Crate launchbound_model 

The analytical model behind --backend model (S6).

It estimates relative cost within one kernel’s space from occupancy and wave count — nothing else. Its output is labelled estimated on every surface, and it ships only with its measured Spearman rank correlation against real hardware attached (docs/LIMITATIONS.md): the model is gated on measured quality, not on plausibility.

Structs

DeviceParams

Per-SM limits by compute capability.

Estimate

One candidate’s estimate. cost is a unitless relative score within a kernel’s space — smaller is predicted faster. It is NOT a time.

Enums

ModelError

What can go wrong estimating a candidate.

Constants

DEVICES

Ascending by compute capability. A test enforces both the order and the internal consistency of every row.

Functions

device

Look up the capacity figures for a compute capability.

estimate

Estimate one candidate. Model: blocks-per-SM limited by threads, smem and the block cap; cost = waves / occupancy — a candidate that needs more waves of less-occupied SMs is predicted slower.

known_capabilities

The capabilities device will accept, ascending, for error messages.

smem_bytes

Shared-memory bytes per block for a candidate: the [model] smem_bytes expression in kernel.toml, over the kernel’s dimensions.

spearman

Spearman rank correlation between two paired samples (average ranks for ties). Returns None below 3 pairs — a correlation of two points is noise dressed up as a number.