kernels/rr: staged-image Swin layer engine prototypes

Browse Source

Per-layer staged-image engine for the eleven translated denoiser Swin
layers: descriptors (verified offsets per phase), cut recipes that splice an
isolated native phase into NVIDIA's PTX, wave templates, epilogue notes and
GPU probes. The isolated phase-8 splice produces a byte-identical final
4K frame but is slower than the translated body (14.90 vs 6.59 ms) - a
correctness experiment, not a shipping improvement; docs/rr-engine.md
records the design and the next direct-converter win (B operands built
from transposed accumulators).

...

This commit is contained in:

woutermans committed 2026-10-12 12:47:53 +02:00

1 parent 798e25611c

commit 0fe3a28cb7

58 files changed

+28330 -5

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

kernels/rr/descriptor_enc4.py

+2315

File diff suppressed because it is too large. Load diff