1% Support statement for this release 2 3This document describes the support status 4and in particular the security support status of the Xen branch 5within which you find it. 6 7See the bottom of the file 8for the definitions of the support status levels etc. 9 10# Release Support 11 12 Xen-Version: 4.14 13 Initial-Release: 2020-07-24 14 Supported-Until: 2022-01-24 15 Security-Support-Until: 2023-07-24 16 17Release Notes 18: <a href="https://wiki.xenproject.org/wiki/Xen_Project_4.14_Release_Notes">RN</a> 19 20# Feature Support 21 22## Kconfig 23 24EXPERT and DEBUG Kconfig options are not security supported. Other 25Kconfig options are supported, if the related features are marked as 26supported in this document. 27 28## Host Architecture 29 30### x86-64 31 32 Status: Supported 33 34### ARM v7 + Virtualization Extensions 35 36 Status: Supported 37 38### ARM v8 39 40 Status: Supported 41 42## Host hardware support 43 44### Physical CPU Hotplug 45 46 Status, x86: Supported 47 48### Physical Memory Hotplug 49 50 Status, x86: Supported 51 52### Host ACPI (via Domain 0) 53 54 Status, x86 PV: Supported 55 Status, ARM: Experimental 56 57### x86/Intel Platform QoS Technologies 58 59 Status: Tech Preview 60 61### IOMMU 62 63 Status, AMD IOMMU: Supported 64 Status, Intel VT-d: Supported 65 Status, ARM SMMUv1: Supported 66 Status, ARM SMMUv2: Supported 67 Status, Renesas IPMMU-VMSA: Tech Preview 68 69### ARM/GICv3 ITS 70 71Extension to the GICv3 interrupt controller to support MSI. 72 73 Status: Experimental 74 75## Guest Type 76 77### x86/PV 78 79Traditional Xen PV guest 80 81No hardware requirements 82 83 Status: Supported 84 85### x86/HVM 86 87Fully virtualised guest using hardware virtualisation extensions 88 89Requires hardware virtualisation support (Intel VMX / AMD SVM) 90 91 Status, domU: Supported 92 93### x86/PVH 94 95PVH is a next-generation paravirtualized mode 96designed to take advantage of hardware virtualization support when possible. 97During development this was sometimes called HVMLite or PVHv2. 98 99Requires hardware virtualisation support (Intel VMX / AMD SVM). 100 101Dom0 support requires an IOMMU (Intel VT-d / AMD IOMMU). 102 103 Status, domU: Supported 104 Status, dom0: Experimental 105 106### ARM 107 108ARM only has one guest type at the moment 109 110 Status: Supported 111 112## Hypervisor file system 113 114### Build info 115 116 Status: Supported 117 118### Hypervisor config 119 120 Status: Supported 121 122### Runtime parameters 123 124 Status: Supported 125 126## Toolstack 127 128### xl 129 130 Status: Supported 131 132### Direct-boot kernel image format 133 134Format which the toolstack accepts for direct-boot kernels 135 136 Supported, x86: bzImage, ELF 137 Supported, ARM32: zImage 138 Supported, ARM64: Image 139 140### Dom0 init support for xl 141 142 Status, SysV: Supported 143 Status, systemd: Supported 144 Status, BSD-style: Supported 145 146### JSON output support for xl 147 148Output of information in machine-parseable JSON format 149 150 Status: Experimental 151 152### Open vSwitch integration for xl 153 154 Status, Linux: Supported 155 156### Virtual cpu hotplug 157 158 Status: Supported 159 160### QEMU backend hotplugging for xl 161 162 Status: Supported 163 164### xenlight Go package 165 166Go (golang) bindings for libxl 167 168 Status: Experimental 169 170### Linux device model stubdomains 171 172Support for running qemu-xen device model in a linux stubdomain. 173 174 Status: Tech Preview 175 176## Toolstack/3rd party 177 178### libvirt driver for xl 179 180 Status: Supported, Security support external 181 182## Debugging, analysis, and crash post-mortem 183 184### Host serial console 185 186 Status, NS16550: Supported 187 Status, EHCI: Supported 188 Status, Cadence UART (ARM): Supported 189 Status, PL011 UART (ARM): Supported 190 Status, Exynos 4210 UART (ARM): Supported 191 Status, OMAP UART (ARM): Supported 192 Status, SCI(F) UART: Supported 193 194### Hypervisor 'debug keys' 195 196These are functions triggered either from the host serial console, 197or via the xl 'debug-keys' command, 198which cause Xen to dump various hypervisor state to the console. 199 200 Status: Supported, not security supported 201 202### Hypervisor synchronous console output (sync_console) 203 204Xen command-line flag to force synchronous console output. 205 206 Status: Supported, not security supported 207 208Useful for debugging, but not suitable for production environments 209due to incurred overhead. 