h264: V3D shader for qpel mc02 (vertical half-pel) #30
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: marfrit/daedalus-fourier#30
Reference in New Issue
Block a user
Delete Branch "noether/v3d-shader-h264-qpel-mc02"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Sibling of cycle 9's mc20 shader — same 6-tap H.264 luma half-pel filter, transposed to vertical orientation. Address arithmetic flips from row-major (cols ±N) to col-major (rows ±N*stride). Shader is 58 lines vs mc20's 86 (same WG geometry: 64 lanes per block, 1 lane per output pixel).
Recipe table flips
DAEDALUS_KERNEL_H264_QPEL_MC02from CPU to QPU.Test on hertz:
qpel mc02: 2048/2048 bytes bit-exact (100.0000%)via QPU substrate.Qpel QPU coverage: mc20 ✓ (cycle 9), mc02 ✓ (this PR). 13 other put_ + 15 avg_ stay on CPU NEON. Next-priority by 1080p impact (per PR #24 bench): mc22 (2D half-pel, 2.33 ms/frame worst case).
Sibling of cycle 9's v3d_h264_qpel_mc20.comp. Same 6-tap H.264 luma half-pel filter, transposed to vertical orientation: filter reads rows [-2..+3] of source per output pixel instead of cols. Shader is ~58 lines (vs mc20's 86) — same WG geometry (64 lanes / 1 block per WG / 1 lane per output pixel). The address arithmetic flips: row_base = src_off + r*stride + c (mc20) → col_base = src_off + c, then col_base + (r±N)*stride (mc02). dispatch_h264_qpel_mc02_qpu mirrors the mc20 QPU dispatch; src_max calculation differs since the V kernel reads rows -2..+10 of source (13 rows × stride wide) vs mc20's cols -2..+10 (8 rows × stride+11). For 8x8 blocks: src_max = src_off + 10*stride + 8. Recipe table flips DAEDALUS_KERNEL_H264_QPEL_MC02 from CPU to QPU. Verified on hertz: $ ./build/test_api_h264 | grep qpel H.264 qpel mc20: 1024/1024 bytes bit-exact (100.0000%) H.264 qpel mc02: 2048/2048 bytes bit-exact (100.0000%) QPU coverage for the 30 qpel positions: put_ mc20 ✓ (cycle 9) mc02 ✓ (this PR) all 13 other put_ CPU NEON avg_ all 15 positions CPU NEON Next-priority candidates by per-frame impact (per PR #24 bench): mc22 (2D half-pel) — 71.5 ns/block NEON × 32 640 blocks worst case = 2.33 ms/frame at 1080p. Most-used qpel position in real H.264 streams. mc11/mc13/mc31/mc33 — corner ¼-pel positions, structurally similar to mc20 + mc02 with L2 averaging. The cascaded H+V structure of mc22 means it can either share the existing mc20 + mc02 shaders' L2 (compute mc20 into tmp, then mc02 on tmp) or get a dedicated 2-stage pipeline. Follow-up.