[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
/*
|
|
|
|
* vmx.h: VMX Architecture related definitions
|
|
|
|
* Copyright (c) 2004, Intel Corporation.
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
|
|
* under the terms and conditions of the GNU General Public License,
|
|
|
|
* version 2, as published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope it will be useful, but WITHOUT
|
|
|
|
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
|
|
|
|
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
|
|
|
|
* more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License along with
|
|
|
|
* this program; if not, write to the Free Software Foundation, Inc., 59 Temple
|
|
|
|
* Place - Suite 330, Boston, MA 02111-1307 USA.
|
|
|
|
*
|
|
|
|
* A few random additions are:
|
|
|
|
* Copyright (C) 2006 Qumranet
|
|
|
|
* Avi Kivity <avi@qumranet.com>
|
|
|
|
* Yaniv Kamay <yaniv@qumranet.com>
|
|
|
|
*
|
|
|
|
*/
|
2012-12-15 05:37:13 +07:00
|
|
|
#ifndef VMX_H
|
|
|
|
#define VMX_H
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2012-09-17 15:31:13 +07:00
|
|
|
|
2016-11-30 09:14:07 +07:00
|
|
|
#include <linux/bitops.h>
|
2010-04-28 19:40:31 +07:00
|
|
|
#include <linux/types.h>
|
2012-12-15 05:37:13 +07:00
|
|
|
#include <uapi/asm/vmx.h>
|
2010-04-28 19:40:31 +07:00
|
|
|
|
2007-11-11 17:27:20 +07:00
|
|
|
/*
|
|
|
|
* Definitions of Primary Processor-Based VM-Execution Controls.
|
|
|
|
*/
|
2007-07-25 16:17:06 +07:00
|
|
|
#define CPU_BASED_VIRTUAL_INTR_PENDING 0x00000004
|
|
|
|
#define CPU_BASED_USE_TSC_OFFSETING 0x00000008
|
|
|
|
#define CPU_BASED_HLT_EXITING 0x00000080
|
|
|
|
#define CPU_BASED_INVLPG_EXITING 0x00000200
|
|
|
|
#define CPU_BASED_MWAIT_EXITING 0x00000400
|
|
|
|
#define CPU_BASED_RDPMC_EXITING 0x00000800
|
|
|
|
#define CPU_BASED_RDTSC_EXITING 0x00001000
|
2008-04-25 09:13:16 +07:00
|
|
|
#define CPU_BASED_CR3_LOAD_EXITING 0x00008000
|
|
|
|
#define CPU_BASED_CR3_STORE_EXITING 0x00010000
|
2007-07-25 16:17:06 +07:00
|
|
|
#define CPU_BASED_CR8_LOAD_EXITING 0x00080000
|
|
|
|
#define CPU_BASED_CR8_STORE_EXITING 0x00100000
|
|
|
|
#define CPU_BASED_TPR_SHADOW 0x00200000
|
2008-05-15 17:23:25 +07:00
|
|
|
#define CPU_BASED_VIRTUAL_NMI_PENDING 0x00400000
|
2007-07-25 16:17:06 +07:00
|
|
|
#define CPU_BASED_MOV_DR_EXITING 0x00800000
|
|
|
|
#define CPU_BASED_UNCOND_IO_EXITING 0x01000000
|
|
|
|
#define CPU_BASED_USE_IO_BITMAPS 0x02000000
|
2015-07-06 00:08:57 +07:00
|
|
|
#define CPU_BASED_MONITOR_TRAP_FLAG 0x08000000
|
2007-07-25 16:17:06 +07:00
|
|
|
#define CPU_BASED_USE_MSR_BITMAPS 0x10000000
|
|
|
|
#define CPU_BASED_MONITOR_EXITING 0x20000000
|
|
|
|
#define CPU_BASED_PAUSE_EXITING 0x40000000
|
|
|
|
#define CPU_BASED_ACTIVATE_SECONDARY_CONTROLS 0x80000000
|
2014-06-16 18:59:42 +07:00
|
|
|
|
|
|
|
#define CPU_BASED_ALWAYSON_WITHOUT_TRUE_MSR 0x0401e172
|
|
|
|
|
2007-11-11 17:27:20 +07:00
|
|
|
/*
|
|
|
|
* Definitions of Secondary Processor-Based VM-Execution Controls.
|
|
|
|
*/
|
|
|
|
#define SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES 0x00000001
|
2008-04-25 09:13:16 +07:00
|
|
|
#define SECONDARY_EXEC_ENABLE_EPT 0x00000002
|
2016-10-25 21:06:30 +07:00
|
|
|
#define SECONDARY_EXEC_DESC 0x00000004
|
2009-12-18 15:48:47 +07:00
|
|
|
#define SECONDARY_EXEC_RDTSCP 0x00000008
|
2013-01-25 09:18:50 +07:00
|
|
|
#define SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE 0x00000010
|
2008-01-17 14:14:33 +07:00
|
|
|
#define SECONDARY_EXEC_ENABLE_VPID 0x00000020
|
2007-11-11 17:28:35 +07:00
|
|
|
#define SECONDARY_EXEC_WBINVD_EXITING 0x00000040
|
2009-06-09 01:34:16 +07:00
|
|
|
#define SECONDARY_EXEC_UNRESTRICTED_GUEST 0x00000080
|
2013-01-25 09:18:49 +07:00
|
|
|
#define SECONDARY_EXEC_APIC_REGISTER_VIRT 0x00000100
|
2013-01-25 09:18:51 +07:00
|
|
|
#define SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY 0x00000200
|
2009-10-09 17:03:20 +07:00
|
|
|
#define SECONDARY_EXEC_PAUSE_LOOP_EXITING 0x00000400
|
2017-08-25 01:51:37 +07:00
|
|
|
#define SECONDARY_EXEC_RDRAND_EXITING 0x00000800
|
2012-07-02 08:18:48 +07:00
|
|
|
#define SECONDARY_EXEC_ENABLE_INVPCID 0x00001000
|
2017-08-04 02:54:41 +07:00
|
|
|
#define SECONDARY_EXEC_ENABLE_VMFUNC 0x00002000
|
2013-04-18 18:34:55 +07:00
|
|
|
#define SECONDARY_EXEC_SHADOW_VMCS 0x00004000
|
2018-08-14 23:33:33 +07:00
|
|
|
