375 lines · plain
1# SPDX-License-Identifier: GPL-2.0-only2menu "Xen driver support"3 depends on XEN4 5config XEN_BALLOON6 bool "Xen memory balloon driver"7 default y8 help9 The balloon driver allows the Xen domain to request more memory from10 the system to expand the domain's memory allocation, or alternatively11 return unneeded memory to the system.12 13config XEN_BALLOON_MEMORY_HOTPLUG14 bool "Memory hotplug support for Xen balloon driver"15 depends on XEN_BALLOON && MEMORY_HOTPLUG16 default y17 help18 Memory hotplug support for Xen balloon driver allows expanding memory19 available for the system above limit declared at system startup.20 It is very useful on critical systems which require long21 run without rebooting.22 23 It's also very useful for non PV domains to obtain unpopulated physical24 memory ranges to use in order to map foreign memory or grants.25 26 Memory could be hotplugged in following steps:27 28 1) target domain: ensure that memory auto online policy is in29 effect by checking /sys/devices/system/memory/auto_online_blocks30 file (should be 'online').31 32 2) control domain: xl mem-max <target-domain> <maxmem>33 where <maxmem> is >= requested memory size,34 35 3) control domain: xl mem-set <target-domain> <memory>36 where <memory> is requested memory size; alternatively memory37 could be added by writing proper value to38 /sys/devices/system/xen_memory/xen_memory0/target or39 /sys/devices/system/xen_memory/xen_memory0/target_kb on the40 target domain.41 42 Alternatively, if memory auto onlining was not requested at step 143 the newly added memory can be manually onlined in the target domain44 by doing the following:45 46 for i in /sys/devices/system/memory/memory*/state; do \47 [ "`cat "$i"`" = offline ] && echo online > "$i"; done48 49 or by adding the following line to udev rules:50 51 SUBSYSTEM=="memory", ACTION=="add", RUN+="/bin/sh -c '[ -f /sys$devpath/state ] && echo online > /sys$devpath/state'"52 53config XEN_MEMORY_HOTPLUG_LIMIT54 int "Hotplugged memory limit (in GiB) for a PV guest"55 default 51256 depends on XEN_HAVE_PVMMU57 depends on MEMORY_HOTPLUG58 help59 Maximum amount of memory (in GiB) that a PV guest can be60 expanded to when using memory hotplug.61 62 A PV guest can have more memory than this limit if is63 started with a larger maximum.64 65 This value is used to allocate enough space in internal66 tables needed for physical memory administration.67 68config XEN_SCRUB_PAGES_DEFAULT69 bool "Scrub pages before returning them to system by default"70 depends on XEN_BALLOON71 default y72 help73 Scrub pages before returning them to the system for reuse by74 other domains. This makes sure that any confidential data75 is not accidentally visible to other domains. It is more76 secure, but slightly less efficient. This can be controlled with77 xen_scrub_pages=0 parameter and78 /sys/devices/system/xen_memory/xen_memory0/scrub_pages.79 This option only sets the default value.80 81 If in doubt, say yes.82 83config XEN_DEV_EVTCHN84 tristate "Xen /dev/xen/evtchn device"85 default y86 help87 The evtchn driver allows a userspace process to trigger event88 channels and to receive notification of an event channel89 firing.90 If in doubt, say yes.91 92config XEN_BACKEND93 bool "Backend driver support"94 default XEN_DOM095 help96 Support for backend device drivers that provide I/O services97 to other virtual machines.98 99config XENFS100 tristate "Xen filesystem"101 select XEN_PRIVCMD102 default y103 help104 The xen filesystem provides a way for domains to share105 information with each other and with the hypervisor.106 For example, by reading and writing the "xenbus" file, guests107 may pass arbitrary information to the initial domain.108 If in doubt, say yes.109 110config XEN_COMPAT_XENFS111 bool "Create compatibility mount point /proc/xen"112 depends on XENFS113 default y114 help115 The old xenstore userspace tools expect to find "xenbus"116 under /proc/xen, but "xenbus" is now found at the root of the117 xenfs filesystem. Selecting this causes the kernel to create118 the compatibility mount point /proc/xen if it is running on119 a xen platform.120 If in doubt, say yes.121 122config XEN_SYS_HYPERVISOR123 bool "Create xen entries under /sys/hypervisor"124 depends on SYSFS125 