Kconfig 11 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338
  1. # SPDX-License-Identifier: GPL-2.0-only
  2. menu "Xen driver support"
  3. depends on XEN
  4. config XEN_BALLOON
  5. bool "Xen memory balloon driver"
  6. default y
  7. help
  8. The balloon driver allows the Xen domain to request more memory from
  9. the system to expand the domain's memory allocation, or alternatively
  10. return unneeded memory to the system.
  11. config XEN_BALLOON_MEMORY_HOTPLUG
  12. bool "Memory hotplug support for Xen balloon driver"
  13. depends on XEN_BALLOON && MEMORY_HOTPLUG
  14. default y
  15. help
  16. Memory hotplug support for Xen balloon driver allows expanding memory
  17. available for the system above limit declared at system startup.
  18. It is very useful on critical systems which require long
  19. run without rebooting.
  20. It's also very useful for non PV domains to obtain unpopulated physical
  21. memory ranges to use in order to map foreign memory or grants.
  22. Memory could be hotplugged in following steps:
  23. 1) target domain: ensure that memory auto online policy is in
  24. effect by checking /sys/devices/system/memory/auto_online_blocks
  25. file (should be 'online').
  26. 2) control domain: xl mem-max <target-domain> <maxmem>
  27. where <maxmem> is >= requested memory size,
  28. 3) control domain: xl mem-set <target-domain> <memory>
  29. where <memory> is requested memory size; alternatively memory
  30. could be added by writing proper value to
  31. /sys/devices/system/xen_memory/xen_memory0/target or
  32. /sys/devices/system/xen_memory/xen_memory0/target_kb on the
  33. target domain.
  34. Alternatively, if memory auto onlining was not requested at step 1
  35. the newly added memory can be manually onlined in the target domain
  36. by doing the following:
  37. for i in /sys/devices/system/memory/memory*/state; do \
  38. [ "`cat "$i"`" = offline ] && echo online > "$i"; done
  39. or by adding the following line to udev rules:
  40. SUBSYSTEM=="memory", ACTION=="add", RUN+="/bin/sh -c '[ -f /sys$devpath/state ] && echo online > /sys$devpath/state'"
  41. config XEN_MEMORY_HOTPLUG_LIMIT
  42. int "Hotplugged memory limit (in GiB) for a PV guest"
  43. default 512
  44. depends on XEN_HAVE_PVMMU
  45. depends on MEMORY_HOTPLUG
  46. help
  47. Maxmium amount of memory (in GiB) that a PV guest can be
  48. expanded to when using memory hotplug.
  49. A PV guest can have more memory than this limit if is
  50. started with a larger maximum.
  51. This value is used to allocate enough space in internal
  52. tables needed for physical memory administration.
  53. config XEN_SCRUB_PAGES_DEFAULT
  54. bool "Scrub pages before returning them to system by default"
  55. depends on XEN_BALLOON
  56. default y
  57. help
  58. Scrub pages before returning them to the system for reuse by
  59. other domains. This makes sure that any confidential data
  60. is not accidentally visible to other domains. It is more
  61. secure, but slightly less efficient. This can be controlled with
  62. xen_scrub_pages=0 parameter and
  63. /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
  64. This option only sets the default value.
  65. If in doubt, say yes.
  66. config XEN_DEV_EVTCHN
  67. tristate "Xen /dev/xen/evtchn device"
  68. default y
  69. help
  70. The evtchn driver allows a userspace process to trigger event
  71. channels and to receive notification of an event channel
  72. firing.
  73. If in doubt, say yes.
  74. config XEN_BACKEND
  75. bool "Backend driver support"
  76. default XEN_DOM0
  77. help
  78. Support for backend device drivers that provide I/O services
  79. to other virtual machines.
  80. config XENFS
  81. tristate "Xen filesystem"
  82. select XEN_PRIVCMD
  83. default y
  84. help
  85. The xen filesystem provides a way for domains to share
  86. information with each other and with the hypervisor.
  87. For example, by reading and writing the "xenbus" file, guests
  88. may pass arbitrary information to the initial domain.
  89. If in doubt, say yes.
  90. config XEN_COMPAT_XENFS
  91. bool "Create compatibility mount point /proc/xen"
  92. depends on XENFS
  93. default y
  94. help
  95. The old xenstore userspace tools expect to find "xenbus"
  96. under /proc/xen, but "xenbus" is now found at the root of the
  97. xenfs filesystem. Selecting this causes the kernel to create
  98. the compatibility mount point /proc/xen if it is running on
  99. a xen platform.