#define SECONDARY_EXEC_ENCLS_EXITING 0x00008000
|
2017-08-25 01:51:37 +07:00
|
|
|
#define SECONDARY_EXEC_RDSEED_EXITING 0x00010000
|
2015-01-28 09:54:28 +07:00
|
|
|
#define SECONDARY_EXEC_ENABLE_PML 0x00020000
|
2014-12-02 18:21:30 +07:00
|
|
|
#define SECONDARY_EXEC_XSAVES 0x00100000
|
2015-10-20 14:39:09 +07:00
|
|
|
#define SECONDARY_EXEC_TSC_SCALING 0x02000000
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2007-07-25 16:17:06 +07:00
|
|
|
#define PIN_BASED_EXT_INTR_MASK 0x00000001
|
|
|
|
#define PIN_BASED_NMI_EXITING 0x00000008
|
|
|
|
#define PIN_BASED_VIRTUAL_NMIS 0x00000020
|
2013-03-13 17:31:24 +07:00
|
|
|
#define PIN_BASED_VMX_PREEMPTION_TIMER 0x00000040
|
2013-04-11 18:25:12 +07:00
|
|
|
#define PIN_BASED_POSTED_INTR 0x00000080
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2013-03-13 17:30:50 +07:00
|
|
|
#define PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR 0x00000016
|
|
|
|
|
2014-06-13 00:40:32 +07:00
|
|
|
#define VM_EXIT_SAVE_DEBUG_CONTROLS 0x00000004
|
2007-07-25 16:17:06 +07:00
|
|
|
#define VM_EXIT_HOST_ADDR_SPACE_SIZE 0x00000200
|
2010-12-21 17:54:19 +07:00
|
|
|
#define VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL 0x00001000
|
2007-07-25 16:17:06 +07:00
|
|
|
#define VM_EXIT_ACK_INTR_ON_EXIT 0x00008000
|
2008-10-09 15:01:55 +07:00
|
|
|
#define VM_EXIT_SAVE_IA32_PAT 0x00040000
|
|
|
|
#define VM_EXIT_LOAD_IA32_PAT 0x00080000
|
2010-12-21 17:54:19 +07:00
|
|
|
#define VM_EXIT_SAVE_IA32_EFER 0x00100000
|
|
|
|
#define VM_EXIT_LOAD_IA32_EFER 0x00200000
|
|
|
|
#define VM_EXIT_SAVE_VMX_PREEMPTION_TIMER 0x00400000
|
2014-02-24 17:55:46 +07:00
|
|
|
#define VM_EXIT_CLEAR_BNDCFGS 0x00800000
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2013-03-06 21:44:03 +07:00
|
|
|
#define VM_EXIT_ALWAYSON_WITHOUT_TRUE_MSR 0x00036dff
|
|
|
|
|
2014-06-13 00:40:32 +07:00
|
|
|
#define VM_ENTRY_LOAD_DEBUG_CONTROLS 0x00000004
|
2007-07-25 16:17:06 +07:00
|
|
|
#define VM_ENTRY_IA32E_MODE 0x00000200
|
|
|
|
#define VM_ENTRY_SMM 0x00000400
|
|
|
|
#define VM_ENTRY_DEACT_DUAL_MONITOR 0x00000800
|
2010-12-21 17:54:19 +07:00
|
|
|
#define VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL 0x00002000
|
2008-10-09 15:01:55 +07:00
|
|
|
#define VM_ENTRY_LOAD_IA32_PAT 0x00004000
|
2010-12-21 17:54:19 +07:00
|
|
|
#define VM_ENTRY_LOAD_IA32_EFER 0x00008000
|
2014-02-24 17:55:46 +07:00
|
|
|
#define VM_ENTRY_LOAD_BNDCFGS 0x00010000
|
2007-07-25 16:17:06 +07:00
|
|
|
|
2013-03-06 21:44:03 +07:00
|
|
|
#define VM_ENTRY_ALWAYSON_WITHOUT_TRUE_MSR 0x000011ff
|
|
|
|
|
2013-03-13 17:31:24 +07:00
|
|
|
#define VMX_MISC_PREEMPTION_TIMER_RATE_MASK 0x0000001f
|
2013-03-13 22:06:41 +07:00
|
|
|
#define VMX_MISC_SAVE_EFER_LMA 0x00000020
|
2013-12-04 14:58:54 +07:00
|
|
|
#define VMX_MISC_ACTIVITY_HLT 0x00000040
|
2018-06-21 07:21:29 +07:00
|
|
|
#define VMX_MISC_ZERO_LEN_INS 0x40000000
|
2013-03-13 22:06:41 +07:00
|
|
|
|
2017-08-04 02:54:43 +07:00
|
|
|
/* VMFUNC functions */
|
|
|
|
#define VMX_VMFUNC_EPTP_SWITCHING 0x00000001
|
|
|
|
#define VMFUNC_EPTP_ENTRIES 512
|
|
|
|
|
2016-11-30 09:14:07 +07:00
|
|
|
static inline u32 vmx_basic_vmcs_revision_id(u64 vmx_basic)
|
|
|
|
{
|
|
|
|
return vmx_basic & GENMASK_ULL(30, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline u32 vmx_basic_vmcs_size(u64 vmx_basic)
|
|
|
|
{
|
|
|
|
return (vmx_basic & GENMASK_ULL(44, 32)) >> 32;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int vmx_misc_preemption_timer_rate(u64 vmx_misc)
|
|
|
|
{
|
|
|
|
return vmx_misc & VMX_MISC_PREEMPTION_TIMER_RATE_MASK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int vmx_misc_cr3_count(u64 vmx_misc)
|
|
|
|
{
|
|
|
|
return (vmx_misc & GENMASK_ULL(24, 16)) >> 16;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int vmx_misc_max_msr(u64 vmx_misc)
|
|
|
|
{
|
|
|
|
return (vmx_misc & GENMASK_ULL(27, 25)) >> 25;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int vmx_misc_mseg_revid(u64 vmx_misc)
|
|
|
|
{
|
|
|
|
return (vmx_misc & GENMASK_ULL(63, 32)) >> 32;
|
|
|
|
}
|
|
|
|
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
/* VMCS Encodings */
|
|
|
|
enum vmcs_field {
|
2008-01-17 14:14:33 +07:00
|
|
|
VIRTUAL_PROCESSOR_ID = 0x00000000,
|
2013-04-11 18:25:12 +07:00
|
|
|
POSTED_INTR_NV = 0x00000002,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
GUEST_ES_SELECTOR = 0x00000800,
|
|
|
|
