dma_buf_unpin(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… dma_buf_unpin(9) β€” like, the Kernel API Manual one, bestie

okay so real talk, this is the <pink>dma_buf_unpin</pink> page from the Kernel API Manual and it is genuinely short but SO important, don't skip it πŸŽ€


✨ NAME ✨

dma_buf_unpin β€” the vibe is: unpin a DMA-buf πŸ¦‹


🌸 SYNOPSIS 🌸

void dma_buf_unpin (struct dma_buf_attachment *attach );

like, it returns void, it takes one thing, it's giving "I do one job and I do it well" energy πŸ’…


πŸŽ€ ARGUMENTS πŸŽ€

  • attach β€” [in] the attachment that should get unpinned 🫢 literally just hand it the attachment and it handles it, no cap.

πŸ’– DESCRIPTION πŸ’–

bestie. listen. this is the undo button for dma_buf_pin. πŸ”₯

So like, dma_buf_unpin unpins a buffer that was previously pinned by dma_buf_pin, and once that pin is gone it allows the exporter to move any mapping of attach again 🧚 β€” yes, they were being held in place and now they're free to move, main character energy honestly.

And while it's doing that, it also informs the importer through dma_buf_attach_ops.invalidate_mappings so the importer knows the mappings are no longer valid. communication! we love to see it πŸ˜­πŸ’œ

ngl the dma_buf_pin / dma_buf_unpin pair is a set β€” if you pin it, you unpin it. periodt.

⭐ SEE ALSO ⭐

so you can go down the whole rabbit hole, here's every single one, no skips, we don't gatekeep here πŸ‘‘

go read dma_buf_pin(9) first if you haven't, iykyk πŸ’€


okay and the lil sign-off, because the original footer said so: this one lives under ./drivers/dma-buf, it's from October 2026, and it's the dma_buf_unpin(9) page from the Kernel API Manual β€” bye bestie, go unpin something responsibly πŸŒˆβœ¨πŸ’–

πŸ“¦ where this page comes from