[PATCH 2/2] nvme: fix unmatched id's under delayed path deletion

[Christoph Hellwig]

On Thu, Feb 26, 2026 at 11:35:39AM -0700, Keith Busch wrote:
> On Fri, Feb 27, 2026 at 12:01:48AM +0530, Nilay Shroff wrote:
> > On 2/26/26 10:21 PM, Keith Busch wrote:
> > > This is looking pretty tricky to resolve. The best solution I'm coming
> > > up with so far is to have the scan_work synthesize a
> > > NVME_AER_NOTICE_NS_CHANGED event for every controller in the subsystem,
> > > then re-kick their scan work if the scan_work removed anything.
> > 
> > So does your disk when reuse NSID, changes ns ids such as NGUID/UUID/EUI64?
> 
> Yes. The NSID is put back into the available pool when it's deleted, but
> the UID associated with it is newly generated upon creation.

Yeah, that is one of the two allowed nvme behaviors, and there's a bit
to detect if the persistent ids can be reused or not.