↑ ↓ to navigate
    ↵ to select
    esc to close
    Archmage

    Memory

    Load, store, gather, scatter, and data layout patterns

    Moving data between memory and SIMD registers efficiently. Data layout, arithmetic, and target-feature context all affect performance.

    1. Load & Store — Array loads/stores, length proofs, scalar tails
    2. Gather & Scatter — Checked access and the limits of the current API
    3. Interleaved Data — deinterleave_4ch, interleave_4ch for RGBA and similar
    4. Chunked Processing — Processing large arrays in SIMD-sized chunks, alignment, performance
    Edit this page