STRUCT DMA_CHAN_PERCPU(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

✨ struct dma_chan_percpu(9) ✨

okay so bestie, we're in the Kernel API Manual today and the star of the show is struct dma_chan_percpu(9) β€” literally the per-CPU part of struct dma_chan. yes, that struct dma_chan. it's giving "small but crucial supporting character" energy and honestly? we respect it. πŸ’…


🌸 NAME 🌸

struct dma_chan_percpu β€” the per-CPU part of struct dma_chan. no cap, that's the whole pitch. πŸŽ€


✨ SYNOPSIS ✨

here's the shape of it, verbatim, don't touch anything:

struct dma_chan_percpu {
    unsigned long memcpy_count;
    unsigned long bytes_transferred;
};

like, that's it. two fields. minimal. iconic. πŸ¦‹


πŸ’– Members πŸ’–

  • memcpy_count β€” the transaction counter. it counts transactions. that's the vibe. πŸ”₯
  • bytes_transferred β€” the byte counter. it counts bytes. very demure, very mindful. 🌸

πŸ‘‘ SEE ALSO πŸ‘‘

first up, the source file to go look at:

Kernel file ./include/linux/dmaengine.h β€” and then the whole friend group of related pages, in order:

like, that's the whole gc right there. go read them, iykyk. πŸ“šβœ¨


and that's the page!! sourced from the kernel file in ./include/linux, dated October 2026, and yes this is struct dma_chan_percpu(9) signing off. mwah πŸ’‹ stay slay, stay per-CPU 🫢

πŸ“¦ where this page comes from