  100. If in doubt, say yes.
  101. config XEN_SYS_HYPERVISOR
  102. bool "Create xen entries under /sys/hypervisor"
  103. depends on SYSFS
  104. select SYS_HYPERVISOR
  105. default y
  106. help
  107. Create entries under /sys/hypervisor describing the Xen
  108. hypervisor environment. When running native or in another
  109. virtual environment, /sys/hypervisor will still be present,
  110. but will have no xen contents.
  111. config XEN_XENBUS_FRONTEND
  112. tristate
  113. config XEN_GNTDEV
  114. tristate "userspace grant access device driver"
  115. depends on XEN
  116. default m
  117. select MMU_NOTIFIER
  118. help
  119. Allows userspace processes to use grants.
  120. config XEN_GNTDEV_DMABUF
  121. bool "Add support for dma-buf grant access device driver extension"
  122. depends on XEN_GNTDEV && XEN_GRANT_DMA_ALLOC
  123. select DMA_SHARED_BUFFER
  124. help
  125. Allows userspace processes and kernel modules to use Xen backed
  126. dma-buf implementation. With this extension grant references to
  127. the pages of an imported dma-buf can be exported for other domain
  128. use and grant references coming from a foreign domain can be
  129. converted into a local dma-buf for local export.
  130. config XEN_GRANT_DEV_ALLOC
  131. tristate "User-space grant reference allocator driver"
  132. depends on XEN
  133. default m
  134. help
  135. Allows userspace processes to create pages with access granted
  136. to other domains. This can be used to implement frontend drivers
  137. or as part of an inter-domain shared memory channel.
  138. config XEN_GRANT_DMA_ALLOC
  139. bool "Allow allocating DMA capable buffers with grant reference module"
  140. depends on XEN && HAS_DMA
  141. help
  142. Extends grant table module API to allow allocating DMA capable
  143. buffers and mapping foreign grant references on top of it.
  144. The resulting buffer is similar to one allocated by the balloon
  145. driver in that proper memory reservation is made by
  146. ({increase|decrease}_reservation and VA mappings are updated if
  147. needed).
  148. This is useful for sharing foreign buffers with HW drivers which
  149. cannot work with scattered buffers provided by the balloon driver,
  150. but require DMAable memory instead.
  151. config SWIOTLB_XEN
  152. def_bool y
  153. select DMA_OPS
  154. select SWIOTLB
  155. config XEN_PCIDEV_BACKEND
  156. tristate "Xen PCI-device backend driver"
  157. depends on PCI && X86 && XEN
  158. depends on XEN_BACKEND
  159. default m
  160. help
  161. The PCI device backend driver allows the kernel to export arbitrary
  162. PCI devices to other guests. If you select this to be a module, you
  163. will need to make sure no other driver has bound to the device(s)
  164. you want to make visible to other guests.
  165. The parameter "passthrough" allows you specify how you want the PCI
  166. devices to appear in the guest. You can choose the default (0) where
  167. PCI topology starts at 00.00.0, or (1) for passthrough if you want
  168. the PCI devices topology appear the same as in the host.
  169. The "hide" parameter (only applicable if backend driver is compiled
  170. into the kernel) allows you to bind the PCI devices to this module
  171. from the default device drivers. The argument is the list of PCI BDFs:
  172. xen-pciback.hide=(03:00.0)(04:00.0)
  173. If in doubt, say m.
  174. config XEN_PVCALLS_FRONTEND
  175. tristate "XEN PV Calls frontend driver"
  176. depends on INET && XEN
  177. select XEN_XENBUS_FRONTEND
  178. help
  179. Experimental frontend for the Xen PV Calls protocol
  180. (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It
  181. sends a small set of POSIX calls to the backend, which
  182. implements them.
  183. config XEN_PVCALLS_BACKEND
  184. bool "XEN PV Calls backend driver"
  185. depends on INET && XEN && XEN_BACKEND
  186. help
  187. Experimental backend for the Xen PV Calls protocol
  188. (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It
  189. allows PV Calls frontends to send POSIX calls to the backend,
  190. which implements them.
