[vfio-users] Want to Confirm ACS on Root Ports i7-5930K "High End", and Understand "PCH root port ACS Workaround Enabled "

[Alex Williamson]

On Thu, 2015-12-03 at 09:44 -0600, Jeff wrote:
> I'm pretty certain that ACS on the CPU root ports are working from the
> output of sudo lspci -vvv for the 3 CPU PCI bridges
> 
> (Heavily snipped)
> 00:01.0 PCI bridge: Intel Corporation Xeon E5 v3/Core i7 PCI Express Root
> Port 1
>  Capabilities: [110 v1] Access Control Services
>                 ACSCap: SrcValid+ TransBlk+ ReqRedir+ CmpltRedir+
> UpstreamFwd+ EgressCtrl- DirectTrans-
>                 ACSCtl: SrcValid+ TransBlk- ReqRedir+ CmpltRedir+
> UpstreamFwd+ EgressCtrl- DirectTrans-
> 
> However I don't understand the dmesg | grep output:
> pci 0000:00:1c.0: Intel PCH root port ACS workaround enabled
> pci 0000:00:1c.6: Intel PCH root port ACS workaround enabled
> pci 0000:00:1c.7: Intel PCH root port ACS workaround enabled
> 
> I understand that is from the PCH, but don't understand the "workaround"
> being used for these ports.
> 
> The IOMMU listing outputs proper isolation per device, but I wanted to make
> sure there is no issues for expansion slots coming from the PCH and details
> of the workaround.
> 
> My brief understanding is that is has to do with properly grouping the
> devices or something of that nature.

The kernel log indicates that ACS-equivalent isolation has been enabled
for those PCH root ports:

http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=d99321b63b1f2cd386cdf20aefe4d8d3d414cb5b

For more on IOMMU groups, see:

http://vfio.blogspot.com/2014/08/iommu-groups-inside-and-out.html