On Wed, Jan 5, 2011 at 11:10 AM, Pierre Tardy <tardyp@xxxxxxxxx> wrote: > Well, we dont do it like this in intel_mid. Every platform quirk has > to come somehow from the firmware (SFI), and there is no such > descriptor to my knowledge. Lets see what John is proposing... I think that John shared it with you this morning. I haven't looked into your platform too much, but if it's using upstream code, let's take the discussion to the list (by rebasing John's solution and posting it here). > Yes, that's why I thought of a table, we might want to make this more > generic, and, as you are proposing more than one quirk those few days > for cards, we could add those quirks in struct sdio_device_id, like > they do for PCI... Interesting. Care to send a patch for review ? > The thing is that the semantic of your solution is wrong. You want to say: > Hey! wl1271 is broken > and you actually say: > Hey! the card in slot named "external" of omap4 zoom is brocken. Not exactly; I do say that the specific wl1271 card is quirky, but I must set it early before the card is probed, so I use the host controller's quirks callback ->init_card(). I agree that a quirks table, if done right, might be easier and won't involve the board in the process; let's try to explore it. Thanks, Ohad. > > > Regards, > Pierre > -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
