zpci_clear_error_state(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… zpci_clear_error_state β€” the Kernel API Manual said what it said

Okay bestie, so this one is zpci_clear_error_state from the Kernel API Manual, and honestly? It does exactly one thing and it does it with main character energy: it wipes the zPCI error state off a device. Clean slate, fresh start, aura restored. ✨


πŸŽ€ SYNOPSIS πŸŽ€

int zpci_clear_error_state (struct zpci_dev *zdev );

Yep. That's the whole signature. One argument, returns an int. Iconic, lowkey.


🌸 ARGUMENTS 🌸

  • zdev β€” the zdev whose zPCI error state you want reset. Like, that's your girl right there. She's the one getting un-errored. 🫢

✨ DESCRIPTION ✨

So here's the vibe: you call this, and it clears the zPCI error state of the device. Bada bing, bada boom. πŸ’«

But listen β€” and this is the important part, so don't skip it β€” if clearing the zPCI error state fails, the device is just... left in the error state. No rescue, no glow-up. It stays in its flop era. 😭

And in that case, it may make sense to call zpci_io_perm_failure on the associated pdev, but only if it exists. That's the "may make sense" β€” not a promise, just a suggestion, iykyk. πŸ™ˆ

Not me writing this with zero % guarantee: fail = device stays errored, and that's just how it is. Plan accordingly, bestie. πŸ’€

⭐ RETURN ⭐

  • 0 on success β€” slay, we ate πŸ“
  • -EIO otherwise β€” it's giving "the I/O said no" πŸ’”

🌈 SEE ALSO 🌈

Look at this friend group, literally all of them:


k bye bestie πŸ’– β€” this page came to you from ./arch/s390/pci, dated October 2026, and yes it's still zpci_clear_error_state(9) because we don't do rebranding around here. πŸ‘‘

πŸ“¦ where this page comes from