210 211### gdbsx 212 213 Status, x86: Supported, not security supported 214 215Debugger to debug ELF guests 216 217### Soft-reset for PV guests 218 219Soft-reset allows a new kernel to start 'from scratch' with a fresh VM state, 220but with all the memory from the previous state of the VM intact. 221This is primarily designed to allow "crash kernels", 222which can do core dumps of memory to help with debugging in the event of a crash. 223 224 Status: Supported 225 226### xentrace 227 228Tool to capture Xen trace buffer data 229 230 Status, x86: Supported 231 232### gcov 233 234Export hypervisor coverage data suitable for analysis by gcov or lcov. 235 236 Status: Supported, Not security supported 237 238## Memory Management 239 240### Dynamic memory control 241 242Allows a guest to add or remove memory after boot-time. 243This is typically done by a guest kernel agent known as a "balloon driver". 244 245 Status: Supported 246 247### Populate-on-demand memory 248 249This is a mechanism that allows normal operating systems with only a balloon driver 250to boot with memory < maxmem. 251 252 Status, x86 HVM: Supported 253 254### Memory Sharing 255 256Allow sharing of identical pages between guests 257 258 Status, x86 HVM: Experimental 259 260### Memory Paging 261 262Allow pages belonging to guests to be paged to disk 263 264 Status, x86 HVM: Experimental 265 266### Alternative p2m 267 268Alternative p2m (altp2m) allows external monitoring of guest memory 269by maintaining multiple physical to machine (p2m) memory mappings. 270 271 Status, x86 HVM: Tech Preview 272 Status, ARM: Tech Preview 273 274## Resource Management 275 276### CPU Pools 277 278Groups physical cpus into distinct groups called "cpupools", 279with each pool having the capability 280of using different schedulers and scheduling properties. 281 282 Status: Supported 283 284### Core Scheduling 285 286Allows to group virtual cpus into virtual cores which are scheduled on the 287physical cores. This results in never running different guests at the same 288time on the same physical core. 289 290 Status, x86: Experimental 291 292### Credit Scheduler 293 294A weighted proportional fair share virtual CPU scheduler. 295This is the default scheduler. 296 297 Status: Supported 298 299### Credit2 Scheduler 300 301A general purpose scheduler for Xen, 302designed with particular focus on fairness, responsiveness, and scalability 303 304 Status: Supported 305 306### RTDS based Scheduler 307 308A soft real-time CPU scheduler 309built to provide guaranteed CPU capacity to guest VMs on SMP hosts 310 311 Status: Experimental 312 313### ARINC653 Scheduler 314 315A periodically repeating fixed timeslice scheduler. 316 317 Status: Supported 318 319Currently only single-vcpu domains are supported. 320 321### Null Scheduler 322 323A very simple, very static scheduling policy 324that always schedules the same vCPU(s) on the same pCPU(s). 325It is designed for maximum determinism and minimum overhead 326on embedded platforms. 327 328 Status: Experimental 329 330### NUMA scheduler affinity 331 332Enables NUMA aware scheduling in Xen 333 334 Status, x86: Supported 335 336## Scalability 337 338### Super page support 339 340NB that this refers to the ability of guests 341to have higher-level page table entries point directly to memory, 342improving TLB performance. 343On ARM, and on x86 in HAP mode, 344the guest has whatever support is enabled by the hardware. 345 346This feature is independent 347of the ARM "page granularity" feature (see below). 348 349 Status, x86 HVM/PVH, HAP: Supported 350 Status, x86 HVM/PVH, Shadow, 2MiB: Supported 351 Status, ARM: Supported 352 353On x86 in shadow mode, only 2MiB (L2) superpages are available; 354furthermore, they do not have the performance characteristics 355of hardware superpages. 356 357### x86/PVHVM 358 359This is a useful label for a set of hypervisor features 360which add paravirtualized functionality to HVM guests 361for improved performance and scalability. 362This includes exposing event channels to HVM guests. 