On 04/11/16 16:34, Ard Biesheuvel wrote: > On 11 April 2016 at 16:12, Laszlo Ersek <[email protected]> wrote: >> On 04/11/16 15:43, Ard Biesheuvel wrote:
[snip] > I simply want to check that the 'pci-ecam-generic' DT node we end up > consuming is the only one that exists in the device tree. In fact, I > think it makes sense to assert that PcdPciExpressBaseAddress equals > the DT node reg property when we encounter it. Yes, that definitely makes sense (at least if I understand it correctly). That is, the library constructor would do the "reg" parsing / PCD setting / caching etc (for any DXE_DRIVER that uses this instance, really), and *specifically* PciHostBridgeDxe, when it parses the same node (for those other two properties), could also look at "reg", and assert that PcdPciExpressBaseAddress is *already set* the way it expects it to be. I think that's a good idea. >> Do you have a fresh v3-wip branch somewhere? I'd like to try expressing >> my proposal for patches #13 and #14 in code. (If you are okay with that.) >> > > Sure, give me 30 mins. Let's do the following instead (as we discussed off-list): Please commit patches v2 1-10 (with the fixes I requested) to the master branch, and then we can collaborate, on the list, with patches for fw_cfg and PciExpressLib / PciHostBridgeDxe, directly against the master branch. Cheers! Laszlo _______________________________________________ edk2-devel mailing list [email protected] https://lists.01.org/mailman/listinfo/edk2-devel
