Re: [PATCH 6/7] isofs: stop using ioctl_by_bdev

[Christoph Hellwig]

On Thu, Apr 23, 2020 at 01:03:47PM +0200, Jan Kara wrote:
> There's no error handling in the caller and this function actually returns
> unsigned int... So I believe you need to return 0 here to maintain previous
> behavior (however suspicious it may be)?

Indeed, and I don't think it is suspicious at all - if we have no CDROM
info we should assume session 0, which is the same as for non-CDROM
devices.  Fixed for the next version.