[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 10:21:36 +00:00
|
|
|
/*
|
|
|
|
* Kernel-based Virtual Machine driver for Linux
|
|
|
|
*
|
|
|
|
* AMD SVM support
|
|
|
|
*
|
|
|
|
* Copyright (C) 2006 Qumranet, Inc.
|
2010-10-06 12:23:22 +00:00
|
|
|
* Copyright 2010 Red Hat, Inc. and/or its affiliates.
|
[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 10:21:36 +00:00
|
|
|
*
|
|
|
|
* Authors:
|
|
|
|
* Yaniv Kamay <yaniv@qumranet.com>
|
|
|
|
* Avi Kivity <avi@qumranet.com>
|
|
|
|
*
|
|
|
|
* This work is licensed under the terms of the GNU GPL, version 2. See
|
|
|
|
* the COPYING file in the top-level directory.
|
|
|
|
*
|
|
|
|
*/
|
2007-12-16 09:02:48 +00:00
|
|
|
#include <linux/kvm_host.h>
|
|
|
|
|
2007-07-06 09:20:49 +00:00
|
|
|
#include "irq.h"
|
2007-12-14 01:35:10 +00:00
|
|
|
#include "mmu.h"
|
2008-06-27 17:58:02 +00:00
|
|
|
#include "kvm_cache_regs.h"
|
2009-03-23 09:23:18 +00:00
|
|
|
#include "x86.h"
|
2012-12-05 14:26:19 +00:00
|
|
|
#include "cpuid.h"
|
2007-06-28 18:15:57 +00:00
|
|
|
|
[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 10:21:36 +00:00
|
|
|
#include <linux/module.h>
|
2012-03-28 18:32:28 +00:00
|
|
|
#include <linux/mod_devicetable.h>
|
2007-02-19 12:37:46 +00:00
|
|
|
#include <linux/kernel.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 10:21:36 +00:00
|
|
|
#include <linux/vmalloc.h>
|
|
|
|
#include <linux/highmem.h>
|
Detach sched.h from mm.h
First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.
This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
getting them indirectly
Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
they don't need sched.h
b) sched.h stops being dependency for significant number of files:
on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
after patch it's only 3744 (-8.3%).
Cross-compile tested on
all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
alpha alpha-up
arm
i386 i386-up i386-defconfig i386-allnoconfig
ia64 ia64-up
m68k
mips
parisc parisc-up
powerpc powerpc-up
s390 s390-up
sparc sparc-up
sparc64 sparc64-up
um-x86_64
x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig
as well as my two usual configs.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-20 21:22:52 +00:00
|
|
|
#include <linux/sched.h>
|
2009-06-17 12:22:14 +00:00
|
|
|
#include <linux/ftrace_event.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
|
|
|
#include <linux/slab.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 10:21:36 +00:00
|
|
|
|
2012-02-29 13:57:32 +00:00
|
|
|
#include <asm/perf_event.h>
|
2010-05-17 12:43:35 +00:00
|
|
|
#include <asm/tlbflush.h>
|
2007-06-28 18:15:57 +00:00
|
|
|
#include <asm/desc.h>
|
2014-02-21 09:32:27 +00:00
|
|
|
#include <asm/debugreg.h>
|
2010-10-14 09:22:52 +00:00
|
|
|
#include <asm/kvm_para.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 10:21:36 +00:00
|
|
|
|
2008-11-17 21:03:20 +00:00
|
|
|
#include <asm/virtext.h>
|
2009-06-17 12:22:14 +00:00
|
|
|
#include "trace.h"
|
2008-11-17 21:03:20 +00:00
|
|
|
|
2008-05-13 10:23:38 +00:00
|
|
|
#define __ex(x) __kvm_handle_fault_on_reboot(x)
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
MODULE_AUTHOR("Qumranet");
|
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
|
2012-03-28 18:32:28 +00:00
|
|
|
static const struct x86_cpu_id svm_cpu_id[] = {
|
|
|
|
X86_FEATURE_MATCH(X86_FEATURE_SVM),
|
|
|
|
{}
|
|
|
|
};
|
|
|
|
MODULE_DEVICE_TABLE(x86cpu, svm_cpu_id);
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
#define IOPM_ALLOC_ORDER 2
|
|
|
|
#define MSRPM_ALLOC_ORDER 1
|
|
|
|
|
|
|
|
#define SEG_TYPE_LDT 2
|
|
|
|
#define SEG_TYPE_BUSY_TSS16 3
|
|
|
|
|
KVM: SVM: implement NEXTRIPsave SVM feature
On SVM we set the instruction length of skipped instructions
to hard-coded, well known values, which could be wrong when (bogus,
but valid) prefixes (REX, segment override) are used.
Newer AMD processors (Fam10h 45nm and better, aka. PhenomII or
AthlonII) have an explicit NEXTRIP field in the VMCB containing the
desired information.
Since it is cheap to do so, we use this field to override the guessed
value on newer processors.
A fix for older CPUs would be rather expensive, as it would require
to fetch and partially decode the instruction. As the problem is not
a security issue and needs special, handcrafted code to trigger
(no compiler will ever generate such code), I omit a fix for older
CPUs.
If someone is interested, I have both a patch for these CPUs as well as
demo code triggering this issue: It segfaults under KVM, but runs
perfectly on native Linux.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-04-11 21:07:28 +00:00
|
|
|
#define SVM_FEATURE_NPT (1 << 0)
|
|
|
|
#define SVM_FEATURE_LBRV (1 << 1)
|
|
|
|
#define SVM_FEATURE_SVML (1 << 2)
|
|
|
|
#define SVM_FEATURE_NRIP (1 << 3)
|
2010-12-21 10:12:03 +00:00
|
|
|
#define SVM_FEATURE_TSC_RATE (1 << 4)
|
|
|
|
#define SVM_FEATURE_VMCB_CLEAN (1 << 5)
|
|
|
|
#define SVM_FEATURE_FLUSH_ASID (1 << 6)
|
|
|
|
#define SVM_FEATURE_DECODE_ASSIST (1 << 7)
|
KVM: SVM: implement NEXTRIPsave SVM feature
On SVM we set the instruction length of skipped instructions
to hard-coded, well known values, which could be wrong when (bogus,
but valid) prefixes (REX, segment override) are used.
Newer AMD processors (Fam10h 45nm and better, aka. PhenomII or
AthlonII) have an explicit NEXTRIP field in the VMCB containing the
desired information.
Since it is cheap to do so, we use this field to override the guessed
value on newer processors.
A fix for older CPUs would be rather expensive, as it would require
to fetch and partially decode the instruction. As the problem is not
a security issue and needs special, handcrafted code to trigger
(no compiler will ever generate such code), I omit a fix for older
CPUs.
If someone is interested, I have both a patch for these CPUs as well as
demo code triggering this issue: It segfaults under KVM, but runs
perfectly on native Linux.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-04-11 21:07:28 +00:00
|
|
|
#define SVM_FEATURE_PAUSE_FILTER (1 << 10)
|
2007-03-30 14:02:14 +00:00
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
#define NESTED_EXIT_HOST 0 /* Exit handled on host level */
|
|
|
|
#define NESTED_EXIT_DONE 1 /* Exit caused nested vmexit */
|
|
|
|
#define NESTED_EXIT_CONTINUE 2 /* Further checks needed */
|
|
|
|
|
2008-02-13 17:58:47 +00:00
|
|
|
#define DEBUGCTL_RESERVED_BITS (~(0x3fULL))
|
|
|
|
|
2011-03-25 08:44:46 +00:00
|
|
|
#define TSC_RATIO_RSVD 0xffffff0000000000ULL
|
2011-03-25 08:44:51 +00:00
|
|
|
#define TSC_RATIO_MIN 0x0000000000000001ULL
|
|
|
|
#define TSC_RATIO_MAX 0x000000ffffffffffULL
|
2011-03-25 08:44:46 +00:00
|
|
|
|
2010-05-17 12:43:35 +00:00
|
|
|
static bool erratum_383_found __read_mostly;
|
|
|
|
|
2009-05-31 15:15:37 +00:00
|
|
|
static const u32 host_save_user_msrs[] = {
|
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
MSR_STAR, MSR_LSTAR, MSR_CSTAR, MSR_SYSCALL_MASK, MSR_KERNEL_GS_BASE,
|
|
|
|
MSR_FS_BASE,
|
|
|
|
#endif
|
|
|
|
MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
|
|
|
|
};
|
|
|
|
|
|
|
|
#define NR_HOST_SAVE_USER_MSRS ARRAY_SIZE(host_save_user_msrs)
|
|
|
|
|
|
|
|
struct kvm_vcpu;
|
|
|
|
|
2009-08-07 09:49:33 +00:00
|
|
|
struct nested_state {
|
|
|
|
struct vmcb *hsave;
|
|
|
|
u64 hsave_msr;
|
2010-02-24 17:59:15 +00:00
|
|
|
u64 vm_cr_msr;
|
2009-08-07 09:49:33 +00:00
|
|
|
u64 vmcb;
|
|
|
|
|
|
|
|
/* These are the merged vectors */
|
|
|
|
u32 *msrpm;
|
|
|
|
|
|
|
|
/* gpa pointers to the real vectors */
|
|
|
|
u64 vmcb_msrpm;
|
2010-03-01 14:34:39 +00:00
|
|
|
u64 vmcb_iopm;
|
2009-08-07 09:49:34 +00:00
|
|
|
|
2009-10-09 14:08:26 +00:00
|
|
|
/* A VMEXIT is required but not yet emulated */
|
|
|
|
bool exit_required;
|
|
|
|
|
2009-08-07 09:49:34 +00:00
|
|
|
/* cache for intercepts of the guest */
|
2010-12-03 09:50:51 +00:00
|
|
|
u32 intercept_cr;
|
2010-11-30 17:03:58 +00:00
|
|
|
u32 intercept_dr;
|
2009-08-07 09:49:34 +00:00
|
|
|
u32 intercept_exceptions;
|
|
|
|
u64 intercept;
|
|
|
|
|
2010-09-10 15:31:02 +00:00
|
|
|
/* Nested Paging related state */
|
|
|
|
u64 nested_cr3;
|
2009-08-07 09:49:33 +00:00
|
|
|
};
|
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
#define MSRPM_OFFSETS 16
|
|
|
|
static u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly;
|
|
|
|
|
2012-01-09 19:00:35 +00:00
|
|
|
/*
|
|
|
|
* Set osvw_len to higher value when updated Revision Guides
|
|
|
|
* are published and we know what the new status bits are
|
|
|
|
*/
|
|
|
|
static uint64_t osvw_len = 4, osvw_status;
|
|
|
|
|
2009-05-31 15:15:37 +00:00
|
|
|
struct vcpu_svm {
|
|
|
|
struct kvm_vcpu vcpu;
|
|
|
|
struct vmcb *vmcb;
|
|
|
|
unsigned long vmcb_pa;
|
|
|
|
struct svm_cpu_data *svm_data;
|
|
|
|
uint64_t asid_generation;
|
|
|
|
uint64_t sysenter_esp;
|
|
|
|
uint64_t sysenter_eip;
|
|
|
|
|
|
|
|
u64 next_rip;
|
|
|
|
|
|
|
|
u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS];
|
2010-10-21 10:20:32 +00:00
|
|
|
struct {
|
2010-10-21 10:20:33 +00:00
|
|
|
u16 fs;
|
|
|
|
u16 gs;
|
|
|
|
u16 ldt;
|
2010-10-21 10:20:32 +00:00
|
|
|
u64 gs_base;
|
|
|
|
} host;
|
2009-05-31 15:15:37 +00:00
|
|
|
|
|
|
|
u32 *msrpm;
|
|
|
|
|
KVM: SVM: check for progress after IRET interception
When we enable an NMI window, we ask for an IRET intercept, since
the IRET re-enables NMIs. However, the IRET intercept happens before
the instruction executes, while the NMI window architecturally opens
afterwards.
To compensate for this mismatch, we only open the NMI window in the
following exit, assuming that the IRET has by then executed; however,
this assumption is not always correct; we may exit due to a host interrupt
or page fault, without having executed the instruction.
Fix by checking for forward progress by recording and comparing the IRET's
rip. This is somewhat of a hack, since an unchaging rip does not mean that
no forward progress has been made, but is the simplest fix for now.
Signed-off-by: Avi Kivity <avi@redhat.com>
2011-02-03 13:29:52 +00:00
|
|
|
ulong nmi_iret_rip;
|
|
|
|
|
2009-08-07 09:49:33 +00:00
|
|
|
struct nested_state nested;
|
2009-10-18 11:24:54 +00:00
|
|
|
|
|
|
|
bool nmi_singlestep;
|
2010-02-23 16:47:56 +00:00
|
|
|
|
|
|
|
unsigned int3_injected;
|
|
|
|
unsigned long int3_rip;
|
2010-10-14 09:22:52 +00:00
|
|
|
u32 apf_reason;
|
2011-03-25 08:44:46 +00:00
|
|
|
|
|
|
|
u64 tsc_ratio;
|
2009-05-31 15:15:37 +00:00
|
|
|
};
|
|
|
|
|
2011-03-25 08:44:46 +00:00
|
|
|
static DEFINE_PER_CPU(u64, current_tsc_ratio);
|
|
|
|
#define TSC_RATIO_DEFAULT 0x0100000000ULL
|
|
|
|
|
2010-03-01 14:34:35 +00:00
|
|
|
#define MSR_INVALID 0xffffffffU
|
|
|
|
|
2012-08-29 23:30:20 +00:00
|
|
|
static const struct svm_direct_access_msrs {
|
2010-03-01 14:34:36 +00:00
|
|
|
u32 index; /* Index of the MSR */
|
|
|
|
bool always; /* True if intercept is always on */
|
|
|
|
} direct_access_msrs[] = {
|
2010-07-17 13:03:26 +00:00
|
|
|
{ .index = MSR_STAR, .always = true },
|
2010-03-01 14:34:36 +00:00
|
|
|
{ .index = MSR_IA32_SYSENTER_CS, .always = true },
|
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
{ .index = MSR_GS_BASE, .always = true },
|
|
|
|
{ .index = MSR_FS_BASE, .always = true },
|
|
|
|
{ .index = MSR_KERNEL_GS_BASE, .always = true },
|
|
|
|
{ .index = MSR_LSTAR, .always = true },
|
|
|
|
{ .index = MSR_CSTAR, .always = true },
|
|
|
|
{ .index = MSR_SYSCALL_MASK, .always = true },
|
|
|
|
#endif
|
|
|
|
{ .index = MSR_IA32_LASTBRANCHFROMIP, .always = false },
|
|
|
|
{ .index = MSR_IA32_LASTBRANCHTOIP, .always = false },
|
|
|
|
{ .index = MSR_IA32_LASTINTFROMIP, .always = false },
|
|
|
|
{ .index = MSR_IA32_LASTINTTOIP, .always = false },
|
|
|
|
{ .index = MSR_INVALID, .always = false },
|
2009-05-31 15:15:37 +00:00
|
|
|
};
|
|
|
|
|
2008-02-07 12:47:45 +00:00
|
|
|
/* enable NPT for AMD64 and X86 with PAE */
|
|
|
|
#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
|
|
|
|
static bool npt_enabled = true;
|
|
|
|
#else
|
2010-02-24 17:59:10 +00:00
|
|
|
static bool npt_enabled;
|
2008-02-07 12:47:45 +00:00
|
|
|
#endif
|
2008-02-07 12:47:40 +00:00
|
|
|
|
2012-01-17 13:09:50 +00:00
|
|
|
/* allow nested paging (virtualized MMU) for all guests */
|
|
|
|
static int npt = true;
|
2008-02-07 12:47:40 +00:00
|
|
|
module_param(npt, int, S_IRUGO);
|
2008-02-07 12:47:39 +00:00
|
|
|
|
2012-01-17 13:09:50 +00:00
|
|
|
/* allow nested virtualization in KVM/SVM */
|
|
|
|
static int nested = true;
|
2008-11-25 19:17:10 +00:00
|
|
|
module_param(nested, int, S_IRUGO);
|
|
|
|
|
2008-08-27 12:18:43 +00:00
|
|
|
static void svm_flush_tlb(struct kvm_vcpu *vcpu);
|
2009-08-07 09:49:32 +00:00
|
|
|
static void svm_complete_interrupts(struct vcpu_svm *svm);
|
2007-09-10 15:10:54 +00:00
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
static int nested_svm_exit_handled(struct vcpu_svm *svm);
|
2010-02-19 15:23:02 +00:00
|
|
|
static int nested_svm_intercept(struct vcpu_svm *svm);
|
2008-11-25 19:17:08 +00:00
|
|
|
static int nested_svm_vmexit(struct vcpu_svm *svm);
|
|
|
|
static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
|
|
|
|
bool has_error_code, u32 error_code);
|
2011-03-25 08:44:51 +00:00
|
|
|
static u64 __scale_tsc(u64 ratio, u64 tsc);
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2010-12-03 12:15:21 +00:00
|
|
|
enum {
|
2010-12-03 10:45:49 +00:00
|
|
|
VMCB_INTERCEPTS, /* Intercept vectors, TSC offset,
|
|
|
|
pause filter count */
|
2010-12-03 10:45:50 +00:00
|
|
|
VMCB_PERM_MAP, /* IOPM Base and MSRPM Base */
|
2010-12-03 10:45:51 +00:00
|
|
|
VMCB_ASID, /* ASID */
|
2010-12-03 10:45:52 +00:00
|
|
|
VMCB_INTR, /* int_ctl, int_vector */
|
2010-12-03 10:45:53 +00:00
|
|
|
VMCB_NPT, /* npt_en, nCR3, gPAT */
|
2010-12-03 10:45:54 +00:00
|
|
|
VMCB_CR, /* CR0, CR3, CR4, EFER */
|
2010-12-03 10:45:55 +00:00
|
|
|
VMCB_DR, /* DR6, DR7 */
|
2010-12-03 10:45:56 +00:00
|
|
|
VMCB_DT, /* GDT, IDT */
|
2010-12-03 10:45:57 +00:00
|
|
|
VMCB_SEG, /* CS, DS, SS, ES, CPL */
|
2010-12-03 10:45:58 +00:00
|
|
|
VMCB_CR2, /* CR2 only */
|
2010-12-03 10:45:59 +00:00
|
|
|
VMCB_LBR, /* DBGCTL, BR_FROM, BR_TO, LAST_EX_FROM, LAST_EX_TO */
|
2010-12-03 12:15:21 +00:00
|
|
|
VMCB_DIRTY_MAX,
|
|
|
|
};
|
|
|
|
|
2010-12-03 10:45:58 +00:00
|
|
|
/* TPR and CR2 are always written before VMRUN */
|
|
|
|
#define VMCB_ALWAYS_DIRTY_MASK ((1U << VMCB_INTR) | (1U << VMCB_CR2))
|
2010-12-03 12:15:21 +00:00
|
|
|
|
|
|
|
static inline void mark_all_dirty(struct vmcb *vmcb)
|
|
|
|
{
|
|
|
|
vmcb->control.clean = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void mark_all_clean(struct vmcb *vmcb)
|
|
|
|
{
|
|
|
|
vmcb->control.clean = ((1 << VMCB_DIRTY_MAX) - 1)
|
|
|
|
& ~VMCB_ALWAYS_DIRTY_MASK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void mark_dirty(struct vmcb *vmcb, int bit)
|
|
|
|
{
|
|
|
|
vmcb->control.clean &= ~(1 << bit);
|
|
|
|
}
|
|
|
|
|
2007-07-27 12:13:10 +00:00
|
|
|
static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2007-07-27 07:16:56 +00:00
|
|
|
return container_of(vcpu, struct vcpu_svm, vcpu);
|
2007-07-27 12:13:10 +00:00
|
|
|
}
|
|
|
|
|
2010-11-30 17:03:56 +00:00
|
|
|
static void recalc_intercepts(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
struct vmcb_control_area *c, *h;
|
|
|
|
struct nested_state *g;
|
|
|
|
|
2010-12-03 10:45:49 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
|
|
|
|
|
2010-11-30 17:03:56 +00:00
|
|
|
if (!is_guest_mode(&svm->vcpu))
|
|
|
|
return;
|
|
|
|
|
|
|
|
c = &svm->vmcb->control;
|
|
|
|
h = &svm->nested.hsave->control;
|
|
|
|
g = &svm->nested;
|
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
c->intercept_cr = h->intercept_cr | g->intercept_cr;
|
2010-11-30 17:03:58 +00:00
|
|
|
c->intercept_dr = h->intercept_dr | g->intercept_dr;
|
2010-11-30 17:03:56 +00:00
|
|
|
c->intercept_exceptions = h->intercept_exceptions | g->intercept_exceptions;
|
|
|
|
c->intercept = h->intercept | g->intercept;
|
|
|
|
}
|
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
static inline struct vmcb *get_host_vmcb(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
if (is_guest_mode(&svm->vcpu))
|
|
|
|
return svm->nested.hsave;
|
|
|
|
else
|
|
|
|
return svm->vmcb;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_cr_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept_cr |= (1U << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void clr_cr_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept_cr &= ~(1U << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool is_cr_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
return vmcb->control.intercept_cr & (1U << bit);
|
|
|
|
}
|
|
|
|
|
2014-03-03 12:08:29 +00:00
|
|
|
static inline void set_dr_intercepts(struct vcpu_svm *svm)
|
2010-11-30 17:03:58 +00:00
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
2014-03-03 12:08:29 +00:00
|
|
|
vmcb->control.intercept_dr = (1 << INTERCEPT_DR0_READ)
|
|
|
|
| (1 << INTERCEPT_DR1_READ)
|
|
|
|
| (1 << INTERCEPT_DR2_READ)
|
|
|
|
| (1 << INTERCEPT_DR3_READ)
|
|
|
|
| (1 << INTERCEPT_DR4_READ)
|
|
|
|
| (1 << INTERCEPT_DR5_READ)
|
|
|
|
| (1 << INTERCEPT_DR6_READ)
|
|
|
|
| (1 << INTERCEPT_DR7_READ)
|
|
|
|
| (1 << INTERCEPT_DR0_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR1_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR2_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR3_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR4_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR5_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR6_WRITE)
|
|
|
|
| (1 << INTERCEPT_DR7_WRITE);
|
2010-11-30 17:03:58 +00:00
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
2014-03-03 12:08:29 +00:00
|
|
|
static inline void clr_dr_intercepts(struct vcpu_svm *svm)
|
2010-11-30 17:03:58 +00:00
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
2014-03-03 12:08:29 +00:00
|
|
|
vmcb->control.intercept_dr = 0;
|
2010-11-30 17:03:58 +00:00
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
2010-11-30 17:03:59 +00:00
|
|
|
static inline void set_exception_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept_exceptions |= (1U << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void clr_exception_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept_exceptions &= ~(1U << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
2010-11-30 17:04:00 +00:00
|
|
|
static inline void set_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept |= (1ULL << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void clr_intercept(struct vcpu_svm *svm, int bit)
|
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(svm);
|
|
|
|
|
|
|
|
vmcb->control.intercept &= ~(1ULL << bit);
|
|
|
|
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
static inline void enable_gif(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
svm->vcpu.arch.hflags |= HF_GIF_MASK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void disable_gif(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
svm->vcpu.arch.hflags &= ~HF_GIF_MASK;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool gif_set(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
return !!(svm->vcpu.arch.hflags & HF_GIF_MASK);
|
|
|
|
}
|
|
|
|
|
2008-02-19 18:32:02 +00:00
|
|
|
static unsigned long iopm_base;
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
struct kvm_ldttss_desc {
|
|
|
|
u16 limit0;
|
|
|
|
u16 base0;
|
2010-02-24 17:59:10 +00:00
|
|
|
unsigned base1:8, type:5, dpl:2, p:1;
|
|
|
|
unsigned limit1:4, zero0:3, g:1, base2:8;
|
[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 10:21:36 +00:00
|
|
|
u32 base3;
|
|
|
|
u32 zero1;
|
|
|
|
} __attribute__((packed));
|
|
|
|
|
|
|
|
struct svm_cpu_data {
|
|
|
|
int cpu;
|
|
|
|
|
2007-04-02 10:05:50 +00:00
|
|
|
u64 asid_generation;
|
|
|
|
u32 max_asid;
|
|
|
|
u32 next_asid;
|
[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 10:21:36 +00:00
|
|
|
struct kvm_ldttss_desc *tss_desc;
|
|
|
|
|
|
|
|
struct page *save_area;
|
|
|
|
};
|
|
|
|
|
|
|
|
static DEFINE_PER_CPU(struct svm_cpu_data *, svm_data);
|
|
|
|
|
|
|
|
struct svm_init_data {
|
|
|
|
int cpu;
|
|
|
|
int r;
|
|
|
|
};
|
|
|
|
|
2012-08-29 23:30:20 +00:00
|
|
|
static const u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000};
|
[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 10:21:36 +00:00
|
|
|
|
2007-02-19 12:37:46 +00:00
|
|
|
#define NUM_MSR_MAPS ARRAY_SIZE(msrpm_ranges)
|
[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 10:21:36 +00:00
|
|
|
#define MSRS_RANGE_SIZE 2048
|
|
|
|
#define MSRS_IN_RANGE (MSRS_RANGE_SIZE * 8 / 2)
|
|
|
|
|
2010-03-01 14:34:35 +00:00
|
|
|
static u32 svm_msrpm_offset(u32 msr)
|
|
|
|
{
|
|
|
|
u32 offset;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < NUM_MSR_MAPS; i++) {
|
|
|
|
if (msr < msrpm_ranges[i] ||
|
|
|
|
msr >= msrpm_ranges[i] + MSRS_IN_RANGE)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
offset = (msr - msrpm_ranges[i]) / 4; /* 4 msrs per u8 */
|
|
|
|
offset += (i * MSRS_RANGE_SIZE); /* add range offset */
|
|
|
|
|
|
|
|
/* Now we have the u8 offset - but need the u32 offset */
|
|
|
|
return offset / 4;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* MSR not in any range */
|
|
|
|
return MSR_INVALID;
|
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
#define MAX_INST_SIZE 15
|
|
|
|
|
|
|
|
static inline void clgi(void)
|
|
|
|
{
|
2008-05-13 10:23:38 +00:00
|
|
|
asm volatile (__ex(SVM_CLGI));
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void stgi(void)
|
|
|
|
{
|
2008-05-13 10:23:38 +00:00
|
|
|
asm volatile (__ex(SVM_STGI));
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void invlpga(unsigned long addr, u32 asid)
|
|
|
|
{
|
2010-02-24 17:59:10 +00:00
|
|
|
asm volatile (__ex(SVM_INVLPGA) : : "a"(addr), "c"(asid));
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-09-10 15:31:03 +00:00
|
|
|
static int get_npt_level(void)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
return PT64_ROOT_LEVEL;
|
|
|
|
#else
|
|
|
|
return PT32E_ROOT_LEVEL;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer)
|
|
|
|
{
|
2010-05-27 01:09:43 +00:00
|
|
|
vcpu->arch.efer = efer;
|
2008-02-07 12:47:45 +00:00
|
|
|
if (!npt_enabled && !(efer & EFER_LMA))
|
2007-12-01 12:17:11 +00:00
|
|
|
efer &= ~EFER_LME;
|
[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 10:21:36 +00:00
|
|
|
|
2008-11-25 19:17:02 +00:00
|
|
|
to_svm(vcpu)->vmcb->save.efer = efer | EFER_SVME;
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(to_svm(vcpu)->vmcb, VMCB_CR);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int is_external_interrupt(u32 info)
|
|
|
|
{
|
|
|
|
info &= SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID;
|
|
|
|
return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR);
|
|
|
|
}
|
|
|
|
|
2009-05-12 20:21:05 +00:00
|
|
|
static u32 svm_get_interrupt_shadow(struct kvm_vcpu *vcpu, int mask)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u32 ret = 0;
|
|
|
|
|
|
|
|
if (svm->vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK)
|
2010-02-19 18:38:07 +00:00
|
|
|
ret |= KVM_X86_SHADOW_INT_STI | KVM_X86_SHADOW_INT_MOV_SS;
|
2009-05-12 20:21:05 +00:00
|
|
|
return ret & mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_interrupt_shadow(struct kvm_vcpu *vcpu, int mask)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
if (mask == 0)
|
|
|
|
svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
|
|
|
|
else
|
|
|
|
svm->vmcb->control.int_state |= SVM_INTERRUPT_SHADOW_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 10:21:36 +00:00
|
|
|
static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
KVM: SVM: implement NEXTRIPsave SVM feature
On SVM we set the instruction length of skipped instructions
to hard-coded, well known values, which could be wrong when (bogus,
but valid) prefixes (REX, segment override) are used.
