> -----Original Message-----
> From: Christoph Hellwig <[email protected]>
> Sent: Tuesday, September 04, 2018 10:20 PM
> To: Avri Altman <[email protected]>
> Cc: Christoph Hellwig <[email protected]>; Johannes Thumshirn
> <[email protected]>; Hannes Reinecke <[email protected]>; Bart Van Assche
> <[email protected]>; James E.J. Bottomley
> <[email protected]>; Martin K. Petersen
> <[email protected]>; [email protected]; Stanislav Nijnikov
> <[email protected]>; Avi Shchislowski
> <[email protected]>; Alex Lemberg <[email protected]>;
> Subhash Jadavani <[email protected]>; Vinayak Holikatti
> <[email protected]>
> Subject: Re: [PATCH v3 1/7] scsi: ufs: Add ufs-bsg module
>
> > +config SCSI_UFS_BSG
> > + bool "Universal Flash Storage BSG device node"
>
> So this a bool,
>
> > obj-$(CONFIG_SCSI_UFSHCD_PCI) += ufshcd-pci.o
> > obj-$(CONFIG_SCSI_UFSHCD_PLATFORM) += ufshcd-pltfrm.o
> > obj-$(CONFIG_SCSI_UFS_HISI) += ufs-hisi.o
> > +obj-$(CONFIG_SCSI_UFS_BSG) += ufs_bsg.o
>
> But built as a separate module, which is rather odd. I think
> you wan to build this into the main ufshcd-core driver as a conditionally
> compiled object instead.
Done.
>
> > +
> > +struct ufs_bsg {
> > + struct ufs_hba *hba;
> > + struct ufs_bsg_node *node;
> > +};
> > +static struct ufs_bsg *bsg_host;
>
> This looks odd. Why would you want a global variable like this?
> I'd expect every ufs host to have a ufs_bsg_node, and in fact
> we should probably just merge the ufs_bsg_node into the ufs_hba
> structure.
Done.
Thanks,
Avri
