On Tue, 2018-09-04 at 11:06 +0100, Bruce Richardson wrote: > On Tue, Sep 04, 2018 at 04:59:07PM +0800, tone.zhang wrote: > > When binding the devices used by DPDK to the "uio-pci" kernel > > module, > > the IOMMU should be disabled in order not to break the IO > > transmission > > because of the virtual / physical address mapping. > > > > The patch clarifies the IOMMU configuration on both x86_64 and > > arm64 > > systems. > > > > Signed-off-by: tone.zhang <tone.zh...@arm.com> > > --- > > doc/guides/linux_gsg/linux_drivers.rst | 7 +++++++ > > 1 file changed, 7 insertions(+) > > > > diff --git a/doc/guides/linux_gsg/linux_drivers.rst > > b/doc/guides/linux_gsg/linux_drivers.rst > > index 371a817..8f9ec8f 100644 > > --- a/doc/guides/linux_gsg/linux_drivers.rst > > +++ b/doc/guides/linux_gsg/linux_drivers.rst > > @@ -48,6 +48,13 @@ be loaded as shown below: > > ``vfio-pci`` kernel module rather than ``igb_uio`` or > > ``uio_pci_generic``. > > For more details see :ref:`linux_gsg_binding_kernel` below. > > > > +.. note:: > > + > > + If the devices for used DPDK bound to the ``uio-pci`` kernel > > module, please make > > + sure that the IOMMU is disabled. We can add ``intel_iommu=off`` > > or ``amd_iommu=off`` > > + in ``GRUB_CMDLINE_LINUX`` in grub on x86_64 systems, or add > > ``iommu.passthrough=1`` > > + on arm64 system. > > + > > I think passthrough mode should work on x86 too. I remember running > with > iommu=pt setting in the kernel in the past. > > /Bruce
It does, can confirm. -- Kind regards, Luca Boccassi
