Expand description
Launch planning for explicit small-matrix warp-shared kernels. Dependency-free launch geometry for the opt-in warp-shared direct solvers. This describes source-level storage/work, NOT measured bandwidth or latency.
Structs§
- Warp
Direct Limits - Capability snapshot supplied by a real runtime, not inferred from its name.
- Warp
Direct Plan - Immutable validated plan; one 32-thread block owns one system.
Enums§
- Warp
Direct Kind - Which scratch layout is needed. Both algorithms preserve the serial APIs.
- Warp
Plan Error