363 364 Status: Supported 365 366## High Availability and Fault Tolerance 367 368### Remus Fault Tolerance 369 370 Status: Experimental 371 372### COLO Manager 373 374 Status: Experimental 375 376### x86/vMCE 377 378Forward Machine Check Exceptions to appropriate guests 379 380 Status: Supported 381 382## Virtual driver support, guest side 383 384### Blkfront 385 386Guest-side driver capable of speaking the Xen PV block protocol 387 388 Status, Linux: Supported 389 Status, FreeBSD: Supported, Security support external 390 Status, NetBSD: Supported, Security support external 391 Status, OpenBSD: Supported, Security support external 392 Status, Windows: Supported 393 394### Netfront 395 396Guest-side driver capable of speaking the Xen PV networking protocol 397 398 Status, Linux: Supported 399 Status, FreeBSD: Supported, Security support external 400 Status, NetBSD: Supported, Security support external 401 Status, OpenBSD: Supported, Security support external 402 Status, Windows: Supported 403 404### PV Framebuffer (frontend) 405 406Guest-side driver capable of speaking the Xen PV Framebuffer protocol 407 408 Status, Linux (xen-fbfront): Supported 409 410### PV display (frontend) 411 412Guest-side driver capable of speaking the Xen PV display protocol 413 414 Status, Linux: Supported 415 416### PV Console (frontend) 417 418Guest-side driver capable of speaking the Xen PV console protocol 419 420 Status, Linux (hvc_xen): Supported 421 Status, FreeBSD: Supported, Security support external 422 Status, NetBSD: Supported, Security support external 423 Status, Windows: Supported 424 425### PV keyboard (frontend) 426 427Guest-side driver capable of speaking the Xen PV keyboard protocol. 428Note that the "keyboard protocol" includes mouse / pointer / 429multi-touch support as well. 430 431 Status, Linux (xen-kbdfront): Supported 432 433### PV USB (frontend) 434 435 Status, Linux: Supported 436 437### PV SCSI protocol (frontend) 438 439 Status, Linux: Supported, with caveats 440 441NB that while the PV SCSI frontend is in Linux and tested regularly, 442there is currently no xl support. 443 444### PV TPM (frontend) 445 446Guest-side driver capable of speaking the Xen PV TPM protocol 447 448 Status, Linux (xen-tpmfront): Tech Preview 449 450### PV 9pfs frontend 451 452Guest-side driver capable of speaking the Xen 9pfs protocol 453 454 Status, Linux: Tech Preview 455 456### PVCalls (frontend) 457 458Guest-side driver capable of making pv system calls 459 460 Status, Linux: Tech Preview 461 462### PV sound (frontend) 463 464Guest-side driver capable of speaking the Xen PV sound protocol 465 466 Status, Linux: Supported 467 468## Virtual device support, host side 469 470For host-side virtual device support, 471"Supported" and "Tech preview" include xl/libxl support 472unless otherwise noted. 473 474### Blkback 475 476Host-side implementations of the Xen PV block protocol. 477 478 Status, Linux (xen-blkback): Supported 479 Status, QEMU (xen_disk), raw format: Supported 480 Status, QEMU (xen_disk), qcow format: Supported 481 Status, QEMU (xen_disk), qcow2 format: Supported 482 Status, QEMU (xen_disk), vhd format: Supported 483 Status, FreeBSD (blkback): Supported, Security support external 484 Status, NetBSD (xbdback): Supported, security support external 485 Status, Blktap2, raw format: Deprecated 486 Status, Blktap2, vhd format: Deprecated 487 488Backends only support raw format unless otherwise specified. 489 490### Netback 491 492Host-side implementations of Xen PV network protocol 493 494 Status, Linux (xen-netback): Supported 495 Status, FreeBSD (netback): Supported, Security support external 496 Status, NetBSD (xennetback): Supported, Security support external 497 498### PV Framebuffer (backend) 499 500Host-side implementation of the Xen PV framebuffer protocol 501 502 Status, QEMU: Supported 503 504### PV Console (xenconsoled) 505 506Host-side implementation of the Xen PV console protocol 507 508 Status: Supported 509 510### PV keyboard (backend) 511 512Host-side implementation of the Xen PV keyboard protocol. 513Note that the "keyboard protocol" includes mouse / pointer support as well. 514 515 Status, QEMU: Supported 516 517### PV USB (backend) 518 519Host-side implementation of the Xen PV USB protocol 520 521 Status, QEMU: Supported 522 523### PV SCSI protocol (backend) 524 525 Status, Linux: Experimental 526 527NB that while the PV SCSI backend is in Linux and tested regularly, 528there is currently no xl support. 529 530### PV TPM (backend) 531 532 Status: Tech Preview 533 534### PV 9pfs (backend) 535 536 Status, QEMU: Tech Preview 537 538### PVCalls (backend) 539 540 Status, Linux: Experimental 541 542PVCalls backend has been checked into Linux, 543but has no xl support. 