Newer AMD processors (Fam10h 45nm and better, aka. PhenomII or
AthlonII) have an explicit NEXTRIP field in the VMCB containing the
desired information.
Since it is cheap to do so, we use this field to override the guessed
value on newer processors.
A fix for older CPUs would be rather expensive, as it would require
to fetch and partially decode the instruction. As the problem is not
a security issue and needs special, handcrafted code to trigger
(no compiler will ever generate such code), I omit a fix for older
CPUs.
If someone is interested, I have both a patch for these CPUs as well as
demo code triggering this issue: It segfaults under KVM, but runs
perfectly on native Linux.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-04-11 21:07:28 +00:00
|
|
|
if (svm->vmcb->control.next_rip != 0)
|
|
|
|
svm->next_rip = svm->vmcb->control.next_rip;
|
|
|
|
|
2007-07-27 12:13:10 +00:00
|
|
|
if (!svm->next_rip) {
|
2010-12-21 10:12:02 +00:00
|
|
|
if (emulate_instruction(vcpu, EMULTYPE_SKIP) !=
|
2009-05-11 10:35:49 +00:00
|
|
|
EMULATE_DONE)
|
|
|
|
printk(KERN_DEBUG "%s: NOP\n", __func__);
|
[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 10:21:36 +00:00
|
|
|
return;
|
|
|
|
}
|
2008-06-27 17:58:02 +00:00
|
|
|
if (svm->next_rip - kvm_rip_read(vcpu) > MAX_INST_SIZE)
|
|
|
|
printk(KERN_ERR "%s: ip 0x%lx next 0x%llx\n",
|
|
|
|
__func__, kvm_rip_read(vcpu), svm->next_rip);
|
[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 10:21:36 +00:00
|
|
|
|
2008-06-27 17:58:02 +00:00
|
|
|
kvm_rip_write(vcpu, svm->next_rip);
|
2009-05-12 20:21:05 +00:00
|
|
|
svm_set_interrupt_shadow(vcpu, 0);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-02-23 16:47:54 +00:00
|
|
|
static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
|
2010-04-22 10:33:13 +00:00
|
|
|
bool has_error_code, u32 error_code,
|
|
|
|
bool reinject)
|
2010-02-23 16:47:54 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* If we are within a nested VM we'd better #VMEXIT and let the guest
|
|
|
|
* handle the exception
|
|
|
|
*/
|
2010-04-22 10:33:13 +00:00
|
|
|
if (!reinject &&
|
|
|
|
nested_svm_check_exception(svm, nr, has_error_code, error_code))
|
2010-02-23 16:47:54 +00:00
|
|
|
return;
|
|
|
|
|
2010-11-09 14:15:42 +00:00
|
|
|
if (nr == BP_VECTOR && !static_cpu_has(X86_FEATURE_NRIPS)) {
|
2010-02-23 16:47:56 +00:00
|
|
|
unsigned long rip, old_rip = kvm_rip_read(&svm->vcpu);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* For guest debugging where we have to reinject #BP if some
|
|
|
|
* INT3 is guest-owned:
|
|
|
|
* Emulate nRIP by moving RIP forward. Will fail if injection
|
|
|
|
* raises a fault that is not intercepted. Still better than
|
|
|
|
* failing in all cases.
|
|
|
|
*/
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
rip = kvm_rip_read(&svm->vcpu);
|
|
|
|
svm->int3_rip = rip + svm->vmcb->save.cs.base;
|
|
|
|
svm->int3_injected = rip - old_rip;
|
|
|
|
}
|
|
|
|
|
2010-02-23 16:47:54 +00:00
|
|
|
svm->vmcb->control.event_inj = nr
|
|
|
|
| SVM_EVTINJ_VALID
|
|
|
|
| (has_error_code ? SVM_EVTINJ_VALID_ERR : 0)
|
|
|
|
| SVM_EVTINJ_TYPE_EXEPT;
|
|
|
|
svm->vmcb->control.event_inj_err = error_code;
|
|
|
|
}
|
|
|
|
|
2010-05-17 12:43:35 +00:00
|
|
|
static void svm_init_erratum_383(void)
|
|
|
|
{
|
|
|
|
u32 low, high;
|
|
|
|
int err;
|
|
|
|
u64 val;
|
|
|
|
|
2013-03-20 14:07:27 +00:00
|
|
|
if (!static_cpu_has_bug(X86_BUG_AMD_TLB_MMATCH))
|
2010-05-17 12:43:35 +00:00
|
|
|
return;
|
|
|
|
|
|
|
|
/* Use _safe variants to not break nested virtualization */
|
|
|
|
val = native_read_msr_safe(MSR_AMD64_DC_CFG, &err);
|
|
|
|
if (err)
|
|
|
|
return;
|
|
|
|
|
|
|
|
val |= (1ULL << 47);
|
|
|
|
|
|
|
|
low = lower_32_bits(val);
|
|
|
|
high = upper_32_bits(val);
|
|
|
|
|
|
|
|
native_write_msr_safe(MSR_AMD64_DC_CFG, low, high);
|
|
|
|
|
|
|
|
erratum_383_found = true;
|
|
|
|
}
|
|
|
|
|
2012-01-09 19:00:35 +00:00
|
|
|
static void svm_init_osvw(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Guests should see errata 400 and 415 as fixed (assuming that
|
|
|
|
* HLT and IO instructions are intercepted).
|
|
|
|
*/
|
|
|
|
vcpu->arch.osvw.length = (osvw_len >= 3) ? (osvw_len) : 3;
|
|
|
|
vcpu->arch.osvw.status = osvw_status & ~(6ULL);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* By increasing VCPU's osvw.length to 3 we are telling the guest that
|
|
|
|
* all osvw.status bits inside that length, including bit 0 (which is
|
|
|
|
* reserved for erratum 298), are valid. However, if host processor's
|
|
|
|
* osvw_len is 0 then osvw_status[0] carries no information. We need to
|
|
|
|
* be conservative here and therefore we tell the guest that erratum 298
|
|
|
|
* is present (because we really don't know).
|
|
|
|
*/
|
|
|
|
if (osvw_len == 0 && boot_cpu_data.x86 == 0x10)
|
|
|
|
vcpu->arch.osvw.status |= 1;
|
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static int has_svm(void)
|
|
|
|
{
|
2008-11-17 21:03:20 +00:00
|
|
|
const char *msg;
|
[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 10:21:36 +00:00
|
|
|
|
2008-11-17 21:03:20 +00:00
|
|
|
if (!cpu_has_svm(&msg)) {
|
2009-01-08 19:05:17 +00:00
|
|
|
printk(KERN_INFO "has_svm: %s\n", msg);
|
[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 10:21:36 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_hardware_disable(void *garbage)
|
|
|
|
{
|
2011-03-25 08:44:46 +00:00
|
|
|
/* Make sure we clean up behind us */
|
|
|
|
if (static_cpu_has(X86_FEATURE_TSCRATEMSR))
|
|
|
|
wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT);
|
|
|
|
|
2008-11-17 21:03:21 +00:00
|
|
|
cpu_svm_disable();
|
2012-02-29 13:57:32 +00:00
|
|
|
|
|
|
|
amd_pmu_disable_virt();
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-09-15 09:37:46 +00:00
|
|
|
static int svm_hardware_enable(void *garbage)
|
[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 10:21:36 +00:00
|
|
|
{
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
struct svm_cpu_data *sd;
|
[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 10:21:36 +00:00
|
|
|
uint64_t efer;
|
2010-02-16 08:51:48 +00:00
|
|
|
struct desc_ptr gdt_descr;
|
[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 10:21:36 +00:00
|
|
|
struct desc_struct *gdt;
|
|
|
|
int me = raw_smp_processor_id();
|
|
|
|
|
2009-09-15 09:37:46 +00:00
|
|
|
rdmsrl(MSR_EFER, efer);
|
|
|
|
if (efer & EFER_SVME)
|
|
|
|
return -EBUSY;
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
if (!has_svm()) {
|
2012-10-20 18:20:04 +00:00
|
|
|
pr_err("%s: err EOPNOTSUPP on %d\n", __func__, me);
|
2009-09-15 09:37:46 +00:00
|
|
|
return -EINVAL;
|
[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 10:21:36 +00:00
|
|
|
}
|
2009-10-29 13:34:14 +00:00
|
|
|
sd = per_cpu(svm_data, me);
|
|
|
|
if (!sd) {
|
2012-10-20 18:20:04 +00:00
|
|
|
pr_err("%s: svm_data is NULL on %d\n", __func__, me);
|
2009-09-15 09:37:46 +00:00
|
|
|
return -EINVAL;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
sd->asid_generation = 1;
|
|
|
|
sd->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1;
|
|
|
|
sd->next_asid = sd->max_asid + 1;
|
[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 10:21:36 +00:00
|
|
|
|
2010-02-25 10:43:07 +00:00
|
|
|
native_store_gdt(&gdt_descr);
|
2010-02-16 08:51:48 +00:00
|
|
|
gdt = (struct desc_struct *)gdt_descr.address;
|
2009-10-29 13:34:14 +00:00
|
|
|
sd->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS);
|
[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 10:21:36 +00:00
|
|
|
|
2008-11-25 19:17:02 +00:00
|
|
|
wrmsrl(MSR_EFER, efer | EFER_SVME);
|
[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 10:21:36 +00:00
|
|
|
|
2009-12-14 17:58:24 +00:00
|
|
|
wrmsrl(MSR_VM_HSAVE_PA, page_to_pfn(sd->save_area) << PAGE_SHIFT);
|
2009-09-15 09:37:46 +00:00
|
|
|
|
2011-03-25 08:44:46 +00:00
|
|
|
if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) {
|
|
|
|
wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT);
|
|
|
|
__get_cpu_var(current_tsc_ratio) = TSC_RATIO_DEFAULT;
|
|
|
|
}
|
|
|
|
|
2012-01-09 19:00:35 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Get OSVW bits.
|
|
|
|
*
|
|
|
|
* Note that it is possible to have a system with mixed processor
|
|
|
|
* revisions and therefore different OSVW bits. If bits are not the same
|
|
|
|
* on different processors then choose the worst case (i.e. if erratum
|
|
|
|
* is present on one processor and not on another then assume that the
|
|
|
|
* erratum is present everywhere).
|
|
|
|
*/
|
|
|
|
if (cpu_has(&boot_cpu_data, X86_FEATURE_OSVW)) {
|
|
|
|
uint64_t len, status = 0;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
len = native_read_msr_safe(MSR_AMD64_OSVW_ID_LENGTH, &err);
|
|
|
|
if (!err)
|
|
|
|
status = native_read_msr_safe(MSR_AMD64_OSVW_STATUS,
|
|
|
|
&err);
|
|
|
|
|
|
|
|
if (err)
|
|
|
|
osvw_status = osvw_len = 0;
|
|
|
|
else {
|
|
|
|
if (len < osvw_len)
|
|
|
|
osvw_len = len;
|
|
|
|
osvw_status |= status;
|
|
|
|
osvw_status &= (1ULL << osvw_len) - 1;
|
|
|
|
}
|
|
|
|
} else
|
|
|
|
osvw_status = osvw_len = 0;
|
|
|
|
|
2010-05-17 12:43:35 +00:00
|
|
|
svm_init_erratum_383();
|
|
|
|
|
2012-02-29 13:57:32 +00:00
|
|
|
amd_pmu_enable_virt();
|
|
|
|
|
2009-09-15 09:37:46 +00:00
|
|
|
return 0;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2008-07-02 14:02:11 +00:00
|
|
|
static void svm_cpu_uninit(int cpu)
|
|
|
|
{
|
2009-10-29 13:34:14 +00:00
|
|
|
struct svm_cpu_data *sd = per_cpu(svm_data, raw_smp_processor_id());
|
2008-07-02 14:02:11 +00:00
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
if (!sd)
|
2008-07-02 14:02:11 +00:00
|
|
|
return;
|
|
|
|
|
|
|
|
per_cpu(svm_data, raw_smp_processor_id()) = NULL;
|
2009-10-29 13:34:14 +00:00
|
|
|
__free_page(sd->save_area);
|
|
|
|
kfree(sd);
|
2008-07-02 14:02:11 +00:00
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static int svm_cpu_init(int cpu)
|
|
|
|
{
|
2009-10-29 13:34:14 +00:00
|
|
|
struct svm_cpu_data *sd;
|
[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 10:21:36 +00:00
|
|
|
int r;
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
sd = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL);
|
|
|
|
if (!sd)
|
[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 10:21:36 +00:00
|
|
|
return -ENOMEM;
|
2009-10-29 13:34:14 +00:00
|
|
|
sd->cpu = cpu;
|
|
|
|
sd->save_area = alloc_page(GFP_KERNEL);
|
[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 10:21:36 +00:00
|
|
|
r = -ENOMEM;
|
2009-10-29 13:34:14 +00:00
|
|
|
if (!sd->save_area)
|
[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 10:21:36 +00:00
|
|
|
goto err_1;
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
per_cpu(svm_data, cpu) = sd;
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
err_1:
|
2009-10-29 13:34:14 +00:00
|
|
|
kfree(sd);
|
[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 10:21:36 +00:00
|
|
|
return r;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2010-03-01 14:34:36 +00:00
|
|
|
static bool valid_msr_intercept(u32 index)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++)
|
|
|
|
if (direct_access_msrs[i].index == index)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2007-07-31 10:42:42 +00:00
|
|
|
static void set_msr_interception(u32 *msrpm, unsigned msr,
|
|
|
|
int read, int write)
|
[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 10:21:36 +00:00
|
|
|
{
|
2010-03-01 14:34:35 +00:00
|
|
|
u8 bit_read, bit_write;
|
|
|
|
unsigned long tmp;
|
|
|
|
u32 offset;
|
[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 10:21:36 +00:00
|
|
|
|
2010-03-01 14:34:36 +00:00
|
|
|
/*
|
|
|
|
* If this warning triggers extend the direct_access_msrs list at the
|
|
|
|
* beginning of the file
|
|
|
|
*/
|
|
|
|
WARN_ON(!valid_msr_intercept(msr));
|
|
|
|
|
2010-03-01 14:34:35 +00:00
|
|
|
offset = svm_msrpm_offset(msr);
|
|
|
|
bit_read = 2 * (msr & 0x0f);
|
|
|
|
bit_write = 2 * (msr & 0x0f) + 1;
|
|
|
|
tmp = msrpm[offset];
|
|
|
|
|
|
|
|
BUG_ON(offset == MSR_INVALID);
|
|
|
|
|
|
|
|
read ? clear_bit(bit_read, &tmp) : set_bit(bit_read, &tmp);
|
|
|
|
write ? clear_bit(bit_write, &tmp) : set_bit(bit_write, &tmp);
|
|
|
|
|
|
|
|
msrpm[offset] = tmp;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2008-02-13 17:58:46 +00:00
|
|
|
static void svm_vcpu_init_msrpm(u32 *msrpm)
|
[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 10:21:36 +00:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
2008-02-13 17:58:46 +00:00
|
|
|
memset(msrpm, 0xff, PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER));
|
|
|
|
|
2010-03-01 14:34:36 +00:00
|
|
|
for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) {
|
|
|
|
if (!direct_access_msrs[i].always)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
set_msr_interception(msrpm, direct_access_msrs[i].index, 1, 1);
|
|
|
|
}
|
2008-02-13 17:58:46 +00:00
|
|
|
}
|
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
static void add_msr_offset(u32 offset)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < MSRPM_OFFSETS; ++i) {
|
|
|
|
|
|
|
|
/* Offset already in list? */
|
|
|
|
if (msrpm_offsets[i] == offset)
|
2007-07-31 10:42:42 +00:00
|
|
|
return;
|
2010-03-01 14:34:37 +00:00
|
|
|
|
|
|
|
/* Slot used by another offset? */
|
|
|
|
if (msrpm_offsets[i] != MSR_INVALID)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/* Add offset to list */
|
|
|
|
msrpm_offsets[i] = offset;
|
|
|
|
|
|
|
|
return;
|
[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 10:21:36 +00:00
|
|
|
}
|
2010-03-01 14:34:37 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If this BUG triggers the msrpm_offsets table has an overflow. Just
|
|
|
|
* increase MSRPM_OFFSETS in this case.
|
|
|
|
*/
|
2007-07-31 10:42:42 +00:00
|
|
|
BUG();
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
static void init_msrpm_offsets(void)
|
2008-02-13 17:58:46 +00:00
|
|
|
{
|
2010-03-01 14:34:37 +00:00
|
|
|
int i;
|
2008-02-13 17:58:46 +00:00
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
memset(msrpm_offsets, 0xff, sizeof(msrpm_offsets));
|
|
|
|
|
|
|
|
for (i = 0; direct_access_msrs[i].index != MSR_INVALID; i++) {
|
|
|
|
u32 offset;
|
|
|
|
|
|
|
|
offset = svm_msrpm_offset(direct_access_msrs[i].index);
|
|
|
|
BUG_ON(offset == MSR_INVALID);
|
|
|
|
|
|
|
|
add_msr_offset(offset);
|
|
|
|
}
|
2008-02-13 17:58:46 +00:00
|
|
|
}
|
|
|
|
|
2008-02-13 17:58:47 +00:00
|
|
|
static void svm_enable_lbrv(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
u32 *msrpm = svm->msrpm;
|
|
|
|
|
|
|
|
svm->vmcb->control.lbr_ctl = 1;
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTBRANCHFROMIP, 1, 1);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTINTFROMIP, 1, 1);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTINTTOIP, 1, 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_disable_lbrv(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
u32 *msrpm = svm->msrpm;
|
|
|
|
|
|
|
|
svm->vmcb->control.lbr_ctl = 0;
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTBRANCHFROMIP, 0, 0);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTBRANCHTOIP, 0, 0);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTINTFROMIP, 0, 0);
|
|
|
|
set_msr_interception(msrpm, MSR_IA32_LASTINTTOIP, 0, 0);
|
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static __init int svm_hardware_setup(void)
|
|
|
|
{
|
|
|
|
int cpu;
|
|
|
|
struct page *iopm_pages;
|
2008-02-13 17:58:46 +00:00
|
|
|
void *iopm_va;
|
[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 10:21:36 +00:00
|
|
|
int r;
|
|
|
|
|
|
|
|
iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER);
|
|
|
|
|
|
|
|
if (!iopm_pages)
|
|
|
|
return -ENOMEM;
|
2007-04-30 06:48:11 +00:00
|
|
|
|
|
|
|
iopm_va = page_address(iopm_pages);
|
|
|
|
memset(iopm_va, 0xff, PAGE_SIZE * (1 << IOPM_ALLOC_ORDER));
|
[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 10:21:36 +00:00
|
|
|
iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT;
|
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
init_msrpm_offsets();
|
|
|
|
|
2008-01-31 13:57:38 +00:00
|
|
|
if (boot_cpu_has(X86_FEATURE_NX))
|
|
|
|
kvm_enable_efer_bits(EFER_NX);
|
|
|
|
|
2009-02-02 15:23:51 +00:00
|
|
|
if (boot_cpu_has(X86_FEATURE_FXSR_OPT))
|
|
|
|
kvm_enable_efer_bits(EFER_FFXSR);
|
|
|
|
|
2011-03-25 08:44:51 +00:00
|
|
|
if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) {
|
|
|
|
u64 max;
|
|
|
|
|
|
|
|
kvm_has_tsc_control = true;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Make sure the user can only configure tsc_khz values that
|
|
|
|
* fit into a signed integer.
|
|
|
|
* A min value is not calculated needed because it will always
|
|
|
|
* be 1 on all machines and a value of 0 is used to disable
|
|
|
|
* tsc-scaling for the vcpu.
|
|
|
|
*/
|
|
|
|
max = min(0x7fffffffULL, __scale_tsc(tsc_khz, TSC_RATIO_MAX));
|
|
|
|
|
|
|
|
kvm_max_guest_tsc_khz = max;
|
|
|
|
}
|
|
|
|
|
2008-11-25 19:17:10 +00:00
|
|
|
if (nested) {
|
|
|
|
printk(KERN_INFO "kvm: Nested Virtualization enabled\n");
|
2010-05-05 14:04:44 +00:00
|
|
|
kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE);
|
2008-11-25 19:17:10 +00:00
|
|
|
}
|
|
|
|
|
2009-09-29 21:38:37 +00:00
|
|
|
for_each_possible_cpu(cpu) {
|
[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 10:21:36 +00:00
|
|
|
r = svm_cpu_init(cpu);
|
|
|
|
if (r)
|
2008-02-13 17:58:46 +00:00
|
|
|
goto err;
|
[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 10:21:36 +00:00
|
|
|
}
|
2008-02-07 12:47:38 +00:00
|
|
|
|
2010-11-09 14:15:42 +00:00
|
|
|
if (!boot_cpu_has(X86_FEATURE_NPT))
|
2008-02-07 12:47:39 +00:00
|
|
|
npt_enabled = false;
|
|
|
|
|
2008-02-07 12:47:40 +00:00
|
|
|
if (npt_enabled && !npt) {
|
|
|
|
printk(KERN_INFO "kvm: Nested Paging disabled\n");
|
|
|
|
npt_enabled = false;
|
|
|
|
}
|
|
|
|
|
2008-02-07 12:47:41 +00:00
|
|
|
if (npt_enabled) {
|
2008-02-07 12:47:39 +00:00
|
|
|
printk(KERN_INFO "kvm: Nested Paging enabled\n");
|
2008-02-07 12:47:41 +00:00
|
|
|
kvm_enable_tdp();
|
2008-07-14 18:36:36 +00:00
|
|
|
} else
|
|
|
|
kvm_disable_tdp();
|
2008-02-07 12:47:39 +00:00
|
|
|
|
[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 10:21:36 +00:00
|
|
|
return 0;
|
|
|
|
|
2008-02-13 17:58:46 +00:00
|
|
|
err:
|
[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 10:21:36 +00:00
|
|
|
__free_pages(iopm_pages, IOPM_ALLOC_ORDER);
|
|
|
|
iopm_base = 0;
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __exit void svm_hardware_unsetup(void)
|
|
|
|
{
|
2008-07-02 14:02:11 +00:00
|
|
|
int cpu;
|
|
|
|
|
2009-09-29 21:38:37 +00:00
|
|
|
for_each_possible_cpu(cpu)
|
2008-07-02 14:02:11 +00:00
|
|
|
svm_cpu_uninit(cpu);
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
__free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), IOPM_ALLOC_ORDER);
|
2008-02-13 17:58:46 +00:00
|
|
|
iopm_base = 0;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void init_seg(struct vmcb_seg *seg)
|
|
|
|
{
|
|
|
|
seg->selector = 0;
|
|
|
|
seg->attrib = SVM_SELECTOR_P_MASK | SVM_SELECTOR_S_MASK |
|
2010-02-24 17:59:10 +00:00
|
|
|
SVM_SELECTOR_WRITE_MASK; /* Read/Write Data Segment */
|
[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 10:21:36 +00:00
|
|
|
seg->limit = 0xffff;
|
|
|
|
seg->base = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void init_sys_seg(struct vmcb_seg *seg, uint32_t type)
|
|
|
|
{
|
|
|
|
seg->selector = 0;
|
|
|
|
seg->attrib = SVM_SELECTOR_P_MASK | type;
|
|
|
|
seg->limit = 0xffff;
|
|
|
|
seg->base = 0;
|
|
|
|
}
|
|
|
|
|
2011-03-25 08:44:46 +00:00
|
|
|
static u64 __scale_tsc(u64 ratio, u64 tsc)
|
|
|
|
{
|
|
|
|
u64 mult, frac, _tsc;
|
|
|
|
|
|
|
|
mult = ratio >> 32;
|
|
|
|
frac = ratio & ((1ULL << 32) - 1);
|
|
|
|
|
|
|
|
_tsc = tsc;
|
|
|
|
_tsc *= mult;
|
|
|
|
_tsc += (tsc >> 32) * frac;
|
|
|
|
_tsc += ((tsc & ((1ULL << 32) - 1)) * frac) >> 32;
|
|
|
|
|
|
|
|
return _tsc;
|
|
|
|
}
|
|
|
|
|
|
|
|
static u64 svm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u64 _tsc = tsc;
|
|
|
|
|
|
|
|
if (svm->tsc_ratio != TSC_RATIO_DEFAULT)
|
|
|
|
_tsc = __scale_tsc(svm->tsc_ratio, tsc);
|
|
|
|
|
|
|
|
return _tsc;
|
|
|
|
}
|
|
|
|
|
KVM: Infrastructure for software and hardware based TSC rate scaling
This requires some restructuring; rather than use 'virtual_tsc_khz'
to indicate whether hardware rate scaling is in effect, we consider
each VCPU to always have a virtual TSC rate. Instead, there is new
logic above the vendor-specific hardware scaling that decides whether
it is even necessary to use and updates all rate variables used by
common code. This means we can simply query the virtual rate at
any point, which is needed for software rate scaling.