GUEST_CS_SELECTOR = 0x00000802,
|
|
|
|
GUEST_SS_SELECTOR = 0x00000804,
|
|
|
|
GUEST_DS_SELECTOR = 0x00000806,
|
|
|
|
GUEST_FS_SELECTOR = 0x00000808,
|
|
|
|
GUEST_GS_SELECTOR = 0x0000080a,
|
|
|
|
GUEST_LDTR_SELECTOR = 0x0000080c,
|
|
|
|
GUEST_TR_SELECTOR = 0x0000080e,
|
2013-01-25 09:18:51 +07:00
|
|
|
GUEST_INTR_STATUS = 0x00000810,
|
2015-01-28 09:54:28 +07:00
|
|
|
GUEST_PML_INDEX = 0x00000812,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
HOST_ES_SELECTOR = 0x00000c00,
|
|
|
|
HOST_CS_SELECTOR = 0x00000c02,
|
|
|
|
HOST_SS_SELECTOR = 0x00000c04,
|
|
|
|
HOST_DS_SELECTOR = 0x00000c06,
|
|
|
|
HOST_FS_SELECTOR = 0x00000c08,
|
|
|
|
HOST_GS_SELECTOR = 0x00000c0a,
|
|
|
|
HOST_TR_SELECTOR = 0x00000c0c,
|
|
|
|
IO_BITMAP_A = 0x00002000,
|
|
|
|
IO_BITMAP_A_HIGH = 0x00002001,
|
|
|
|
IO_BITMAP_B = 0x00002002,
|
|
|
|
IO_BITMAP_B_HIGH = 0x00002003,
|
|
|
|
MSR_BITMAP = 0x00002004,
|
|
|
|
MSR_BITMAP_HIGH = 0x00002005,
|
|
|
|
VM_EXIT_MSR_STORE_ADDR = 0x00002006,
|
|
|
|
VM_EXIT_MSR_STORE_ADDR_HIGH = 0x00002007,
|
|
|
|
VM_EXIT_MSR_LOAD_ADDR = 0x00002008,
|
|
|
|
VM_EXIT_MSR_LOAD_ADDR_HIGH = 0x00002009,
|
|
|
|
VM_ENTRY_MSR_LOAD_ADDR = 0x0000200a,
|
|
|
|
VM_ENTRY_MSR_LOAD_ADDR_HIGH = 0x0000200b,
|
2015-01-28 09:54:28 +07:00
|
|
|
PML_ADDRESS = 0x0000200e,
|
|
|
|
PML_ADDRESS_HIGH = 0x0000200f,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
TSC_OFFSET = 0x00002010,
|
|
|
|
TSC_OFFSET_HIGH = 0x00002011,
|
|
|
|
VIRTUAL_APIC_PAGE_ADDR = 0x00002012,
|
|
|
|
VIRTUAL_APIC_PAGE_ADDR_HIGH = 0x00002013,
|
2007-10-29 08:40:42 +07:00
|
|
|
APIC_ACCESS_ADDR = 0x00002014,
|
|
|
|
APIC_ACCESS_ADDR_HIGH = 0x00002015,
|
2013-04-11 18:25:12 +07:00
|
|
|
POSTED_INTR_DESC_ADDR = 0x00002016,
|
|
|
|
POSTED_INTR_DESC_ADDR_HIGH = 0x00002017,
|
2017-08-04 02:54:41 +07:00
|
|
|
VM_FUNCTION_CONTROL = 0x00002018,
|
|
|
|
VM_FUNCTION_CONTROL_HIGH = 0x00002019,
|
2008-04-25 09:13:16 +07:00
|
|
|
EPT_POINTER = 0x0000201a,
|
|
|
|
EPT_POINTER_HIGH = 0x0000201b,
|
2013-01-25 09:18:51 +07:00
|
|
|
EOI_EXIT_BITMAP0 = 0x0000201c,
|
|
|
|
EOI_EXIT_BITMAP0_HIGH = 0x0000201d,
|
|
|
|
EOI_EXIT_BITMAP1 = 0x0000201e,
|
|
|
|
EOI_EXIT_BITMAP1_HIGH = 0x0000201f,
|
|
|
|
EOI_EXIT_BITMAP2 = 0x00002020,
|
|
|
|
EOI_EXIT_BITMAP2_HIGH = 0x00002021,
|
|
|
|
EOI_EXIT_BITMAP3 = 0x00002022,
|
|
|
|
EOI_EXIT_BITMAP3_HIGH = 0x00002023,
|
2017-08-04 02:54:43 +07:00
|
|
|
EPTP_LIST_ADDRESS = 0x00002024,
|
|
|
|
EPTP_LIST_ADDRESS_HIGH = 0x00002025,
|
2013-04-18 18:34:55 +07:00
|
|
|
VMREAD_BITMAP = 0x00002026,
|
2018-05-02 05:40:27 +07:00
|
|
|
VMREAD_BITMAP_HIGH = 0x00002027,
|
2013-04-18 18:34:55 +07:00
|
|
|
VMWRITE_BITMAP = 0x00002028,
|
2018-05-02 05:40:27 +07:00
|
|
|
VMWRITE_BITMAP_HIGH = 0x00002029,
|
2014-12-02 18:14:58 +07:00
|
|
|
XSS_EXIT_BITMAP = 0x0000202C,
|
|
|
|
XSS_EXIT_BITMAP_HIGH = 0x0000202D,
|
2018-08-14 23:33:33 +07:00
|
|
|
ENCLS_EXITING_BITMAP = 0x0000202E,
|
|
|
|
ENCLS_EXITING_BITMAP_HIGH = 0x0000202F,
|
2015-10-20 14:39:09 +07:00
|
|
|
TSC_MULTIPLIER = 0x00002032,
|
|
|
|
TSC_MULTIPLIER_HIGH = 0x00002033,
|
2008-04-25 09:13:16 +07:00
|
|
|
GUEST_PHYSICAL_ADDRESS = 0x00002400,
|
|
|
|
GUEST_PHYSICAL_ADDRESS_HIGH = 0x00002401,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
VMCS_LINK_POINTER = 0x00002800,
|
|
|
|
VMCS_LINK_POINTER_HIGH = 0x00002801,
|
|
|
|
GUEST_IA32_DEBUGCTL = 0x00002802,
|
|
|
|
GUEST_IA32_DEBUGCTL_HIGH = 0x00002803,
|
2008-10-09 15:01:55 +07:00
|
|
|
GUEST_IA32_PAT = 0x00002804,
|
|
|
|
GUEST_IA32_PAT_HIGH = 0x00002805,
|
2010-04-28 19:41:03 +07:00
|
|
|
GUEST_IA32_EFER = 0x00002806,
|
|
|
|
GUEST_IA32_EFER_HIGH = 0x00002807,
|
2011-05-26 03:11:34 +07:00
|
|
|
GUEST_IA32_PERF_GLOBAL_CTRL = 0x00002808,
|
|
|
|
GUEST_IA32_PERF_GLOBAL_CTRL_HIGH= 0x00002809,
|
2008-04-25 09:13:16 +07:00
|
|
|
GUEST_PDPTR0 = 0x0000280a,
|
|
|
|
GUEST_PDPTR0_HIGH = 0x0000280b,
|
|
|
|
GUEST_PDPTR1 = 0x0000280c,
|
|
|
|
GUEST_PDPTR1_HIGH = 0x0000280d,
|
|
|
|
GUEST_PDPTR2 = 0x0000280e,
|
|
|
|
GUEST_PDPTR2_HIGH = 0x0000280f,
|
|
|
|
GUEST_PDPTR3 = 0x00002810,
|
|
|
|
GUEST_PDPTR3_HIGH = 0x00002811,
|
2014-02-24 17:55:46 +07:00
|
|
|
GUEST_BNDCFGS = 0x00002812,
|
|
|
|
GUEST_BNDCFGS_HIGH = 0x00002813,
|
2008-10-09 15:01:55 +07:00
|
|
|
HOST_IA32_PAT = 0x00002c00,
|
|
|
|
HOST_IA32_PAT_HIGH = 0x00002c01,
|
2010-04-28 19:41:03 +07:00
|
|
|
HOST_IA32_EFER = 0x00002c02,
|
|
|
|