544 545### Online resize of virtual disks 546 547 Status: Supported 548 549## Security 550 551### Driver Domains 552 553"Driver domains" means allowing non-Domain 0 domains 554with access to physical devices to act as back-ends. 555 556 Status: Supported, with caveats 557 558See the appropriate "Device Passthrough" section 559for more information about security support. 560 561### Device Model Stub Domains 562 563 Status: Supported, with caveats 564 565Vulnerabilities of a device model stub domain 566to a hostile driver domain (either compromised or untrusted) 567are excluded from security support. 568 569### Device Model Deprivileging 570 571 Status, Linux dom0: Tech Preview, with limited support 572 573This means adding extra restrictions to a device model in order to 574prevent a compromised device model from attacking the rest of the 575domain it's running in (normally dom0). 576 577"Tech preview with limited support" means we will not issue XSAs for 578the _additional_ functionality provided by the feature; but we will 579issue XSAs in the event that enabling this feature opens up a security 580hole that would not be present without the feature disabled. 581 582For example, while this is classified as tech preview, a bug in libxl 583which failed to change the user ID of QEMU would not receive an XSA, 584since without this feature the user ID wouldn't be changed. But a 585change which made it possible for a compromised guest to read 586arbitrary files on the host filesystem without compromising QEMU would 587be issued an XSA, since that does weaken security. 588 589### KCONFIG Expert 590 591 Status: Experimental 592 593### Live Patching 594 595 Status, x86: Supported 596 Status, ARM: Experimental 597 598Compile time disabled for ARM by default. 599 600### Virtual Machine Introspection 601 602 Status, x86: Supported, not security supported 603 604### XSM & FLASK 605 606 Status: Experimental 607 608Compile time disabled by default. 609 610Also note that using XSM 611to delegate various domain control hypercalls 612to particular other domains, rather than only permitting use by dom0, 613is also specifically excluded from security support for many hypercalls. 614Please see XSA-77 for more details. 615 616### FLASK default policy 617 618 Status: Experimental 619 620The default policy includes FLASK labels and roles for a "typical" Xen-based system 621with dom0, driver domains, stub domains, domUs, and so on. 622 623## Virtual Hardware, Hypervisor 624 625### x86/Nested PV 626 627This means running a Xen hypervisor inside an HVM domain on a Xen system, 628with support for PV L2 guests only 629(i.e., hardware virtualization extensions not provided 630to the guest). 631 632 Status, x86 Xen HVM: Tech Preview 633 634This works, but has performance limitations 635because the L1 dom0 can only access emulated L1 devices. 636 637Xen may also run inside other hypervisors (KVM, Hyper-V, VMWare), 638but nobody has reported on performance. 639 640### x86/Nested HVM 641 642This means providing hardware virtulization support to guest VMs 643allowing, for instance, a nested Xen to support both PV and HVM guests. 644It also implies support for other hypervisors, 645such as KVM, Hyper-V, Bromium, and so on as guests. 646 647 Status, x86 HVM: Experimental 648 649### vPMU 650 651Virtual Performance Management Unit for HVM guests 652 653 Status, x86: Supported, Not security supported 654 655Disabled by default (enable with hypervisor command line option). 656This feature is not security supported: see https://xenbits.xen.org/xsa/advisory-163.html 657 658### Argo: Inter-domain message delivery by hypercall 659 660 Status: Experimental 661 662### x86/PCI Device Passthrough 663 664 Status, x86 PV: Supported, with caveats 665 Status, x86 HVM: Supported, with caveats 666 667Only systems using IOMMUs are supported. 668 669Not compatible with migration, populate-on-demand, altp2m, 670introspection, memory sharing, or memory paging. 671 672Because of hardware limitations 673(affecting any operating system or hypervisor), 674it is generally not safe to use this feature 675to expose a physical device to completely untrusted guests. 676However, this feature can still confer significant security benefit 677when used to remove drivers and backends from domain 0 678(i.e., Driver Domains). 679 680### x86/Multiple IOREQ servers 681 682An IOREQ server provides emulated devices to HVM and PVH guests. 683QEMU is normally the only IOREQ server, 684but Xen has support for multiple IOREQ servers. 