Route NanoVDB builder scratch through PyTorch's active CUDA allocator via upstream memory-resource seams - #732

Merged

swahtz merged 9 commits into

openvdb:mainfrom

swahtz:feat/nanovdb-torch-resource

Aug 20, 2026

Commits on Aug 7, 2026

Pin nanovdb to the injectable-memory-resource stack

swahtz
and
claude
committed
44d0375
Route nanovdb builder scratch through PyTorch's caching allocator

swahtz
and
claude
committed
8aac07b
Format fix
swahtz
committed
2ef632b

Commits on Aug 18, 2026

Merge branch 'main' into feat/nanovdb-torch-resource
swahtz
committed
9bd4e9f
Update NanoVDB code to latest on master to pickup all needed memory resource work
swahtz
committed
8b0289a
Route PadGrid and CUB scratch through PyTorch's caching allocator

swahtz
and
claude
committed
6e04022

Commits on Aug 19, 2026

Route the SaveNanoVDB CUDA path through PyTorch's caching allocator

swahtz
and
claude
committed
f43839a

Commits on Aug 20, 2026

Clarify TorchResource docs: dispatches to Torch's currently active CUDA allocator

swahtz
and
claude
committed
d87e852
Centralize the builder scratch allocator choice behind a BuilderResource alias

swahtz
and
claude
committed
416fa43