There is also now a threshold added to the TSC rate scaling; minor
differences and variations of measured TSC rate can accidentally
provoke rate scaling to be used when it is not needed. Instead,
we have a tolerance variable called tsc_tolerance_ppm, which is
the maximum variation from user requested rate at which scaling
will be used. The default is 250ppm, which is the half the
threshold for NTP adjustment, allowing for some hardware variation.
In the event that hardware rate scaling is not available, we can
kludge a bit by forcing TSC catchup to turn on when a faster than
hardware speed has been requested, but there is nothing available
yet for the reverse case; this requires a trap and emulate software
implementation for RDTSC, which is still forthcoming.
[avi: fix 64-bit division on i386]
Signed-off-by: Zachary Amsden <zamsden@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-02-03 17:43:50 +00:00
|
|
|
static void svm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 user_tsc_khz, bool scale)
|
2011-03-25 08:44:49 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u64 ratio;
|
|
|
|
u64 khz;
|
|
|
|
|
KVM: Infrastructure for software and hardware based TSC rate scaling
This requires some restructuring; rather than use 'virtual_tsc_khz'
to indicate whether hardware rate scaling is in effect, we consider
each VCPU to always have a virtual TSC rate. Instead, there is new
logic above the vendor-specific hardware scaling that decides whether
it is even necessary to use and updates all rate variables used by
common code. This means we can simply query the virtual rate at
any point, which is needed for software rate scaling.
There is also now a threshold added to the TSC rate scaling; minor
differences and variations of measured TSC rate can accidentally
provoke rate scaling to be used when it is not needed. Instead,
we have a tolerance variable called tsc_tolerance_ppm, which is
the maximum variation from user requested rate at which scaling
will be used. The default is 250ppm, which is the half the
threshold for NTP adjustment, allowing for some hardware variation.
In the event that hardware rate scaling is not available, we can
kludge a bit by forcing TSC catchup to turn on when a faster than
hardware speed has been requested, but there is nothing available
yet for the reverse case; this requires a trap and emulate software
implementation for RDTSC, which is still forthcoming.
[avi: fix 64-bit division on i386]
Signed-off-by: Zachary Amsden <zamsden@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-02-03 17:43:50 +00:00
|
|
|
/* Guest TSC same frequency as host TSC? */
|
|
|
|
if (!scale) {
|
|
|
|
svm->tsc_ratio = TSC_RATIO_DEFAULT;
|
2011-03-25 08:44:49 +00:00
|
|
|
return;
|
KVM: Infrastructure for software and hardware based TSC rate scaling
This requires some restructuring; rather than use 'virtual_tsc_khz'
to indicate whether hardware rate scaling is in effect, we consider
each VCPU to always have a virtual TSC rate. Instead, there is new
logic above the vendor-specific hardware scaling that decides whether
it is even necessary to use and updates all rate variables used by
common code. This means we can simply query the virtual rate at
any point, which is needed for software rate scaling.
There is also now a threshold added to the TSC rate scaling; minor
differences and variations of measured TSC rate can accidentally
provoke rate scaling to be used when it is not needed. Instead,
we have a tolerance variable called tsc_tolerance_ppm, which is
the maximum variation from user requested rate at which scaling
will be used. The default is 250ppm, which is the half the
threshold for NTP adjustment, allowing for some hardware variation.
In the event that hardware rate scaling is not available, we can
kludge a bit by forcing TSC catchup to turn on when a faster than
hardware speed has been requested, but there is nothing available
yet for the reverse case; this requires a trap and emulate software
implementation for RDTSC, which is still forthcoming.
[avi: fix 64-bit division on i386]
Signed-off-by: Zachary Amsden <zamsden@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-02-03 17:43:50 +00:00
|
|
|
}
|
2011-03-25 08:44:49 +00:00
|
|
|
|
KVM: Infrastructure for software and hardware based TSC rate scaling
This requires some restructuring; rather than use 'virtual_tsc_khz'
to indicate whether hardware rate scaling is in effect, we consider
each VCPU to always have a virtual TSC rate. Instead, there is new
logic above the vendor-specific hardware scaling that decides whether
it is even necessary to use and updates all rate variables used by
common code. This means we can simply query the virtual rate at
any point, which is needed for software rate scaling.
There is also now a threshold added to the TSC rate scaling; minor
differences and variations of measured TSC rate can accidentally
provoke rate scaling to be used when it is not needed. Instead,
we have a tolerance variable called tsc_tolerance_ppm, which is
the maximum variation from user requested rate at which scaling
will be used. The default is 250ppm, which is the half the
threshold for NTP adjustment, allowing for some hardware variation.
In the event that hardware rate scaling is not available, we can
kludge a bit by forcing TSC catchup to turn on when a faster than
hardware speed has been requested, but there is nothing available
yet for the reverse case; this requires a trap and emulate software
implementation for RDTSC, which is still forthcoming.
[avi: fix 64-bit division on i386]
Signed-off-by: Zachary Amsden <zamsden@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-02-03 17:43:50 +00:00
|
|
|
/* TSC scaling supported? */
|
|
|
|
if (!boot_cpu_has(X86_FEATURE_TSCRATEMSR)) {
|
|
|
|
if (user_tsc_khz > tsc_khz) {
|
|
|
|
vcpu->arch.tsc_catchup = 1;
|
|
|
|
vcpu->arch.tsc_always_catchup = 1;
|
|
|
|
} else
|
|
|
|
WARN(1, "user requested TSC rate below hardware speed\n");
|
2011-03-25 08:44:49 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
khz = user_tsc_khz;
|
|
|
|
|
|
|
|
/* TSC scaling required - calculate ratio */
|
|
|
|
ratio = khz << 32;
|
|
|
|
do_div(ratio, tsc_khz);
|
|
|
|
|
|
|
|
if (ratio == 0 || ratio & TSC_RATIO_RSVD) {
|
|
|
|
WARN_ONCE(1, "Invalid TSC ratio - virtual-tsc-khz=%u\n",
|
|
|
|
user_tsc_khz);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
svm->tsc_ratio = ratio;
|
|
|
|
}
|
|
|
|
|
2012-11-29 20:42:50 +00:00
|
|
|
static u64 svm_read_tsc_offset(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
return svm->vmcb->control.tsc_offset;
|
|
|
|
}
|
|
|
|
|
2010-08-20 08:07:16 +00:00
|
|
|
static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u64 g_tsc_offset = 0;
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu)) {
|
2010-08-20 08:07:16 +00:00
|
|
|
g_tsc_offset = svm->vmcb->control.tsc_offset -
|
|
|
|
svm->nested.hsave->control.tsc_offset;
|
|
|
|
svm->nested.hsave->control.tsc_offset = offset;
|
2013-06-12 07:43:44 +00:00
|
|
|
} else
|
|
|
|
trace_kvm_write_tsc_offset(vcpu->vcpu_id,
|
|
|
|
svm->vmcb->control.tsc_offset,
|
|
|
|
offset);
|
2010-08-20 08:07:16 +00:00
|
|
|
|
|
|
|
svm->vmcb->control.tsc_offset = offset + g_tsc_offset;
|
2010-12-03 10:45:49 +00:00
|
|
|
|
|
|
|
mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
|
2010-08-20 08:07:16 +00:00
|
|
|
}
|
|
|
|
|
2012-02-03 17:43:55 +00:00
|
|
|
static void svm_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool host)
|
2010-08-20 08:07:23 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2012-02-03 17:43:55 +00:00
|
|
|
WARN_ON(adjustment < 0);
|
|
|
|
if (host)
|
|
|
|
adjustment = svm_scale_tsc(vcpu, adjustment);
|
|
|
|
|
2010-08-20 08:07:23 +00:00
|
|
|
svm->vmcb->control.tsc_offset += adjustment;
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu))
|
2010-08-20 08:07:23 +00:00
|
|
|
svm->nested.hsave->control.tsc_offset += adjustment;
|
2013-06-12 07:43:44 +00:00
|
|
|
else
|
|
|
|
trace_kvm_write_tsc_offset(vcpu->vcpu_id,
|
|
|
|
svm->vmcb->control.tsc_offset - adjustment,
|
|
|
|
svm->vmcb->control.tsc_offset);
|
|
|
|
|
2010-12-03 10:45:49 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
|
2010-08-20 08:07:23 +00:00
|
|
|
}
|
|
|
|
|
2011-03-25 08:44:50 +00:00
|
|
|
static u64 svm_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc)
|
|
|
|
{
|
|
|
|
u64 tsc;
|
|
|
|
|
|
|
|
tsc = svm_scale_tsc(vcpu, native_read_tsc());
|
|
|
|
|
|
|
|
return target_tsc - tsc;
|
|
|
|
}
|
|
|
|
|
2008-02-13 17:58:45 +00:00
|
|
|
static void init_vmcb(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-02-13 17:58:45 +00:00
|
|
|
struct vmcb_control_area *control = &svm->vmcb->control;
|
|
|
|
struct vmcb_save_area *save = &svm->vmcb->save;
|
[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 10:21:36 +00:00
|
|
|
|
2010-01-07 11:16:08 +00:00
|
|
|
svm->vcpu.fpu_active = 1;
|
2010-12-03 09:50:51 +00:00
|
|
|
svm->vcpu.arch.hflags = 0;
|
2010-01-07 11:16:08 +00:00
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
set_cr_intercept(svm, INTERCEPT_CR0_READ);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR3_READ);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR4_READ);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR0_WRITE);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR3_WRITE);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR4_WRITE);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR8_WRITE);
|
[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 10:21:36 +00:00
|
|
|
|
2014-03-03 12:08:29 +00:00
|
|
|
set_dr_intercepts(svm);
|
[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 10:21:36 +00:00
|
|
|
|
2010-11-30 17:03:59 +00:00
|
|
|
set_exception_intercept(svm, PF_VECTOR);
|
|
|
|
set_exception_intercept(svm, UD_VECTOR);
|
|
|
|
set_exception_intercept(svm, MC_VECTOR);
|
[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 10:21:36 +00:00
|
|
|
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_INTR);
|
|
|
|
set_intercept(svm, INTERCEPT_NMI);
|
|
|
|
set_intercept(svm, INTERCEPT_SMI);
|
|
|
|
set_intercept(svm, INTERCEPT_SELECTIVE_CR0);
|
2011-11-10 12:57:24 +00:00
|
|
|
set_intercept(svm, INTERCEPT_RDPMC);
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_CPUID);
|
|
|
|
set_intercept(svm, INTERCEPT_INVD);
|
|
|
|
set_intercept(svm, INTERCEPT_HLT);
|
|
|
|
set_intercept(svm, INTERCEPT_INVLPG);
|
|
|
|
set_intercept(svm, INTERCEPT_INVLPGA);
|
|
|
|
set_intercept(svm, INTERCEPT_IOIO_PROT);
|
|
|
|
set_intercept(svm, INTERCEPT_MSR_PROT);
|
|
|
|
set_intercept(svm, INTERCEPT_TASK_SWITCH);
|
|
|
|
set_intercept(svm, INTERCEPT_SHUTDOWN);
|
|
|
|
set_intercept(svm, INTERCEPT_VMRUN);
|
|
|
|
set_intercept(svm, INTERCEPT_VMMCALL);
|
|
|
|
set_intercept(svm, INTERCEPT_VMLOAD);
|
|
|
|
set_intercept(svm, INTERCEPT_VMSAVE);
|
|
|
|
set_intercept(svm, INTERCEPT_STGI);
|
|
|
|
set_intercept(svm, INTERCEPT_CLGI);
|
|
|
|
set_intercept(svm, INTERCEPT_SKINIT);
|
|
|
|
set_intercept(svm, INTERCEPT_WBINVD);
|
|
|
|
set_intercept(svm, INTERCEPT_MONITOR);
|
|
|
|
set_intercept(svm, INTERCEPT_MWAIT);
|
2010-12-07 16:15:06 +00:00
|
|
|
set_intercept(svm, INTERCEPT_XSETBV);
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
control->iopm_base_pa = iopm_base;
|
2008-02-13 17:58:46 +00:00
|
|
|
control->msrpm_base_pa = __pa(svm->msrpm);
|
[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 10:21:36 +00:00
|
|
|
control->int_ctl = V_INTR_MASKING_MASK;
|
|
|
|
|
|
|
|
init_seg(&save->es);
|
|
|
|
init_seg(&save->ss);
|
|
|
|
init_seg(&save->ds);
|
|
|
|
init_seg(&save->fs);
|
|
|
|
init_seg(&save->gs);
|
|
|
|
|
|
|
|
save->cs.selector = 0xf000;
|
2013-03-19 15:30:26 +00:00
|
|
|
save->cs.base = 0xffff0000;
|
[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 10:21:36 +00:00
|
|
|
/* Executable/Readable Code Segment */
|
|
|
|
save->cs.attrib = SVM_SELECTOR_READ_MASK | SVM_SELECTOR_P_MASK |
|
|
|
|
SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK;
|
|
|
|
save->cs.limit = 0xffff;
|
|
|
|
|
|
|
|
save->gdtr.limit = 0xffff;
|
|
|
|
save->idtr.limit = 0xffff;
|
|
|
|
|
|
|
|
init_sys_seg(&save->ldtr, SEG_TYPE_LDT);
|
|
|
|
init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16);
|
|
|
|
|
2010-08-31 22:13:14 +00:00
|
|
|
svm_set_efer(&svm->vcpu, 0);
|
2007-10-08 13:02:08 +00:00
|
|
|
save->dr6 = 0xffff0ff0;
|
2010-08-02 12:30:20 +00:00
|
|
|
kvm_set_rflags(&svm->vcpu, 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 10:21:36 +00:00
|
|
|
save->rip = 0x0000fff0;
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip;
|
[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 10:21:36 +00:00
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* This is the guest-visible cr0 value.
|
2009-10-24 04:49:59 +00:00
|
|
|
* svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0.
|
[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 10:21:36 +00:00
|
|
|
*/
|
2010-08-31 22:13:13 +00:00
|
|
|
svm->vcpu.arch.cr0 = 0;
|
|
|
|
(void)kvm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET);
|
2009-10-24 04:49:59 +00:00
|
|
|
|
2007-07-17 13:34:16 +00:00
|
|
|
save->cr4 = X86_CR4_PAE;
|
[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 10:21:36 +00:00
|
|
|
/* rdx = ?? */
|
2008-02-07 12:47:45 +00:00
|
|
|
|
|
|
|
if (npt_enabled) {
|
|
|
|
/* Setup VMCB for Nested Paging */
|
|
|
|
control->nested_ctl = 1;
|
2010-11-30 17:04:00 +00:00
|
|
|
clr_intercept(svm, INTERCEPT_INVLPG);
|
2010-11-30 17:03:59 +00:00
|
|
|
clr_exception_intercept(svm, PF_VECTOR);
|
2010-12-03 09:50:51 +00:00
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR3_READ);
|
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR3_WRITE);
|
2008-02-07 12:47:45 +00:00
|
|
|
save->g_pat = 0x0007040600070406ULL;
|
|
|
|
save->cr3 = 0;
|
|
|
|
save->cr4 = 0;
|
|
|
|
}
|
2010-12-03 14:25:15 +00:00
|
|
|
svm->asid_generation = 0;
|
2008-11-25 19:17:04 +00:00
|
|
|
|
2009-08-07 09:49:33 +00:00
|
|
|
svm->nested.vmcb = 0;
|
2009-08-07 09:49:28 +00:00
|
|
|
svm->vcpu.arch.hflags = 0;
|
|
|
|
|
2010-11-09 14:15:42 +00:00
|
|
|
if (boot_cpu_has(X86_FEATURE_PAUSEFILTER)) {
|
2009-10-06 19:25:02 +00:00
|
|
|
control->pause_filter_count = 3000;
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_PAUSE);
|
2009-10-06 19:25:02 +00:00
|
|
|
}
|
|
|
|
|
2010-12-03 12:15:21 +00:00
|
|
|
mark_all_dirty(svm->vmcb);
|
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
enable_gif(svm);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2013-03-12 09:20:24 +00:00
|
|
|
static void svm_vcpu_reset(struct kvm_vcpu *vcpu)
|
2007-09-10 15:10:54 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2012-12-05 14:26:19 +00:00
|
|
|
u32 dummy;
|
|
|
|
u32 eax = 1;
|
2007-09-10 15:10:54 +00:00
|
|
|
|
2008-02-13 17:58:45 +00:00
|
|
|
init_vmcb(svm);
|
2007-11-07 10:57:23 +00:00
|
|
|
|
2012-12-05 14:26:19 +00:00
|
|
|
kvm_cpuid(vcpu, &eax, &dummy, &dummy, &dummy);
|
|
|
|
kvm_register_write(vcpu, VCPU_REGS_RDX, eax);
|
2007-09-10 15:10:54 +00:00
|
|
|
}
|
|
|
|
|
2007-07-27 07:16:56 +00:00
|
|
|
static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm;
|
[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 10:21:36 +00:00
|
|
|
struct page *page;
|
2008-02-13 17:58:46 +00:00
|
|
|
struct page *msrpm_pages;
|
2008-11-25 19:17:05 +00:00
|
|
|
struct page *hsave_page;
|
2008-11-25 19:17:07 +00:00
|
|
|
struct page *nested_msrpm_pages;
|
2007-07-27 07:16:56 +00:00
|
|
|
int err;
|
[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 10:21:36 +00:00
|
|
|
|
2007-07-30 11:12:19 +00:00
|
|
|
svm = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
|
2007-07-27 07:16:56 +00:00
|
|
|
if (!svm) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2011-03-25 08:44:46 +00:00
|
|
|
svm->tsc_ratio = TSC_RATIO_DEFAULT;
|
|
|
|
|
2007-07-27 07:16:56 +00:00
|
|
|
err = kvm_vcpu_init(&svm->vcpu, kvm, id);
|
|
|
|
if (err)
|
|
|
|
goto free_svm;
|
|
|
|
|
2010-03-09 05:55:19 +00:00
|
|
|
err = -ENOMEM;
|
[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 10:21:36 +00:00
|
|
|
page = alloc_page(GFP_KERNEL);
|
2010-03-09 05:55:19 +00:00
|
|
|
if (!page)
|
2007-07-27 07:16:56 +00:00
|
|
|
goto uninit;
|
[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 10:21:36 +00:00
|
|
|
|
2008-02-13 17:58:46 +00:00
|
|
|
msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
|
|
|
|
if (!msrpm_pages)
|
2010-03-09 05:55:19 +00:00
|
|
|
goto free_page1;
|
2008-11-25 19:17:07 +00:00
|
|
|
|
|
|
|
nested_msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER);
|
|
|
|
if (!nested_msrpm_pages)
|
2010-03-09 05:55:19 +00:00
|
|
|
goto free_page2;
|
2008-02-13 17:58:46 +00:00
|
|
|
|
2008-11-25 19:17:05 +00:00
|
|
|
hsave_page = alloc_page(GFP_KERNEL);
|
|
|
|
if (!hsave_page)
|
2010-03-09 05:55:19 +00:00
|
|
|
goto free_page3;
|
|
|
|
|
2009-08-07 09:49:33 +00:00
|
|
|
svm->nested.hsave = page_address(hsave_page);
|
2008-11-25 19:17:05 +00:00
|
|
|
|
2010-03-09 05:55:19 +00:00
|
|
|
svm->msrpm = page_address(msrpm_pages);
|
|
|
|
svm_vcpu_init_msrpm(svm->msrpm);
|
|
|
|
|
2009-08-07 09:49:33 +00:00
|
|
|
svm->nested.msrpm = page_address(nested_msrpm_pages);
|
2010-03-01 14:34:37 +00:00
|
|
|
svm_vcpu_init_msrpm(svm->nested.msrpm);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb = page_address(page);
|
|
|
|
clear_page(svm->vmcb);
|
|
|
|
svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT;
|
|
|
|
svm->asid_generation = 0;
|
2008-02-13 17:58:45 +00:00
|
|
|
init_vmcb(svm);
|
2007-07-27 12:13:10 +00:00
|
|
|
|
2007-12-13 15:50:52 +00:00
|
|
|
svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
|
2009-06-09 12:56:26 +00:00
|
|
|
if (kvm_vcpu_is_bsp(&svm->vcpu))
|
2007-12-13 15:50:52 +00:00
|
|
|
svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP;
|
[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 10:21:36 +00:00
|
|
|
|
2012-01-09 19:00:35 +00:00
|
|
|
svm_init_osvw(&svm->vcpu);
|
|
|
|
|
2007-07-27 07:16:56 +00:00
|
|
|
return &svm->vcpu;
|
2006-12-22 09:05:20 +00:00
|
|
|
|
2010-03-09 05:55:19 +00:00
|
|
|
free_page3:
|
|
|
|
__free_pages(nested_msrpm_pages, MSRPM_ALLOC_ORDER);
|
|
|
|
free_page2:
|
|
|
|
__free_pages(msrpm_pages, MSRPM_ALLOC_ORDER);
|
|
|
|
free_page1:
|
|
|
|
__free_page(page);
|
2007-07-27 07:16:56 +00:00
|
|
|
uninit:
|
|
|
|
kvm_vcpu_uninit(&svm->vcpu);
|
|
|
|
free_svm:
|
2007-08-01 04:46:11 +00:00
|
|
|
kmem_cache_free(kvm_vcpu_cache, svm);
|
2007-07-27 07:16:56 +00:00
|
|
|
out:
|
|
|
|
return ERR_PTR(err);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_free_vcpu(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2007-07-27 07:16:56 +00:00
|
|
|
__free_page(pfn_to_page(svm->vmcb_pa >> PAGE_SHIFT));
|
2008-02-13 17:58:46 +00:00
|
|
|
__free_pages(virt_to_page(svm->msrpm), MSRPM_ALLOC_ORDER);
|
2009-08-07 09:49:33 +00:00
|
|
|
__free_page(virt_to_page(svm->nested.hsave));
|
|
|
|
__free_pages(virt_to_page(svm->nested.msrpm), MSRPM_ALLOC_ORDER);
|
2007-07-27 07:16:56 +00:00
|
|
|
kvm_vcpu_uninit(vcpu);
|
2007-08-01 04:46:11 +00:00
|
|
|
kmem_cache_free(kvm_vcpu_cache, svm);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2007-07-11 15:17:21 +00:00
|
|
|
static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2007-07-11 15:17:21 +00:00
|
|
|
int i;
|
2007-03-25 10:07:27 +00:00
|
|
|
|
|
|
|
if (unlikely(cpu != vcpu->cpu)) {
|
2009-07-21 15:47:45 +00:00
|
|
|
svm->asid_generation = 0;
|
2010-12-03 12:15:21 +00:00
|
|
|
mark_all_dirty(svm->vmcb);
|
2007-03-25 10:07:27 +00:00
|
|
|
}
|
2007-04-29 08:56:06 +00:00
|
|
|
|
2010-10-21 10:20:34 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base);
|
|
|
|
#endif
|
2010-10-21 10:20:33 +00:00
|
|
|
savesegment(fs, svm->host.fs);
|
|
|
|
savesegment(gs, svm->host.gs);
|
|
|
|
svm->host.ldt = kvm_read_ldt();
|
|
|
|
|
2007-04-29 08:56:06 +00:00
|
|
|
for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
|
2007-07-27 12:13:10 +00:00
|
|
|
rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
|
2011-03-25 08:44:46 +00:00
|
|
|
|
|
|
|
if (static_cpu_has(X86_FEATURE_TSCRATEMSR) &&
|
|
|
|
svm->tsc_ratio != __get_cpu_var(current_tsc_ratio)) {
|
|
|
|
__get_cpu_var(current_tsc_ratio) = svm->tsc_ratio;
|
|
|
|
wrmsrl(MSR_AMD64_TSC_RATIO, svm->tsc_ratio);
|
|
|
|
}
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_vcpu_put(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2007-04-29 08:56:06 +00:00
|
|
|
int i;
|
|
|
|
|
2007-11-18 11:50:24 +00:00
|
|
|
++vcpu->stat.host_state_reload;
|
2010-10-21 10:20:33 +00:00
|
|
|
kvm_load_ldt(svm->host.ldt);
|
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
loadsegment(fs, svm->host.fs);
|
|
|
|
wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs);
|
2011-01-14 15:45:01 +00:00
|
|
|
load_gs_index(svm->host.gs);
|
2010-10-21 10:20:33 +00:00
|
|
|
#else
|
2011-03-08 14:09:51 +00:00
|
|
|
#ifdef CONFIG_X86_32_LAZY_GS
|
2010-10-21 10:20:33 +00:00
|
|
|
loadsegment(gs, svm->host.gs);
|
2011-03-08 14:09:51 +00:00
|
|
|
#endif
|
2010-10-21 10:20:33 +00:00
|
|
|
#endif
|
2007-04-29 08:56:06 +00:00
|
|
|
for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
|
2007-07-27 12:13:10 +00:00
|
|
|
wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
return to_svm(vcpu)->vmcb->save.rflags;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
|
|
|
|
{
|
2014-05-14 07:39:49 +00:00
|
|
|
/*
|
|
|
|
* Any change of EFLAGS.VM is accompained by a reload of SS
|
|
|
|
* (caused by either a task switch or an inter-privilege IRET),
|
|
|
|
* so we do not need to update the CPL here.
