On Wed, Oct 09, 2019 at 05:55:04PM -0400, Sasha Levin wrote: > On Thu, Oct 10, 2019 at 01:29:10AM +0900, Keith Busch wrote: > > commit 733e4b69d508d03c20adfdcf4bd27abc60fae9cc upstream > > > > The namespace disk names must be unique for the lifetime of the > > subsystem. This was accomplished by using their parent subsystems' > > instances which were allocated independently from the controllers > > connected to that subsystem. This allowed name prefixes assigned to > > namespaces to match a controller from an unrelated subsystem, and has > > created confusion among users examining device nodes. > > > > Ensure a namespace's subsystem instance never clashes with a controller > > instance of another subsystem by transferring the instance ownership > > to the parent subsystem from the first controller discovered in that > > subsystem. > > > > Reviewed-by: Logan Gunthorpe <logang@xxxxxxxxxxxx> > > Reviewed-by: Christoph Hellwig <hch@xxxxxx> > > Reviewed-by: Minwoo Im <minwoo.im@xxxxxxxxxxx> > > Reviewed-by: Hannes Reinecke <hare@xxxxxxxx> > > Reviewed-by: Sagi Grimberg <sagi@xxxxxxxxxxx> > > Signed-off-by: Keith Busch <kbusch@xxxxxxxxxx> > > Signed-off-by: Sagi Grimberg <sagi@xxxxxxxxxxx> > > What are your thoughts about taking the following for 4.19 instead? > > 733e4b69d508d nvme: Assign subsys instance from first ctrl > e654dfd38c1ec nvme: fix memory leak caused by incorrect subsystem free > 32fd90c407680 nvme: change locking for the per-subsystem controller list > 092ff0520070f nvme: fix kernel paging oops > cb5b7262b011c nvme: provide fallback for discard alloc failure Thank you for the suggestion. I have reviewed the proposed sequence, and I would prefer to use this for stable instead of this back-ported patch. > -- > Thanks, > Sasha
