Re: satalink at cardbus

[Johan Ihren <johani%johani.org@localhost>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 5 Aug 2008, at 18:53, KIYOHARA Takahshi wrote:

> I wrote the satalink_cardbus and the pciide_cardbus layer.
> Current pciide depends to pci.  I added some members to struct  
> pciide_softc.
> And member cardbus_chip_map added to struct pciide_product_desc.

> And I want testers.  ;-)


Wonderful! I'd be happy to play with this.

On 5 Aug 2008, at 20:05, Tom Spindler wrote:

> In order to get this to compile on my system, I had to adjust all the
> initalizers in piixide.c; presumably, this would be required in other
> ide chipset drivers, as well.

I get this:

/usr/store/source/netbsd/current/src/sys/dev/pci/acardide.c:56:  
warning: missing initializer
/usr/store/source/netbsd/current/src/sys/dev/pci/acardide.c:56:  
warning: (near initialization for  
'pciide_acard_products[0].cardbus_chip_map')
/usr/store/source/netbsd/current/src/sys/dev/pci/acardide.c:61:  
warning: missing initializer
...

It seems obvious that I'm hitting the same problem you did, except I  
don't understand how to fix this in piixide.c. Can you send me a diff?

> It'd be nice if the cardbus and pci
> bits were unified, but that's definitely outside the scope of he
> driver.

I agree. Also it would be *really* nice with some basic expresscard  
support. Nearly every laptop eSATA-card on the market is for  
expresscard rather than cardbus these days, I had to hunt a bit to  
find a cardbus card with the SiI3512 in it.

I understand that the right way to do expresscard is to not implement  
it like cardbus was implemented, but to be perfectly honest I'd rather  
have the cardbus equivalent, if that's easy to do, than nothing at all.

Regards,

Johan

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFIqJ36KJmr+nqSTbYRAp+oAJ9JNFdvTeA21VKP5sMDVP3sOLECYACeMwWB
bm/6ZRNq+bh++fJzJNkvuAc=
=FCd7
-----END PGP SIGNATURE-----