Re: [PATCH] pciehp: Add pciehp_surprise module option

[Matthew Garrett]

On Wed, Mar 20, 2013 at 08:23:39PM +0100, Takashi Iwai wrote:
> At Wed, 20 Mar 2013 19:12:50 +0000,
> Matthew Garrett wrote:
> > On Wed, Mar 20, 2013 at 08:11:15PM +0100, Takashi Iwai wrote:
> > > I'm afraid that it's too radical to enable always.
> > 
> > Why?
> 
> Because I'm conservative :)
> Well, dunno.  It's just my feeling without deep thought.

Adding a parameter means it's broken until people discover that they 
need to pass a parameter to make it work, and many people will just be 
stuck with it not working. It really should be the last resort.

> > > Or, what about to check this bit only for disable path?
> > 
> > What happens if you unplug the device on one of the affected machines?
> 
> It continues working.  I mean, the PCI device is still there after
> unplug, but the Realtek driver unmounts it smoothly.  Re-plugging also
> works, too.  So, the handling via pciehp is needed only for the very
> first time to register the PCI device.

But if I plug in a different device, things will be broken, right?

-- 
Matthew Garrett | mjg59@xxxxxxxxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/