On Mon, Jan 30, 2012 at 12:49 PM, Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx> wrote: > On Mon, 2012-01-30 at 09:58 -0700, Bjorn Helgaas wrote: >> Tell the PCI core about host bridge address translation so it can take >> care of bus-to-resource conversion for us. >> >> N.B. powerpc used a 32-bit mask when converting IO addresses. If we >> figure out the right offset, I think simple addition should do the >> same thing, but I might not have it exactly right yet. > > Wow that's going to be a big change :-) A quick look... seems like you > do a per-resource offset right ? I will need that (was about to change > powerpc) as we have hardware that provides us with two different windows > to MMIO space with different offsets unfortunately. Exactly ... the offset is per-aperture. > One thing I also want to generalize is the whole "REASSIGN_ALL_RSRC" > flag, ie, avoid reading & printing all those resources we know we are > just going to kill off. > > I'll try to test your stuff asap (I'm doing 3 things at once right now > so it might take a few days unless I find somebody to look at it). Thanks! No particular rush on my side; I'm expecting to have to iterate a few times to get this right. Bjorn -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