HOST_IA32_EFER_HIGH = 0x00002c03,
|
2011-05-26 03:11:34 +07:00
|
|
|
HOST_IA32_PERF_GLOBAL_CTRL = 0x00002c04,
|
|
|
|
HOST_IA32_PERF_GLOBAL_CTRL_HIGH = 0x00002c05,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
PIN_BASED_VM_EXEC_CONTROL = 0x00004000,
|
|
|
|
CPU_BASED_VM_EXEC_CONTROL = 0x00004002,
|
|
|
|
EXCEPTION_BITMAP = 0x00004004,
|
|
|
|
PAGE_FAULT_ERROR_CODE_MASK = 0x00004006,
|
|
|
|
PAGE_FAULT_ERROR_CODE_MATCH = 0x00004008,
|
|
|
|
CR3_TARGET_COUNT = 0x0000400a,
|
|
|
|
VM_EXIT_CONTROLS = 0x0000400c,
|
|
|
|
VM_EXIT_MSR_STORE_COUNT = 0x0000400e,
|
|
|
|
VM_EXIT_MSR_LOAD_COUNT = 0x00004010,
|
|
|
|
VM_ENTRY_CONTROLS = 0x00004012,
|
|
|
|
VM_ENTRY_MSR_LOAD_COUNT = 0x00004014,
|
|
|
|
VM_ENTRY_INTR_INFO_FIELD = 0x00004016,
|
|
|
|
VM_ENTRY_EXCEPTION_ERROR_CODE = 0x00004018,
|
|
|
|
VM_ENTRY_INSTRUCTION_LEN = 0x0000401a,
|
|
|
|
TPR_THRESHOLD = 0x0000401c,
|
|
|
|
SECONDARY_VM_EXEC_CONTROL = 0x0000401e,
|
2009-10-09 17:03:20 +07:00
|
|
|
PLE_GAP = 0x00004020,
|
|
|
|
PLE_WINDOW = 0x00004022,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
VM_INSTRUCTION_ERROR = 0x00004400,
|
|
|
|
VM_EXIT_REASON = 0x00004402,
|
|
|
|
VM_EXIT_INTR_INFO = 0x00004404,
|
|
|
|
VM_EXIT_INTR_ERROR_CODE = 0x00004406,
|
|
|
|
IDT_VECTORING_INFO_FIELD = 0x00004408,
|
|
|
|
IDT_VECTORING_ERROR_CODE = 0x0000440a,
|
|
|
|
VM_EXIT_INSTRUCTION_LEN = 0x0000440c,
|
|
|
|
VMX_INSTRUCTION_INFO = 0x0000440e,
|
|
|
|
GUEST_ES_LIMIT = 0x00004800,
|
|
|
|
GUEST_CS_LIMIT = 0x00004802,
|
|
|
|
GUEST_SS_LIMIT = 0x00004804,
|
|
|
|
GUEST_DS_LIMIT = 0x00004806,
|
|
|
|
GUEST_FS_LIMIT = 0x00004808,
|
|
|
|
GUEST_GS_LIMIT = 0x0000480a,
|
|
|
|
GUEST_LDTR_LIMIT = 0x0000480c,
|
|
|
|
GUEST_TR_LIMIT = 0x0000480e,
|
|
|
|
GUEST_GDTR_LIMIT = 0x00004810,
|
|
|
|
GUEST_IDTR_LIMIT = 0x00004812,
|
|
|
|
GUEST_ES_AR_BYTES = 0x00004814,
|
|
|
|
GUEST_CS_AR_BYTES = 0x00004816,
|
|
|
|
GUEST_SS_AR_BYTES = 0x00004818,
|
|
|
|
GUEST_DS_AR_BYTES = 0x0000481a,
|
|
|
|
GUEST_FS_AR_BYTES = 0x0000481c,
|
|
|
|
GUEST_GS_AR_BYTES = 0x0000481e,
|
|
|
|
GUEST_LDTR_AR_BYTES = 0x00004820,
|
|
|
|
GUEST_TR_AR_BYTES = 0x00004822,
|
|
|
|
GUEST_INTERRUPTIBILITY_INFO = 0x00004824,
|
|
|
|
GUEST_ACTIVITY_STATE = 0X00004826,
|
|
|
|
GUEST_SYSENTER_CS = 0x0000482A,
|
2013-03-13 17:31:24 +07:00
|
|
|
VMX_PREEMPTION_TIMER_VALUE = 0x0000482E,
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
HOST_IA32_SYSENTER_CS = 0x00004c00,
|
|
|
|
CR0_GUEST_HOST_MASK = 0x00006000,
|
|
|
|
CR4_GUEST_HOST_MASK = 0x00006002,
|
|
|
|
CR0_READ_SHADOW = 0x00006004,
|
|
|
|
CR4_READ_SHADOW = 0x00006006,
|
|
|
|
CR3_TARGET_VALUE0 = 0x00006008,
|
|
|
|
CR3_TARGET_VALUE1 = 0x0000600a,
|
|
|
|
CR3_TARGET_VALUE2 = 0x0000600c,
|
|
|
|
CR3_TARGET_VALUE3 = 0x0000600e,
|
|
|
|
EXIT_QUALIFICATION = 0x00006400,
|
|
|
|
GUEST_LINEAR_ADDRESS = 0x0000640a,
|
|
|
|
GUEST_CR0 = 0x00006800,
|
|
|
|
GUEST_CR3 = 0x00006802,
|
|
|
|
GUEST_CR4 = 0x00006804,
|
|
|
|
GUEST_ES_BASE = 0x00006806,
|
|
|
|
GUEST_CS_BASE = 0x00006808,
|
|
|
|
GUEST_SS_BASE = 0x0000680a,
|
|
|
|
GUEST_DS_BASE = 0x0000680c,
|
|
|
|
GUEST_FS_BASE = 0x0000680e,
|
|
|
|
GUEST_GS_BASE = 0x00006810,
|
|
|
|
GUEST_LDTR_BASE = 0x00006812,
|
|
|
|
GUEST_TR_BASE = 0x00006814,
|
|
|
|
GUEST_GDTR_BASE = 0x00006816,
|
|
|
|
GUEST_IDTR_BASE = 0x00006818,
|
|
|
|
GUEST_DR7 = 0x0000681a,
|
|
|
|
GUEST_RSP = 0x0000681c,
|
|
|
|
GUEST_RIP = 0x0000681e,
|
|
|
|
GUEST_RFLAGS = 0x00006820,
|
|
|
|
GUEST_PENDING_DBG_EXCEPTIONS = 0x00006822,
|
|
|
|
GUEST_SYSENTER_ESP = 0x00006824,
|
|
|
|
GUEST_SYSENTER_EIP = 0x00006826,
|
|
|
|
HOST_CR0 = 0x00006c00,
|
|
|
|
HOST_CR3 = 0x00006c02,
|
|
|
|
HOST_CR4 = 0x00006c04,
|
|
|
|
HOST_FS_BASE = 0x00006c06,
|
|
|
|
HOST_GS_BASE = 0x00006c08,
|
|
|
|
HOST_TR_BASE = 0x00006c0a,
|
|
|
|
HOST_GDTR_BASE = 0x00006c0c,
|
|
|
|
HOST_IDTR_BASE = 0x00006c0e,
|
|
|
|
HOST_IA32_SYSENTER_ESP = 0x00006c10,
|
|
|
|
HOST_IA32_SYSENTER_EIP = 0x00006c12,
|
|
|
|
HOST_RSP = 0x00006c14,
|
|
|
|
HOST_RIP = 0x00006c16,
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Interruption-information format
|
|
|
|
*/
|
|
|
|
#define INTR_INFO_VECTOR_MASK 0xff /* 7:0 */
|
|
|
|
#define INTR_INFO_INTR_TYPE_MASK 0x700 /* 10:8 */
|
2008-02-11 23:26:38 +07:00
|
|
|
#define INTR_INFO_DELIVER_CODE_MASK 0x800 /* 11 */
|
2008-05-15 17:23:25 +07:00
|
|
|
#define INTR_INFO_UNBLOCK_NMI 0x1000 /* 12 */
