Re: [BK PATCH] PCI Express patches for 2.4.27-pre3

[Marcelo Tosatti]

On Tue, May 25, 2004 at 07:40:56AM -0700, Greg KH wrote:
> On Tue, May 25, 2004 at 09:54:53AM -0300, Marcelo Tosatti wrote:
> > 
> > Hi kernel fellows,
> > 
> > On Tue, May 25, 2004 at 12:32:31PM +0100, Matthew Wilcox wrote:
> > > On Tue, May 25, 2004 at 01:00:06AM -0700, Greg KH wrote:
> > > > > how does this mesh with the "2.4 is now feature frozen"?
> > > > 
> > > > As the major chunk of ACPI support just got added to the tree, and the
> > > > only reason that went in was for this patch, I assumed that it was
> > > > acceptable.
> > 
> > major? the MMConfig support is minimal as I can see? 
> 
> It isn't that big of a patch, but it is make to core PCI code.
> 
> > > > Marcelo, feel free to tell me otherwise if you do not want
> > > > this in the 2.4 tree. 
> > 
> > Is this code necessary for PCI-Express devices/busses to work properly?
> 
> Not that I can tell, the main point is accessing the extended config
> space, and speeding up the access to the device to its natural speed.
> 
> > > I assume it was added because Len tries to keep ACPI in 2.4 and 2.6 as
> > > close to identical as possible.  It certainly doesn't hurt anyone to add
> > > the ACPI functionality without the MMConfig support.
> > 
> > I've humbly asked Len to stop doing big updates whenever possible on the 
> > v2.4 ACPI code, and do bugfixes only instead. Is that a pain in the ass for you, Len?    
> > 
> > I asked that because it is common to see new bugs introduced by an ACPI update, 
> > and you know that more than I do.
> 
> Yes, I know that quite well :)
> 
> So, because of this, you are saying that we should not apply these
> patches at this time? 

Yeap, I would prefer not to apply them at this time. For one, Arjan told
me privately it can break XFree86 which accesses the PCI config space directly.
Right?

> If so, that's fine with me, as now any distro that wants to add this to
> their 2.4 kernel can, as the patches are public.

Yeap. 

> > PS: Greg, about the PCI-Express hotplug drivers, I assume they are independant 
> > on any of this?
> 
> Yes, that is independant of these changes (that is required as there is
> no other way to control the pci hotplug controller of those systems,
> except with the driver I had submitted in the past to you.).
> 
> thanks,
> 
> greg k-h
-
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/