1# SPDX-License-Identifier: GPL-2.0-only 2menu "Xen driver support" 3 depends on XEN 4 5config XEN_BALLOON 6 bool "Xen memory balloon driver" 7 default y 8 help 9 The balloon driver allows the Xen domain to request more memory from 10 the system to expand the domain's memory allocation, or alternatively 11 return unneeded memory to the system. 12 13config XEN_BALLOON_MEMORY_HOTPLUG 14 bool "Memory hotplug support for Xen balloon driver" 15 depends on XEN_BALLOON && MEMORY_HOTPLUG 16 help 17 Memory hotplug support for Xen balloon driver allows expanding memory 18 available for the system above limit declared at system startup. 19 It is very useful on critical systems which require long 20 run without rebooting. 21 22 Memory could be hotplugged in following steps: 23 24 1) target domain: ensure that memory auto online policy is in 25 effect by checking /sys/devices/system/memory/auto_online_blocks 26 file (should be 'online'). 27 28 2) control domain: xl mem-max <target-domain> <maxmem> 29 where <maxmem> is >= requested memory size, 30 31 3) control domain: xl mem-set <target-domain> <memory> 32 where <memory> is requested memory size; alternatively memory 33 could be added by writing proper value to 34 /sys/devices/system/xen_memory/xen_memory0/target or 35 /sys/devices/system/xen_memory/xen_memory0/target_kb on the 36 target domain. 37 38 Alternatively, if memory auto onlining was not requested at step 1 39 the newly added memory can be manually onlined in the target domain 40 by doing the following: 41 42 for i in /sys/devices/system/memory/memory*/state; do \ 43 [ "`cat "$i"`" = offline ] && echo online > "$i"; done 44 45 or by adding the following line to udev rules: 46 47 SUBSYSTEM=="memory", ACTION=="add", RUN+="/bin/sh -c '[ -f /sys$devpath/state ] && echo online > /sys$devpath/state'" 48 49config XEN_BALLOON_MEMORY_HOTPLUG_LIMIT 50 int "Hotplugged memory limit (in GiB) for a PV guest" 51 default 512 if X86_64 52 default 4 if X86_32 53 range 0 64 if X86_32 54 depends on XEN_HAVE_PVMMU 55 depends on XEN_BALLOON_MEMORY_HOTPLUG 56 help 57 Maxmium amount of memory (in GiB) that a PV guest can be 58 expanded to when using memory hotplug. 59 60 A PV guest can have more memory than this limit if is 61 started with a larger maximum. 62 63 This value is used to allocate enough space in internal 64 tables needed for physical memory administration. 65 66config XEN_SCRUB_PAGES_DEFAULT 67 bool "Scrub pages before returning them to system by default" 68 depends on XEN_BALLOON 69 default y 70 help 71 Scrub pages before returning them to the system for reuse by 72 other domains. This makes sure that any confidential data 73 is not accidentally visible to other domains. It is more 74 secure, but slightly less efficient. This can be controlled with 75 xen_scrub_pages=0 parameter and 76 /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 77 This option only sets the default value. 78 79 If in doubt, say yes. 80 81config XEN_DEV_EVTCHN 82 tristate "Xen /dev/xen/evtchn device" 83 default y 84 help 85 The evtchn driver allows a userspace process to trigger event 86 channels and to receive notification of an event channel 87 firing. 88 If in doubt, say yes. 89 90config XEN_BACKEND 91 bool "Backend driver support" 92 default XEN_DOM0 93 help 94 Support for backend device drivers that provide I/O services 95 to other virtual machines. 96 97config XENFS 98 tristate "Xen filesystem" 99 select XEN_PRIVCMD 100 default y 101 help 102 The xen filesystem provides a way for domains to share 103 information with each other and with the hypervisor. 104 For example, by reading and writing the "xenbus" file, guests 105 may pass arbitrary information to the initial domain. 