|
|
|
|
*/
|
2007-07-27 12:13:10 +00:00
|
|
|
to_svm(vcpu)->vmcb->save.rflags = rflags;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-05-31 19:58:47 +00:00
|
|
|
static void svm_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg)
|
|
|
|
{
|
|
|
|
switch (reg) {
|
|
|
|
case VCPU_EXREG_PDPTR:
|
|
|
|
BUG_ON(!npt_enabled);
|
2010-12-05 15:30:00 +00:00
|
|
|
load_pdptrs(vcpu, vcpu->arch.walk_mmu, kvm_read_cr3(vcpu));
|
2009-05-31 19:58:47 +00:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
BUG();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-11-25 19:17:01 +00:00
|
|
|
static void svm_set_vintr(struct vcpu_svm *svm)
|
|
|
|
{
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_VINTR);
|
2008-11-25 19:17:01 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_clear_vintr(struct vcpu_svm *svm)
|
|
|
|
{
|
2010-11-30 17:04:00 +00:00
|
|
|
clr_intercept(svm, INTERCEPT_VINTR);
|
2008-11-25 19:17:01 +00:00
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save;
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
switch (seg) {
|
|
|
|
case VCPU_SREG_CS: return &save->cs;
|
|
|
|
case VCPU_SREG_DS: return &save->ds;
|
|
|
|
case VCPU_SREG_ES: return &save->es;
|
|
|
|
case VCPU_SREG_FS: return &save->fs;
|
|
|
|
case VCPU_SREG_GS: return &save->gs;
|
|
|
|
case VCPU_SREG_SS: return &save->ss;
|
|
|
|
case VCPU_SREG_TR: return &save->tr;
|
|
|
|
case VCPU_SREG_LDTR: return &save->ldtr;
|
|
|
|
}
|
|
|
|
BUG();
|
2007-02-09 16:38:40 +00:00
|
|
|
return NULL;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static u64 svm_get_segment_base(struct kvm_vcpu *vcpu, int seg)
|
|
|
|
{
|
|
|
|
struct vmcb_seg *s = svm_seg(vcpu, seg);
|
|
|
|
|
|
|
|
return s->base;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_get_segment(struct kvm_vcpu *vcpu,
|
|
|
|
struct kvm_segment *var, int seg)
|
|
|
|
{
|
|
|
|
struct vmcb_seg *s = svm_seg(vcpu, seg);
|
|
|
|
|
|
|
|
var->base = s->base;
|
|
|
|
var->limit = s->limit;
|
|
|
|
var->selector = s->selector;
|
|
|
|
var->type = s->attrib & SVM_SELECTOR_TYPE_MASK;
|
|
|
|
var->s = (s->attrib >> SVM_SELECTOR_S_SHIFT) & 1;
|
|
|
|
var->dpl = (s->attrib >> SVM_SELECTOR_DPL_SHIFT) & 3;
|
|
|
|
var->present = (s->attrib >> SVM_SELECTOR_P_SHIFT) & 1;
|
|
|
|
var->avl = (s->attrib >> SVM_SELECTOR_AVL_SHIFT) & 1;
|
|
|
|
var->l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1;
|
|
|
|
var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1;
|
|
|
|
var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1;
|
2008-10-27 09:04:17 +00:00
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* AMD's VMCB does not have an explicit unusable field, so emulate it
|
2009-04-28 10:45:30 +00:00
|
|
|
* for cross vendor migration purposes by "not present"
|
|
|
|
*/
|
|
|
|
var->unusable = !var->present || (var->type == 0);
|
|
|
|
|
2009-01-11 21:39:44 +00:00
|
|
|
switch (seg) {
|
|
|
|
case VCPU_SREG_CS:
|
|
|
|
/*
|
|
|
|
* SVM always stores 0 for the 'G' bit in the CS selector in
|
|
|
|
* the VMCB on a VMEXIT. This hurts cross-vendor migration:
|
|
|
|
* Intel's VMENTRY has a check on the 'G' bit.
|
|
|
|
*/
|
2008-10-27 09:04:17 +00:00
|
|
|
var->g = s->limit > 0xfffff;
|
2009-01-11 21:39:44 +00:00
|
|
|
break;
|
|
|
|
case VCPU_SREG_TR:
|
|
|
|
/*
|
|
|
|
* Work around a bug where the busy flag in the tr selector
|
|
|
|
* isn't exposed
|
|
|
|
*/
|
2008-10-27 09:04:18 +00:00
|
|
|
var->type |= 0x2;
|
2009-01-11 21:39:44 +00:00
|
|
|
break;
|
|
|
|
case VCPU_SREG_DS:
|
|
|
|
case VCPU_SREG_ES:
|
|
|
|
case VCPU_SREG_FS:
|
|
|
|
case VCPU_SREG_GS:
|
|
|
|
/*
|
|
|
|
* The accessed bit must always be set in the segment
|
|
|
|
* descriptor cache, although it can be cleared in the
|
|
|
|
* descriptor, the cached bit always remains at 1. Since
|
|
|
|
* Intel has a check on this, set it here to support
|
|
|
|
* cross-vendor migration.
|
|
|
|
*/
|
|
|
|
if (!var->unusable)
|
|
|
|
var->type |= 0x1;
|
|
|
|
break;
|
2009-04-28 10:45:43 +00:00
|
|
|
case VCPU_SREG_SS:
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* On AMD CPUs sometimes the DB bit in the segment
|
2009-04-28 10:45:43 +00:00
|
|
|
* descriptor is left as 1, although the whole segment has
|
|
|
|
* been made unusable. Clear it here to pass an Intel VMX
|
|
|
|
* entry check when cross vendor migrating.
|
|
|
|
*/
|
|
|
|
if (var->unusable)
|
|
|
|
var->db = 0;
|
2014-06-29 15:12:43 +00:00
|
|
|
var->dpl = to_svm(vcpu)->vmcb->save.cpl;
|
2009-04-28 10:45:43 +00:00
|
|
|
break;
|
2009-01-11 21:39:44 +00:00
|
|
|
}
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2008-03-24 17:38:34 +00:00
|
|
|
static int svm_get_cpl(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save;
|
|
|
|
|
|
|
|
return save->cpl;
|
|
|
|
}
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
static void svm_get_idt(struct kvm_vcpu *vcpu, struct desc_ptr *dt)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
dt->size = svm->vmcb->save.idtr.limit;
|
|
|
|
dt->address = svm->vmcb->save.idtr.base;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
static void svm_set_idt(struct kvm_vcpu *vcpu, struct desc_ptr *dt)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
svm->vmcb->save.idtr.limit = dt->size;
|
|
|
|
svm->vmcb->save.idtr.base = dt->address ;
|
2010-12-03 10:45:56 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_DT);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
static void svm_get_gdt(struct kvm_vcpu *vcpu, struct desc_ptr *dt)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
dt->size = svm->vmcb->save.gdtr.limit;
|
|
|
|
dt->address = svm->vmcb->save.gdtr.base;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
static void svm_set_gdt(struct kvm_vcpu *vcpu, struct desc_ptr *dt)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-16 08:51:48 +00:00
|
|
|
svm->vmcb->save.gdtr.limit = dt->size;
|
|
|
|
svm->vmcb->save.gdtr.base = dt->address ;
|
2010-12-03 10:45:56 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_DT);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-12-29 16:43:06 +00:00
|
|
|
static void svm_decache_cr0_guest_bits(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2010-12-05 16:56:11 +00:00
|
|
|
static void svm_decache_cr3(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2007-04-27 06:29:21 +00:00
|
|
|
static void svm_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
|
2007-01-06 00:36:38 +00:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2010-01-06 08:55:27 +00:00
|
|
|
static void update_cr0_intercept(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
ulong gcr0 = svm->vcpu.arch.cr0;
|
|
|
|
u64 *hcr0 = &svm->vmcb->save.cr0;
|
|
|
|
|
|
|
|
if (!svm->vcpu.fpu_active)
|
|
|
|
*hcr0 |= SVM_CR0_SELECTIVE_MASK;
|
|
|
|
else
|
|
|
|
*hcr0 = (*hcr0 & ~SVM_CR0_SELECTIVE_MASK)
|
|
|
|
| (gcr0 & SVM_CR0_SELECTIVE_MASK);
|
|
|
|
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_CR);
|
2010-01-06 08:55:27 +00:00
|
|
|
|
|
|
|
if (gcr0 == *hcr0 && svm->vcpu.fpu_active) {
|
2010-12-03 09:50:51 +00:00
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR0_READ);
|
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR0_WRITE);
|
2010-01-06 08:55:27 +00:00
|
|
|
} else {
|
2010-12-03 09:50:51 +00:00
|
|
|
set_cr_intercept(svm, INTERCEPT_CR0_READ);
|
|
|
|
set_cr_intercept(svm, INTERCEPT_CR0_WRITE);
|
2010-01-06 08:55:27 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2006-12-13 08:33:45 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
2010-01-21 13:31:50 +00:00
|
|
|
if (vcpu->arch.efer & EFER_LME) {
|
2007-07-17 13:19:08 +00:00
|
|
|
if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
|
2010-01-21 13:31:50 +00:00
|
|
|
vcpu->arch.efer |= EFER_LMA;
|
2007-12-01 12:17:11 +00:00
|
|
|
svm->vmcb->save.efer |= EFER_LMA | EFER_LME;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2007-10-08 13:02:08 +00:00
|
|
|
if (is_paging(vcpu) && !(cr0 & X86_CR0_PG)) {
|
2010-01-21 13:31:50 +00:00
|
|
|
vcpu->arch.efer &= ~EFER_LMA;
|
2007-12-01 12:17:11 +00:00
|
|
|
svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|
2007-12-13 15:50:52 +00:00
|
|
|
vcpu->arch.cr0 = cr0;
|
2010-01-10 10:14:04 +00:00
|
|
|
|
|
|
|
if (!npt_enabled)
|
|
|
|
cr0 |= X86_CR0_PG | X86_CR0_WP;
|
2009-12-30 10:40:26 +00:00
|
|
|
|
|
|
|
if (!vcpu->fpu_active)
|
2008-01-21 12:09:33 +00:00
|
|
|
cr0 |= X86_CR0_TS;
|
2008-02-07 12:47:45 +00:00
|
|
|
/*
|
|
|
|
* re-enable caching here because the QEMU bios
|
|
|
|
* does not do it - this results in some delay at
|
|
|
|
* reboot
|
|
|
|
*/
|
|
|
|
cr0 &= ~(X86_CR0_CD | X86_CR0_NW);
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.cr0 = cr0;
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_CR);
|
2010-01-06 08:55:27 +00:00
|
|
|
update_cr0_intercept(svm);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2011-05-25 20:03:24 +00:00
|
|
|
static int svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-04-09 12:15:29 +00:00
|
|
|
unsigned long host_cr4_mce = read_cr4() & X86_CR4_MCE;
|
2008-09-09 17:11:51 +00:00
|
|
|
unsigned long old_cr4 = to_svm(vcpu)->vmcb->save.cr4;
|
|
|
|
|
2011-05-25 20:03:24 +00:00
|
|
|
if (cr4 & X86_CR4_VMXE)
|
|
|
|
return 1;
|
|
|
|
|
2008-09-09 17:11:51 +00:00
|
|
|
if (npt_enabled && ((old_cr4 ^ cr4) & X86_CR4_PGE))
|
2010-12-03 14:25:15 +00:00
|
|
|
svm_flush_tlb(vcpu);
|
2008-04-09 12:15:29 +00:00
|
|
|
|
2008-04-09 12:15:28 +00:00
|
|
|
vcpu->arch.cr4 = cr4;
|
|
|
|
if (!npt_enabled)
|
|
|
|
cr4 |= X86_CR4_PAE;
|
2008-04-09 12:15:29 +00:00
|
|
|
cr4 |= host_cr4_mce;
|
2008-04-09 12:15:28 +00:00
|
|
|
to_svm(vcpu)->vmcb->save.cr4 = cr4;
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(to_svm(vcpu)->vmcb, VMCB_CR);
|
2011-05-25 20:03:24 +00:00
|
|
|
return 0;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_segment(struct kvm_vcpu *vcpu,
|
|
|
|
struct kvm_segment *var, int seg)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
[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 10:21:36 +00:00
|
|
|
struct vmcb_seg *s = svm_seg(vcpu, seg);
|
|
|
|
|
|
|
|
s->base = var->base;
|
|
|
|
s->limit = var->limit;
|
|
|
|
s->selector = var->selector;
|
|
|
|
if (var->unusable)
|
|
|
|
s->attrib = 0;
|
|
|
|
else {
|
|
|
|
s->attrib = (var->type & SVM_SELECTOR_TYPE_MASK);
|
|
|
|
s->attrib |= (var->s & 1) << SVM_SELECTOR_S_SHIFT;
|
|
|
|
s->attrib |= (var->dpl & 3) << SVM_SELECTOR_DPL_SHIFT;
|
|
|
|
s->attrib |= (var->present & 1) << SVM_SELECTOR_P_SHIFT;
|
|
|
|
s->attrib |= (var->avl & 1) << SVM_SELECTOR_AVL_SHIFT;
|
|
|
|
s->attrib |= (var->l & 1) << SVM_SELECTOR_L_SHIFT;
|
|
|
|
s->attrib |= (var->db & 1) << SVM_SELECTOR_DB_SHIFT;
|
|
|
|
s->attrib |= (var->g & 1) << SVM_SELECTOR_G_SHIFT;
|
|
|
|
}
|
2014-05-14 07:39:49 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* This is always accurate, except if SYSRET returned to a segment
|
|
|
|
* with SS.DPL != 3. Intel does not have this quirk, and always
|
|
|
|
* forces SS.DPL to 3 on sysret, so we ignore that case; fixing it
|
|
|
|
* would entail passing the CPL to userspace and back.
|
|
|
|
*/
|
|
|
|
if (seg == VCPU_SREG_SS)
|
|
|
|
svm->vmcb->save.cpl = (s->attrib >> SVM_SELECTOR_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 10:21:36 +00:00
|
|
|
|
2010-12-03 10:45:57 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_SEG);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2012-09-21 03:42:55 +00:00
|
|
|
static void update_db_bp_intercept(struct kvm_vcpu *vcpu)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-12-15 12:52:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-11-30 17:03:59 +00:00
|
|
|
clr_exception_intercept(svm, DB_VECTOR);
|
|
|
|
clr_exception_intercept(svm, BP_VECTOR);
|
2009-05-11 10:35:52 +00:00
|
|
|
|
2009-10-18 11:24:54 +00:00
|
|
|
if (svm->nmi_singlestep)
|
2010-11-30 17:03:59 +00:00
|
|
|
set_exception_intercept(svm, DB_VECTOR);
|
2009-05-11 10:35:52 +00:00
|
|
|
|
2008-12-15 12:52:10 +00:00
|
|
|
if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
|
|
|
|
if (vcpu->guest_debug &
|
|
|
|
(KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
|
2010-11-30 17:03:59 +00:00
|
|
|
set_exception_intercept(svm, DB_VECTOR);
|
2008-12-15 12:52:10 +00:00
|
|
|
if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
|
2010-11-30 17:03:59 +00:00
|
|
|
set_exception_intercept(svm, BP_VECTOR);
|
2008-12-15 12:52:10 +00:00
|
|
|
} else
|
|
|
|
vcpu->guest_debug = 0;
|
2009-05-11 10:35:52 +00:00
|
|
|
}
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd)
|
[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 10:21:36 +00:00
|
|
|
{
|
2009-10-29 13:34:14 +00:00
|
|
|
if (sd->next_asid > sd->max_asid) {
|
|
|
|
++sd->asid_generation;
|
|
|
|
sd->next_asid = 1;
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ALL_ASID;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
svm->asid_generation = sd->asid_generation;
|
|
|
|
svm->vmcb->control.asid = sd->next_asid++;
|
2010-12-03 10:45:51 +00:00
|
|
|
|
|
|
|
mark_dirty(svm->vmcb, VMCB_ASID);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2014-01-04 17:47:16 +00:00
|
|
|
static u64 svm_get_dr6(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
return to_svm(vcpu)->vmcb->save.dr6;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_dr6(struct kvm_vcpu *vcpu, unsigned long value)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->save.dr6 = value;
|
|
|
|
mark_dirty(svm->vmcb, VMCB_DR);
|
|
|
|
}
|
|
|
|
|
2014-02-21 09:32:27 +00:00
|
|
|
static void svm_sync_dirty_debug_regs(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
get_debugreg(vcpu->arch.db[0], 0);
|
|
|
|
get_debugreg(vcpu->arch.db[1], 1);
|
|
|
|
get_debugreg(vcpu->arch.db[2], 2);
|
|
|
|
get_debugreg(vcpu->arch.db[3], 3);
|
|
|
|
vcpu->arch.dr6 = svm_get_dr6(vcpu);
|
|
|
|
vcpu->arch.dr7 = svm->vmcb->save.dr7;
|
|
|
|
|
|
|
|
vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_WONT_EXIT;
|
|
|
|
set_dr_intercepts(svm);
|
|
|
|
}
|
|
|
|
|
2010-04-13 07:05:23 +00:00
|
|
|
static void svm_set_dr7(struct kvm_vcpu *vcpu, unsigned long value)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-12-15 12:52:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-04-13 07:05:23 +00:00
|
|
|
svm->vmcb->save.dr7 = value;
|
2010-12-03 10:45:55 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_DR);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int pf_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2010-10-14 09:22:52 +00:00
|
|
|
u64 fault_address = svm->vmcb->control.exit_info_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 10:21:36 +00:00
|
|
|
u32 error_code;
|
2010-10-14 09:22:52 +00:00
|
|
|
int r = 1;
|
[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 10:21:36 +00:00
|
|
|
|
2010-10-14 09:22:52 +00:00
|
|
|
switch (svm->apf_reason) {
|
|
|
|
default:
|
|
|
|
error_code = svm->vmcb->control.exit_info_1;
|
2008-04-30 15:56:03 +00:00
|
|
|
|
2010-10-14 09:22:52 +00:00
|
|
|
trace_kvm_page_fault(fault_address, error_code);
|
|
|
|
if (!npt_enabled && kvm_event_needs_reinjection(&svm->vcpu))
|
|
|
|
kvm_mmu_unprotect_page_virt(&svm->vcpu, fault_address);
|
2010-12-21 10:12:07 +00:00
|
|
|
r = kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code,
|
|
|
|
svm->vmcb->control.insn_bytes,
|
|
|
|
svm->vmcb->control.insn_len);
|
2010-10-14 09:22:52 +00:00
|
|
|
break;
|
|
|
|
case KVM_PV_REASON_PAGE_NOT_PRESENT:
|
|
|
|
svm->apf_reason = 0;
|
|
|
|
local_irq_disable();
|
|
|
|
kvm_async_pf_task_wait(fault_address);
|
|
|
|
local_irq_enable();
|
|
|
|
break;
|
|
|
|
case KVM_PV_REASON_PAGE_READY:
|
|
|
|
svm->apf_reason = 0;
|
|
|
|
local_irq_disable();
|
|
|
|
kvm_async_pf_task_wake(fault_address);
|
|
|
|
local_irq_enable();
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return r;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int db_interception(struct vcpu_svm *svm)
|
2008-12-15 12:52:10 +00:00
|
|
|
{
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = svm->vcpu.run;
|
|
|
|
|
2008-12-15 12:52:10 +00:00
|
|
|
if (!(svm->vcpu.guest_debug &
|
2009-05-11 10:35:52 +00:00
|
|
|
(KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) &&
|
2009-10-18 11:24:54 +00:00
|
|
|
!svm->nmi_singlestep) {
|
2008-12-15 12:52:10 +00:00
|
|
|
kvm_queue_exception(&svm->vcpu, DB_VECTOR);
|
|
|
|
return 1;
|
|
|
|
}
|
2009-05-11 10:35:52 +00:00
|
|
|
|
2009-10-18 11:24:54 +00:00
|
|
|
if (svm->nmi_singlestep) {
|
|
|
|
svm->nmi_singlestep = false;
|
2009-05-11 10:35:52 +00:00
|
|
|
if (!(svm->vcpu.guest_debug & KVM_GUESTDBG_SINGLESTEP))
|
|
|
|
svm->vmcb->save.rflags &=
|
|
|
|
~(X86_EFLAGS_TF | X86_EFLAGS_RF);
|
2012-09-21 03:42:55 +00:00
|
|
|
update_db_bp_intercept(&svm->vcpu);
|
2009-05-11 10:35:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (svm->vcpu.guest_debug &
|
2010-02-24 17:59:10 +00:00
|
|
|
(KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) {
|
2009-05-11 10:35:52 +00:00
|
|
|
kvm_run->exit_reason = KVM_EXIT_DEBUG;
|
|
|
|
kvm_run->debug.arch.pc =
|
|
|
|
svm->vmcb->save.cs.base + svm->vmcb->save.rip;
|
|
|
|
kvm_run->debug.arch.exception = DB_VECTOR;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
2008-12-15 12:52:10 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int bp_interception(struct vcpu_svm *svm)
|
2008-12-15 12:52:10 +00:00
|
|
|
{
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = svm->vcpu.run;
|
|
|
|
|
2008-12-15 12:52:10 +00:00
|
|
|
kvm_run->exit_reason = KVM_EXIT_DEBUG;
|
|
|
|
kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip;
|
|
|
|
kvm_run->debug.arch.exception = BP_VECTOR;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int ud_interception(struct vcpu_svm *svm)
|
2007-09-17 19:57:50 +00:00
|
|
|
{
|
|
|
|
int er;
|
|
|
|
|
2010-12-21 10:12:02 +00:00
|
|
|
er = emulate_instruction(&svm->vcpu, EMULTYPE_TRAP_UD);
|
2007-09-17 19:57:50 +00:00
|
|
|
if (er != EMULATE_DONE)
|
2007-11-25 13:22:50 +00:00
|
|
|
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
|
2007-09-17 19:57:50 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-01-21 13:31:47 +00:00
|
|
|
static void svm_fpu_activate(struct kvm_vcpu *vcpu)
|
2007-04-23 14:17:21 +00:00
|
|
|
{
|
2010-01-21 13:31:47 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2010-02-19 15:23:08 +00:00
|
|
|
|
2010-11-30 17:03:59 +00:00
|
|
|
clr_exception_intercept(svm, NM_VECTOR);
|
2010-02-19 15:23:08 +00:00
|
|
|
|
2007-07-30 10:07:08 +00:00
|
|
|
svm->vcpu.fpu_active = 1;
|
2010-01-06 08:55:27 +00:00
|
|
|
update_cr0_intercept(svm);
|
2010-01-21 13:31:47 +00:00
|
|
|
}
|
2007-07-27 12:13:10 +00:00
|
|
|
|
2010-01-21 13:31:47 +00:00
|
|
|
static int nm_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
svm_fpu_activate(&svm->vcpu);
|
2007-07-27 12:13:10 +00:00
|
|
|
return 1;
|
2007-04-23 14:17:21 +00:00
|
|
|
}
|
|
|
|
|
2010-05-17 12:43:35 +00:00
|
|
|
static bool is_erratum_383(void)
|
|
|
|
{
|
|
|
|
int err, i;
|
|
|
|
u64 value;
|
|
|
|
|
|
|
|
if (!erratum_383_found)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
value = native_read_msr_safe(MSR_IA32_MC0_STATUS, &err);
|
|
|
|
if (err)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
/* Bit 62 may or may not be set for this mce */
|
|
|
|
value &= ~(1ULL << 62);
|
|
|
|
|
|
|
|
if (value != 0xb600000000010015ULL)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
/* Clear MCi_STATUS registers */
|
|
|
|
for (i = 0; i < 6; ++i)
|
|
|
|
native_write_msr_safe(MSR_IA32_MCx_STATUS(i), 0, 0);
|
|
|
|
|
|
|
|
value = native_read_msr_safe(MSR_IA32_MCG_STATUS, &err);
|
|
|
|
if (!err) {
|
|
|
|
u32 low, high;
|
|
|
|
|
|
|
|
value &= ~(1ULL << 2);
|
|
|
|
low = lower_32_bits(value);
|
|
|
|
high = upper_32_bits(value);
|
|
|
|
|
|
|
|
native_write_msr_safe(MSR_IA32_MCG_STATUS, low, high);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Flush tlb to evict multi-match entries */
|
|
|
|
__flush_tlb_all();
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2010-05-17 12:43:34 +00:00
|
|
|
static void svm_handle_mce(struct vcpu_svm *svm)
|
2008-04-09 12:15:30 +00:00
|
|
|
{
|
2010-05-17 12:43:35 +00:00
|
|
|
if (is_erratum_383()) {
|
|
|
|
/*
|
|
|
|
* Erratum 383 triggered. Guest state is corrupt so kill the
|
|
|
|
* guest.
