Hello! I investigating the virtualisation abilities of current AMD FirePro hardware; specifically how to securely pass a share of a GPU to one application (a Rkt/ Docker container). I already read vfio.txt and pci-iov-howto.txt, but they do not answer my questions. I was unable to find documentation on the process online, either. 1. How do I create VFs from a AMD FirePro PF? I saw an example using VMWare ESXi tools, where it is possible to specify RAM size and time-share, but I have not seen how that translates to modifying /sys/bus/pci/..., which is the method recommended for SR-IOV devices in pci-iov-howto.txt. 2. After having created VFs, is it possible to bind these to the host's amdgpu driver, so that I get multiple /dev/dri/card* device nodes? 3. Will passing these /dev/dri/card* device nodes into a Rkt/Docker container be enough for the AMD Radeon Pro OpenGL runtime to be able to use that card? Best regards, Dennis Schridde -- Heidelberg University Computing Centre Service division: Future IT - Research & Education Tel. +49 6221 54-4519, Fax +49 6221 54-5581 dennis.schridde at uni-heidelberg.de http://www.urz.uni-heidelberg.de/ Ruprecht-Karls-Universität Heidelberg Universitätsrechenzentrum Im Neuenheimer Feld 293, 69120 Heidelberg, Germany -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 681 bytes Desc: This is a digitally signed message part. URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20160825/930e8f4a/attachment.sig>
