Re: [PATCH] DRM: add missing pci_enable_device()

[Bjorn Helgaas <bjorn.helgaas@hp.com>]

On Tuesday 14 September 2004 5:12 pm, Dave Airlie wrote:
> > OK, I'll assume you understand the issue and will resolve it.  In the
> > meantime, users of DRM will have to supply "pci=routeirq".
> 
> is this -mm only or is it mainline kernel stuff now?

It's been in -mm for about a month so far, and it still
needs some cooking before it's ready for mainline.  The
remaining issues are:

	- nvidia: Nvidia posted a patch for the open-source part
		of their driver, but we'll likely have to keep
		the "pci=routeirq" option longer than I originally
		hoped.
	- swsusp: Some devices like prism54, USB don't
		work after suspend/resume.  Prototype patch
		being tested.  I suspect we'll trip over
		more issues here, because the resume hooks
		are poorly documented and inconsistently
		implemented.
	- DRI: Sounds like you can do the trivial "enable-only"
		change that will make things work.

I'm a little surprised that I've only heard one report about
DRI, actually.

> I'll throw an enable in to the bk tree later on....

Great, thanks!