diff options
| author | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-18 15:03:17 -0400 |
|---|---|---|
| committer | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-18 15:03:17 -0400 |
| commit | e398de4cf68ffd0f43e0b1692b08b2726457fad4 (patch) | |
| tree | 6f6f247d01c17523a05ed9e36ea3d650ba580305 /scripts/windows_guest_automation.sh | |
| parent | 5df50190fad0fb57174acd30c89dc090c037f315 (diff) | |
| download | open-iov-e398de4cf68ffd0f43e0b1692b08b2726457fad4.tar.gz open-iov-e398de4cf68ffd0f43e0b1692b08b2726457fad4.zip | |
cuda-ab: ctaid_probe (CTA-coordinate + A-row reader) + GEO geometry print in transformer_mt
The vGPU GR engine's failing chains are correct for the by=0 band and
replicate one tile for rows >= 16 with the QMD geometry and the input recs
proven clean, so the remaining hypothesis is engine-side CTA-coordinate
behaviour. ctaid_probe reads it directly in two phases on the exact
(16,4,1)/(16,16,1) geometry:
Phase 1 coord_map writes by*10000+bx*100+ty*10+tx per element (int32
64x256) and reports per-band row0 plus a bad count - a collapsed or
wrong CTAID.Y shows instantly.
Phase 2 arow_read writes A[(by*16+ty)*256+(bx*16+tx)] with A[i]=i, so a
correct engine reads each band's own rows; a collapsed row register reads
one band's rows everywhere.
transformer_mt prints its per-chain geometry (GEO lines) for the host-side
GRID-WIT comparison. Runtime <10 s. Built with
nvcc -O2 --cudart=static -arch=sm_75 (deterministic static runtime).
Diffstat (limited to 'scripts/windows_guest_automation.sh')
0 files changed, 0 insertions, 0 deletions