|
|
|
|
*/
|
|
|
|
pr_err("KVM: Guest triggered AMD Erratum 383\n");
|
|
|
|
|
2010-05-10 09:34:53 +00:00
|
|
|
kvm_make_request(KVM_REQ_TRIPLE_FAULT, &svm->vcpu);
|
2010-05-17 12:43:35 +00:00
|
|
|
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2008-04-09 12:15:30 +00:00
|
|
|
/*
|
|
|
|
* On an #MC intercept the MCE handler is not called automatically in
|
|
|
|
* the host. So do it by hand here.
|
|
|
|
*/
|
|
|
|
asm volatile (
|
|
|
|
"int $0x12\n");
|
|
|
|
/* not sure if we ever come back to this point */
|
|
|
|
|
2010-05-17 12:43:34 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int mc_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
2008-04-09 12:15:30 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int shutdown_interception(struct vcpu_svm *svm)
|
2007-01-26 08:56:42 +00:00
|
|
|
{
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = svm->vcpu.run;
|
|
|
|
|
2007-01-26 08:56:42 +00:00
|
|
|
/*
|
|
|
|
* VMCB is undefined after a SHUTDOWN intercept
|
|
|
|
* so reinitialize it.
|
|
|
|
*/
|
2007-07-27 12:13:10 +00:00
|
|
|
clear_page(svm->vmcb);
|
2008-02-13 17:58:45 +00:00
|
|
|
init_vmcb(svm);
|
2007-01-26 08:56:42 +00:00
|
|
|
|
|
|
|
kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int io_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2010-03-18 13:20:23 +00:00
|
|
|
struct kvm_vcpu *vcpu = &svm->vcpu;
|
2007-10-08 13:02:08 +00:00
|
|
|
u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */
|
2009-02-08 12:28:15 +00:00
|
|
|
int size, in, string;
|
2007-03-20 10:46:50 +00:00
|
|
|
unsigned port;
|
[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 10:21:36 +00:00
|
|
|
|
2007-07-30 10:07:08 +00:00
|
|
|
++svm->vcpu.stat.io_exits;
|
2007-08-05 07:36:40 +00:00
|
|
|
string = (io_info & SVM_IOIO_STR_MASK) != 0;
|
2007-03-20 10:46:50 +00:00
|
|
|
in = (io_info & SVM_IOIO_TYPE_MASK) != 0;
|
2010-03-18 13:20:23 +00:00
|
|
|
if (string || in)
|
2010-12-21 10:12:02 +00:00
|
|
|
return emulate_instruction(vcpu, 0) == EMULATE_DONE;
|
2010-03-18 13:20:23 +00:00
|
|
|
|
2007-03-20 10:46:50 +00:00
|
|
|
port = io_info >> 16;
|
|
|
|
size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT;
|
2010-03-18 13:20:23 +00:00
|
|
|
svm->next_rip = svm->vmcb->control.exit_info_2;
|
2008-10-28 09:51:30 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
2010-03-18 13:20:23 +00:00
|
|
|
|
|
|
|
return kvm_fast_pio_out(vcpu, size, port);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int nmi_interception(struct vcpu_svm *svm)
|
2008-04-30 15:56:00 +00:00
|
|
|
{
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int intr_interception(struct vcpu_svm *svm)
|
2008-04-30 15:56:01 +00:00
|
|
|
{
|
|
|
|
++svm->vcpu.stat.irq_exits;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int nop_on_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int halt_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 1;
|
2007-07-30 10:07:08 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
return kvm_emulate_halt(&svm->vcpu);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int vmmcall_interception(struct vcpu_svm *svm)
|
2007-02-19 12:37:47 +00:00
|
|
|
{
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
2007-07-30 10:07:08 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
2007-09-17 19:57:50 +00:00
|
|
|
kvm_emulate_hypercall(&svm->vcpu);
|
|
|
|
return 1;
|
2007-02-19 12:37:47 +00:00
|
|
|
}
|
|
|
|
|
2010-09-10 15:31:02 +00:00
|
|
|
static unsigned long nested_svm_get_tdp_cr3(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
return svm->nested.nested_cr3;
|
|
|
|
}
|
|
|
|
|
2011-07-28 08:36:17 +00:00
|
|
|
static u64 nested_svm_get_tdp_pdptr(struct kvm_vcpu *vcpu, int index)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u64 cr3 = svm->nested.nested_cr3;
|
|
|
|
u64 pdpte;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
ret = kvm_read_guest_page(vcpu->kvm, gpa_to_gfn(cr3), &pdpte,
|
|
|
|
offset_in_page(cr3) + index * 8, 8);
|
|
|
|
if (ret)
|
|
|
|
return 0;
|
|
|
|
return pdpte;
|
|
|
|
}
|
|
|
|
|
2010-09-10 15:31:02 +00:00
|
|
|
static void nested_svm_set_tdp_cr3(struct kvm_vcpu *vcpu,
|
|
|
|
unsigned long root)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->control.nested_cr3 = root;
|
2010-12-03 10:45:53 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_NPT);
|
2010-12-03 14:25:15 +00:00
|
|
|
svm_flush_tlb(vcpu);
|
2010-09-10 15:31:02 +00:00
|
|
|
}
|
|
|
|
|
2010-11-29 14:12:30 +00:00
|
|
|
static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu,
|
|
|
|
struct x86_exception *fault)
|
2010-09-10 15:31:02 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_NPF;
|
|
|
|
svm->vmcb->control.exit_code_hi = 0;
|
2010-11-29 14:12:30 +00:00
|
|
|
svm->vmcb->control.exit_info_1 = fault->error_code;
|
|
|
|
svm->vmcb->control.exit_info_2 = fault->address;
|
2010-09-10 15:31:02 +00:00
|
|
|
|
|
|
|
nested_svm_vmexit(svm);
|
|
|
|
}
|
|
|
|
|
2013-10-02 14:56:13 +00:00
|
|
|
static void nested_svm_init_mmu_context(struct kvm_vcpu *vcpu)
|
2010-09-10 15:31:03 +00:00
|
|
|
{
|
2013-10-02 14:56:13 +00:00
|
|
|
kvm_init_shadow_mmu(vcpu, &vcpu->arch.mmu);
|
2010-09-10 15:31:03 +00:00
|
|
|
|
|
|
|
vcpu->arch.mmu.set_cr3 = nested_svm_set_tdp_cr3;
|
|
|
|
vcpu->arch.mmu.get_cr3 = nested_svm_get_tdp_cr3;
|
2011-07-28 08:36:17 +00:00
|
|
|
vcpu->arch.mmu.get_pdptr = nested_svm_get_tdp_pdptr;
|
2010-09-10 15:31:03 +00:00
|
|
|
vcpu->arch.mmu.inject_page_fault = nested_svm_inject_npf_exit;
|
|
|
|
vcpu->arch.mmu.shadow_root_level = get_npt_level();
|
|
|
|
vcpu->arch.walk_mmu = &vcpu->arch.nested_mmu;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void nested_svm_uninit_mmu_context(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
vcpu->arch.walk_mmu = &vcpu->arch.mmu;
|
|
|
|
}
|
|
|
|
|
2008-11-25 19:17:03 +00:00
|
|
|
static int nested_svm_check_permissions(struct vcpu_svm *svm)
|
|
|
|
{
|
2010-01-21 13:31:50 +00:00
|
|
|
if (!(svm->vcpu.arch.efer & EFER_SVME)
|
2008-11-25 19:17:03 +00:00
|
|
|
|| !is_paging(&svm->vcpu)) {
|
|
|
|
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (svm->vmcb->save.cpl) {
|
|
|
|
kvm_inject_gp(&svm->vcpu, 0);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-11-25 19:17:08 +00:00
|
|
|
static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
|
|
|
|
bool has_error_code, u32 error_code)
|
|
|
|
{
|
2010-02-19 15:23:02 +00:00
|
|
|
int vmexit;
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (!is_guest_mode(&svm->vcpu))
|
2009-08-07 09:49:37 +00:00
|
|
|
return 0;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2009-08-07 09:49:37 +00:00
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr;
|
|
|
|
svm->vmcb->control.exit_code_hi = 0;
|
|
|
|
svm->vmcb->control.exit_info_1 = error_code;
|
|
|
|
svm->vmcb->control.exit_info_2 = svm->vcpu.arch.cr2;
|
|
|
|
|
2010-02-19 15:23:02 +00:00
|
|
|
vmexit = nested_svm_intercept(svm);
|
|
|
|
if (vmexit == NESTED_EXIT_DONE)
|
|
|
|
svm->nested.exit_required = true;
|
|
|
|
|
|
|
|
return vmexit;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2010-02-19 15:23:01 +00:00
|
|
|
/* This function returns true if it is save to enable the irq window */
|
|
|
|
static inline bool nested_svm_intr(struct vcpu_svm *svm)
|
2008-11-25 19:17:08 +00:00
|
|
|
{
|
2010-11-29 16:51:48 +00:00
|
|
|
if (!is_guest_mode(&svm->vcpu))
|
2010-02-19 15:23:01 +00:00
|
|
|
return true;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2009-08-07 09:49:46 +00:00
|
|
|
if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK))
|
2010-02-19 15:23:01 +00:00
|
|
|
return true;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2009-08-07 09:49:46 +00:00
|
|
|
if (!(svm->vcpu.arch.hflags & HF_HIF_MASK))
|
2010-02-19 15:23:01 +00:00
|
|
|
return false;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2010-09-20 08:15:32 +00:00
|
|
|
/*
|
|
|
|
* if vmexit was already requested (by intercepted exception
|
|
|
|
* for instance) do not overwrite it with "external interrupt"
|
|
|
|
* vmexit.
|
|
|
|
*/
|
|
|
|
if (svm->nested.exit_required)
|
|
|
|
return false;
|
|
|
|
|
2010-02-24 17:59:19 +00:00
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_INTR;
|
|
|
|
svm->vmcb->control.exit_info_1 = 0;
|
|
|
|
svm->vmcb->control.exit_info_2 = 0;
|
2009-08-07 09:49:46 +00:00
|
|
|
|
2009-10-09 14:08:26 +00:00
|
|
|
if (svm->nested.intercept & 1ULL) {
|
|
|
|
/*
|
|
|
|
* The #vmexit can't be emulated here directly because this
|
2012-06-28 07:16:43 +00:00
|
|
|
* code path runs with irqs and preemption disabled. A
|
2009-10-09 14:08:26 +00:00
|
|
|
* #vmexit emulation might sleep. Only signal request for
|
|
|
|
* the #vmexit here.
|
|
|
|
*/
|
|
|
|
svm->nested.exit_required = true;
|
2009-10-09 14:08:30 +00:00
|
|
|
trace_kvm_nested_intr_vmexit(svm->vmcb->save.rip);
|
2010-02-19 15:23:01 +00:00
|
|
|
return false;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2010-02-19 15:23:01 +00:00
|
|
|
return true;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2010-02-24 17:59:12 +00:00
|
|
|
/* This function returns true if it is save to enable the nmi window */
|
|
|
|
static inline bool nested_svm_nmi(struct vcpu_svm *svm)
|
|
|
|
{
|
2010-11-29 16:51:48 +00:00
|
|
|
if (!is_guest_mode(&svm->vcpu))
|
2010-02-24 17:59:12 +00:00
|
|
|
return true;
|
|
|
|
|
|
|
|
if (!(svm->nested.intercept & (1ULL << INTERCEPT_NMI)))
|
|
|
|
return true;
|
|
|
|
|
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_NMI;
|
|
|
|
svm->nested.exit_required = true;
|
|
|
|
|
|
|
|
return false;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
static void *nested_svm_map(struct vcpu_svm *svm, u64 gpa, struct page **_page)
|
2009-08-07 09:49:38 +00:00
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
|
2010-02-19 15:23:04 +00:00
|
|
|
might_sleep();
|
|
|
|
|
2009-08-07 09:49:38 +00:00
|
|
|
page = gfn_to_page(svm->vcpu.kvm, gpa >> PAGE_SHIFT);
|
|
|
|
if (is_error_page(page))
|
|
|
|
goto error;
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
*_page = page;
|
|
|
|
|
|
|
|
return kmap(page);
|
2009-08-07 09:49:38 +00:00
|
|
|
|
|
|
|
error:
|
|
|
|
kvm_inject_gp(&svm->vcpu, 0);
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
static void nested_svm_unmap(struct page *page)
|
2009-08-07 09:49:38 +00:00
|
|
|
{
|
2010-02-19 15:23:00 +00:00
|
|
|
kunmap(page);
|
2009-08-07 09:49:38 +00:00
|
|
|
kvm_release_page_dirty(page);
|
|
|
|
}
|
|
|
|
|
2010-03-01 14:34:39 +00:00
|
|
|
static int nested_svm_intercept_ioio(struct vcpu_svm *svm)
|
|
|
|
{
|
2014-06-30 08:54:17 +00:00
|
|
|
unsigned port, size, iopm_len;
|
|
|
|
u16 val, mask;
|
|
|
|
u8 start_bit;
|
2010-03-01 14:34:39 +00:00
|
|
|
u64 gpa;
|
2009-08-07 09:49:38 +00:00
|
|
|
|
2010-03-01 14:34:39 +00:00
|
|
|
if (!(svm->nested.intercept & (1ULL << INTERCEPT_IOIO_PROT)))
|
|
|
|
return NESTED_EXIT_HOST;
|
2009-08-07 09:49:38 +00:00
|
|
|
|
2010-03-01 14:34:39 +00:00
|
|
|
port = svm->vmcb->control.exit_info_1 >> 16;
|
2014-06-30 08:54:17 +00:00
|
|
|
size = (svm->vmcb->control.exit_info_1 & SVM_IOIO_SIZE_MASK) >>
|
|
|
|
SVM_IOIO_SIZE_SHIFT;
|
2010-03-01 14:34:39 +00:00
|
|
|
gpa = svm->nested.vmcb_iopm + (port / 8);
|
2014-06-30 08:54:17 +00:00
|
|
|
start_bit = port % 8;
|
|
|
|
iopm_len = (start_bit + size > 8) ? 2 : 1;
|
|
|
|
mask = (0xf >> (4 - size)) << start_bit;
|
|
|
|
val = 0;
|
2010-03-01 14:34:39 +00:00
|
|
|
|
2014-06-30 08:54:17 +00:00
|
|
|
if (kvm_read_guest(svm->vcpu.kvm, gpa, &val, iopm_len))
|
|
|
|
return NESTED_EXIT_DONE;
|
2010-03-01 14:34:39 +00:00
|
|
|
|
2014-06-30 08:54:17 +00:00
|
|
|
return (val & mask) ? NESTED_EXIT_DONE : NESTED_EXIT_HOST;
|
2009-08-07 09:49:38 +00:00
|
|
|
}
|
|
|
|
|
2010-03-01 14:34:34 +00:00
|
|
|
static int nested_svm_exit_handled_msr(struct vcpu_svm *svm)
|
2009-08-07 09:49:35 +00:00
|
|
|
{
|
2010-03-01 14:34:38 +00:00
|
|
|
u32 offset, msr, value;
|
|
|
|
int write, mask;
|
2009-08-07 09:49:35 +00:00
|
|
|
|
2009-08-07 09:49:39 +00:00
|
|
|
if (!(svm->nested.intercept & (1ULL << INTERCEPT_MSR_PROT)))
|
2010-03-01 14:34:34 +00:00
|
|
|
return NESTED_EXIT_HOST;
|
2009-08-07 09:49:39 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
msr = svm->vcpu.arch.regs[VCPU_REGS_RCX];
|
|
|
|
offset = svm_msrpm_offset(msr);
|
|
|
|
write = svm->vmcb->control.exit_info_1 & 1;
|
|
|
|
mask = 1 << ((2 * (msr & 0xf)) + write);
|
2009-08-07 09:49:39 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
if (offset == MSR_INVALID)
|
|
|
|
return NESTED_EXIT_DONE;
|
2009-08-07 09:49:35 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
/* Offset is in 32 bit units but need in 8 bit units */
|
|
|
|
offset *= 4;
|
2009-08-07 09:49:35 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
if (kvm_read_guest(svm->vcpu.kvm, svm->nested.vmcb_msrpm + offset, &value, 4))
|
|
|
|
return NESTED_EXIT_DONE;
|
2009-08-07 09:49:39 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
return (value & mask) ? NESTED_EXIT_DONE : NESTED_EXIT_HOST;
|
2009-08-07 09:49:35 +00:00
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
static int nested_svm_exit_special(struct vcpu_svm *svm)
|
2008-11-25 19:17:08 +00:00
|
|
|
{
|
|
|
|
u32 exit_code = svm->vmcb->control.exit_code;
|
2009-08-07 09:49:35 +00:00
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
switch (exit_code) {
|
|
|
|
case SVM_EXIT_INTR:
|
|
|
|
case SVM_EXIT_NMI:
|
2010-04-22 10:33:14 +00:00
|
|
|
case SVM_EXIT_EXCP_BASE + MC_VECTOR:
|
2009-08-07 09:49:44 +00:00
|
|
|
return NESTED_EXIT_HOST;
|
|
|
|
case SVM_EXIT_NPF:
|
2010-02-24 17:59:10 +00:00
|
|
|
/* For now we are always handling NPFs when using them */
|
2009-08-07 09:49:44 +00:00
|
|
|
if (npt_enabled)
|
|
|
|
return NESTED_EXIT_HOST;
|
|
|
|
break;
|
|
|
|
case SVM_EXIT_EXCP_BASE + PF_VECTOR:
|
2010-10-14 09:22:52 +00:00
|
|
|
/* When we're shadowing, trap PFs, but not async PF */
|
|
|
|
if (!npt_enabled && svm->apf_reason == 0)
|
2009-08-07 09:49:44 +00:00
|
|
|
return NESTED_EXIT_HOST;
|
|
|
|
break;
|
2010-02-19 15:23:08 +00:00
|
|
|
case SVM_EXIT_EXCP_BASE + NM_VECTOR:
|
|
|
|
nm_interception(svm);
|
|
|
|
break;
|
2009-08-07 09:49:44 +00:00
|
|
|
default:
|
|
|
|
break;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
return NESTED_EXIT_CONTINUE;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If this function returns true, this #vmexit was already handled
|
|
|
|
*/
|
2010-02-19 15:23:02 +00:00
|
|
|
static int nested_svm_intercept(struct vcpu_svm *svm)
|
2009-08-07 09:49:44 +00:00
|
|
|
{
|
|
|
|
u32 exit_code = svm->vmcb->control.exit_code;
|
|
|
|
int vmexit = NESTED_EXIT_HOST;
|
|
|
|
|
2008-11-25 19:17:08 +00:00
|
|
|
switch (exit_code) {
|
2009-08-07 09:49:36 +00:00
|
|
|
case SVM_EXIT_MSR:
|
2009-08-07 09:49:39 +00:00
|
|
|
vmexit = nested_svm_exit_handled_msr(svm);
|
2009-08-07 09:49:36 +00:00
|
|
|
break;
|
2010-03-01 14:34:39 +00:00
|
|
|
case SVM_EXIT_IOIO:
|
|
|
|
vmexit = nested_svm_intercept_ioio(svm);
|
|
|
|
break;
|
2010-12-03 09:50:51 +00:00
|
|
|
case SVM_EXIT_READ_CR0 ... SVM_EXIT_WRITE_CR8: {
|
|
|
|
u32 bit = 1U << (exit_code - SVM_EXIT_READ_CR0);
|
|
|
|
if (svm->nested.intercept_cr & bit)
|
2009-08-07 09:49:44 +00:00
|
|
|
vmexit = NESTED_EXIT_DONE;
|
2008-11-25 19:17:08 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-11-30 17:03:58 +00:00
|
|
|
case SVM_EXIT_READ_DR0 ... SVM_EXIT_WRITE_DR7: {
|
|
|
|
u32 bit = 1U << (exit_code - SVM_EXIT_READ_DR0);
|
|
|
|
if (svm->nested.intercept_dr & bit)
|
2009-08-07 09:49:44 +00:00
|
|
|
vmexit = NESTED_EXIT_DONE;
|
2008-11-25 19:17:08 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: {
|
|
|
|
u32 excp_bits = 1 << (exit_code - SVM_EXIT_EXCP_BASE);
|
2009-08-07 09:49:34 +00:00
|
|
|
if (svm->nested.intercept_exceptions & excp_bits)
|
2009-08-07 09:49:44 +00:00
|
|
|
vmexit = NESTED_EXIT_DONE;
|
2010-10-14 09:22:52 +00:00
|
|
|
/* async page fault always cause vmexit */
|
|
|
|
else if ((exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR) &&
|
|
|
|
svm->apf_reason != 0)
|
|
|
|
vmexit = NESTED_EXIT_DONE;
|
2008-11-25 19:17:08 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-04-22 10:33:10 +00:00
|
|
|
case SVM_EXIT_ERR: {
|
|
|
|
vmexit = NESTED_EXIT_DONE;
|
|
|
|
break;
|
|
|
|
}
|
2008-11-25 19:17:08 +00:00
|
|
|
default: {
|
|
|
|
u64 exit_bits = 1ULL << (exit_code - SVM_EXIT_INTR);
|
2009-08-07 09:49:34 +00:00
|
|
|
if (svm->nested.intercept & exit_bits)
|
2009-08-07 09:49:44 +00:00
|
|
|
vmexit = NESTED_EXIT_DONE;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-02-19 15:23:02 +00:00
|
|
|
return vmexit;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int nested_svm_exit_handled(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
int vmexit;
|
|
|
|
|
|
|
|
vmexit = nested_svm_intercept(svm);
|
|
|
|
|
|
|
|
if (vmexit == NESTED_EXIT_DONE)
|
2009-08-07 09:49:36 +00:00
|
|
|
nested_svm_vmexit(svm);
|
|
|
|
|
|
|
|
return vmexit;
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:31 +00:00
|
|
|
static inline void copy_vmcb_control_area(struct vmcb *dst_vmcb, struct vmcb *from_vmcb)
|
|
|
|
{
|
|
|
|
struct vmcb_control_area *dst = &dst_vmcb->control;
|
|
|
|
struct vmcb_control_area *from = &from_vmcb->control;
|
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
dst->intercept_cr = from->intercept_cr;
|
2010-11-30 17:03:58 +00:00
|
|
|
dst->intercept_dr = from->intercept_dr;
|
2009-08-07 09:49:31 +00:00
|
|
|
dst->intercept_exceptions = from->intercept_exceptions;
|
|
|
|
dst->intercept = from->intercept;
|
|
|
|
dst->iopm_base_pa = from->iopm_base_pa;
|
|
|
|
dst->msrpm_base_pa = from->msrpm_base_pa;
|
|
|
|
dst->tsc_offset = from->tsc_offset;
|
|
|
|
dst->asid = from->asid;
|
|
|
|
dst->tlb_ctl = from->tlb_ctl;
|
|
|
|
dst->int_ctl = from->int_ctl;
|
|
|
|
dst->int_vector = from->int_vector;
|
|
|
|
dst->int_state = from->int_state;
|
|
|
|
dst->exit_code = from->exit_code;
|
|
|
|
dst->exit_code_hi = from->exit_code_hi;
|
|
|
|
dst->exit_info_1 = from->exit_info_1;
|
|
|
|
dst->exit_info_2 = from->exit_info_2;
|
|
|
|
dst->exit_int_info = from->exit_int_info;
|
|
|
|
dst->exit_int_info_err = from->exit_int_info_err;
|
|
|
|
dst->nested_ctl = from->nested_ctl;
|
|
|
|
dst->event_inj = from->event_inj;
|
|
|
|
dst->event_inj_err = from->event_inj_err;
|
|
|
|
dst->nested_cr3 = from->nested_cr3;
|
|
|
|
dst->lbr_ctl = from->lbr_ctl;
|
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:38 +00:00
|
|
|
static int nested_svm_vmexit(struct vcpu_svm *svm)
|
2008-11-25 19:17:08 +00:00
|
|
|
{
|
2009-08-07 09:49:38 +00:00
|
|
|
struct vmcb *nested_vmcb;
|
2009-08-07 09:49:33 +00:00
|
|
|
struct vmcb *hsave = svm->nested.hsave;
|
2009-08-07 09:49:29 +00:00
|
|
|
struct vmcb *vmcb = svm->vmcb;
|
2010-02-19 15:23:00 +00:00
|
|
|
struct page *page;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2009-10-09 14:08:29 +00:00
|
|
|
trace_kvm_nested_vmexit_inject(vmcb->control.exit_code,
|
|
|
|
vmcb->control.exit_info_1,
|
|
|
|
vmcb->control.exit_info_2,
|
|
|
|
vmcb->control.exit_int_info,
|
2011-07-22 11:46:52 +00:00
|
|
|
vmcb->control.exit_int_info_err,
|
|
|
|
KVM_ISA_SVM);
|
2009-10-09 14:08:29 +00:00
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_vmcb = nested_svm_map(svm, svm->nested.vmcb, &page);
|
2009-08-07 09:49:38 +00:00
|
|
|
if (!nested_vmcb)
|
|
|
|
return 1;
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
/* Exit Guest-Mode */
|
|
|
|
leave_guest_mode(&svm->vcpu);
|
2010-02-19 15:23:07 +00:00
|
|
|
svm->nested.vmcb = 0;
|
|
|
|
|
2008-11-25 19:17:08 +00:00
|
|
|
/* Give the current vmcb to the guest */
|
2009-08-07 09:49:29 +00:00
|
|
|
disable_gif(svm);
|
|
|
|
|
|
|
|
nested_vmcb->save.es = vmcb->save.es;
|
|
|
|
nested_vmcb->save.cs = vmcb->save.cs;
|
|
|
|
nested_vmcb->save.ss = vmcb->save.ss;
|
|
|
|
nested_vmcb->save.ds = vmcb->save.ds;
|
|
|
|
nested_vmcb->save.gdtr = vmcb->save.gdtr;
|
|
|
|
nested_vmcb->save.idtr = vmcb->save.idtr;
|
2010-07-27 16:14:20 +00:00
|
|
|
nested_vmcb->save.efer = svm->vcpu.arch.efer;
|
2010-02-19 15:23:03 +00:00
|
|
|
nested_vmcb->save.cr0 = kvm_read_cr0(&svm->vcpu);
|
2010-12-05 15:30:00 +00:00
|
|
|
nested_vmcb->save.cr3 = kvm_read_cr3(&svm->vcpu);
|
2009-08-07 09:49:29 +00:00
|
|
|
nested_vmcb->save.cr2 = vmcb->save.cr2;
|
2010-02-19 15:23:03 +00:00
|
|
|
nested_vmcb->save.cr4 = svm->vcpu.arch.cr4;
|
2010-08-02 12:30:20 +00:00
|
|
|
nested_vmcb->save.rflags = kvm_get_rflags(&svm->vcpu);
|
2009-08-07 09:49:29 +00:00
|
|
|
nested_vmcb->save.rip = vmcb->save.rip;
|
|
|
|
nested_vmcb->save.rsp = vmcb->save.rsp;
|
|
|
|
nested_vmcb->save.rax = vmcb->save.rax;
|
|
|
|
nested_vmcb->save.dr7 = vmcb->save.dr7;
|
|
|
|
nested_vmcb->save.dr6 = vmcb->save.dr6;
|
|
|
|
nested_vmcb->save.cpl = vmcb->save.cpl;
|
|
|
|
|
|
|
|
nested_vmcb->control.int_ctl = vmcb->control.int_ctl;
|
|
|
|
nested_vmcb->control.int_vector = vmcb->control.int_vector;
|
|
|
|
nested_vmcb->control.int_state = vmcb->control.int_state;
|
|
|
|
nested_vmcb->control.exit_code = vmcb->control.exit_code;
|
|
|
|
nested_vmcb->control.exit_code_hi = vmcb->control.exit_code_hi;
|
|
|
|
nested_vmcb->control.exit_info_1 = vmcb->control.exit_info_1;
|
|
|
|
nested_vmcb->control.exit_info_2 = vmcb->control.exit_info_2;
|
|
|
|
nested_vmcb->control.exit_int_info = vmcb->control.exit_int_info;
|
|
|
|
nested_vmcb->control.exit_int_info_err = vmcb->control.exit_int_info_err;
|
2010-07-27 16:14:21 +00:00
|
|
|
nested_vmcb->control.next_rip = vmcb->control.next_rip;
|
2009-10-09 14:08:25 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If we emulate a VMRUN/#VMEXIT in the same host #vmexit cycle we have
|
|
|
|
* to make sure that we do not lose injected events. So check event_inj
|
|
|
|
* here and copy it to exit_int_info if it is valid.
