On 2/15/22 12:28 PM, Huang Adrian wrote:
On Tue, Feb 15, 2022 at 9:26 AM Lu Baolu <baolu...@linux.intel.com> wrote:
On 2/14/22 5:32 PM, Huang Adrian wrote:
Hi Baolu,
On Mon, Feb 14, 2022 at 8:35 AM Lu Baolu <baolu...@linux.intel.com> wrote:
Hi Adrian,
The solution is to prevent from allocating pasid table if those
devices are subdevices of the VMD device.
Thanks for your patch!
Is this the only patch that is needed to make VMD devices work in VT-d
scalable mode?
Yes, it is. With this patch, the system can boot successfully and the
RAID device connected to the VMD device can be recognized by the host
OS. I also ran a simple IO stress on the RAID device by compiling
kernel `$ make -j $(nproc)`. Everything looks good.
Thank you! So how far should this patch be back ported?
Thanks for the feedback.
Do you mean the CC stable kernel tag? Something like: "Cc:
sta...@vger.kernel.org # v5.8+"?
Can you please add a Fixes tag?
Fixes: 0bbeb01a4faf ("iommu/vt-d: Manage scalalble mode PASID tables")
Do you want me to send a v2 with the CC stable kernel tag and Fixes
tag? Or, do you just need to take the Fixes tag directly without
sending a v2?
I don't think we can back port it as far as that. Probably,
Fixes: 2b0140c69637e ("iommu/vt-d: Use pci_real_dma_dev() for mapping")
Cc: sta...@vger.kernel.org # v5.6+
Please verify whether v5.6 works with your patch back ported. If it
works, please send a v2 with above tags appended.
-- Adrian
Best regards,
baolu
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