685This allows for custom or proprietary device emulators 686to be used in addition to QEMU. 687 688 Status: Experimental 689 690### ARM/Non-PCI device passthrough 691 692 Status: Supported, not security supported 693 694Note that this still requires an IOMMU 695that covers the DMA of the device to be passed through. 696 697### ARM: 16K and 64K page granularity in guests 698 699 Status: Supported, with caveats 700 701No support for QEMU backends in a 16K or 64K domain. 702 703### ARM: Guest Device Tree support 704 705 Status: Supported 706 707### ARM: Guest ACPI support 708 709 Status: Supported 710 711### Arm: OP-TEE Mediator 712 713 Status: Tech Preview 714 715## Virtual Hardware, QEMU 716 717This section describes supported devices available in HVM mode using a 718qemu devicemodel (the default). 719 720 Status: Support scope restricted 721 722Note that other devices are available but not security supported. 723 724### x86/Emulated platform devices (QEMU): 725 726 Status, piix3: Supported 727 728### x86/Emulated network (QEMU): 729 730 Status, e1000: Supported 731 Status, rtl8193: Supported 732 Status, virtio-net: Supported 733 734### x86/Emulated storage (QEMU): 735 736 Status, piix3 ide: Supported 737 Status, ahci: Supported 738 739See the section **Blkback** for image formats supported by QEMU. 740 741### x86/Emulated graphics (QEMU): 742 743 Status, cirrus-vga: Supported 744 Status, stdvga: Supported 745 746### x86/Emulated audio (QEMU): 747 748 Status, sb16: Supported 749 Status, es1370: Supported 750 Status, ac97: Supported 751 752### x86/Emulated input (QEMU): 753 754 Status, usbmouse: Supported 755 Status, usbtablet: Supported 756 Status, ps/2 keyboard: Supported 757 Status, ps/2 mouse: Supported 758 759### x86/Emulated serial card (QEMU): 760 761 Status, UART 16550A: Supported 762 763### x86/Host USB passthrough (QEMU): 764 765 Status: Supported, not security supported 766 767### qemu-xen-traditional ### 768 769The Xen Project provides an old version of qemu with modifications 770which enable use as a device model stub domain. The old version is 771normally selected by default only in a stub dm configuration, but it 772can be requested explicitly in other configurations, for example in 773`xl` with `device_model_version="QEMU_XEN_TRADITIONAL"`. 774 775 Status, Device Model Stub Domains: Supported, with caveats 776 Status, as host process device model: No security support, not recommended 777 778qemu-xen-traditional is security supported only for those available 779devices which are supported for mainstream QEMU (see above), with 780trusted driver domains (see Device Model Stub Domains). 781 782## Virtual Firmware 783 784### x86/HVM iPXE 785 786Booting a guest via PXE. 787 788 Status: Supported, with caveats 789 790PXE inherently places full trust of the guest in the network, 791and so should only be used 792when the guest network is under the same administrative control 793as the guest itself. 794 795### x86/HVM BIOS 796 797Booting a guest via guest BIOS firmware 798 799 Status, SeaBIOS (qemu-xen): Supported 800 Status, ROMBIOS (qemu-xen-traditional): Supported 801 802### x86/HVM OVMF 803 804OVMF firmware implements the UEFI boot protocol. 805 806 Status, qemu-xen: Supported 807 808# Format and definitions 809 810This file contains prose, and machine-readable fragments. 811The data in a machine-readable fragment relate to 812the section and subsection in which it is found. 813 814The file is in markdown format. 815The machine-readable fragments are markdown literals 816containing RFC-822-like (deb822-like) data. 817 818In each case, descriptions which expand on the name of a feature as 819provided in the section heading, precede the Status indications. 820Any paragraphs which follow the Status indication are caveats or 821qualifications of the information provided in Status fields. 822 823## Keys found in the Feature Support subsections 824 825### Status 826 827This gives the overall status of the feature, 828including security support status, functional completeness, etc. 829Refer to the detailed definitions below. 830 831If support differs based on implementation 832(for instance, x86 / ARM, Linux / QEMU / FreeBSD), 833one line for each set of implementations will be listed. 