|
|
|
|
* Exit_int_info and event_inj can't be both valid because the case
|
|
|
|
* below only happens on a VMRUN instruction intercept which has
|
|
|
|
* no valid exit_int_info set.
|
|
|
|
*/
|
|
|
|
if (vmcb->control.event_inj & SVM_EVTINJ_VALID) {
|
|
|
|
struct vmcb_control_area *nc = &nested_vmcb->control;
|
|
|
|
|
|
|
|
nc->exit_int_info = vmcb->control.event_inj;
|
|
|
|
nc->exit_int_info_err = vmcb->control.event_inj_err;
|
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:29 +00:00
|
|
|
nested_vmcb->control.tlb_ctl = 0;
|
|
|
|
nested_vmcb->control.event_inj = 0;
|
|
|
|
nested_vmcb->control.event_inj_err = 0;
|
2008-11-25 19:17:08 +00:00
|
|
|
|
|
|
|
/* We always set V_INTR_MASKING and remember the old value in hflags */
|
|
|
|
if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK))
|
|
|
|
nested_vmcb->control.int_ctl &= ~V_INTR_MASKING_MASK;
|
|
|
|
|
|
|
|
/* Restore the original control entries */
|
2009-08-07 09:49:31 +00:00
|
|
|
copy_vmcb_control_area(vmcb, hsave);
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2009-06-15 13:21:25 +00:00
|
|
|
kvm_clear_exception_queue(&svm->vcpu);
|
|
|
|
kvm_clear_interrupt_queue(&svm->vcpu);
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2010-09-10 15:31:03 +00:00
|
|
|
svm->nested.nested_cr3 = 0;
|
|
|
|
|
2008-11-25 19:17:08 +00:00
|
|
|
/* Restore selected save entries */
|
|
|
|
svm->vmcb->save.es = hsave->save.es;
|
|
|
|
svm->vmcb->save.cs = hsave->save.cs;
|
|
|
|
svm->vmcb->save.ss = hsave->save.ss;
|
|
|
|
svm->vmcb->save.ds = hsave->save.ds;
|
|
|
|
svm->vmcb->save.gdtr = hsave->save.gdtr;
|
|
|
|
svm->vmcb->save.idtr = hsave->save.idtr;
|
2010-08-02 12:30:20 +00:00
|
|
|
kvm_set_rflags(&svm->vcpu, hsave->save.rflags);
|
2008-11-25 19:17:08 +00:00
|
|
|
svm_set_efer(&svm->vcpu, hsave->save.efer);
|
|
|
|
svm_set_cr0(&svm->vcpu, hsave->save.cr0 | X86_CR0_PE);
|
|
|
|
svm_set_cr4(&svm->vcpu, hsave->save.cr4);
|
|
|
|
if (npt_enabled) {
|
|
|
|
svm->vmcb->save.cr3 = hsave->save.cr3;
|
|
|
|
svm->vcpu.arch.cr3 = hsave->save.cr3;
|
|
|
|
} else {
|
2010-06-10 14:02:16 +00:00
|
|
|
(void)kvm_set_cr3(&svm->vcpu, hsave->save.cr3);
|
2008-11-25 19:17:08 +00:00
|
|
|
}
|
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, hsave->save.rax);
|
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, hsave->save.rsp);
|
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, hsave->save.rip);
|
|
|
|
svm->vmcb->save.dr7 = 0;
|
|
|
|
svm->vmcb->save.cpl = 0;
|
|
|
|
svm->vmcb->control.exit_int_info = 0;
|
|
|
|
|
2010-12-03 12:15:21 +00:00
|
|
|
mark_all_dirty(svm->vmcb);
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_svm_unmap(page);
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2010-09-10 15:31:03 +00:00
|
|
|
nested_svm_uninit_mmu_context(&svm->vcpu);
|
2008-11-25 19:17:08 +00:00
|
|
|
kvm_mmu_reset_context(&svm->vcpu);
|
|
|
|
kvm_mmu_load(&svm->vcpu);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2009-08-07 09:49:41 +00:00
|
|
|
static bool nested_svm_vmrun_msrpm(struct vcpu_svm *svm)
|
2008-11-25 19:17:07 +00:00
|
|
|
{
|
2010-03-01 14:34:37 +00:00
|
|
|
/*
|
|
|
|
* This function merges the msr permission bitmaps of kvm and the
|
2012-06-28 07:16:43 +00:00
|
|
|
* nested vmcb. It is optimized in that it only merges the parts where
|
2010-03-01 14:34:37 +00:00
|
|
|
* the kvm msr permission bitmap may contain zero bits
|
|
|
|
*/
|
2008-11-25 19:17:07 +00:00
|
|
|
int i;
|
2009-08-07 09:49:41 +00:00
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
if (!(svm->nested.intercept & (1ULL << INTERCEPT_MSR_PROT)))
|
|
|
|
return true;
|
2009-08-07 09:49:41 +00:00
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
for (i = 0; i < MSRPM_OFFSETS; i++) {
|
|
|
|
u32 value, p;
|
|
|
|
u64 offset;
|
2009-08-07 09:49:41 +00:00
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
if (msrpm_offsets[i] == 0xffffffff)
|
|
|
|
break;
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-03-01 14:34:38 +00:00
|
|
|
p = msrpm_offsets[i];
|
|
|
|
offset = svm->nested.vmcb_msrpm + (p * 4);
|
2010-03-01 14:34:37 +00:00
|
|
|
|
|
|
|
if (kvm_read_guest(svm->vcpu.kvm, offset, &value, 4))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
svm->nested.msrpm[p] = svm->msrpm[p] | value;
|
|
|
|
}
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-03-01 14:34:37 +00:00
|
|
|
svm->vmcb->control.msrpm_base_pa = __pa(svm->nested.msrpm);
|
2009-08-07 09:49:41 +00:00
|
|
|
|
|
|
|
return true;
|
2008-11-25 19:17:07 +00:00
|
|
|
}
|
|
|
|
|
2010-08-02 14:46:44 +00:00
|
|
|
static bool nested_vmcb_checks(struct vmcb *vmcb)
|
|
|
|
{
|
|
|
|
if ((vmcb->control.intercept & (1ULL << INTERCEPT_VMRUN)) == 0)
|
|
|
|
return false;
|
|
|
|
|
2010-08-02 14:46:45 +00:00
|
|
|
if (vmcb->control.asid == 0)
|
|
|
|
return false;
|
|
|
|
|
2010-09-10 15:31:03 +00:00
|
|
|
if (vmcb->control.nested_ctl && !npt_enabled)
|
|
|
|
return false;
|
|
|
|
|
2010-08-02 14:46:44 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:41 +00:00
|
|
|
static bool nested_svm_vmrun(struct vcpu_svm *svm)
|
2008-11-25 19:17:07 +00:00
|
|
|
{
|
2009-08-07 09:49:41 +00:00
|
|
|
struct vmcb *nested_vmcb;
|
2009-08-07 09:49:33 +00:00
|
|
|
struct vmcb *hsave = svm->nested.hsave;
|
2009-08-07 09:49:30 +00:00
|
|
|
struct vmcb *vmcb = svm->vmcb;
|
2010-02-19 15:23:00 +00:00
|
|
|
struct page *page;
|
2010-02-19 15:23:07 +00:00
|
|
|
u64 vmcb_gpa;
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-02-19 15:23:07 +00:00
|
|
|
vmcb_gpa = svm->vmcb->save.rax;
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_vmcb = nested_svm_map(svm, svm->vmcb->save.rax, &page);
|
2009-08-07 09:49:41 +00:00
|
|
|
if (!nested_vmcb)
|
|
|
|
return false;
|
|
|
|
|
2010-08-02 14:46:44 +00:00
|
|
|
if (!nested_vmcb_checks(nested_vmcb)) {
|
|
|
|
nested_vmcb->control.exit_code = SVM_EXIT_ERR;
|
|
|
|
nested_vmcb->control.exit_code_hi = 0;
|
|
|
|
nested_vmcb->control.exit_info_1 = 0;
|
|
|
|
nested_vmcb->control.exit_info_2 = 0;
|
|
|
|
|
|
|
|
nested_svm_unmap(page);
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2010-09-03 12:21:40 +00:00
|
|
|
trace_kvm_nested_vmrun(svm->vmcb->save.rip, vmcb_gpa,
|
2009-10-09 14:08:27 +00:00
|
|
|
nested_vmcb->save.rip,
|
|
|
|
nested_vmcb->control.int_ctl,
|
|
|
|
nested_vmcb->control.event_inj,
|
|
|
|
nested_vmcb->control.nested_ctl);
|
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
trace_kvm_nested_intercepts(nested_vmcb->control.intercept_cr & 0xffff,
|
|
|
|
nested_vmcb->control.intercept_cr >> 16,
|
2010-02-24 17:59:14 +00:00
|
|
|
nested_vmcb->control.intercept_exceptions,
|
|
|
|
nested_vmcb->control.intercept);
|
|
|
|
|
2008-11-25 19:17:07 +00:00
|
|
|
/* Clear internal status */
|
2009-06-15 13:21:25 +00:00
|
|
|
kvm_clear_exception_queue(&svm->vcpu);
|
|
|
|
kvm_clear_interrupt_queue(&svm->vcpu);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* Save the old vmcb, so we don't need to pick what we save, but can
|
|
|
|
* restore everything when a VMEXIT occurs
|
|
|
|
*/
|
2009-08-07 09:49:30 +00:00
|
|
|
hsave->save.es = vmcb->save.es;
|
|
|
|
hsave->save.cs = vmcb->save.cs;
|
|
|
|
hsave->save.ss = vmcb->save.ss;
|
|
|
|
hsave->save.ds = vmcb->save.ds;
|
|
|
|
hsave->save.gdtr = vmcb->save.gdtr;
|
|
|
|
hsave->save.idtr = vmcb->save.idtr;
|
2010-01-21 13:31:50 +00:00
|
|
|
hsave->save.efer = svm->vcpu.arch.efer;
|
2009-12-29 16:07:30 +00:00
|
|
|
hsave->save.cr0 = kvm_read_cr0(&svm->vcpu);
|
2009-08-07 09:49:30 +00:00
|
|
|
hsave->save.cr4 = svm->vcpu.arch.cr4;
|
2010-08-02 12:30:20 +00:00
|
|
|
hsave->save.rflags = kvm_get_rflags(&svm->vcpu);
|
2010-09-03 12:21:40 +00:00
|
|
|
hsave->save.rip = kvm_rip_read(&svm->vcpu);
|
2009-08-07 09:49:30 +00:00
|
|
|
hsave->save.rsp = vmcb->save.rsp;
|
|
|
|
hsave->save.rax = vmcb->save.rax;
|
|
|
|
if (npt_enabled)
|
|
|
|
hsave->save.cr3 = vmcb->save.cr3;
|
|
|
|
else
|
2010-12-05 15:30:00 +00:00
|
|
|
hsave->save.cr3 = kvm_read_cr3(&svm->vcpu);
|
2009-08-07 09:49:30 +00:00
|
|
|
|
2009-08-07 09:49:31 +00:00
|
|
|
copy_vmcb_control_area(hsave, vmcb);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-08-02 12:30:20 +00:00
|
|
|
if (kvm_get_rflags(&svm->vcpu) & X86_EFLAGS_IF)
|
2008-11-25 19:17:07 +00:00
|
|
|
svm->vcpu.arch.hflags |= HF_HIF_MASK;
|
|
|
|
else
|
|
|
|
svm->vcpu.arch.hflags &= ~HF_HIF_MASK;
|
|
|
|
|
2010-09-10 15:31:03 +00:00
|
|
|
if (nested_vmcb->control.nested_ctl) {
|
|
|
|
kvm_mmu_unload(&svm->vcpu);
|
|
|
|
svm->nested.nested_cr3 = nested_vmcb->control.nested_cr3;
|
|
|
|
nested_svm_init_mmu_context(&svm->vcpu);
|
|
|
|
}
|
|
|
|
|
2008-11-25 19:17:07 +00:00
|
|
|
/* Load the nested guest state */
|
|
|
|
svm->vmcb->save.es = nested_vmcb->save.es;
|
|
|
|
svm->vmcb->save.cs = nested_vmcb->save.cs;
|
|
|
|
svm->vmcb->save.ss = nested_vmcb->save.ss;
|
|
|
|
svm->vmcb->save.ds = nested_vmcb->save.ds;
|
|
|
|
svm->vmcb->save.gdtr = nested_vmcb->save.gdtr;
|
|
|
|
svm->vmcb->save.idtr = nested_vmcb->save.idtr;
|
2010-08-02 12:30:20 +00:00
|
|
|
kvm_set_rflags(&svm->vcpu, nested_vmcb->save.rflags);
|
2008-11-25 19:17:07 +00:00
|
|
|
svm_set_efer(&svm->vcpu, nested_vmcb->save.efer);
|
|
|
|
svm_set_cr0(&svm->vcpu, nested_vmcb->save.cr0);
|
|
|
|
svm_set_cr4(&svm->vcpu, nested_vmcb->save.cr4);
|
|
|
|
if (npt_enabled) {
|
|
|
|
svm->vmcb->save.cr3 = nested_vmcb->save.cr3;
|
|
|
|
svm->vcpu.arch.cr3 = nested_vmcb->save.cr3;
|
2010-02-24 17:59:11 +00:00
|
|
|
} else
|
2010-06-10 14:02:16 +00:00
|
|
|
(void)kvm_set_cr3(&svm->vcpu, nested_vmcb->save.cr3);
|
2010-02-24 17:59:11 +00:00
|
|
|
|
|
|
|
/* Guest paging mode is active - reset mmu */
|
|
|
|
kvm_mmu_reset_context(&svm->vcpu);
|
|
|
|
|
2009-08-07 09:49:30 +00:00
|
|
|
svm->vmcb->save.cr2 = svm->vcpu.arch.cr2 = nested_vmcb->save.cr2;
|
2008-11-25 19:17:07 +00:00
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, nested_vmcb->save.rax);
|
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, nested_vmcb->save.rsp);
|
|
|
|
kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, nested_vmcb->save.rip);
|
2010-02-24 17:59:10 +00:00
|
|
|
|
2008-11-25 19:17:07 +00:00
|
|
|
/* In case we don't even reach vcpu_run, the fields are not updated */
|
|
|
|
svm->vmcb->save.rax = nested_vmcb->save.rax;
|
|
|
|
svm->vmcb->save.rsp = nested_vmcb->save.rsp;
|
|
|
|
svm->vmcb->save.rip = nested_vmcb->save.rip;
|
|
|
|
svm->vmcb->save.dr7 = nested_vmcb->save.dr7;
|
|
|
|
svm->vmcb->save.dr6 = nested_vmcb->save.dr6;
|
|
|
|
svm->vmcb->save.cpl = nested_vmcb->save.cpl;
|
|
|
|
|
2010-03-01 14:34:40 +00:00
|
|
|
svm->nested.vmcb_msrpm = nested_vmcb->control.msrpm_base_pa & ~0x0fffULL;
|
2010-03-01 14:34:39 +00:00
|
|
|
svm->nested.vmcb_iopm = nested_vmcb->control.iopm_base_pa & ~0x0fffULL;
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2009-08-07 09:49:34 +00:00
|
|
|
/* cache intercepts */
|
2010-12-03 09:50:51 +00:00
|
|
|
svm->nested.intercept_cr = nested_vmcb->control.intercept_cr;
|
2010-11-30 17:03:58 +00:00
|
|
|
svm->nested.intercept_dr = nested_vmcb->control.intercept_dr;
|
2009-08-07 09:49:34 +00:00
|
|
|
svm->nested.intercept_exceptions = nested_vmcb->control.intercept_exceptions;
|
|
|
|
svm->nested.intercept = nested_vmcb->control.intercept;
|
|
|
|
|
2010-12-03 14:25:15 +00:00
|
|
|
svm_flush_tlb(&svm->vcpu);
|
2008-11-25 19:17:07 +00:00
|
|
|
svm->vmcb->control.int_ctl = nested_vmcb->control.int_ctl | V_INTR_MASKING_MASK;
|
|
|
|
if (nested_vmcb->control.int_ctl & V_INTR_MASKING_MASK)
|
|
|
|
svm->vcpu.arch.hflags |= HF_VINTR_MASK;
|
|
|
|
else
|
|
|
|
svm->vcpu.arch.hflags &= ~HF_VINTR_MASK;
|
|
|
|
|
2010-02-19 15:23:06 +00:00
|
|
|
if (svm->vcpu.arch.hflags & HF_VINTR_MASK) {
|
|
|
|
/* We only want the cr8 intercept bits of the guest */
|
2010-12-03 09:50:51 +00:00
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR8_READ);
|
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR8_WRITE);
|
2010-02-19 15:23:06 +00:00
|
|
|
}
|
|
|
|
|
2010-05-05 14:04:45 +00:00
|
|
|
/* We don't want to see VMMCALLs from a nested guest */
|
2010-11-30 17:04:00 +00:00
|
|
|
clr_intercept(svm, INTERCEPT_VMMCALL);
|
2010-05-05 14:04:45 +00:00
|
|
|
|
2010-02-19 15:23:06 +00:00
|
|
|
svm->vmcb->control.lbr_ctl = nested_vmcb->control.lbr_ctl;
|
2008-11-25 19:17:07 +00:00
|
|
|
svm->vmcb->control.int_vector = nested_vmcb->control.int_vector;
|
|
|
|
svm->vmcb->control.int_state = nested_vmcb->control.int_state;
|
|
|
|
svm->vmcb->control.tsc_offset += nested_vmcb->control.tsc_offset;
|
|
|
|
svm->vmcb->control.event_inj = nested_vmcb->control.event_inj;
|
|
|
|
svm->vmcb->control.event_inj_err = nested_vmcb->control.event_inj_err;
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_svm_unmap(page);
|
2009-08-07 09:49:41 +00:00
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
/* Enter Guest-Mode */
|
|
|
|
enter_guest_mode(&svm->vcpu);
|
|
|
|
|
2010-11-30 17:03:56 +00:00
|
|
|
/*
|
|
|
|
* Merge guest and host intercepts - must be called with vcpu in
|
|
|
|
* guest-mode to take affect here
|
|
|
|
*/
|
|
|
|
recalc_intercepts(svm);
|
|
|
|
|
2010-02-19 15:23:07 +00:00
|
|
|
svm->nested.vmcb = vmcb_gpa;
|
2009-08-07 09:49:41 +00:00
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
enable_gif(svm);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2010-12-03 12:15:21 +00:00
|
|
|
mark_all_dirty(svm->vmcb);
|
|
|
|
|
2009-08-07 09:49:41 +00:00
|
|
|
return true;
|
2008-11-25 19:17:07 +00:00
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:40 +00:00
|
|
|
static void nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb)
|
2008-11-25 19:17:06 +00:00
|
|
|
{
|
|
|
|
to_vmcb->save.fs = from_vmcb->save.fs;
|
|
|
|
to_vmcb->save.gs = from_vmcb->save.gs;
|
|
|
|
to_vmcb->save.tr = from_vmcb->save.tr;
|
|
|
|
to_vmcb->save.ldtr = from_vmcb->save.ldtr;
|
|
|
|
to_vmcb->save.kernel_gs_base = from_vmcb->save.kernel_gs_base;
|
|
|
|
to_vmcb->save.star = from_vmcb->save.star;
|
|
|
|
to_vmcb->save.lstar = from_vmcb->save.lstar;
|
|
|
|
to_vmcb->save.cstar = from_vmcb->save.cstar;
|
|
|
|
to_vmcb->save.sfmask = from_vmcb->save.sfmask;
|
|
|
|
to_vmcb->save.sysenter_cs = from_vmcb->save.sysenter_cs;
|
|
|
|
to_vmcb->save.sysenter_esp = from_vmcb->save.sysenter_esp;
|
|
|
|
to_vmcb->save.sysenter_eip = from_vmcb->save.sysenter_eip;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int vmload_interception(struct vcpu_svm *svm)
|
2008-11-25 19:17:06 +00:00
|
|
|
{
|
2009-08-07 09:49:40 +00:00
|
|
|
struct vmcb *nested_vmcb;
|
2010-02-19 15:23:00 +00:00
|
|
|
struct page *page;
|
2009-08-07 09:49:40 +00:00
|
|
|
|
2008-11-25 19:17:06 +00:00
|
|
|
if (nested_svm_check_permissions(svm))
|
|
|
|
return 1;
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_vmcb = nested_svm_map(svm, svm->vmcb->save.rax, &page);
|
2009-08-07 09:49:40 +00:00
|
|
|
if (!nested_vmcb)
|
|
|
|
return 1;
|
|
|
|
|
2011-04-06 10:30:03 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
|
2009-08-07 09:49:40 +00:00
|
|
|
nested_svm_vmloadsave(nested_vmcb, svm->vmcb);
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_svm_unmap(page);
|
2008-11-25 19:17:06 +00:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int vmsave_interception(struct vcpu_svm *svm)
|
2008-11-25 19:17:06 +00:00
|
|
|
{
|
2009-08-07 09:49:40 +00:00
|
|
|
struct vmcb *nested_vmcb;
|
2010-02-19 15:23:00 +00:00
|
|
|
struct page *page;
|
2009-08-07 09:49:40 +00:00
|
|
|
|
2008-11-25 19:17:06 +00:00
|
|
|
if (nested_svm_check_permissions(svm))
|
|
|
|
return 1;
|
|
|
|
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_vmcb = nested_svm_map(svm, svm->vmcb->save.rax, &page);
|
2009-08-07 09:49:40 +00:00
|
|
|
if (!nested_vmcb)
|
|
|
|
return 1;
|
|
|
|
|
2011-04-06 10:30:03 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
|
2009-08-07 09:49:40 +00:00
|
|
|
nested_svm_vmloadsave(svm->vmcb, nested_vmcb);
|
2010-02-19 15:23:00 +00:00
|
|
|
nested_svm_unmap(page);
|
2008-11-25 19:17:06 +00:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int vmrun_interception(struct vcpu_svm *svm)
|
2008-11-25 19:17:07 +00:00
|
|
|
{
|
|
|
|
if (nested_svm_check_permissions(svm))
|
|
|
|
return 1;
|
|
|
|
|
2010-09-03 12:21:40 +00:00
|
|
|
/* Save rip after vmrun instruction */
|
|
|
|
kvm_rip_write(&svm->vcpu, kvm_rip_read(&svm->vcpu) + 3);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
2009-08-07 09:49:41 +00:00
|
|
|
if (!nested_svm_vmrun(svm))
|
2008-11-25 19:17:07 +00:00
|
|
|
return 1;
|
|
|
|
|
2009-08-07 09:49:41 +00:00
|
|
|
if (!nested_svm_vmrun_msrpm(svm))
|
2009-08-07 09:49:43 +00:00
|
|
|
goto failed;
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
failed:
|
|
|
|
|
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_ERR;
|
|
|
|
svm->vmcb->control.exit_code_hi = 0;
|
|
|
|
svm->vmcb->control.exit_info_1 = 0;
|
|
|
|
svm->vmcb->control.exit_info_2 = 0;
|
|
|
|
|
|
|
|
nested_svm_vmexit(svm);
|
2008-11-25 19:17:07 +00:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int stgi_interception(struct vcpu_svm *svm)
|
2008-11-25 19:17:04 +00:00
|
|
|
{
|
|
|
|
if (nested_svm_check_permissions(svm))
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
2010-07-27 09:30:24 +00:00
|
|
|
kvm_make_request(KVM_REQ_EVENT, &svm->vcpu);
|
2008-11-25 19:17:04 +00:00
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
enable_gif(svm);
|
2008-11-25 19:17:04 +00:00
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int clgi_interception(struct vcpu_svm *svm)
|
2008-11-25 19:17:04 +00:00
|
|
|
{
|
|
|
|
if (nested_svm_check_permissions(svm))
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
disable_gif(svm);
|
2008-11-25 19:17:04 +00:00
|
|
|
|
|
|
|
/* After a CLGI no interrupts should come */
|
|
|
|
svm_clear_vintr(svm);
|
|
|
|
svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
|
|
|
|
|
2010-12-03 10:45:52 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_INTR);
|
|
|
|
|
2008-11-25 19:17:04 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int invlpga_interception(struct vcpu_svm *svm)
|
2009-06-15 13:21:24 +00:00
|
|
|
{
|
|
|
|
struct kvm_vcpu *vcpu = &svm->vcpu;
|
|
|
|
|
2009-10-09 14:08:31 +00:00
|
|
|
trace_kvm_invlpga(svm->vmcb->save.rip, vcpu->arch.regs[VCPU_REGS_RCX],
|
|
|
|
vcpu->arch.regs[VCPU_REGS_RAX]);
|
|
|
|
|
2009-06-15 13:21:24 +00:00
|
|
|
/* Let's treat INVLPGA the same as INVLPG (can be optimized!) */
|
|
|
|
kvm_mmu_invlpg(vcpu, vcpu->arch.regs[VCPU_REGS_RAX]);
|
|
|
|
|
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-10-09 14:08:32 +00:00
|
|
|
static int skinit_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
