Messages in this thread | | | Subject | Re: ohci1394 problem (MMIO broken) (was 2.6.25-rc6-git6: Reported regressions from 2.6.24) | From | Benjamin Herrenschmidt <> | Date | Wed, 26 Mar 2008 23:17:44 +1100 |
| |
On Wed, 2008-03-26 at 13:14 +0300, Ivan Kokshaysky wrote: > > Yeah, it's resource issue for sure. The 00:1e.0 is a transparent > bridge, > so I blame commit 8fa5913d54f3b1e09948e6a0db34da887e05ff1f > (PCI: remove transparent bridge sizing). It's wrong for two reasons: > - we cannot ignore standard windows of a transparent bridge as they > always positive decode, so they are potential source of address > conflicts; > - that patch just broke whole bridge setup logic in unpredictable way.
The root of the problem is again HW vendors getting wild I think...
Some bridges are fully transparent, some have explicit decode windows but are also transparent for whatever else and we definitely don't deal with it.
Ben.
| |