select SYS_HYPERVISOR126 default y127 help128 Create entries under /sys/hypervisor describing the Xen129 hypervisor environment. When running native or in another130 virtual environment, /sys/hypervisor will still be present,131 but will have no xen contents.132 133config XEN_XENBUS_FRONTEND134 tristate135 136config XEN_GNTDEV137 tristate "userspace grant access device driver"138 depends on XEN139 default m140 select MMU_NOTIFIER141 help142 Allows userspace processes to use grants.143 144config XEN_GNTDEV_DMABUF145 bool "Add support for dma-buf grant access device driver extension"146 depends on XEN_GNTDEV && XEN_GRANT_DMA_ALLOC147 select DMA_SHARED_BUFFER148 help149 Allows userspace processes and kernel modules to use Xen backed150 dma-buf implementation. With this extension grant references to151 the pages of an imported dma-buf can be exported for other domain152 use and grant references coming from a foreign domain can be153 converted into a local dma-buf for local export.154 155config XEN_GRANT_DEV_ALLOC156 tristate "User-space grant reference allocator driver"157 depends on XEN158 default m159 help160 Allows userspace processes to create pages with access granted161 to other domains. This can be used to implement frontend drivers162 or as part of an inter-domain shared memory channel.163 164config XEN_GRANT_DMA_ALLOC165 bool "Allow allocating DMA capable buffers with grant reference module"166 depends on XEN && HAS_DMA167 help168 Extends grant table module API to allow allocating DMA capable169 buffers and mapping foreign grant references on top of it.170 The resulting buffer is similar to one allocated by the balloon171 driver in that proper memory reservation is made by172 ({increase|decrease}_reservation and VA mappings are updated if173 needed).174 This is useful for sharing foreign buffers with HW drivers which175 cannot work with scattered buffers provided by the balloon driver,176 but require DMAable memory instead.177 178config SWIOTLB_XEN179 def_bool y180 depends on ARCH_HAS_DMA_OPS181 depends on XEN_PV || ARM || ARM64182 select SWIOTLB183 184config XEN_PCI_STUB185 bool186 187config XEN_PCIDEV_STUB188 tristate "Xen PCI-device stub driver"189 depends on PCI && !X86 && XEN190 depends on XEN_BACKEND191 select XEN_PCI_STUB192 default m193 help194 The PCI device stub driver provides limited version of the PCI195 device backend driver without para-virtualized support for guests.196 If you select this to be a module, you will need to make sure no197 other driver has bound to the device(s) you want to make visible to198 other guests.199 200 The "hide" parameter (only applicable if backend driver is compiled201 into the kernel) allows you to bind the PCI devices to this module202 from the default device drivers. The argument is the list of PCI BDFs:203 xen-pciback.hide=(03:00.0)(04:00.0)204 205 If in doubt, say m.206 207config XEN_PCIDEV_BACKEND208 tristate "Xen PCI-device backend driver"209 depends on PCI && X86 && XEN210 depends on XEN_BACKEND211 select XEN_PCI_STUB212 default m213 help214 The PCI device backend driver allows the kernel to export arbitrary215 PCI devices to other guests. If you select this to be a module, you216 will need to make sure no other driver has bound to the device(s)217 you want to make visible to other guests.218 219 The parameter "passthrough" allows you specify how you want the PCI220 devices to appear in the guest. You can choose the default (0) where221 PCI topology starts at 00.00.0, or (1) for passthrough if you want222 the PCI devices topology appear the same as in the host.223 224 The "hide" parameter (only applicable if backend driver is compiled225 into the kernel) allows you to bind the PCI devices to this module226 from the default device drivers. The argument is the list of PCI BDFs:227 xen-pciback.hide=(03:00.0)(04:00.0)228 229 If in doubt, say m.230 231config XEN_PVCALLS_FRONTEND232 tristate "XEN PV Calls frontend driver"233 depends on INET && XEN234 select XEN_XENBUS_FRONTEND235 help236 Experimental frontend for the Xen PV Calls protocol237 (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It238 sends a small set of POSIX calls to the backend, which239 implements them.240 241config XEN_PVCALLS_BACKEND242 tristate "XEN PV Calls backend