iommufd_device_unbind(9)
β¨ iommufd_device_unbind(9) β¨
okay bestie, this page is from the Kernel API Manual, and it's literally the cleanup bestie to iommufd_device_bind() β the βundoβ moment for binding a device to iommufd. π«Ά
π NAME π
iommufd_device_unbind - Undo iommufd_device_bind()
π SYNOPSIS π
void iommufd_device_unbind (struct iommufd_device *idev );
πΈ ARGUMENTS πΈ
idevβ Device returned byiommufd_device_bind.
π¦ DESCRIPTION π¦
So what's the vibe? It releases the device from iommufd control. The DMA ownership will return back to unowned with DMA controlled by the DMA API. Slay. β¨
Big huge important thing: this invalidates the iommufd_device pointer, so other APIs that consume it must not be called concurrently. Not me warning you twice, butβ¦ do not do that. π
π₯ SEE ALSO π₯
Kernel file ./drivers/iommu/iommufd/device.c, plus these besties:
iommufd_device_bind(9)iommufd_ctx_has_group(9)iommufd_device_attach(9)iommufd_device_replace(9)iommufd_device_detach(9)iommufd_access_create(9)iommufd_access_destroy(9)iommufd_access_notify_unmap(9)iommufd_access_unpin_pages(9)iommufd_access_pin_pages(9)iommufd_access_rw(9)
xoxo, from ./drivers/iommu/iommufd, October 2026, with all the iommufd_device_unbind(9) energy π
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms