X-Git-Url: https://git.proxmox.com/?a=blobdiff_plain;f=qm-pci-passthrough.adoc;h=28bb4fc39dff30b3b8b2ff4e6b1846b5a4e512cb;hb=a4c6084830b7f5e32fae9f080683e041f48834cc;hp=abb9075abd163ad59761834435a3b771e6f0e855;hpb=eebb350680c6367c08ac82e827d8098d7f9a73ba;p=pve-docs.git

diff --git a/qm-pci-passthrough.adoc b/qm-pci-passthrough.adoc
index abb9075..28bb4fc 100644
--- a/qm-pci-passthrough.adoc
+++ b/qm-pci-passthrough.adoc
@@ -45,8 +45,13 @@ some configuration to enable PCI(e) passthrough.
 
 .IOMMU
 
-The IOMMU has to be activated on the
+First, the IOMMU support has to be enabled in your BIOS/UEFI. Most often, that
+options is named `IOMMU` or `VT-d`, but check the manual for your motherboard
+for the exact option you need to enable.
+
+Then, the IOMMU might need to be activated on the
 xref:sysboot_edit_kernel_cmdline[kernel commandline].
+(On newer kernels, this should not be necessary.)
 
 The command line parameters are:
 
@@ -55,12 +60,25 @@ The command line parameters are:
 ----
  intel_iommu=on
 ----
-* for AMD CPUs:
-+
+* for AMD CPUs it should be enabled automatically.
+
+
+If your hardware supports it, enabling IOMMU passthrough mode might increase
+performance, because then the VMs bypass the (default) DMA translation
+which is normally done by the hypervisor, before handing DMA requests off to
+the hardware IOMMU. You can enable it with adding
+
+----
+ iommu.passthrough=1
+----
+
+or
+
 ----
- amd_iommu=on
+ iommu=pt
 ----
 
+to the kernel commandline.
 
 .Kernel Modules
 
@@ -206,7 +224,9 @@ configuration, for example by executing:
 ----
 
 If your device has multiple functions (e.g., ``00:02.0`' and ``00:02.1`' ),
-you can pass them through all together with the shortened syntax ``00:02`'
+you can pass them through all together with the shortened syntax ``00:02`'.
+This is equivalent with checking the ``All Functions`' checkbox in the
+web-interface.
 
 There are some options to which may be necessary, depending on the device
 and guest OS:
@@ -232,6 +252,24 @@ An example of PCIe passthrough with a GPU set to primary:
 # qm set VMID -hostpci0 02:00,pcie=on,x-vga=on
 ----
 
+.PCI ID overrides
+
+You can override the PCI vendor ID, device ID, and subsystem IDs that will be
+seen by the guest. This is useful if your device is a variant with an ID that
+your guest's drivers don't recognize, but you want to force those drivers to be
+loaded anyway (e.g. if you know your device shares the same chipset as a
+supported variant).
+
+The available options are `vendor-id`, `device-id`, `sub-vendor-id`, and
+`sub-device-id`. You can set any or all of these to override your device's
+default IDs.
+
+For example:
+
+----
+# qm set VMID -hostpci0 02:00,device-id=0x10f6,sub-vendor-id=0x0000
+----
+
 
 Other considerations
 ^^^^^^^^^^^^^^^^^^^^
@@ -309,7 +347,7 @@ Mediated Devices (vGPU, GVT-g)
 
 Mediated devices are another method to reuse features and performance from
 physical hardware for virtualized hardware. These are found most common in
-virtualized GPU setups such as Intels GVT-g and Nvidias vGPUs used in their
+virtualized GPU setups such as Intel's GVT-g and NVIDIA's vGPUs used in their
 GRID technology.
 
 With this, a physical Card is able to create virtual cards, similar to SR-IOV.
@@ -324,7 +362,7 @@ In general your card's driver must support that feature, otherwise it will
 not work. So please refer to your vendor for compatible drivers and how to
 configure them.
 
-Intels drivers for GVT-g are integrated in the Kernel and should work
+Intel's drivers for GVT-g are integrated in the Kernel and should work
 with 5th, 6th and 7th generation Intel Core Processors, as well as E3 v4, E3
 v5 and E3 v6 Xeon Processors.