Here is an attempt to isolate NVLink interconnects between GPU to let them be passed through individually.
At the moment I mostly wonder about the sanity of the appoach. Please comment. Thanks. Alexey Kardashevskiy (2): vfio_pci: Allow device specific error handlers vfio-pci-nvlink2: Implement interconnect isolation drivers/vfio/pci/vfio_pci_private.h | 5 ++ arch/powerpc/platforms/powernv/npu-dma.c | 24 +++++- drivers/vfio/pci/vfio_pci.c | 17 ++++ drivers/vfio/pci/vfio_pci_nvlink2.c | 98 ++++++++++++++++++++++++ 4 files changed, 142 insertions(+), 2 deletions(-) -- 2.17.1
