ib_unregister_device_queued(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… ib_unregister_device_queued β€” unregister a device using a work queue ✨

okay bestie, real talk: this page is from the Kernel API Manual, section 9, and it’s all about ib_unregister_device_queued. no cap, it’s short but important 🫢


🌸 NAME 🌸

ib_unregister_device_queued β€” unregister a device using a work queue πŸ’–


✨ SYNOPSIS ✨

void ib_unregister_device_queued (struct ib_device *ib_dev );

πŸŽ€ ARGUMENTS πŸŽ€

  • ib_dev β€” the device to unregister πŸ‘‘

πŸ’– DESCRIPTION πŸ’–

So this schedules an asynchronous unregistration using a WQ for the device. it’s giving β€œhandle it later, but make it safe” energy ✨

A driver should use this to avoid holding locks while doing unregistration, such as holding the RTNL lock. literally the whole point, bestie β€” don’t be out here clutching locks while unregistering if you can avoid it πŸ’…

🚨 important caveat, no cap: drivers using this API must use ib_unregister_driver before module unload to ensure that all scheduled unregistrations have completed. periodt.

πŸ¦‹ SEE ALSO πŸ¦‹

kernel file: ./drivers/infiniband/core/device.c


source/version: ./drivers/infiniband/core Β· date: October 2026 Β· page: ib_unregister_device_queued(9) β€” mwah, go slay those unregistrations πŸ’‹πŸŒˆ

πŸ“¦ where this page comes from