  191. If in doubt, say n.
  192. config XEN_SCSI_BACKEND
  193. tristate "XEN SCSI backend driver"
  194. depends on XEN && XEN_BACKEND && TARGET_CORE
  195. help
  196. The SCSI backend driver allows the kernel to export its SCSI Devices
  197. to other guests via a high-performance shared-memory interface.
  198. Only needed for systems running as XEN driver domains (e.g. Dom0) and
  199. if guests need generic access to SCSI devices.
  200. config XEN_PRIVCMD
  201. tristate
  202. depends on XEN
  203. default m
  204. config XEN_STUB
  205. bool "Xen stub drivers"
  206. depends on XEN && X86_64 && BROKEN
  207. help
  208. Allow kernel to install stub drivers, to reserve space for Xen drivers,
  209. i.e. memory hotplug and cpu hotplug, and to block native drivers loaded,
  210. so that real Xen drivers can be modular.
  211. To enable Xen features like cpu and memory hotplug, select Y here.
  212. config XEN_ACPI_HOTPLUG_MEMORY
  213. tristate "Xen ACPI memory hotplug"
  214. depends on XEN_DOM0 && XEN_STUB && ACPI
  215. help
  216. This is Xen ACPI memory hotplug.
  217. Currently Xen only support ACPI memory hot-add. If you want
  218. to hot-add memory at runtime (the hot-added memory cannot be
  219. removed until machine stop), select Y/M here, otherwise select N.
  220. config XEN_ACPI_HOTPLUG_CPU
  221. tristate "Xen ACPI cpu hotplug"
  222. depends on XEN_DOM0 && XEN_STUB && ACPI
  223. select ACPI_CONTAINER
  224. help
  225. Xen ACPI cpu enumerating and hotplugging
  226. For hotplugging, currently Xen only support ACPI cpu hotadd.
  227. If you want to hotadd cpu at runtime (the hotadded cpu cannot
  228. be removed until machine stop), select Y/M here.
  229. config XEN_ACPI_PROCESSOR
  230. tristate "Xen ACPI processor"
  231. depends on XEN && XEN_DOM0 && X86 && ACPI_PROCESSOR && CPU_FREQ
  232. default m
  233. help
  234. This ACPI processor uploads Power Management information to the Xen
  235. hypervisor.
  236. To do that the driver parses the Power Management data and uploads
  237. said information to the Xen hypervisor. Then the Xen hypervisor can
  238. select the proper Cx and Pxx states. It also registers itself as the
  239. SMM so that other drivers (such as ACPI cpufreq scaling driver) will
  240. not load.
  241. To compile this driver as a module, choose M here: the module will be
  242. called xen_acpi_processor If you do not know what to choose, select
  243. M here. If the CPUFREQ drivers are built in, select Y here.
  244. config XEN_MCE_LOG
  245. bool "Xen platform mcelog"
  246. depends on XEN_DOM0 && X86_MCE
  247. help
  248. Allow kernel fetching MCE error from Xen platform and
  249. converting it into Linux mcelog format for mcelog tools
  250. config XEN_HAVE_PVMMU
  251. bool
  252. config XEN_EFI
  253. def_bool y
  254. depends on (ARM || ARM64 || X86_64) && EFI
  255. config XEN_AUTO_XLATE
  256. def_bool y
  257. depends on ARM || ARM64 || XEN_PVHVM
  258. help
  259. Support for auto-translated physmap guests.
  260. config XEN_ACPI
  261. def_bool y
  262. depends on X86 && ACPI
  263. config XEN_SYMS
  264. bool "Xen symbols"
  265. depends on X86 && XEN_DOM0 && XENFS
  266. default y if KALLSYMS
  267. help
  268. Exports hypervisor symbols (along with their types and addresses) via
  269. /proc/xen/xensyms file, similar to /proc/kallsyms
  270. config XEN_HAVE_VPMU
  271. bool
  272. config XEN_FRONT_PGDIR_SHBUF
  273. tristate
  274. config XEN_UNPOPULATED_ALLOC
  275. bool "Use unpopulated memory ranges for guest mappings"
  276. depends on X86 && ZONE_DEVICE
  277. default XEN_BACKEND || XEN_GNTDEV || XEN_DOM0
  278. help
  279. Use unpopulated memory ranges in order to create mappings for guest
  280. memory regions, including grant maps and foreign pages. This avoids
  281. having to balloon out RAM regions in order to obtain physical memory
  282. space to create such mappings.
  283. endmenu