driver"243 depends on INET && XEN && XEN_BACKEND244 help245 Experimental backend for the Xen PV Calls protocol246 (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It247 allows PV Calls frontends to send POSIX calls to the backend,248 which implements them.249 250 If in doubt, say n.251 252config XEN_SCSI_BACKEND253 tristate "XEN SCSI backend driver"254 depends on XEN && XEN_BACKEND && TARGET_CORE255 help256 The SCSI backend driver allows the kernel to export its SCSI Devices257 to other guests via a high-performance shared-memory interface.258 Only needed for systems running as XEN driver domains (e.g. Dom0) and259 if guests need generic access to SCSI devices.260 261config XEN_PRIVCMD262 tristate "Xen hypercall passthrough driver"263 depends on XEN264 default m265 help266 The hypercall passthrough driver allows privileged user programs to267 perform Xen hypercalls. This driver is normally required for systems268 running as Dom0 to perform privileged operations, but in some269 disaggregated Xen setups this driver might be needed for other270 domains, too.271 272config XEN_PRIVCMD_EVENTFD273 bool "Xen Ioeventfd and irqfd support"274 depends on XEN_PRIVCMD && XEN_VIRTIO && EVENTFD275 help276 Using the ioeventfd / irqfd mechanism a virtio backend running in a277 daemon can speed up interrupt delivery from / to a guest.278 279config XEN_ACPI_PROCESSOR280 tristate "Xen ACPI processor"281 depends on XEN && XEN_PV_DOM0 && X86 && ACPI_PROCESSOR && CPU_FREQ282 default m283 help284 This ACPI processor uploads Power Management information to the Xen285 hypervisor.286 287 To do that the driver parses the Power Management data and uploads288 said information to the Xen hypervisor. Then the Xen hypervisor can289 select the proper Cx and Pxx states. It also registers itself as the290 SMM so that other drivers (such as ACPI cpufreq scaling driver) will291 not load.292 293 To compile this driver as a module, choose M here: the module will be294 called xen_acpi_processor If you do not know what to choose, select295 M here. If the CPUFREQ drivers are built in, select Y here.296 297config XEN_MCE_LOG298 bool "Xen platform mcelog"299 depends on XEN_PV_DOM0 && X86_MCE300 help301 Allow kernel fetching MCE error from Xen platform and302 converting it into Linux mcelog format for mcelog tools303 304config XEN_HAVE_PVMMU305 bool306 307config XEN_EFI308 def_bool y309 depends on (ARM || ARM64 || X86_64) && EFI310 311config XEN_AUTO_XLATE312 def_bool y313 depends on ARM || ARM64 || XEN_PVHVM314 help315 Support for auto-translated physmap guests.316 317config XEN_ACPI318 def_bool y319 depends on X86 && ACPI320 321config XEN_SYMS322 bool "Xen symbols"323 depends on X86 && XEN_DOM0 && XENFS324 default y if KALLSYMS325 help326 Exports hypervisor symbols (along with their types and addresses) via327 /proc/xen/xensyms file, similar to /proc/kallsyms328 329config XEN_HAVE_VPMU330 bool331 332config XEN_FRONT_PGDIR_SHBUF333 tristate334 335config XEN_UNPOPULATED_ALLOC336 bool "Use unpopulated memory ranges for guest mappings"337 depends on ZONE_DEVICE338 default XEN_BACKEND || XEN_GNTDEV || XEN_DOM0339 help340 Use unpopulated memory ranges in order to create mappings for guest341 memory regions, including grant maps and foreign pages. This avoids342 having to balloon out RAM regions in order to obtain physical memory343 space to create such mappings.344 345config XEN_GRANT_DMA_IOMMU346 bool347 select IOMMU_API348 349config XEN_GRANT_DMA_OPS350 bool351 352config XEN_VIRTIO353 bool "Xen virtio support"354 depends on ARCH_HAS_DMA_OPS355 depends on VIRTIO356 select XEN_GRANT_DMA_OPS357 select XEN_GRANT_DMA_IOMMU if OF358 help359 Enable virtio support for running as Xen guest. Depending on the360 guest type this will require special support on the backend side361 (qemu or kernel, depending on the virtio device types used).362 363 If in doubt, say n.364 365config XEN_VIRTIO_FORCE_GRANT366 bool "Require Xen virtio support to use grants"367 depends on XEN_VIRTIO368 help369 Require virtio for Xen guests to use grant mappings.370 This will avoid the need to give the backend the right to map all371 of the guest memory. This will need support on the backend side372 (e.g. qemu or kernel, depending on the virtio device types used).373 374endmenu375