trace_kvm_skinit(svm->vmcb->save.rip, svm->vcpu.arch.regs[VCPU_REGS_RAX]);
|
|
|
|
|
|
|
|
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-12-07 16:15:06 +00:00
|
|
|
static int xsetbv_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
u64 new_bv = kvm_read_edx_eax(&svm->vcpu);
|
|
|
|
u32 index = kvm_register_read(&svm->vcpu, VCPU_REGS_RCX);
|
|
|
|
|
|
|
|
if (kvm_set_xcr(&svm->vcpu, index, new_bv) == 0) {
|
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 3;
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int task_switch_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-03-24 21:14:53 +00:00
|
|
|
u16 tss_selector;
|
2009-03-30 13:03:29 +00:00
|
|
|
int reason;
|
|
|
|
int int_type = svm->vmcb->control.exit_int_info &
|
|
|
|
SVM_EXITINTINFO_TYPE_MASK;
|
2009-04-12 10:37:02 +00:00
|
|
|
int int_vec = svm->vmcb->control.exit_int_info & SVM_EVTINJ_VEC_MASK;
|
2009-04-23 14:03:48 +00:00
|
|
|
uint32_t type =
|
|
|
|
svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_TYPE_MASK;
|
|
|
|
uint32_t idt_v =
|
|
|
|
svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_VALID;
|
2010-04-14 13:51:09 +00:00
|
|
|
bool has_error_code = false;
|
|
|
|
u32 error_code = 0;
|
2008-03-24 21:14:53 +00:00
|
|
|
|
|
|
|
tss_selector = (u16)svm->vmcb->control.exit_info_1;
|
2009-03-30 13:03:29 +00:00
|
|
|
|
2008-03-24 21:14:53 +00:00
|
|
|
if (svm->vmcb->control.exit_info_2 &
|
|
|
|
(1ULL << SVM_EXITINFOSHIFT_TS_REASON_IRET))
|
2009-03-30 13:03:29 +00:00
|
|
|
reason = TASK_SWITCH_IRET;
|
|
|
|
else if (svm->vmcb->control.exit_info_2 &
|
|
|
|
(1ULL << SVM_EXITINFOSHIFT_TS_REASON_JMP))
|
|
|
|
reason = TASK_SWITCH_JMP;
|
2009-04-23 14:03:48 +00:00
|
|
|
else if (idt_v)
|
2009-03-30 13:03:29 +00:00
|
|
|
reason = TASK_SWITCH_GATE;
|
|
|
|
else
|
|
|
|
reason = TASK_SWITCH_CALL;
|
|
|
|
|
2009-04-23 14:03:48 +00:00
|
|
|
if (reason == TASK_SWITCH_GATE) {
|
|
|
|
switch (type) {
|
|
|
|
case SVM_EXITINTINFO_TYPE_NMI:
|
|
|
|
svm->vcpu.arch.nmi_injected = false;
|
|
|
|
break;
|
|
|
|
case SVM_EXITINTINFO_TYPE_EXEPT:
|
2010-04-14 13:51:09 +00:00
|
|
|
if (svm->vmcb->control.exit_info_2 &
|
|
|
|
(1ULL << SVM_EXITINFOSHIFT_TS_HAS_ERROR_CODE)) {
|
|
|
|
has_error_code = true;
|
|
|
|
error_code =
|
|
|
|
(u32)svm->vmcb->control.exit_info_2;
|
|
|
|
}
|
2009-04-23 14:03:48 +00:00
|
|
|
kvm_clear_exception_queue(&svm->vcpu);
|
|
|
|
break;
|
|
|
|
case SVM_EXITINTINFO_TYPE_INTR:
|
|
|
|
kvm_clear_interrupt_queue(&svm->vcpu);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2009-03-30 13:03:29 +00:00
|
|
|
|
2009-04-12 10:37:02 +00:00
|
|
|
if (reason != TASK_SWITCH_GATE ||
|
|
|
|
int_type == SVM_EXITINTINFO_TYPE_SOFT ||
|
|
|
|
(int_type == SVM_EXITINTINFO_TYPE_EXEPT &&
|
2009-05-11 10:35:49 +00:00
|
|
|
(int_vec == OF_VECTOR || int_vec == BP_VECTOR)))
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
2009-03-30 13:03:29 +00:00
|
|
|
|
2012-02-08 13:34:38 +00:00
|
|
|
if (int_type != SVM_EXITINTINFO_TYPE_SOFT)
|
|
|
|
int_vec = -1;
|
|
|
|
|
|
|
|
if (kvm_task_switch(&svm->vcpu, tss_selector, int_vec, reason,
|
2010-04-15 18:03:50 +00:00
|
|
|
has_error_code, error_code) == EMULATE_FAIL) {
|
|
|
|
svm->vcpu.run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
|
|
|
|
svm->vcpu.run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
|
|
|
|
svm->vcpu.run->internal.ndata = 0;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return 1;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int cpuid_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 2;
|
2007-07-30 10:07:08 +00:00
|
|
|
kvm_emulate_cpuid(&svm->vcpu);
|
2007-02-28 18:46:53 +00:00
|
|
|
return 1;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int iret_interception(struct vcpu_svm *svm)
|
2009-04-21 14:45:08 +00:00
|
|
|
{
|
|
|
|
++svm->vcpu.stat.nmi_window_exits;
|
2010-11-30 17:04:00 +00:00
|
|
|
clr_intercept(svm, INTERCEPT_IRET);
|
2009-05-11 10:35:52 +00:00
|
|
|
svm->vcpu.arch.hflags |= HF_IRET_MASK;
|
KVM: SVM: check for progress after IRET interception
When we enable an NMI window, we ask for an IRET intercept, since
the IRET re-enables NMIs. However, the IRET intercept happens before
the instruction executes, while the NMI window architecturally opens
afterwards.
To compensate for this mismatch, we only open the NMI window in the
following exit, assuming that the IRET has by then executed; however,
this assumption is not always correct; we may exit due to a host interrupt
or page fault, without having executed the instruction.
Fix by checking for forward progress by recording and comparing the IRET's
rip. This is somewhat of a hack, since an unchaging rip does not mean that
no forward progress has been made, but is the simplest fix for now.
Signed-off-by: Avi Kivity <avi@redhat.com>
2011-02-03 13:29:52 +00:00
|
|
|
svm->nmi_iret_rip = kvm_rip_read(&svm->vcpu);
|
2014-01-17 19:52:42 +00:00
|
|
|
kvm_make_request(KVM_REQ_EVENT, &svm->vcpu);
|
2009-04-21 14:45:08 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int invlpg_interception(struct vcpu_svm *svm)
|
2008-09-23 16:18:35 +00:00
|
|
|
{
|
2010-12-21 10:12:06 +00:00
|
|
|
if (!static_cpu_has(X86_FEATURE_DECODEASSISTS))
|
|
|
|
return emulate_instruction(&svm->vcpu, 0) == EMULATE_DONE;
|
|
|
|
|
|
|
|
kvm_mmu_invlpg(&svm->vcpu, svm->vmcb->control.exit_info_1);
|
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
return 1;
|
2008-09-23 16:18:35 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int emulate_on_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2010-12-21 10:12:02 +00:00
|
|
|
return emulate_instruction(&svm->vcpu, 0) == EMULATE_DONE;
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2011-11-10 12:57:24 +00:00
|
|
|
static int rdpmc_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
|
|
|
if (!static_cpu_has(X86_FEATURE_NRIPS))
|
|
|
|
return emulate_on_interception(svm);
|
|
|
|
|
|
|
|
err = kvm_rdpmc(&svm->vcpu);
|
|
|
|
kvm_complete_insn_gp(&svm->vcpu, err);
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2011-04-04 10:39:36 +00:00
|
|
|
bool check_selective_cr0_intercepted(struct vcpu_svm *svm, unsigned long val)
|
|
|
|
{
|
|
|
|
unsigned long cr0 = svm->vcpu.arch.cr0;
|
|
|
|
bool ret = false;
|
|
|
|
u64 intercept;
|
|
|
|
|
|
|
|
intercept = svm->nested.intercept;
|
|
|
|
|
|
|
|
if (!is_guest_mode(&svm->vcpu) ||
|
|
|
|
(!(intercept & (1ULL << INTERCEPT_SELECTIVE_CR0))))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
cr0 &= ~SVM_CR0_SELECTIVE_MASK;
|
|
|
|
val &= ~SVM_CR0_SELECTIVE_MASK;
|
|
|
|
|
|
|
|
if (cr0 ^ val) {
|
|
|
|
svm->vmcb->control.exit_code = SVM_EXIT_CR0_SEL_WRITE;
|
|
|
|
ret = (nested_svm_exit_handled(svm) == NESTED_EXIT_DONE);
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2010-12-21 10:12:04 +00:00
|
|
|
#define CR_VALID (1ULL << 63)
|
|
|
|
|
|
|
|
static int cr_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
int reg, cr;
|
|
|
|
unsigned long val;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
if (!static_cpu_has(X86_FEATURE_DECODEASSISTS))
|
|
|
|
return emulate_on_interception(svm);
|
|
|
|
|
|
|
|
if (unlikely((svm->vmcb->control.exit_info_1 & CR_VALID) == 0))
|
|
|
|
return emulate_on_interception(svm);
|
|
|
|
|
|
|
|
reg = svm->vmcb->control.exit_info_1 & SVM_EXITINFO_REG_MASK;
|
|
|
|
cr = svm->vmcb->control.exit_code - SVM_EXIT_READ_CR0;
|
|
|
|
|
|
|
|
err = 0;
|
|
|
|
if (cr >= 16) { /* mov to cr */
|
|
|
|
cr -= 16;
|
|
|
|
val = kvm_register_read(&svm->vcpu, reg);
|
|
|
|
switch (cr) {
|
|
|
|
case 0:
|
2011-04-04 10:39:36 +00:00
|
|
|
if (!check_selective_cr0_intercepted(svm, val))
|
|
|
|
err = kvm_set_cr0(&svm->vcpu, val);
|
2011-04-18 09:42:52 +00:00
|
|
|
else
|
|
|
|
return 1;
|
|
|
|
|
2010-12-21 10:12:04 +00:00
|
|
|
break;
|
|
|
|
case 3:
|
|
|
|
err = kvm_set_cr3(&svm->vcpu, val);
|
|
|
|
break;
|
|
|
|
case 4:
|
|
|
|
err = kvm_set_cr4(&svm->vcpu, val);
|
|
|
|
break;
|
|
|
|
case 8:
|
|
|
|
err = kvm_set_cr8(&svm->vcpu, val);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
WARN(1, "unhandled write to CR%d", cr);
|
|
|
|
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
} else { /* mov from cr */
|
|
|
|
switch (cr) {
|
|
|
|
case 0:
|
|
|
|
val = kvm_read_cr0(&svm->vcpu);
|
|
|
|
break;
|
|
|
|
case 2:
|
|
|
|
val = svm->vcpu.arch.cr2;
|
|
|
|
break;
|
|
|
|
case 3:
|
2010-12-05 15:30:00 +00:00
|
|
|
val = kvm_read_cr3(&svm->vcpu);
|
2010-12-21 10:12:04 +00:00
|
|
|
break;
|
|
|
|
case 4:
|
|
|
|
val = kvm_read_cr4(&svm->vcpu);
|
|
|
|
break;
|
|
|
|
case 8:
|
|
|
|
val = kvm_get_cr8(&svm->vcpu);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
WARN(1, "unhandled read from CR%d", cr);
|
|
|
|
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
kvm_register_write(&svm->vcpu, reg, val);
|
|
|
|
}
|
|
|
|
kvm_complete_insn_gp(&svm->vcpu, err);
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-12-21 10:12:05 +00:00
|
|
|
static int dr_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
int reg, dr;
|
|
|
|
unsigned long val;
|
|
|
|
int err;
|
|
|
|
|
2014-02-21 09:32:27 +00:00
|
|
|
if (svm->vcpu.guest_debug == 0) {
|
|
|
|
/*
|
|
|
|
* No more DR vmexits; force a reload of the debug registers
|
|
|
|
* and reenter on this instruction. The next vmexit will
|
|
|
|
* retrieve the full state of the debug registers.
|
|
|
|
*/
|
|
|
|
clr_dr_intercepts(svm);
|
|
|
|
svm->vcpu.arch.switch_db_regs |= KVM_DEBUGREG_WONT_EXIT;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-12-21 10:12:05 +00:00
|
|
|
if (!boot_cpu_has(X86_FEATURE_DECODEASSISTS))
|
|
|
|
return emulate_on_interception(svm);
|
|
|
|
|
|
|
|
reg = svm->vmcb->control.exit_info_1 & SVM_EXITINFO_REG_MASK;
|
|
|
|
dr = svm->vmcb->control.exit_code - SVM_EXIT_READ_DR0;
|
|
|
|
|
|
|
|
if (dr >= 16) { /* mov to DRn */
|
|
|
|
val = kvm_register_read(&svm->vcpu, reg);
|
|
|
|
kvm_set_dr(&svm->vcpu, dr - 16, val);
|
|
|
|
} else {
|
|
|
|
err = kvm_get_dr(&svm->vcpu, dr, &val);
|
|
|
|
if (!err)
|
|
|
|
kvm_register_write(&svm->vcpu, reg, val);
|
|
|
|
}
|
|
|
|
|
2011-02-09 17:29:39 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
|
|
|
|
2010-12-21 10:12:05 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int cr8_write_interception(struct vcpu_svm *svm)
|
2007-12-06 20:02:25 +00:00
|
|
|
{
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = svm->vcpu.run;
|
2010-12-21 10:12:00 +00:00
|
|
|
int r;
|
2009-08-24 08:10:17 +00:00
|
|
|
|
2009-04-21 14:45:06 +00:00
|
|
|
u8 cr8_prev = kvm_get_cr8(&svm->vcpu);
|
|
|
|
/* instruction emulation calls kvm_set_cr8() */
|
2010-12-21 10:12:04 +00:00
|
|
|
r = cr_interception(svm);
|
2014-03-11 18:11:18 +00:00
|
|
|
if (irqchip_in_kernel(svm->vcpu.kvm))
|
2010-12-21 10:12:04 +00:00
|
|
|
return r;
|
2009-04-21 14:45:06 +00:00
|
|
|
if (cr8_prev <= kvm_get_cr8(&svm->vcpu))
|
2010-12-21 10:12:04 +00:00
|
|
|
return r;
|
2007-12-06 20:02:25 +00:00
|
|
|
kvm_run->exit_reason = KVM_EXIT_SET_TPR;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-11-28 01:28:58 +00:00
|
|
|
u64 svm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
|
2011-08-02 12:54:20 +00:00
|
|
|
{
|
|
|
|
struct vmcb *vmcb = get_host_vmcb(to_svm(vcpu));
|
|
|
|
return vmcb->control.tsc_offset +
|
2012-11-28 01:28:58 +00:00
|
|
|
svm_scale_tsc(vcpu, host_tsc);
|
2011-08-02 12:54:20 +00:00
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
switch (ecx) {
|
2009-05-15 13:12:05 +00:00
|
|
|
case MSR_IA32_TSC: {
|
2011-08-02 12:55:23 +00:00
|
|
|
*data = svm->vmcb->control.tsc_offset +
|
2011-03-25 08:44:46 +00:00
|
|
|
svm_scale_tsc(vcpu, native_read_tsc());
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-07-17 13:03:26 +00:00
|
|
|
case MSR_STAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.star;
|
[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 10:21:36 +00:00
|
|
|
break;
|
2006-12-22 09:05:08 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
[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 10:21:36 +00:00
|
|
|
case MSR_LSTAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.lstar;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_CSTAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.cstar;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_KERNEL_GS_BASE:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.kernel_gs_base;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_SYSCALL_MASK:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.sfmask;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
#endif
|
|
|
|
case MSR_IA32_SYSENTER_CS:
|
2007-07-27 12:13:10 +00:00
|
|
|
*data = svm->vmcb->save.sysenter_cs;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_IA32_SYSENTER_EIP:
|
2009-05-28 09:56:31 +00:00
|
|
|
*data = svm->sysenter_eip;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_IA32_SYSENTER_ESP:
|
2009-05-28 09:56:31 +00:00
|
|
|
*data = svm->sysenter_esp;
|
[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 10:21:36 +00:00
|
|
|
break;
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* Nobody will change the following 5 values in the VMCB so we can
|
|
|
|
* safely return them on rdmsr. They will always be 0 until LBRV is
|
|
|
|
* implemented.
|
|
|
|
*/
|
2008-02-13 15:30:28 +00:00
|
|
|
case MSR_IA32_DEBUGCTLMSR:
|
|
|
|
*data = svm->vmcb->save.dbgctl;
|
|
|
|
break;
|
|
|
|
case MSR_IA32_LASTBRANCHFROMIP:
|
|
|
|
*data = svm->vmcb->save.br_from;
|
|
|
|
break;
|
|
|
|
case MSR_IA32_LASTBRANCHTOIP:
|
|
|
|
*data = svm->vmcb->save.br_to;
|
|
|
|
break;
|
|
|
|
case MSR_IA32_LASTINTFROMIP:
|
|
|
|
*data = svm->vmcb->save.last_excp_from;
|
|
|
|
break;
|
|
|
|
case MSR_IA32_LASTINTTOIP:
|
|
|
|
*data = svm->vmcb->save.last_excp_to;
|
|
|
|
break;
|
2008-11-25 19:17:05 +00:00
|
|
|
case MSR_VM_HSAVE_PA:
|
2009-08-07 09:49:33 +00:00
|
|
|
*data = svm->nested.hsave_msr;
|
2008-11-25 19:17:05 +00:00
|
|
|
break;
|
2008-11-25 19:17:09 +00:00
|
|
|
case MSR_VM_CR:
|
2010-02-24 17:59:15 +00:00
|
|
|
*data = svm->nested.vm_cr_msr;
|
2008-11-25 19:17:09 +00:00
|
|
|
break;
|
2009-01-05 15:02:47 +00:00
|
|
|
case MSR_IA32_UCODE_REV:
|
|
|
|
*data = 0x01000065;
|
|
|
|
break;
|
[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 10:21:36 +00:00
|
|
|
default:
|
2006-12-30 00:49:48 +00:00
|
|
|
return kvm_get_msr_common(vcpu, ecx, data);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int rdmsr_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-12-13 15:50:52 +00:00
|
|
|
u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
|
[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 10:21:36 +00:00
|
|
|
u64 data;
|
|
|
|
|
2010-01-25 17:47:02 +00:00
|
|
|
if (svm_get_msr(&svm->vcpu, ecx, &data)) {
|
|
|
|
trace_kvm_msr_read_ex(ecx);
|
2007-11-25 12:12:03 +00:00
|
|
|
kvm_inject_gp(&svm->vcpu, 0);
|
2010-01-25 17:47:02 +00:00
|
|
|
} else {
|
2009-06-17 12:22:14 +00:00
|
|
|
trace_kvm_msr_read(ecx, data);
|
2008-04-30 15:56:03 +00:00
|
|
|
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->vcpu.arch.regs[VCPU_REGS_RAX] = data & 0xffffffff;
|
2007-12-13 15:50:52 +00:00
|
|
|
svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32;
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 2;
|
2007-07-30 10:07:08 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-02-24 17:59:15 +00:00
|
|
|
static int svm_set_vm_cr(struct kvm_vcpu *vcpu, u64 data)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
int svm_dis, chg_mask;
|
|
|
|
|
|
|
|
if (data & ~SVM_VM_CR_VALID_MASK)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
chg_mask = SVM_VM_CR_VALID_MASK;
|
|
|
|
|
|
|
|
if (svm->nested.vm_cr_msr & SVM_VM_CR_SVM_DIS_MASK)
|
|
|
|
chg_mask &= ~(SVM_VM_CR_SVM_LOCK_MASK | SVM_VM_CR_SVM_DIS_MASK);
|
|
|
|
|
|
|
|
svm->nested.vm_cr_msr &= ~chg_mask;
|
|
|
|
svm->nested.vm_cr_msr |= (data & chg_mask);
|
|
|
|
|
|
|
|
svm_dis = svm->nested.vm_cr_msr & SVM_VM_CR_SVM_DIS_MASK;
|
|
|
|
|
|
|
|
/* check for svm_disable while efer.svme is set */
|
|
|
|
if (svm_dis && (vcpu->arch.efer & EFER_SVME))
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-11-29 20:42:12 +00:00
|
|
|
static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2012-11-29 20:42:12 +00:00
|
|
|
u32 ecx = msr->index;
|
|
|
|
u64 data = msr->data;
|
[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 10:21:36 +00:00
|
|
|
switch (ecx) {
|
2010-08-20 08:07:16 +00:00
|
|
|
case MSR_IA32_TSC:
|
2012-11-29 20:42:12 +00:00
|
|
|
kvm_write_tsc(vcpu, msr);
|
[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 10:21:36 +00:00
|
|
|
break;
|
2010-07-17 13:03:26 +00:00
|
|
|
case MSR_STAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.star = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
2007-01-29 21:19:50 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
[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 10:21:36 +00:00
|
|
|
case MSR_LSTAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.lstar = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_CSTAR:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.cstar = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_KERNEL_GS_BASE:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.kernel_gs_base = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_SYSCALL_MASK:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.sfmask = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
#endif
|
|
|
|
case MSR_IA32_SYSENTER_CS:
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.sysenter_cs = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_IA32_SYSENTER_EIP:
|
2009-05-28 09:56:31 +00:00
|
|
|
svm->sysenter_eip = data;
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.sysenter_eip = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
|
|
|
case MSR_IA32_SYSENTER_ESP:
|
2009-05-28 09:56:31 +00:00
|
|
|
svm->sysenter_esp = data;
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->vmcb->save.sysenter_esp = data;
|
[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 10:21:36 +00:00
|
|
|
break;
|
2008-02-13 15:30:28 +00:00
|
|
|
case MSR_IA32_DEBUGCTLMSR:
|
2010-11-09 14:15:42 +00:00
|
|
|
if (!boot_cpu_has(X86_FEATURE_LBRV)) {
|
KVM: Cleanup the kvm_print functions and introduce pr_XX wrappers
Introduces a couple of print functions, which are essentially wrappers
around standard printk functions, with a KVM: prefix.
Functions introduced or modified are:
- kvm_err(fmt, ...)
- kvm_info(fmt, ...)
- kvm_debug(fmt, ...)
- kvm_pr_unimpl(fmt, ...)
- pr_unimpl(vcpu, fmt, ...) -> vcpu_unimpl(vcpu, fmt, ...)
