Revision 2 of the integrated GPU passthrough proposal

Signed-off-by: John Else <[email protected]>

Author: johnelse

features/futures/integrated-gpu-passthrough.md

@@ -2,16 +2,17 @@
 title: Integrated GPU passthrough support
 layout: default
 design_doc: true
-revision: 1
+revision: 2
 status: proposed
 ---
 
 Introduction
 ------------
 
 Passthrough of discrete GPUs has been
-[available since XenServer 6.0]({{site.baseurl}}/xapi/design/gpu-passthrough.html). With some
-extensions, we will also be able to support passthrough of integrated GPUs.
+[available since XenServer 6.0]({{site.baseurl}}/xapi/design/gpu-passthrough.html).
+With some extensions, we will also be able to support passthrough of integrated
+GPUs.
 
 - Whether an integrated GPU will be accessible to dom0 or available to
   passthrough to guests must be configurable via XenAPI.
@@ -20,34 +21,37 @@ extensions, we will also be able to support passthrough of integrated GPUs.
 Host Configuration
 ------------------
 
-A new host field will be added:
+A new PGPU field will be added:
 
-- `host.integrated_GPU_passthrough enum(disabled|enable_on_reboot|enabled|disable_on_reboot)`
+- `PGPU.dom0_access enum(enabled|disable_on_reboot|disabled|enable_on_reboot)`
 
 as well as new API calls:
 
-- `host.enable_integrated_GPU_passthrough`
-- `host.disable_integrated_GPU_passthrough`
+- `PGPU.enable_dom0_access`
+- `PGPU.disable_dom0_access`
 
-Enabling integrated GPU passthrough will modify the xen commandline (using the
-xen-cmdline tool) such that dom0 will not be able to access the integrated GPU
-on next boot.
+Disabling dom0 access will modify the xen commandline (using the xen-cmdline
+tool) such that dom0 will not be able to access the GPU on next boot.
 
-A state diagram for the field host.integrated_GPU_passthrough is shown below:
+A state diagram for the field PGPU.dom0_access is shown below:
 
 ![host.integrated_GPU_passthrough flow diagram](integrated-gpu-passthrough.png)
 
-Note that when a client enables or disables integrated GPU passthrough, the
-change can be cancelled until the host is rebooted.
+Note that when a client enables or disables dom0 access for a PGPU, the change
+can be cancelled until the host is rebooted.
 
 Handling vga_arbiter
 --------------------
 
-Normally, xapi will not create a PGPU object for the PCI device with address
+Currently, xapi will not create a PGPU object for the PCI device with address
 reported by `/dev/vga_arbiter`. This is to prevent a GPU in use by dom0 from
-from being passed through to a guest. This will no longer be the case if:
+from being passed through to a guest. This behaviour will be changed - instead
+of not creating a PGPU object at all, xapi will create a PGPU, but its
+supported_VGPU_types field will be empty.
 
-1.  Integrated GPU passthrough is enabled.
+However, the PGPU's supported_VGPU_types will be populated as normal if:
+
+1.  dom0 access to the GPU is disabled.
 2.  The vendor ID of the device is contained in a whitelist provided by xapi's
     config file.