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#define INTR_INFO_VALID_MASK 0x80000000 /* 31 */
|
2008-05-15 17:23:25 +07:00
|
|
|
#define INTR_INFO_RESVD_BITS_MASK 0x7ffff000
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
|
|
|
#define VECTORING_INFO_VECTOR_MASK INTR_INFO_VECTOR_MASK
|
|
|
|
#define VECTORING_INFO_TYPE_MASK INTR_INFO_INTR_TYPE_MASK
|
2008-02-11 23:26:38 +07:00
|
|
|
#define VECTORING_INFO_DELIVER_CODE_MASK INTR_INFO_DELIVER_CODE_MASK
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#define VECTORING_INFO_VALID_MASK INTR_INFO_VALID_MASK
|
|
|
|
|
|
|
|
#define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */
|
2018-06-21 07:21:29 +07:00
|
|
|
#define INTR_TYPE_RESERVED (1 << 8) /* reserved */
|
2008-05-15 17:23:25 +07:00
|
|
|
#define INTR_TYPE_NMI_INTR (2 << 8) /* NMI */
|
2008-12-15 19:52:10 +07:00
|
|
|
#define INTR_TYPE_HARD_EXCEPTION (3 << 8) /* processor exception */
|
2007-11-08 23:19:20 +07:00
|
|
|
#define INTR_TYPE_SOFT_INTR (4 << 8) /* software interrupt */
|
kvm/x86: fix icebp instruction handling
The undocumented 'icebp' instruction (aka 'int1') works pretty much like
'int3' in the absense of in-circuit probing equipment (except,
obviously, that it raises #DB instead of raising #BP), and is used by
some validation test-suites as such.
But Andy Lutomirski noticed that his test suite acted differently in kvm
than on bare hardware.
The reason is that kvm used an inexact test for the icebp instruction:
it just assumed that an all-zero VM exit qualification value meant that
the VM exit was due to icebp.
That is not unlike the guess that do_debug() does for the actual
exception handling case, but it's purely a heuristic, not an absolute
rule. do_debug() does it because it wants to ascribe _some_ reasons to
the #DB that happened, and an empty %dr6 value means that 'icebp' is the
most likely casue and we have no better information.
But kvm can just do it right, because unlike the do_debug() case, kvm
actually sees the real reason for the #DB in the VM-exit interruption
information field.
So instead of relying on an inexact heuristic, just use the actual VM
exit information that says "it was 'icebp'".
Right now the 'icebp' instruction isn't technically documented by Intel,
but that will hopefully change. The special "privileged software
exception" information _is_ actually mentioned in the Intel SDM, even
though the cause of it isn't enumerated.
Reported-by: Andy Lutomirski <luto@kernel.org>
Tested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-03-21 02:16:59 +07:00
|
|
|
#define INTR_TYPE_PRIV_SW_EXCEPTION (5 << 8) /* ICE breakpoint - undocumented */
|
2008-12-15 19:52:10 +07:00
|
|
|
#define INTR_TYPE_SOFT_EXCEPTION (6 << 8) /* software exception */
|
2018-06-21 07:21:29 +07:00
|
|
|
#define INTR_TYPE_OTHER_EVENT (7 << 8) /* other event */
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2008-05-15 17:23:25 +07:00
|
|
|
/* GUEST_INTERRUPTIBILITY_INFO flags. */
|
|
|
|
#define GUEST_INTR_STATE_STI 0x00000001
|
|
|
|
#define GUEST_INTR_STATE_MOV_SS 0x00000002
|
|
|
|
#define GUEST_INTR_STATE_SMI 0x00000004
|
|
|
|
#define GUEST_INTR_STATE_NMI 0x00000008
|
|
|
|
|
2010-12-06 23:53:38 +07:00
|
|
|
/* GUEST_ACTIVITY_STATE flags */
|
|
|
|
#define GUEST_ACTIVITY_ACTIVE 0
|
|
|
|
#define GUEST_ACTIVITY_HLT 1
|
|
|
|
#define GUEST_ACTIVITY_SHUTDOWN 2
|
|
|
|
#define GUEST_ACTIVITY_WAIT_SIPI 3
|
|
|
|
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
/*
|
|
|
|
* Exit Qualifications for MOV for Control Register Access
|
|
|
|
*/
|
2007-10-08 20:02:08 +07:00
|
|
|
#define CONTROL_REG_ACCESS_NUM 0x7 /* 2:0, number of control reg.*/
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#define CONTROL_REG_ACCESS_TYPE 0x30 /* 5:4, access type */
|
2007-10-08 20:02:08 +07:00
|
|
|
#define CONTROL_REG_ACCESS_REG 0xf00 /* 10:8, general purpose reg. */
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#define LMSW_SOURCE_DATA_SHIFT 16
|
|
|
|
#define LMSW_SOURCE_DATA (0xFFFF << LMSW_SOURCE_DATA_SHIFT) /* 16:31 lmsw source */
|
|
|
|
#define REG_EAX (0 << 8)
|
|
|
|
#define REG_ECX (1 << 8)
|
|
|
|
#define REG_EDX (2 << 8)
|
|
|
|
#define REG_EBX (3 << 8)
|
|
|
|
#define REG_ESP (4 << 8)
|
|
|
|
#define REG_EBP (5 << 8)
|
|
|
|
#define REG_ESI (6 << 8)
|
|
|
|
#define REG_EDI (7 << 8)
|
|
|
|
#define REG_R8 (8 << 8)
