1 # SPDX-License-Identifier: GPL-2.0-only
5 This option is selected by any driver which implements the virtio
6 bus, such as CONFIG_VIRTIO_PCI, CONFIG_VIRTIO_MMIO, CONFIG_RPMSG
12 Modern PCI device implementation. This module implements the
13 basic probe and control for devices which are based on modern
14 PCI device with possible vendor specific extensions. Any
15 module that selects this module must depend on PCI.
17 config VIRTIO_PCI_LIB_LEGACY
20 Legacy PCI device (Virtio PCI Card 0.9.x Draft and older device)
22 This module implements the basic probe and control for devices
23 which are based on legacy PCI device. Any module that selects this
24 module must depend on PCI.
26 menuconfig VIRTIO_MENU
32 config VIRTIO_HARDEN_NOTIFICATION
33 bool "Harden virtio notification"
35 Enable this to harden the device notifications and suppress
36 those that happen at a time where notifications are illegal.
38 Experimental: Note that several drivers still have bugs that
39 may cause crashes or hangs when correct handling of
40 notifications is enforced; depending on the subset of
41 drivers and devices you use, this may or may not work.
46 tristate "PCI driver for virtio devices"
51 This driver provides support for virtio based paravirtual device
52 drivers over PCI. This requires that your VMM has appropriate PCI
53 virtio backends. Most QEMU based VMMs should support these devices
58 config VIRTIO_PCI_LEGACY
59 bool "Support for legacy virtio draft 0.9.X and older devices"
62 select VIRTIO_PCI_LIB_LEGACY
64 Virtio PCI Card 0.9.X Draft (circa 2014) and older device support.
66 This option enables building a transitional driver, supporting
67 both devices conforming to Virtio 1 specification, and legacy devices.
68 If disabled, you get a slightly smaller, non-transitional driver,
69 with no legacy compatibility.
71 So look out into your driveway. Do you have a flying car? If
72 so, you can happily disable this option and virtio will not
73 break. Otherwise, leave it set. Unless you're testing what
74 life will be like in The Future.
79 tristate "vDPA driver for virtio devices"
83 This driver provides support for virtio based paravirtual
84 device driver over vDPA bus. For this to be useful, you need
85 an appropriate vDPA device implementation that operates on a
86 physical device to allow the datapath of virtio to be
87 offloaded to hardware.
92 tristate "Support for virtio pmem driver"
96 This driver provides access to virtio-pmem devices, storage devices
97 that are mapped into the physical address space - similar to NVDIMMs
98 - with a virtio-based flushing interface.
102 config VIRTIO_BALLOON
103 tristate "Virtio balloon driver"
105 select MEMORY_BALLOON
106 select PAGE_REPORTING
108 This driver supports increasing and decreasing the amount
109 of memory within a KVM guest.
114 tristate "Virtio mem driver"
115 depends on X86_64 || ARM64
117 depends on MEMORY_HOTPLUG
118 depends on MEMORY_HOTREMOVE
119 depends on CONTIG_ALLOC
120 depends on EXCLUSIVE_SYSTEM_RAM
122 This driver provides access to virtio-mem paravirtualized memory
123 devices, allowing to hotplug and hotunplug memory.
125 This driver was only tested under x86-64 and arm64, but should
126 theoretically work on all architectures that support memory hotplug
132 tristate "Virtio input driver"
136 This driver supports virtio input devices such as
137 keyboards, mice and tablets.
142 tristate "Platform bus driver for memory mapped virtio devices"
143 depends on HAS_IOMEM && HAS_DMA
146 This drivers provides support for memory mapped virtio
147 platform device driver.
151 config VIRTIO_MMIO_CMDLINE_DEVICES
152 bool "Memory mapped virtio devices parameter parsing"
153 depends on VIRTIO_MMIO
155 Allow virtio-mmio devices instantiation via the kernel command line
156 or module parameters. Be aware that using incorrect parameters (base
157 address in particular) can crash your system - you have been warned.
158 See Documentation/admin-guide/kernel-parameters.rst for details.
162 config VIRTIO_DMA_SHARED_BUFFER
164 depends on DMA_SHARED_BUFFER
166 This option adds a flavor of dma buffers that are backed by