<driver> as child element of <disk>

Edit online

Specifies details that are related to the user space process used to implement the block device.

Text content

None.

Selected attributes

name=qemu: Name of the user space process. Use qemu.
type=raw | qcow2: Use subtype raw, except for qcow2 image files, which require the qcow2 subtype.
iothread=<IOthread-ID>: Assigns a certain I/O thread to the user space process. Use this attribute to ensure best performance.
<IOthread-ID> is a value between 1 and the number of I/O threads which is specified by the iothreads element.
cache=none: Optional; controls the cache mechanism.
error_policy=report | stop | ignore | enospace: Optional; the error_policy attribute controls how the host will behave if a disk read or write error occurs.
rerror_policy=report | stop | ignore: Optional; controls the behavior for read errors only. If no rerror_policy is given, error_policy is used for both read and write errors. If rerror_policy is given, it overrides the error_policy for read errors. Also, note that enospace is not a valid policy for read errors. Therefore, if error_policy is set to enospace and no rerror_policy is given, the read error policy is left at its default (report).
io=threads | native: Optional; controls specific policies on I/O. For a better performance, specify native.
ioeventfd=on | off: Optional; allows users to set domain I/O asynchronous handling for the disk device. The default is left to the discretion of the host. Enabling this attribute allows QEMU to run the virtual server while a separate thread handles I/O. Typically virtual servers experiencing high system CPU utilization during I/O will benefit from this. On the other hand, on overloaded host it could increase virtual server I/O latency. Note: Only very experienced users should attempt to use this option!
event_idx=on | off: Optional; controls some aspects of device event processing. If it is on, it will reduce the number of interrupts and exits for the virtual server. The default is determined by QEMU; usually if the feature is supported, the default is on. If the situation occurs where this behavior is suboptimal, this attribute provides a way to force the feature off. Note: Only experienced users should attempt to use this option!
iommu="on": For virtio devices on guests that are to run in IBM® Secure Execution mode: with this attribute, you enable the device to use the guest's bounce buffer. Use this attribute only if the launchSecurity element at the virtual server level is not an option.
With the preferred launchSecurity element, you enable all virtio devices to use the bounce buffer by default, see Preparing the virtual server.

Usage

Parent elements

<disk>

Child elements

None.

Example

<disk type="block" device="disk">
    <driver name="qemu" type="raw" cache="none" io="native" iothread="1"/>
    <source dev="/dev/mapper/36005076305ffc1ae00000000000021d5"/>
    <target dev="vdb" bus="virtio"/>
    <address type="ccw" cssid="0xfe" ssid="0x0" devno="0xd501"/>
</disk>