106 If in doubt, say yes. 107 108config XEN_COMPAT_XENFS 109 bool "Create compatibility mount point /proc/xen" 110 depends on XENFS 111 default y 112 help 113 The old xenstore userspace tools expect to find "xenbus" 114 under /proc/xen, but "xenbus" is now found at the root of the 115 xenfs filesystem. Selecting this causes the kernel to create 116 the compatibility mount point /proc/xen if it is running on 117 a xen platform. 118 If in doubt, say yes. 119 120config XEN_SYS_HYPERVISOR 121 bool "Create xen entries under /sys/hypervisor" 122 depends on SYSFS 123 select SYS_HYPERVISOR 124 default y 125 help 126 Create entries under /sys/hypervisor describing the Xen 127 hypervisor environment. When running native or in another 128 virtual environment, /sys/hypervisor will still be present, 129 but will have no xen contents. 130 131config XEN_XENBUS_FRONTEND 132 tristate 133 134config XEN_GNTDEV 135 tristate "userspace grant access device driver" 136 depends on XEN 137 default m 138 select MMU_NOTIFIER 139 help 140 Allows userspace processes to use grants. 141 142config XEN_GNTDEV_DMABUF 143 bool "Add support for dma-buf grant access device driver extension" 144 depends on XEN_GNTDEV && XEN_GRANT_DMA_ALLOC && DMA_SHARED_BUFFER 145 help 146 Allows userspace processes and kernel modules to use Xen backed 147 dma-buf implementation. With this extension grant references to 148 the pages of an imported dma-buf can be exported for other domain 149 use and grant references coming from a foreign domain can be 150 converted into a local dma-buf for local export. 151 152config XEN_GRANT_DEV_ALLOC 153 tristate "User-space grant reference allocator driver" 154 depends on XEN 155 default m 156 help 157 Allows userspace processes to create pages with access granted 158 to other domains. This can be used to implement frontend drivers 159 or as part of an inter-domain shared memory channel. 160 161config XEN_GRANT_DMA_ALLOC 162 bool "Allow allocating DMA capable buffers with grant reference module" 163 depends on XEN && HAS_DMA 164 help 165 Extends grant table module API to allow allocating DMA capable 166 buffers and mapping foreign grant references on top of it. 167 The resulting buffer is similar to one allocated by the balloon 168 driver in that proper memory reservation is made by 169 ({increase|decrease}_reservation and VA mappings are updated if 170 needed). 171 This is useful for sharing foreign buffers with HW drivers which 172 cannot work with scattered buffers provided by the balloon driver, 173 but require DMAable memory instead. 174 175config SWIOTLB_XEN 176 def_bool y 177 select SWIOTLB 178 179config XEN_PCIDEV_BACKEND 180 tristate "Xen PCI-device backend driver" 181 depends on PCI && X86 && XEN 182 depends on XEN_BACKEND 183 default m 184 help 185 The PCI device backend driver allows the kernel to export arbitrary 186 PCI devices to other guests. If you select this to be a module, you 187 will need to make sure no other driver has bound to the device(s) 188 you want to make visible to other guests. 189 190 The parameter "passthrough" allows you specify how you want the PCI 191 devices to appear in the guest. You can choose the default (0) where 192 PCI topology starts at 00.00.0, or (1) for passthrough if you want 193 the PCI devices topology appear the same as in the host. 194 195 The "hide" parameter (only applicable if backend driver is compiled 196 into the kernel) allows you to bind the PCI devices to this module 197 from the default device drivers. The argument is the list of PCI BDFs: 198 xen-pciback.hide=(03:00.0)(04:00.0) 199 200 If in doubt, say m. 201 202config XEN_PVCALLS_FRONTEND 203 tristate "XEN PV Calls frontend driver" 204 depends on INET && XEN 205 select XEN_XENBUS_FRONTEND 206 help 207 Experimental frontend for the Xen PV Calls protocol 208 (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It 209 sends a small set of POSIX calls to the backend, which 210 implements them. 211 212config XEN_PVCALLS_BACKEND 213 bool "XEN PV Calls backend driver" 214 depends on INET && XEN && XEN_BACKEND 215 help 216 Experimental backend for the Xen PV Calls protocol 217 (https://xenbits.xen.org/docs/unstable/misc/pvcalls.html). It 218 allows PV Calls frontends to send POSIX calls to the backend, 219 which implements them. 220 221 If in doubt, say n. 222 223config XEN_SCSI_BACKEND 224 tristate "XEN SCSI backend driver" 225 depends on XEN && XEN_BACKEND && TARGET_CORE 226 help 227 The SCSI backend driver allows the kernel to export its SCSI Devices 228 to other guests via a high-performance shared-memory interface. 229 Only needed for systems running as XEN driver domains (e.g. Dom0) and 230 if guests need generic access to SCSI devices. 231 232config XEN_PRIVCMD 233 tristate 234 depends on XEN 235 default m 236 237config XEN_STUB 238 bool "Xen stub drivers" 239 depends on XEN && X86_64 && BROKEN 240 help 241 Allow kernel to install stub drivers, to reserve space for Xen drivers, 242 i.e. memory hotplug and cpu hotplug, and to block native drivers loaded, 243 so that real Xen drivers can be modular. 244 245 To enable Xen features like cpu and memory hotplug, select Y here. 246 247config XEN_ACPI_HOTPLUG_MEMORY 248 tristate "Xen ACPI memory hotplug" 249 depends on XEN_DOM0 && XEN_STUB && ACPI 250 help 251 This is Xen ACPI memory hotplug. 252 253 Currently Xen only support ACPI memory hot-add. If you want 254 to hot-add memory at runtime (the hot-added memory cannot be 255 removed until machine stop), select Y/M here, otherwise select N. 256 257config XEN_ACPI_HOTPLUG_CPU 258 tristate "Xen ACPI cpu hotplug" 259 depends on XEN_DOM0 && XEN_STUB && ACPI 260 select ACPI_CONTAINER 261 help 262 Xen ACPI cpu enumerating and hotplugging 263 264 For hotplugging, currently Xen only support ACPI cpu hotadd. 265 If you want to hotadd cpu at runtime (the hotadded cpu cannot 266 be removed until machine stop), select Y/M here. 267 268config XEN_ACPI_PROCESSOR 269 tristate "Xen ACPI processor" 270 depends on XEN && XEN_DOM0 && X86 && ACPI_PROCESSOR && CPU_FREQ 271 default m 272 help 273 This ACPI processor uploads Power Management information to the Xen 274 hypervisor. 275 276 To do that the driver parses the Power Management data and uploads 277 said information to the Xen hypervisor. Then the Xen hypervisor can 278 select the proper Cx and Pxx states. It also registers itself as the 279 SMM so that other drivers (such as ACPI cpufreq scaling driver) will 280 not load. 281 282 To compile this driver as a module, choose M here: the module will be 283 called xen_acpi_processor If you do not know what to choose, select 284 M here. If the CPUFREQ drivers are built in, select Y here. 285 286config XEN_MCE_LOG 287 bool "Xen platform mcelog" 288 depends on XEN_DOM0 && X86_64 && X86_MCE 289 help 290 Allow kernel fetching MCE error from Xen platform and 291 converting it into Linux mcelog format for mcelog tools 292 293config XEN_HAVE_PVMMU 294 bool 295 296config XEN_EFI 297 def_bool y 298 depends on (ARM || ARM64 || X86_64) && EFI 299 300config XEN_AUTO_XLATE 301 def_bool y 302 depends on ARM || ARM64 || XEN_PVHVM 303 help 304 Support for auto-translated physmap guests. 305 306config XEN_ACPI 307 def_bool y 308 depends on X86 && ACPI 309 310config XEN_SYMS 311 bool "Xen symbols" 312 depends on X86 && XEN_DOM0 && XENFS 313 default y if KALLSYMS 314 help 315 Exports hypervisor symbols (along with their types and addresses) via 316 /proc/xen/xensyms file, similar to /proc/kallsyms 317 318config XEN_HAVE_VPMU 319 bool 320 321config XEN_FRONT_PGDIR_SHBUF 322 tristate 323 324endmenu 325