On 16 June 2014 10:18, Claudio Fontana <claudio.font...@huawei.com> wrote: > On 14.06.2014 01:10, Peter Crosthwaite wrote: >> For PCI support it should just be a case of adding a PCI controller to >> the virt board, then -device should just work for pci devs. > > Thanks for your answer; that seems straightforward enough, > so would a patch to add pci to the virt board be acceptable?
You would need to implement a sensible PCI controller first. Actually Rob Herring has some prototype patches which do this (using the kernel's new support for a PCI controller which can be defined entirely using a device tree). They have some bugs still, though (the lsi SCSI controller device doesn't seem to work, for instance). Rob, would you mind posting your PCI controller patch as an RFC? That way Claudio could have a look at the issues with DMA if he wants to move forward here. thanks -- PMM
