Re: [PATCH v2] ata: New PATA driver for Altera CompactFlash.

[Alan Cox <alan@lxorguk.ukuu.org.uk>]

On Thu, 8 Oct 2009 12:22:22 +0100
Ian Abbott <abbotti@mev.co.uk> wrote:

> On 08/10/09 10:55, Alan Cox wrote:
> > On Thu,  8 Oct 2009 09:58:30 +0800
> > Thomas Chou <thomas@wytron.com.tw> wrote:
> > 
> >> This patch adds a new PATA driver to support the Altera SOPC Builder
> >> CompactFlash component as a PATA device with support for insertion
> >> and removal of CF cards.  It uses the platform driver model.
> > 
> > Can you explain why you remove the interface rather than the device on
> > the interface when you detect a card removal ?
> 
> As the originator of the driver, the only answer I have is "I don't know
> how."  I was hoping for inspiration from some of the other PATA CF
> drivers, but alas, I couldn't find it.  I admit that removing the whole
> ATA host is a bit on the extreme side.

I think the bit you want is something like:

	struct ata_eh_info *ehi = &ap->link->eh_info;
	if (mumblemumble & whatever) {
		ata_ehi_clear_desc(ehi);
		ata_ehi_hotplugged(ehi);
		ata_ehi_push_desc(ehi, "CF connection changed");
		ata_port_freeze(ap);
	}

The core code also assumes your freeze method will mask any further
connection change reports until the port is unfrozen (so you don't get
multiple parallel "in/out/in/out" events at once)

Alan