On 2016-10-26 21:53, Jan Kiszka wrote: > Good news is that I'm working on proper PCI support for ARM based on the > Jetson and found a bug some days ago that prevented completing it. > Should be resolved in a couple of days once I'm back at the keyboard.
A follow-up note after digging deeper into the Jetson TK1 and now also a bit the TX1: It seems that there is no way to split the PCIe resources further up on those SOCs. First, the both consider the AFI controller (the gateway of the PCIe root ports to the system) as a single device, thus managing all PCIe devices behind it as one in the SMMU. That means we are unable to split them up between cells. And second, I also failed to use the physical PCI system on the Jetson to inject virtual shared-memory devices (ivshmem) because the Linux host driver has a very static view on the topology and doesn't rescan it (because there can't be any addition physical devices in practice). IOW, I will not enable physical PCI handling on the Jetson and rather try to inject an additional virtual generic host controller for adding ivshmem devices. Jan -- Siemens AG, Corporate Technology, CT RDA ITP SES-DE Corporate Competence Center Embedded Linux -- You received this message because you are subscribed to the Google Groups "Jailhouse" group. To unsubscribe from this group and stop receiving emails from it, send an email to jailhouse-dev+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
