host1x_bo_pin(9)

📦 version 7.3.1, in nixpkgs unstable

💅 host1x_bo_pin(9) — a Kernel API Manual moment

okay so bestie, welcome to host1x_bo_pin(9), straight out of the Kernel API Manual ✨ this one's short and sweet, no cap 🎀

✨ NAME ✨

host1x_bo_pin — literally makes a DMA mapping for the buffer object, that's the whole vibe 🦋

🌸 SYNOPSIS 🌸

struct host1x_bo_mapping * host1x_bo_pin (struct device *dev , struct
host1x_bo *bo , enum dma_data_direction dir , struct host1x_bo_cache
*cache );

four args, one struct host1x_bo_mapping * coming back at you, periodt 💅

🎀 ARGUMENTS 🎀

  • dev — the device you're DMA-mapping onto, like the destination of the whole operation 🫶
  • bo — the buffer object that's getting mapped, main character energy fr fr 👑
  • dir — the DMA direction, that's an enum dma_data_direction situation
  • cache — the cache where the mapping gets stored… or NULL, your call bestie 💜

🔥 DESCRIPTION 🔥

So what it does: it creates a DMA mapping pointing to bo for dev ✨ and the refcount of bo gets incremented — and it stays bumped until host1x_bo_unpin is called. no early exits, that's the deal 😤

and if you pass a cache? the mapping is also stored in the cache, and it is not released until bo is freed (aka refcount drops to zero) 🌀 that's the whole reason it slaps: when a buffer is pinned and unpinned over and over, like in the case of display use, it improves performance. literally the glow-up ✨💀

ngl the caching part is the icon of this whole function — skip it and you're just doing the plain mapping. still valid, just less slay 🙈

🌈 SEE ALSO 🌈

Kernel file ./drivers/gpu/host1x/bus.c and honestly the whole friend group 🫶:

okay that's the page!! 🧚 this one lives in ./drivers/gpu/host1x, dated October 2026, and it's filed under host1x_bo_pin(9) — go pin some buffers bestie, iykyk 💖⭐

📦 where this page comes from