The "config FUSION" entry in Kconfig is handling both mptbase.ko
and mptscsih.ko for some reason?
The mptbase.ko driver is the lower layer driver which configures
the adapters and is transport layer to and from the chip. It doesn't
depend on anything in the scsi mid layer.
The mptscsih.ko does depend on scsi_mod.ko, but not sd.ko
Therefore - maybe we could have separate entries in Kconfig for
both mptbase and mptscsih. mptbase depend on PCI, and mptscsih
depend on SCSI. What do you think?
Eric Moore
LSI Logic
On Tuesday, January 20, 2004 4:36 PM, Christoph Hellwig wrote:
> On Wed, Jan 21, 2004 at 12:25:07AM +0100, Adrian Bunk wrote:
> > With BLK_DEV_SD=n, I see a "Fusion MPT device support" menu I can't
> > enter.
> >
> > The simple patch below removes the "Fusion MPT device
> support" menu if
> > BLK_DEV_SD=n.
>
> I'd rather see an explanation from LSI why a scsi LLDD depens
> on a uper
> driver. This can't be right.
>
> -
> To unsubscribe from this list: send the line "unsubscribe
> linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Wed, Jan 21, 2004 at 12:00:09PM -0500, Moore, Eric Dean wrote:
> The "config FUSION" entry in Kconfig is handling both mptbase.ko
> and mptscsih.ko for some reason?
>
> The mptbase.ko driver is the lower layer driver which configures
> the adapters and is transport layer to and from the chip. It doesn't
> depend on anything in the scsi mid layer.
>
> The mptscsih.ko does depend on scsi_mod.ko, but not sd.ko
>
> Therefore - maybe we could have separate entries in Kconfig for
> both mptbase and mptscsih. mptbase depend on PCI, and mptscsih
> depend on SCSI. What do you think?
Sounds good. (And mptscsi should depend on mptbase of course)