dma_pool_create_node(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’– NAME πŸ’–

okay so this is the dma_pool_create_node(9) page from the Kernel API Manual, and dma_pool_create_node literally creates a pool of coherent DMA memory blocks. iconic behavior, no cap πŸ’…


✨ SYNOPSIS ✨

struct dma_pool * dma_pool_create_node (const char *name , struct device *dev , size_t size , size_t align , size_t boundary , int node );

πŸŽ€ ARGUMENTS πŸŽ€

  • name β€” name of pool, for diagnostics 🩺
  • dev β€” device that will be doing the DMA 🫢
  • size β€” size of the blocks in this pool. πŸ“
  • align β€” alignment requirement for blocks; must be a power of two ⚑
  • boundary β€” returned blocks won't cross this power of two boundary 🚧
  • node β€” optional NUMA node to allocate structs dma_pool and dma_page on 🌐

πŸ¦‹ CONTEXT πŸ¦‹

bestie, the context is not in_interrupt β€” that’s the vibe check here 😭

🌸 DESCRIPTION 🌸

Given one of these pools, dma_pool_alloc may be used to allocate memory. Such memory will all have coherent DMA mappings, accessible by the device and its driver without using cache flushing primitives. The actual size of blocks allocated may be larger than requested because of alignment. lowkey important, highkey useful πŸ’–

If boundary is nonzero, objects returned from dma_pool_alloc won't cross that size boundary. This is useful for devices which have addressing restrictions on individual DMA transfers, such as not crossing boundaries of 4KBytes. 🧚✨


⭐ RETURN ⭐

a dma allocation pool with the requested characteristics, or NULL if one can't be created. that’s it, that’s the return πŸ’€πŸ«Ά


🌈 SEE ALSO 🌈


and that’s the whole page, bestie πŸŽ€ this one came from source/version ./mm, dated October 2026, and it’s the dma_pool_create_node(9) manual page β€” mwah πŸ’‹βœ¨

πŸ“¦ where this page comes from