|
|
|
|
#define REG_R9 (9 << 8)
|
|
|
|
#define REG_R10 (10 << 8)
|
|
|
|
#define REG_R11 (11 << 8)
|
|
|
|
#define REG_R12 (12 << 8)
|
|
|
|
#define REG_R13 (13 << 8)
|
|
|
|
#define REG_R14 (14 << 8)
|
|
|
|
#define REG_R15 (15 << 8)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Exit Qualifications for MOV for Debug Register Access
|
|
|
|
*/
|
2007-10-08 20:02:08 +07:00
|
|
|
#define DEBUG_REG_ACCESS_NUM 0x7 /* 2:0, number of debug reg. */
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#define DEBUG_REG_ACCESS_TYPE 0x10 /* 4, direction of access */
|
|
|
|
#define TYPE_MOV_TO_DR (0 << 4)
|
|
|
|
#define TYPE_MOV_FROM_DR (1 << 4)
|
2008-12-15 19:52:10 +07:00
|
|
|
#define DEBUG_REG_ACCESS_REG(eq) (((eq) >> 8) & 0xf) /* 11:8, general purpose reg. */
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
|
|
|
|
2011-08-30 17:56:17 +07:00
|
|
|
/*
|
|
|
|
* Exit Qualifications for APIC-Access
|
|
|
|
*/
|
|
|
|
#define APIC_ACCESS_OFFSET 0xfff /* 11:0, offset within the APIC page */
|
|
|
|
#define APIC_ACCESS_TYPE 0xf000 /* 15:12, access type */
|
|
|
|
#define TYPE_LINEAR_APIC_INST_READ (0 << 12)
|
|
|
|
#define TYPE_LINEAR_APIC_INST_WRITE (1 << 12)
|
|
|
|
#define TYPE_LINEAR_APIC_INST_FETCH (2 << 12)
|
|
|
|
#define TYPE_LINEAR_APIC_EVENT (3 << 12)
|
|
|
|
#define TYPE_PHYSICAL_APIC_EVENT (10 << 12)
|
|
|
|
#define TYPE_PHYSICAL_APIC_INST (15 << 12)
|
|
|
|
|
2015-08-14 03:18:48 +07:00
|
|
|
/* segment AR in VMCS -- these are different from what LAR reports */
|
|
|
|
#define VMX_SEGMENT_AR_L_MASK (1 << 13)
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2015-08-14 03:18:48 +07:00
|
|
|
#define VMX_AR_TYPE_ACCESSES_MASK 1
|
|
|
|
#define VMX_AR_TYPE_READABLE_MASK (1 << 1)
|
|
|
|
#define VMX_AR_TYPE_WRITEABLE_MASK (1 << 2)
|
|
|
|
#define VMX_AR_TYPE_CODE_MASK (1 << 3)
|
|
|
|
#define VMX_AR_TYPE_MASK 0x0f
|
|
|
|
#define VMX_AR_TYPE_BUSY_64_TSS 11
|
|
|
|
#define VMX_AR_TYPE_BUSY_32_TSS 11
|
|
|
|
#define VMX_AR_TYPE_BUSY_16_TSS 3
|
|
|
|
#define VMX_AR_TYPE_LDT 2
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2015-08-14 03:18:48 +07:00
|
|
|
#define VMX_AR_UNUSABLE_MASK (1 << 16)
|
|
|
|
#define VMX_AR_S_MASK (1 << 4)
|
|
|
|
#define VMX_AR_P_MASK (1 << 7)
|
|
|
|
#define VMX_AR_L_MASK (1 << 13)
|
|
|
|
#define VMX_AR_DB_MASK (1 << 14)
|
|
|
|
#define VMX_AR_G_MASK (1 << 15)
|
|
|
|
#define VMX_AR_DPL_SHIFT 5
|
|
|
|
#define VMX_AR_DPL(ar) (((ar) >> VMX_AR_DPL_SHIFT) & 3)
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2015-08-14 03:18:48 +07:00
|
|
|
#define VMX_AR_RESERVD_MASK 0xfffe0f00
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
|
2012-12-11 00:33:09 +07:00
|
|
|
#define TSS_PRIVATE_MEMSLOT (KVM_USER_MEM_SLOTS + 0)
|
|
|
|
#define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT (KVM_USER_MEM_SLOTS + 1)
|
|
|
|
#define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT (KVM_USER_MEM_SLOTS + 2)
|
2007-10-29 08:40:42 +07:00
|
|
|
|
2008-01-17 14:14:33 +07:00
|
|
|
#define VMX_NR_VPIDS (1 << 16)
|
2016-10-28 11:00:29 +07:00
|
|
|
#define VMX_VPID_EXTENT_INDIVIDUAL_ADDR 0
|
2008-01-17 14:14:33 +07:00
|
|
|
#define VMX_VPID_EXTENT_SINGLE_CONTEXT 1
|
|
|
|
#define VMX_VPID_EXTENT_ALL_CONTEXT 2
|
2016-10-28 11:00:29 +07:00
|
|
|
#define VMX_VPID_EXTENT_SINGLE_NON_GLOBAL 3
|
2008-01-17 14:14:33 +07:00
|
|
|
|
2008-04-25 09:13:16 +07:00
|
|
|
#define VMX_EPT_EXTENT_CONTEXT 1
|
|
|
|
#define VMX_EPT_EXTENT_GLOBAL 2
|
2013-08-05 15:07:17 +07:00
|
|
|
#define VMX_EPT_EXTENT_SHIFT 24
|
2009-06-11 22:07:40 +07:00
|
|
|
|
|
|
|
#define VMX_EPT_EXECUTE_ONLY_BIT (1ull)
|
|
|
|
#define VMX_EPT_PAGE_WALK_4_BIT (1ull << 6)
|
2017-08-24 19:27:55 +07:00
|
|
|
#define VMX_EPT_PAGE_WALK_5_BIT (1ull << 7)
|
2009-06-11 22:07:40 +07:00
|
|
|
#define VMX_EPTP_UC_BIT (1ull << 8)
|
|
|
|
#define VMX_EPTP_WB_BIT (1ull << 14)
|
|
|
|
#define VMX_EPT_2MB_PAGE_BIT (1ull << 16)
|
2010-01-05 18:02:29 +07:00
|
|
|
#define VMX_EPT_1GB_PAGE_BIT (1ull << 17)
|
2013-08-05 15:07:17 +07:00
|
|
|
#define VMX_EPT_INVEPT_BIT (1ull << 20)
|
2012-12-05 00:55:15 +07:00
|
|
|
#define VMX_EPT_AD_BIT (1ull << 21)
|
2008-04-25 09:13:16 +07:00
|
|
|
#define VMX_EPT_EXTENT_CONTEXT_BIT (1ull << 25)
|
|
|
|
#define VMX_EPT_EXTENT_GLOBAL_BIT (1ull << 26)
|
2009-06-11 22:07:40 +07:00
|
|
|
|
2015-10-13 23:12:21 +07:00
|
|
|
#define VMX_VPID_INVVPID_BIT (1ull << 0) /* (32 - 32) */