834 835## Definition of Status labels 836 837Each Status value corresponds to levels of security support, 838testing, stability, etc., as follows: 839 840### Experimental 841 842 Functional completeness: No 843 Functional stability: Here be dragons 844 Interface stability: Not stable 845 Security supported: No 846 847### Tech Preview 848 849 Functional completeness: Yes 850 Functional stability: Quirky 851 Interface stability: Provisionally stable 852 Security supported: No 853 854#### Supported 855 856 Functional completeness: Yes 857 Functional stability: Normal 858 Interface stability: Yes 859 Security supported: Yes 860 861#### Deprecated 862 863 Functional completeness: Yes 864 Functional stability: Quirky 865 Interface stability: No (as in, may disappear the next release) 866 Security supported: Yes 867 868All of these may appear in modified form. 869There are several interfaces, for instance, 870which are officially declared as not stable; 871in such a case this feature may be described as "Stable / Interface not stable". 872 873## Definition of the status label interpretation tags 874 875### Functionally complete 876 877Does it behave like a fully functional feature? 878Does it work on all expected platforms, 879or does it only work for a very specific sub-case? 880Does it have a sensible UI, 881or do you have to have a deep understanding of the internals 882to get it to work properly? 883 884### Functional stability 885 886What is the risk of it exhibiting bugs? 887 888General answers to the above: 889 890 * **Here be dragons** 891 892 Pretty likely to still crash / fail to work. 893 Not recommended unless you like life on the bleeding edge. 894 895 * **Quirky** 896 897 Mostly works but may have odd behavior here and there. 898 Recommended for playing around or for non-production use cases. 899 900 * **Normal** 901 902 Ready for production use 903 904### Interface stability 905 906If I build a system based on the current interfaces, 907will they still work when I upgrade to the next version? 908 909 * **Not stable** 910 911 Interface is still in the early stages and 912 still fairly likely to be broken in future updates. 913 914 * **Provisionally stable** 915 916 We're not yet promising backwards compatibility, 917 but we think this is probably the final form of the interface. 918 It may still require some tweaks. 919 920 * **Stable** 921 922 We will try very hard to avoid breaking backwards compatibility, 923 and to fix any regressions that are reported. 924 925### Security supported 926 927Will XSAs be issued if security-related bugs are discovered 928in the functionality? 929 930If "no", 931anyone who finds a security-related bug in the feature 932will be advised to 933post it publicly to the Xen Project mailing lists 934(or contact another security response team, 935if a relevant one exists). 936 937Bugs found after the end of **Security-Support-Until** 938in the Release Support section will receive an XSA 939if they also affect newer, security-supported, versions of Xen. 940However, the Xen Project will not provide official fixes 941for non-security-supported versions. 942 943Three common 'diversions' from the 'Supported' category 944are given the following labels: 945 946 * **Supported, Not security supported** 947 948 Functionally complete, normal stability, 949 interface stable, but no security support 950 951 * **Supported, Security support external** 952 953 This feature is security supported 954 by a different organization (not the XenProject). 955 See **External security support** below. 956 957 * **Supported, with caveats** 958 959 This feature is security supported only under certain conditions, 960 or support is given only for certain aspects of the feature, 961 or the feature should be used with care 962 because it is easy to use insecurely without knowing it. 963 Additional details will be given in the description. 964 965### Interaction with other features 966 967Not all features interact well with all other features. 968Some features are only for HVM guests; some don't work with migration, &c. 969 970### External security support 971 972The XenProject security team 973provides security support for XenProject projects. 974 975We also provide security support for Xen-related code in Linux, 976which is an external project but doesn't have its own security process. 977 978External projects that provide their own security support for Xen-related features are listed below. 979 980 * QEMU https://wiki.qemu.org/index.php/SecurityProcess 981 982 * Libvirt https://libvirt.org/securityprocess.html 983 984 * FreeBSD https://www.freebsd.org/security/ 985 986 * NetBSD http://www.netbsd.org/support/security/ 987 988 * OpenBSD https://www.openbsd.org/security.html 989