0;
         for (int r = 0; r < 8; r++)
             for (int c = 0; c < 8; c++)
@@ -129,7 +129,7 @@ int main(void)
         int t[16] = {0}, l[8];
         for (int j = 0; j < 8; j++) l[j] = j;
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_8x8l_horizontal_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_8x8l_horizontal(&buf[1][1], STRIDE);
         int diff = 0;
         for (int r = 0; r < 8; r++)
             for (int c = 0; c < 8; c++)
@@ -146,12 +146,12 @@ int main(void)
     {
         typedef void (*pred_fn_t)(uint8_t *dst, ptrdiff_t stride);
         struct { const char *name; pred_fn_t fn; } modes[] = {
-            { "DDL (mode 3, uniform)",        daedalus_h264_pred_8x8l_ddl_ref },
+            { "DDL (mode 3, uniform)",        daedalus_h264_pred_8x8l_ddl },
-            { "DDR (mode 4, uniform)",        daedalus_h264_pred_8x8l_ddr_ref },
+            { "DDR (mode 4, uniform)",        daedalus_h264_pred_8x8l_ddr },
-            { "VR (mode 5, uniform)",         daedalus_h264_pred_8x8l_vr_ref  },
+            { "VR (mode 5, uniform)",         daedalus_h264_pred_8x8l_vr  },
-            { "HD (mode 6, uniform)",         daedalus_h264_pred_8x8l_hd_ref  },
+            { "HD (mode 6, uniform)",         daedalus_h264_pred_8x8l_hd  },
-            { "VL (mode 7, uniform)",         daedalus_h264_pred_8x8l_vl_ref  },
+            { "VL (mode 7, uniform)",         daedalus_h264_pred_8x8l_vl  },
-            { "HU (mode 8, uniform)",         daedalus_h264_pred_8x8l_hu_ref  },
+            { "HU (mode 8, uniform)",         daedalus_h264_pred_8x8l_hu  },
         };
         for (size_t i = 0; i < sizeof(modes)/sizeof(modes[0]); i++) {
             uint8_t buf[ROWS][STRIDE];

tests/test_intra_pred_chroma8x8.c

@@ -16,10 +16,10 @@
 #include <stdio.h>
 #include <string.h>
 
-extern void daedalus_h264_pred_chroma8x8_dc_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_chroma8x8_dc(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_chroma8x8_horizontal_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_chroma8x8_horizontal(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_chroma8x8_vertical_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_chroma8x8_vertical(uint8_t *dst, ptrdiff_t stride);
-extern void daedalus_h264_pred_chroma8x8_plane_ref(uint8_t *dst, ptrdiff_t stride);
+extern void daedalus_h264_pred_chroma8x8_plane(uint8_t *dst, ptrdiff_t stride);
 
 #define STRIDE 9
 #define ROWS   9
@@ -69,7 +69,7 @@ int main(void)
         uint8_t buf[ROWS][STRIDE];
         int t[8] = {0}, l[8] = {10, 20, 30, 40, 50, 60, 70, 80};
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_chroma8x8_horizontal_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_chroma8x8_horizontal(&buf[1][1], STRIDE);
         uint8_t exp[8][8];
         for (int r = 0; r < 8; r++) for (int c = 0; c < 8; c++) exp[r][c] = (uint8_t) l[r];
         fail |= check_per_cell(buf, "Horizontal (mode 1)", exp);
@@ -80,7 +80,7 @@ int main(void)
         uint8_t buf[ROWS][STRIDE];
         int t[8] = {15, 25, 35, 45, 55, 65, 75, 85}, l[8] = {0};
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_chroma8x8_vertical_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_chroma8x8_vertical(&buf[1][1], STRIDE);
         uint8_t exp[8][8];
         for (int r = 0; r < 8; r++) for (int c = 0; c < 8; c++) exp[r][c] = (uint8_t) t[c];
         fail |= check_per_cell(buf, "Vertical (mode 2)", exp);
@@ -104,7 +104,7 @@ int main(void)
         int t[8] = { 8, 8, 8, 8,  16, 16, 16, 16 };
         int l[8] = { 24, 24, 24, 24,  40, 40, 40, 40 };
         set_ctx(buf, 99, t, l);
-        daedalus_h264_pred_chroma8x8_dc_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_chroma8x8_dc(&buf[1][1], STRIDE);
         uint8_t exp[8][8] = {
             {16,16,16,16, 16,16,16,16},
             {16,16,16,16, 16,16,16,16},
@@ -125,7 +125,7 @@ int main(void)
         int t[8], l[8];
         for (int i = 0; i < 8; i++) { t[i] = 100; l[i] = 100; }
         set_ctx(buf, 100, t, l);
-        daedalus_h264_pred_chroma8x8_plane_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_chroma8x8_plane(&buf[1][1], STRIDE);
         uint8_t exp[8][8];
         for (int r = 0; r < 8; r++) for (int c = 0; c < 8; c++) exp[r][c] = 100;
         fail |= check_per_cell(buf, "Plane uniform (mode 3)", exp);
@@ -153,7 +153,7 @@ int main(void)
         int t[8], l[8];
         for (int i = 0; i < 8; i++) { t[i] = i; l[i] = i; }
         set_ctx(buf, 0, t, l);
-        daedalus_h264_pred_chroma8x8_plane_ref(&buf[1][1], STRIDE);
+        daedalus_h264_pred_chroma8x8_plane(&buf[1][1], STRIDE);
         uint8_t tl_actual = buf[1 + 0][1 + 0];
         uint8_t br_actual = buf[1 + 7][1 + 7];
         int spot_fail = 0;