|
2016-10-28 11:00:29 +07:00
|
|
|
#define VMX_VPID_EXTENT_INDIVIDUAL_ADDR_BIT (1ull << 8) /* (40 - 32) */
|
2010-06-04 07:51:39 +07:00
|
|
|
#define VMX_VPID_EXTENT_SINGLE_CONTEXT_BIT (1ull << 9) /* (41 - 32) */
|
2010-06-07 09:32:29 +07:00
|
|
|
#define VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT (1ull << 10) /* (42 - 32) */
|
2016-10-28 11:00:29 +07:00
|
|
|
#define VMX_VPID_EXTENT_SINGLE_NON_GLOBAL_BIT (1ull << 11) /* (43 - 32) */
|
2010-06-04 07:51:39 +07:00
|
|
|
|
2008-04-28 11:24:45 +07:00
|
|
|
#define VMX_EPT_MT_EPTE_SHIFT 3
|
2017-08-11 04:15:28 +07:00
|
|
|
#define VMX_EPTP_PWL_MASK 0x38ull
|
|
|
|
#define VMX_EPTP_PWL_4 0x18ull
|
2017-08-24 19:27:55 +07:00
|
|
|
#define VMX_EPTP_PWL_5 0x20ull
|
2017-08-11 04:15:28 +07:00
|
|
|
#define VMX_EPTP_AD_ENABLE_BIT (1ull << 6)
|
|
|
|
#define VMX_EPTP_MT_MASK 0x7ull
|
|
|
|
#define VMX_EPTP_MT_WB 0x6ull
|
|
|
|
#define VMX_EPTP_MT_UC 0x0ull
|
2008-04-28 11:24:45 +07:00
|
|
|
#define VMX_EPT_READABLE_MASK 0x1ull
|
|
|
|
#define VMX_EPT_WRITABLE_MASK 0x2ull
|
|
|
|
#define VMX_EPT_EXECUTABLE_MASK 0x4ull
|
2010-02-09 15:41:53 +07:00
|
|
|
#define VMX_EPT_IPAT_BIT (1ull << 6)
|
2016-12-07 07:46:15 +07:00
|
|
|
#define VMX_EPT_ACCESS_BIT (1ull << 8)
|
|
|
|
#define VMX_EPT_DIRTY_BIT (1ull << 9)
|
2016-12-07 07:46:16 +07:00
|
|
|
#define VMX_EPT_RWX_MASK (VMX_EPT_READABLE_MASK | \
|
|
|
|
VMX_EPT_WRITABLE_MASK | \
|
|
|
|
VMX_EPT_EXECUTABLE_MASK)
|
|
|
|
#define VMX_EPT_MT_MASK (7ull << VMX_EPT_MT_EPTE_SHIFT)
|
2016-12-07 07:46:15 +07:00
|
|
|
|
|
|
|
/* The mask to use to trigger an EPT Misconfiguration in order to track MMIO */
|
2016-12-07 07:46:16 +07:00
|
|
|
#define VMX_EPT_MISCONFIG_WX_VALUE (VMX_EPT_WRITABLE_MASK | \
|
|
|
|
VMX_EPT_EXECUTABLE_MASK)
|
2008-04-25 09:13:16 +07:00
|
|
|
|
2008-04-25 20:44:52 +07:00
|
|
|
#define VMX_EPT_IDENTITY_PAGETABLE_ADDR 0xfffbc000ul
|
|
|
|
|
2010-04-28 19:40:31 +07:00
|
|
|
struct vmx_msr_entry {
|
|
|
|
u32 index;
|
|
|
|
u32 reserved;
|
|
|
|
u64 value;
|
|
|
|
} __aligned(16);
|
2008-11-18 04:03:15 +07:00
|
|
|
|
2011-05-26 03:12:04 +07:00
|
|
|
/*
|
|
|
|
* Exit Qualifications for entry failure during or after loading guest state
|
|
|
|
*/
|
|
|
|
#define ENTRY_FAIL_DEFAULT 0
|
|
|
|
#define ENTRY_FAIL_PDPTE 2
|
|
|
|
#define ENTRY_FAIL_NMI 3
|
|
|
|
#define ENTRY_FAIL_VMCS_LINK_PTR 4
|
|
|
|
|
2016-12-07 07:46:10 +07:00
|
|
|
/*
|
|
|
|
* Exit Qualifications for EPT Violations
|
|
|
|
*/
|
2016-12-22 11:29:28 +07:00
|
|
|
#define EPT_VIOLATION_ACC_READ_BIT 0
|
|
|
|
#define EPT_VIOLATION_ACC_WRITE_BIT 1
|
|
|
|
#define EPT_VIOLATION_ACC_INSTR_BIT 2
|
2016-12-07 07:46:10 +07:00
|
|
|
#define EPT_VIOLATION_READABLE_BIT 3
|
|
|
|
#define EPT_VIOLATION_WRITABLE_BIT 4
|
|
|
|
#define EPT_VIOLATION_EXECUTABLE_BIT 5
|
2017-03-30 16:55:30 +07:00
|
|
|
#define EPT_VIOLATION_GVA_TRANSLATED_BIT 8
|
2016-12-22 11:29:28 +07:00
|
|
|
#define EPT_VIOLATION_ACC_READ (1 << EPT_VIOLATION_ACC_READ_BIT)
|
|
|
|
#define EPT_VIOLATION_ACC_WRITE (1 << EPT_VIOLATION_ACC_WRITE_BIT)
|
|
|
|
#define EPT_VIOLATION_ACC_INSTR (1 << EPT_VIOLATION_ACC_INSTR_BIT)
|
2016-12-07 07:46:10 +07:00
|
|
|
#define EPT_VIOLATION_READABLE (1 << EPT_VIOLATION_READABLE_BIT)
|
|
|
|
#define EPT_VIOLATION_WRITABLE (1 << EPT_VIOLATION_WRITABLE_BIT)
|
|
|
|
#define EPT_VIOLATION_EXECUTABLE (1 << EPT_VIOLATION_EXECUTABLE_BIT)
|
2017-03-30 16:55:30 +07:00
|
|
|
#define EPT_VIOLATION_GVA_TRANSLATED (1 << EPT_VIOLATION_GVA_TRANSLATED_BIT)
|
2016-12-07 07:46:10 +07:00
|
|
|
|
2011-05-26 03:06:28 +07:00
|
|
|
/*
|
|
|
|
* VM-instruction error numbers
|
|
|
|
*/
|
|
|
|
enum vm_instruction_error_number {
|
|
|
|
VMXERR_VMCALL_IN_VMX_ROOT_OPERATION = 1,
|
|
|
|
VMXERR_VMCLEAR_INVALID_ADDRESS = 2,
|
|
|
|
VMXERR_VMCLEAR_VMXON_POINTER = 3,
|
|
|
|
VMXERR_VMLAUNCH_NONCLEAR_VMCS = 4,
|
|
|
|
VMXERR_VMRESUME_NONLAUNCHED_VMCS = 5,
|
|
|
|
VMXERR_VMRESUME_AFTER_VMXOFF = 6,
|
|
|
|
VMXERR_ENTRY_INVALID_CONTROL_FIELD = 7,
|
|
|
|
VMXERR_ENTRY_INVALID_HOST_STATE_FIELD = 8,
|
|
|
|
VMXERR_VMPTRLD_INVALID_ADDRESS = 9,
|
|
|
|
VMXERR_VMPTRLD_VMXON_POINTER = 10,
|
|
|
|
VMXERR_VMPTRLD_INCORRECT_VMCS_REVISION_ID = 11,
|
|
|
|
VMXERR_UNSUPPORTED_VMCS_COMPONENT = 12,
|
|
|
|
VMXERR_VMWRITE_READ_ONLY_VMCS_COMPONENT = 13,
|
|
|
|
VMXERR_VMXON_IN_VMX_ROOT_OPERATION = 15,
|
|
|
|
