[STF] Drive parallel_for and data placement from structured partitions#9808

Merged

caugonnet merged 18 commits into

NVIDIA:mainfrom

caugonnet:stf-cute-parallel-for

Jul 17, 2026

Commits on Jul 16, 2026

[STF] Drive parallel_for and data placement from a cute_partition

caugonnet
and
claude
committed
[STF] Test parallel_for over grids driven by cute_partition

caugonnet
and
claude
committed
[STF] Predicated iteration: interior regions and uneven covers

caugonnet
and
claude
committed
[STF] Test the boundary pattern: classic iteration, cute placement

caugonnet
and
claude
committed
[STF] Document computing over structured partitions

caugonnet
and
claude
committed
[STF] Route the cute-composite owner through the rank-0-safe delinearize

caugonnet
and
claude
committed
Merge branch 'main' into stf-cute-parallel-for
caugonnet
authored
[STF] Cache localized arrays for structured partitions
caugonnet
committed
[STF] Clarify structured cache dispatch
caugonnet
committed
[STF] Use exported partition names in test
caugonnet
committed
[STF] Fix structured partition edge cases
caugonnet
committed

Commits on Jul 17, 2026