Refactor CUDA graph API: decompose cuda_graph_scope into full_iteration impl, inference scope, and per-layer capture modules - #4292

Merged

buptzyb merged 8 commits into

NVIDIA:mainfrom

buptzyb:main-strict-refactor

May 18, 2026

Commits on May 12, 2026

refactor(cuda_graph): rename scope to modules
buptzyb
committed
5955fda
Address CUDA graph inference review feedback
buptzyb
committed
86f86b0

Commits on May 13, 2026

Address CUDA graph config review feedback
buptzyb
committed
6dc6e5d

Commits on May 14, 2026

Merge remote-tracking branch 'origin/main' into main-strict-refactor
buptzyb
committed
b38a95a

Commits on May 15, 2026

Merge branch 'main' into main-strict-refactor
buptzyb
authored
73b71e7

Commits on May 17, 2026