VMXERR_ENTRY_INVALID_EXECUTIVE_VMCS_POINTER = 16,
|
|
|
|
VMXERR_ENTRY_NONLAUNCHED_EXECUTIVE_VMCS = 17,
|
|
|
|
VMXERR_ENTRY_EXECUTIVE_VMCS_POINTER_NOT_VMXON_POINTER = 18,
|
|
|
|
VMXERR_VMCALL_NONCLEAR_VMCS = 19,
|
|
|
|
VMXERR_VMCALL_INVALID_VM_EXIT_CONTROL_FIELDS = 20,
|
|
|
|
VMXERR_VMCALL_INCORRECT_MSEG_REVISION_ID = 22,
|
|
|
|
VMXERR_VMXOFF_UNDER_DUAL_MONITOR_TREATMENT_OF_SMIS_AND_SMM = 23,
|
|
|
|
VMXERR_VMCALL_INVALID_SMM_MONITOR_FEATURES = 24,
|
|
|
|
VMXERR_ENTRY_INVALID_VM_EXECUTION_CONTROL_FIELDS_IN_EXECUTIVE_VMCS = 25,
|
|
|
|
VMXERR_ENTRY_EVENTS_BLOCKED_BY_MOV_SS = 26,
|
|
|
|
VMXERR_INVALID_OPERAND_TO_INVEPT_INVVPID = 28,
|
|
|
|
};
|
|
|
|
|
2018-07-13 21:23:16 +07:00
|
|
|
enum vmx_l1d_flush_state {
|
|
|
|
VMENTER_L1D_FLUSH_AUTO,
|
|
|
|
VMENTER_L1D_FLUSH_NEVER,
|
|
|
|
VMENTER_L1D_FLUSH_COND,
|
|
|
|
VMENTER_L1D_FLUSH_ALWAYS,
|
2018-07-13 21:23:18 +07:00
|
|
|
VMENTER_L1D_FLUSH_EPT_DISABLED,
|
2018-08-05 21:07:46 +07:00
|
|
|
VMENTER_L1D_FLUSH_NOT_REQUIRED,
|
2018-07-13 21:23:16 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
extern enum vmx_l1d_flush_state l1tf_vmx_mitigation;
|
|
|
|
|
[PATCH] kvm: userspace interface
web site: http://kvm.sourceforge.net
mailing list: kvm-devel@lists.sourceforge.net
(http://lists.sourceforge.net/lists/listinfo/kvm-devel)
The following patchset adds a driver for Intel's hardware virtualization
extensions to the x86 architecture. The driver adds a character device
(/dev/kvm) that exposes the virtualization capabilities to userspace. Using
this driver, a process can run a virtual machine (a "guest") in a fully
virtualized PC containing its own virtual hard disks, network adapters, and
display.
Using this driver, one can start multiple virtual machines on a host.
Each virtual machine is a process on the host; a virtual cpu is a thread in
that process. kill(1), nice(1), top(1) work as expected. In effect, the
driver adds a third execution mode to the existing two: we now have kernel
mode, user mode, and guest mode. Guest mode has its own address space mapping
guest physical memory (which is accessible to user mode by mmap()ing
/dev/kvm). Guest mode has no access to any I/O devices; any such access is
intercepted and directed to user mode for emulation.
The driver supports i386 and x86_64 hosts and guests. All combinations are
allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae
and non-pae paging modes are supported.
SMP hosts and UP guests are supported. At the moment only Intel
hardware is supported, but AMD virtualization support is being worked on.
Performance currently is non-stellar due to the naive implementation of the
mmu virtualization, which throws away most of the shadow page table entries
every context switch. We plan to address this in two ways:
- cache shadow page tables across tlb flushes
- wait until AMD and Intel release processors with nested page tables
Currently a virtual desktop is responsive but consumes a lot of CPU. Under
Windows I tried playing pinball and watching a few flash movies; with a recent
CPU one can hardly feel the virtualization. Linux/X is slower, probably due
to X being in a separate process.
In addition to the driver, you need a slightly modified qemu to provide I/O
device emulation and the BIOS.
Caveats (akpm: might no longer be true):
- The Windows install currently bluescreens due to a problem with the
virtual APIC. We are working on a fix. A temporary workaround is to
use an existing image or install through qemu
- Windows 64-bit does not work. That's also true for qemu, so it's
probably a problem with the device model.
[bero@arklinux.org: build fix]
[simon.kagstrom@bth.se: build fix, other fixes]
[uril@qumranet.com: KVM: Expose interrupt bitmap]
[akpm@osdl.org: i386 build fix]
[mingo@elte.hu: i386 fixes]
[rdreier@cisco.com: add log levels to all printks]
[randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings]
[anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support]
Signed-off-by: Yaniv Kamay <yaniv@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Cc: Simon Kagstrom <simon.kagstrom@bth.se>
Cc: Bernhard Rosenkraenzer <bero@arklinux.org>
Signed-off-by: Uri Lublin <uril@qumranet.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Anthony Liguori <anthony@codemonkey.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-10 17:21:36 +07:00
|
|
|
#endif
|