Re: Fwd: Waking up from resume locks up on sr device

[Christoph Hellwig <hch@lst.de>]

On Thu, Jun 15, 2023 at 01:57:37PM +0900, Damien Le Moal wrote:
> > No.  scsi_device_get just increments a reference count, and thus
> > prevents ->release from beeing called.  ->remove is not in any way
> > affected by the refcount.
> 
> What ->remove cb are you talking about ? The gendev one ?

The one for the device locked.

> I am trying to understand why the use of device_lock() helps in any way given
> that this is not used by any other functions in scsi. And given that

The device model locks the device before calling ->remove.

> scsi_rescan_device() should always be called with a ref on the scsi device (and
> so on the gendev as well) held, why would this function be racy with device remove ?

Because ->remove ould otherwise be called at the same time as ->rescan.

> Note that I did find a couple of places where scsi_rescan_device() seems to not
> be called with a reference to the scsi dev held, e.g.  store_rescan_field() and
> store_state_field().

You need both a valid reference and ensure ->remove is not called at the
same time.