sba_unmap_phys(9)

📦 version 7.3.1, in nixpkgs unstable

✨ sba_unmap_phys(9) ✨

okay so bestie, we're in the Kernel API Manual today and this page is literally about sba_unmap_phys — it's giving "cleanup era" and honestly? we love a function that knows how to let go. 🦋

💅 NAME 💅

sba_unmap_phys — unmap one IOVA and free resources

like yes, that's the whole vibe: one IOVA, unmap it, free the resources, go home. periodt. ✨

🌸 SYNOPSIS 🌸

void sba_unmap_phys (struct device *dev , dma_addr_t iova , size_t size
, enum dma_data_direction direction , unsigned long attrs );

no cap, that's the exact signature — copy it exactly, don't be weird about it 💖

🎀 ARGUMENTS 🎀

  • dev — instance of PCI owned by the driver that's asking. iykyk, it's the driver's own device, bestie, nobody else's 🙈
  • iova — the IOVA of a driver buffer that was previously mapped. like, you gotta have mapped it first, we're not out here unmapping random stuff 💀
  • size — number of bytes mapped in the driver buffer. yes, the byte count, keep it accurate fr fr
  • direction — R/W or both. that's it, that's the whole description, she's concise 👑
  • attrs — attributes 🫶

🌈 DESCRIPTION 🌈

See Documentation/core-api/dma-api-howto.rst

that's genuinely it, the docs said "go read the howto" and honestly? iconic. the lore is in the rst file, bestie, go read it 📖✨

🌟 SEE ALSO 🌟

Kernel file ./drivers/parisc/sba_iommu.c, plus the whole friend group of sba functions — a full roster, no one left out:

💀 reminder bestie: this frees resources. it's a real unmap, not a decorative one. don't go calling it on stuff that was never mapped, ok? 🫶

and that's the page!! mwah 💋 this one lives in ./drivers/parisc, dated October 2026, and it's sba_unmap_phys(9) all the way down — take care of her, she's a legend 🌸🎀✨

📦 where this page comes from