Signed-off-by: Christoffer Dall <c.dall@virtualopensystems.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-06-03 18:17:48 +00:00
|
|
|
vcpu_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTL 0x%llx, nop\n",
|
|
|
|
__func__, data);
|
2008-02-13 17:58:47 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (data & DEBUGCTL_RESERVED_BITS)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
svm->vmcb->save.dbgctl = data;
|
2010-12-03 10:45:59 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_LBR);
|
2008-02-13 17:58:47 +00:00
|
|
|
if (data & (1ULL<<0))
|
|
|
|
svm_enable_lbrv(svm);
|
|
|
|
else
|
|
|
|
svm_disable_lbrv(svm);
|
2008-02-13 15:30:28 +00:00
|
|
|
break;
|
2008-11-25 19:17:05 +00:00
|
|
|
case MSR_VM_HSAVE_PA:
|
2009-08-07 09:49:33 +00:00
|
|
|
svm->nested.hsave_msr = data;
|
2007-12-11 14:36:57 +00:00
|
|
|
break;
|
2009-06-15 13:21:23 +00:00
|
|
|
case MSR_VM_CR:
|
2010-02-24 17:59:15 +00:00
|
|
|
return svm_set_vm_cr(vcpu, data);
|
2009-06-15 13:21:23 +00:00
|
|
|
case MSR_VM_IGNNE:
|
KVM: Cleanup the kvm_print functions and introduce pr_XX wrappers
Introduces a couple of print functions, which are essentially wrappers
around standard printk functions, with a KVM: prefix.
Functions introduced or modified are:
- kvm_err(fmt, ...)
- kvm_info(fmt, ...)
- kvm_debug(fmt, ...)
- kvm_pr_unimpl(fmt, ...)
- pr_unimpl(vcpu, fmt, ...) -> vcpu_unimpl(vcpu, fmt, ...)
Signed-off-by: Christoffer Dall <c.dall@virtualopensystems.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2012-06-03 18:17:48 +00:00
|
|
|
vcpu_unimpl(vcpu, "unimplemented wrmsr: 0x%x data 0x%llx\n", ecx, data);
|
2009-06-15 13:21:23 +00:00
|
|
|
break;
|
[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 10:21:36 +00:00
|
|
|
default:
|
2012-11-29 20:42:12 +00:00
|
|
|
return kvm_set_msr_common(vcpu, msr);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int wrmsr_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2012-11-29 20:42:12 +00:00
|
|
|
struct msr_data msr;
|
2007-12-13 15:50:52 +00:00
|
|
|
u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
|
2008-06-27 17:58:02 +00:00
|
|
|
u64 data = (svm->vcpu.arch.regs[VCPU_REGS_RAX] & -1u)
|
2007-12-13 15:50:52 +00:00
|
|
|
| ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32);
|
2008-04-30 15:56:03 +00:00
|
|
|
|
2012-11-29 20:42:12 +00:00
|
|
|
msr.data = data;
|
|
|
|
msr.index = ecx;
|
|
|
|
msr.host_initiated = false;
|
2008-04-30 15:56:03 +00:00
|
|
|
|
2008-06-27 17:58:02 +00:00
|
|
|
svm->next_rip = kvm_rip_read(&svm->vcpu) + 2;
|
2012-11-29 20:42:12 +00:00
|
|
|
if (svm_set_msr(&svm->vcpu, &msr)) {
|
2010-01-25 17:47:02 +00:00
|
|
|
trace_kvm_msr_write_ex(ecx, data);
|
2007-11-25 12:12:03 +00:00
|
|
|
kvm_inject_gp(&svm->vcpu, 0);
|
2010-01-25 17:47:02 +00:00
|
|
|
} else {
|
|
|
|
trace_kvm_msr_write(ecx, data);
|
2007-07-30 10:07:08 +00:00
|
|
|
skip_emulated_instruction(&svm->vcpu);
|
2010-01-25 17:47:02 +00:00
|
|
|
}
|
[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 10:21:36 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int msr_interception(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-30 10:07:08 +00:00
|
|
|
if (svm->vmcb->control.exit_info_1)
|
2009-08-24 08:10:17 +00:00
|
|
|
return wrmsr_interception(svm);
|
[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 10:21:36 +00:00
|
|
|
else
|
2009-08-24 08:10:17 +00:00
|
|
|
return rdmsr_interception(svm);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int interrupt_window_interception(struct vcpu_svm *svm)
|
2007-01-06 00:36:24 +00:00
|
|
|
{
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = svm->vcpu.run;
|
|
|
|
|
2010-07-27 09:30:24 +00:00
|
|
|
kvm_make_request(KVM_REQ_EVENT, &svm->vcpu);
|
2008-11-25 19:17:01 +00:00
|
|
|
svm_clear_vintr(svm);
|
2007-07-06 09:20:49 +00:00
|
|
|
svm->vmcb->control.int_ctl &= ~V_IRQ_MASK;
|
2010-12-03 10:45:52 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_INTR);
|
2012-03-08 10:07:56 +00:00
|
|
|
++svm->vcpu.stat.irq_window_exits;
|
2007-01-06 00:36:24 +00:00
|
|
|
/*
|
|
|
|
* If the user space waits to inject interrupts, exit as soon as
|
|
|
|
* possible
|
|
|
|
*/
|
2009-04-21 14:44:56 +00:00
|
|
|
if (!irqchip_in_kernel(svm->vcpu.kvm) &&
|
|
|
|
kvm_run->request_interrupt_window &&
|
|
|
|
!kvm_cpu_has_interrupt(&svm->vcpu)) {
|
2007-01-06 00:36:24 +00:00
|
|
|
kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-10-06 19:25:02 +00:00
|
|
|
static int pause_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
kvm_vcpu_on_spin(&(svm->vcpu));
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2014-05-07 20:52:13 +00:00
|
|
|
static int nop_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
skip_emulated_instruction(&(svm->vcpu));
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int monitor_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
printk_once(KERN_WARNING "kvm: MONITOR instruction emulated as NOP!\n");
|
|
|
|
return nop_interception(svm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int mwait_interception(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
printk_once(KERN_WARNING "kvm: MWAIT instruction emulated as NOP!\n");
|
|
|
|
return nop_interception(svm);
|
|
|
|
}
|
|
|
|
|
2012-08-29 23:30:20 +00:00
|
|
|
static int (*const svm_exit_handlers[])(struct vcpu_svm *svm) = {
|
2010-12-21 10:12:04 +00:00
|
|
|
[SVM_EXIT_READ_CR0] = cr_interception,
|
|
|
|
[SVM_EXIT_READ_CR3] = cr_interception,
|
|
|
|
[SVM_EXIT_READ_CR4] = cr_interception,
|
|
|
|
[SVM_EXIT_READ_CR8] = cr_interception,
|
2010-01-06 08:55:27 +00:00
|
|
|
[SVM_EXIT_CR0_SEL_WRITE] = emulate_on_interception,
|
2011-04-04 10:39:36 +00:00
|
|
|
[SVM_EXIT_WRITE_CR0] = cr_interception,
|
2010-12-21 10:12:04 +00:00
|
|
|
[SVM_EXIT_WRITE_CR3] = cr_interception,
|
|
|
|
[SVM_EXIT_WRITE_CR4] = cr_interception,
|
2010-02-24 17:59:10 +00:00
|
|
|
[SVM_EXIT_WRITE_CR8] = cr8_write_interception,
|
2010-12-21 10:12:05 +00:00
|
|
|
[SVM_EXIT_READ_DR0] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR1] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR2] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR3] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR4] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR5] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR6] = dr_interception,
|
|
|
|
[SVM_EXIT_READ_DR7] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR0] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR1] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR2] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR3] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR4] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR5] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR6] = dr_interception,
|
|
|
|
[SVM_EXIT_WRITE_DR7] = dr_interception,
|
2008-12-15 12:52:10 +00:00
|
|
|
[SVM_EXIT_EXCP_BASE + DB_VECTOR] = db_interception,
|
|
|
|
[SVM_EXIT_EXCP_BASE + BP_VECTOR] = bp_interception,
|
2007-09-17 19:57:50 +00:00
|
|
|
[SVM_EXIT_EXCP_BASE + UD_VECTOR] = ud_interception,
|
2010-02-24 17:59:10 +00:00
|
|
|
[SVM_EXIT_EXCP_BASE + PF_VECTOR] = pf_interception,
|
|
|
|
[SVM_EXIT_EXCP_BASE + NM_VECTOR] = nm_interception,
|
|
|
|
[SVM_EXIT_EXCP_BASE + MC_VECTOR] = mc_interception,
|
|
|
|
[SVM_EXIT_INTR] = intr_interception,
|
2008-04-30 15:56:00 +00:00
|
|
|
[SVM_EXIT_NMI] = nmi_interception,
|
[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 10:21:36 +00:00
|
|
|
[SVM_EXIT_SMI] = nop_on_interception,
|
|
|
|
[SVM_EXIT_INIT] = nop_on_interception,
|
2007-01-06 00:36:24 +00:00
|
|
|
[SVM_EXIT_VINTR] = interrupt_window_interception,
|
2011-11-10 12:57:24 +00:00
|
|
|
[SVM_EXIT_RDPMC] = rdpmc_interception,
|
[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 10:21:36 +00:00
|
|
|
[SVM_EXIT_CPUID] = cpuid_interception,
|
2009-04-21 14:45:08 +00:00
|
|
|
[SVM_EXIT_IRET] = iret_interception,
|
2007-10-28 14:11:58 +00:00
|
|
|
[SVM_EXIT_INVD] = emulate_on_interception,
|
2009-10-06 19:25:02 +00:00
|
|
|
[SVM_EXIT_PAUSE] = pause_interception,
|
[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 10:21:36 +00:00
|
|
|
[SVM_EXIT_HLT] = halt_interception,
|
2008-09-23 16:18:35 +00:00
|
|
|
[SVM_EXIT_INVLPG] = invlpg_interception,
|
2009-06-15 13:21:24 +00:00
|
|
|
[SVM_EXIT_INVLPGA] = invlpga_interception,
|
2010-02-24 17:59:10 +00:00
|
|
|
[SVM_EXIT_IOIO] = io_interception,
|
[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 10:21:36 +00:00
|
|
|
[SVM_EXIT_MSR] = msr_interception,
|
|
|
|
[SVM_EXIT_TASK_SWITCH] = task_switch_interception,
|
2007-01-26 08:56:42 +00:00
|
|
|
[SVM_EXIT_SHUTDOWN] = shutdown_interception,
|
2008-11-25 19:17:07 +00:00
|
|
|
[SVM_EXIT_VMRUN] = vmrun_interception,
|
2007-02-19 12:37:47 +00:00
|
|
|
[SVM_EXIT_VMMCALL] = vmmcall_interception,
|
2008-11-25 19:17:06 +00:00
|
|
|
[SVM_EXIT_VMLOAD] = vmload_interception,
|
|
|
|
[SVM_EXIT_VMSAVE] = vmsave_interception,
|
2008-11-25 19:17:04 +00:00
|
|
|
[SVM_EXIT_STGI] = stgi_interception,
|
|
|
|
[SVM_EXIT_CLGI] = clgi_interception,
|
2009-10-09 14:08:32 +00:00
|
|
|
[SVM_EXIT_SKINIT] = skinit_interception,
|
2007-10-28 14:11:58 +00:00
|
|
|
[SVM_EXIT_WBINVD] = emulate_on_interception,
|
2014-05-07 20:52:13 +00:00
|
|
|
[SVM_EXIT_MONITOR] = monitor_interception,
|
|
|
|
[SVM_EXIT_MWAIT] = mwait_interception,
|
2010-12-07 16:15:06 +00:00
|
|
|
[SVM_EXIT_XSETBV] = xsetbv_interception,
|
2008-02-07 12:47:45 +00:00
|
|
|
[SVM_EXIT_NPF] = pf_interception,
|
[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 10:21:36 +00:00
|
|
|
};
|
|
|
|
|
2011-04-25 05:00:50 +00:00
|
|
|
static void dump_vmcb(struct kvm_vcpu *vcpu)
|
2010-05-05 14:04:42 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
struct vmcb_control_area *control = &svm->vmcb->control;
|
|
|
|
struct vmcb_save_area *save = &svm->vmcb->save;
|
|
|
|
|
|
|
|
pr_err("VMCB Control Area:\n");
|
2011-04-25 05:00:50 +00:00
|
|
|
pr_err("%-20s%04x\n", "cr_read:", control->intercept_cr & 0xffff);
|
|
|
|
pr_err("%-20s%04x\n", "cr_write:", control->intercept_cr >> 16);
|
|
|
|
pr_err("%-20s%04x\n", "dr_read:", control->intercept_dr & 0xffff);
|
|
|
|
pr_err("%-20s%04x\n", "dr_write:", control->intercept_dr >> 16);
|
|
|
|
pr_err("%-20s%08x\n", "exceptions:", control->intercept_exceptions);
|
|
|
|
pr_err("%-20s%016llx\n", "intercepts:", control->intercept);
|
|
|
|
pr_err("%-20s%d\n", "pause filter count:", control->pause_filter_count);
|
|
|
|
pr_err("%-20s%016llx\n", "iopm_base_pa:", control->iopm_base_pa);
|
|
|
|
pr_err("%-20s%016llx\n", "msrpm_base_pa:", control->msrpm_base_pa);
|
|
|
|
pr_err("%-20s%016llx\n", "tsc_offset:", control->tsc_offset);
|
|
|
|
pr_err("%-20s%d\n", "asid:", control->asid);
|
|
|
|
pr_err("%-20s%d\n", "tlb_ctl:", control->tlb_ctl);
|
|
|
|
pr_err("%-20s%08x\n", "int_ctl:", control->int_ctl);
|
|
|
|
pr_err("%-20s%08x\n", "int_vector:", control->int_vector);
|
|
|
|
pr_err("%-20s%08x\n", "int_state:", control->int_state);
|
|
|
|
pr_err("%-20s%08x\n", "exit_code:", control->exit_code);
|
|
|
|
pr_err("%-20s%016llx\n", "exit_info1:", control->exit_info_1);
|
|
|
|
pr_err("%-20s%016llx\n", "exit_info2:", control->exit_info_2);
|
|
|
|
pr_err("%-20s%08x\n", "exit_int_info:", control->exit_int_info);
|
|
|
|
pr_err("%-20s%08x\n", "exit_int_info_err:", control->exit_int_info_err);
|
|
|
|
pr_err("%-20s%lld\n", "nested_ctl:", control->nested_ctl);
|
|
|
|
pr_err("%-20s%016llx\n", "nested_cr3:", control->nested_cr3);
|
|
|
|
pr_err("%-20s%08x\n", "event_inj:", control->event_inj);
|
|
|
|
pr_err("%-20s%08x\n", "event_inj_err:", control->event_inj_err);
|
|
|
|
pr_err("%-20s%lld\n", "lbr_ctl:", control->lbr_ctl);
|
|
|
|
pr_err("%-20s%016llx\n", "next_rip:", control->next_rip);
|
2010-05-05 14:04:42 +00:00
|
|
|
pr_err("VMCB State Save Area:\n");
|
2011-04-25 05:00:50 +00:00
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"es:",
|
|
|
|
save->es.selector, save->es.attrib,
|
|
|
|
save->es.limit, save->es.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"cs:",
|
|
|
|
save->cs.selector, save->cs.attrib,
|
|
|
|
save->cs.limit, save->cs.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"ss:",
|
|
|
|
save->ss.selector, save->ss.attrib,
|
|
|
|
save->ss.limit, save->ss.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"ds:",
|
|
|
|
save->ds.selector, save->ds.attrib,
|
|
|
|
save->ds.limit, save->ds.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"fs:",
|
|
|
|
save->fs.selector, save->fs.attrib,
|
|
|
|
save->fs.limit, save->fs.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"gs:",
|
|
|
|
save->gs.selector, save->gs.attrib,
|
|
|
|
save->gs.limit, save->gs.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"gdtr:",
|
|
|
|
save->gdtr.selector, save->gdtr.attrib,
|
|
|
|
save->gdtr.limit, save->gdtr.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"ldtr:",
|
|
|
|
save->ldtr.selector, save->ldtr.attrib,
|
|
|
|
save->ldtr.limit, save->ldtr.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"idtr:",
|
|
|
|
save->idtr.selector, save->idtr.attrib,
|
|
|
|
save->idtr.limit, save->idtr.base);
|
|
|
|
pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n",
|
|
|
|
"tr:",
|
|
|
|
save->tr.selector, save->tr.attrib,
|
|
|
|
save->tr.limit, save->tr.base);
|
2010-05-05 14:04:42 +00:00
|
|
|
pr_err("cpl: %d efer: %016llx\n",
|
|
|
|
save->cpl, save->efer);
|
2011-04-25 05:00:50 +00:00
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"cr0:", save->cr0, "cr2:", save->cr2);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"cr3:", save->cr3, "cr4:", save->cr4);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"dr6:", save->dr6, "dr7:", save->dr7);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"rip:", save->rip, "rflags:", save->rflags);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"rsp:", save->rsp, "rax:", save->rax);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"star:", save->star, "lstar:", save->lstar);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"cstar:", save->cstar, "sfmask:", save->sfmask);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"kernel_gs_base:", save->kernel_gs_base,
|
|
|
|
"sysenter_cs:", save->sysenter_cs);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"sysenter_esp:", save->sysenter_esp,
|
|
|
|
"sysenter_eip:", save->sysenter_eip);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"gpat:", save->g_pat, "dbgctl:", save->dbgctl);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"br_from:", save->br_from, "br_to:", save->br_to);
|
|
|
|
pr_err("%-15s %016llx %-13s %016llx\n",
|
|
|
|
"excp_from:", save->last_excp_from,
|
|
|
|
"excp_to:", save->last_excp_to);
|
2010-05-05 14:04:42 +00:00
|
|
|
}
|
|
|
|
|
2010-11-18 11:09:54 +00:00
|
|
|
static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2)
|
|
|
|
{
|
|
|
|
struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control;
|
|
|
|
|
|
|
|
*info1 = control->exit_info_1;
|
|
|
|
*info2 = control->exit_info_2;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static int handle_exit(struct kvm_vcpu *vcpu)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-09-10 15:10:54 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2009-08-24 08:10:17 +00:00
|
|
|
struct kvm_run *kvm_run = vcpu->run;
|
2007-07-27 12:13:10 +00:00
|
|
|
u32 exit_code = svm->vmcb->control.exit_code;
|
[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 10:21:36 +00:00
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
if (!is_cr_intercept(svm, INTERCEPT_CR0_WRITE))
|
2010-04-22 10:33:09 +00:00
|
|
|
vcpu->arch.cr0 = svm->vmcb->save.cr0;
|
|
|
|
if (npt_enabled)
|
|
|
|
vcpu->arch.cr3 = svm->vmcb->save.cr3;
|
2008-04-30 15:56:03 +00:00
|
|
|
|
2009-10-09 14:08:26 +00:00
|
|
|
if (unlikely(svm->nested.exit_required)) {
|
|
|
|
nested_svm_vmexit(svm);
|
|
|
|
svm->nested.exit_required = false;
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu)) {
|
2009-08-07 09:49:44 +00:00
|
|
|
int vmexit;
|
|
|
|
|
2009-10-09 14:08:28 +00:00
|
|
|
trace_kvm_nested_vmexit(svm->vmcb->save.rip, exit_code,
|
|
|
|
svm->vmcb->control.exit_info_1,
|
|
|
|
svm->vmcb->control.exit_info_2,
|
|
|
|
svm->vmcb->control.exit_int_info,
|
2011-07-22 11:46:52 +00:00
|
|
|
svm->vmcb->control.exit_int_info_err,
|
|
|
|
KVM_ISA_SVM);
|
2009-10-09 14:08:28 +00:00
|
|
|
|
2009-08-07 09:49:44 +00:00
|
|
|
vmexit = nested_svm_exit_special(svm);
|
|
|
|
|
|
|
|
if (vmexit == NESTED_EXIT_CONTINUE)
|
|
|
|
vmexit = nested_svm_exit_handled(svm);
|
|
|
|
|
|
|
|
if (vmexit == NESTED_EXIT_DONE)
|
2008-11-25 19:17:08 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2009-08-07 09:49:32 +00:00
|
|
|
svm_complete_interrupts(svm);
|
|
|
|
|
2007-09-10 15:10:54 +00:00
|
|
|
if (svm->vmcb->control.exit_code == SVM_EXIT_ERR) {
|
|
|
|
kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
|
|
|
|
kvm_run->fail_entry.hardware_entry_failure_reason
|
|
|
|
= svm->vmcb->control.exit_code;
|
2010-05-05 14:04:42 +00:00
|
|
|
pr_err("KVM: FAILED VMRUN WITH VMCB:\n");
|
|
|
|
dump_vmcb(vcpu);
|
2007-09-10 15:10:54 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-07-27 12:13:10 +00:00
|
|
|
if (is_external_interrupt(svm->vmcb->control.exit_int_info) &&
|
2008-02-07 12:47:45 +00:00
|
|
|
exit_code != SVM_EXIT_EXCP_BASE + PF_VECTOR &&
|
2010-09-10 15:31:04 +00:00
|
|
|
exit_code != SVM_EXIT_NPF && exit_code != SVM_EXIT_TASK_SWITCH &&
|
|
|
|
exit_code != SVM_EXIT_INTR && exit_code != SVM_EXIT_NMI)
|
2013-04-25 22:22:01 +00:00
|
|
|
printk(KERN_ERR "%s: unexpected exit_int_info 0x%x "
|
[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 10:21:36 +00:00
|
|
|
"exit_code 0x%x\n",
|
2008-03-03 20:59:56 +00:00
|
|
|
__func__, svm->vmcb->control.exit_int_info,
|
[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 10:21:36 +00:00
|
|
|
exit_code);
|
|
|
|
|
2007-02-19 12:37:46 +00:00
|
|
|
if (exit_code >= ARRAY_SIZE(svm_exit_handlers)
|
2007-11-13 04:06:51 +00:00
|
|
|
|| !svm_exit_handlers[exit_code]) {
|
[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 10:21:36 +00:00
|
|
|
kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
|
2007-04-16 11:28:40 +00:00
|
|
|
kvm_run->hw.hardware_exit_reason = exit_code;
|
[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 10:21:36 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
return svm_exit_handlers[exit_code](svm);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void reload_tss(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
int cpu = raw_smp_processor_id();
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
struct svm_cpu_data *sd = per_cpu(svm_data, cpu);
|
|
|
|
sd->tss_desc->type = 9; /* available 32/64-bit TSS */
|
[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 10:21:36 +00:00
|
|
|
load_TR_desc();
|
|
|
|
}
|
|
|
|
|
2007-07-30 10:07:08 +00:00
|
|
|
static void pre_svm_run(struct vcpu_svm *svm)
|
[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 10:21:36 +00:00
|
|
|
{
|
|
|
|
int cpu = raw_smp_processor_id();
|
|
|
|
|
2009-10-29 13:34:14 +00:00
|
|
|
struct svm_cpu_data *sd = per_cpu(svm_data, cpu);
|
[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 10:21:36 +00:00
|
|
|
|
2009-07-21 15:47:45 +00:00
|
|
|
/* FIXME: handle wraparound of asid_generation */
|
2009-10-29 13:34:14 +00:00
|
|
|
if (svm->asid_generation != sd->asid_generation)
|
|
|
|
new_asid(svm, sd);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-04-21 14:45:08 +00:00
|
|
|
static void svm_inject_nmi(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_NMI;
|
|
|
|
vcpu->arch.hflags |= HF_NMI_MASK;
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_IRET);
|
2009-04-21 14:45:08 +00:00
|
|
|
++vcpu->stat.nmi_injections;
|
|
|
|
}
|
[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 10:21:36 +00:00
|
|
|
|
2007-07-06 09:20:49 +00:00
|
|
|
static inline void svm_inject_irq(struct vcpu_svm *svm, int irq)
|
[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 10:21:36 +00:00
|
|
|
{
|
|
|
|
struct vmcb_control_area *control;
|
|
|
|
|
2007-07-30 10:07:08 +00:00
|
|
|
control = &svm->vmcb->control;
|
2007-07-06 09:20:49 +00:00
|
|
|
control->int_vector = irq;
|
[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 10:21:36 +00:00
|
|
|
control->int_ctl &= ~V_INTR_PRIO_MASK;
|
|
|
|
control->int_ctl |= V_IRQ_MASK |
|
|
|
|
((/*control->int_vector >> 4*/ 0xf) << V_INTR_PRIO_SHIFT);
|
2010-12-03 10:45:52 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_INTR);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2009-05-11 10:35:50 +00:00
|
|
|
static void svm_set_irq(struct kvm_vcpu *vcpu)
|
2007-08-06 13:29:07 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2009-08-07 09:49:28 +00:00
|
|
|
BUG_ON(!(gif_set(svm)));
|
2008-11-25 19:17:08 +00:00
|
|
|
|
2010-05-23 11:28:26 +00:00
|
|
|
trace_kvm_inj_virq(vcpu->arch.interrupt.nr);
|
|
|
|
++vcpu->stat.irq_injections;
|
|
|
|
|
2009-06-15 13:21:25 +00:00
|
|
|
svm->vmcb->control.event_inj = vcpu->arch.interrupt.nr |
|
|
|
|
SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR;
|
2007-08-06 13:29:07 +00:00
|
|
|
}
|
|
|
|
|
2009-04-21 14:45:08 +00:00
|
|
|
static void update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr)
|
2008-04-16 14:51:18 +00:00
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu) && (vcpu->arch.hflags & HF_VINTR_MASK))
|
2010-02-19 15:23:06 +00:00
|
|
|
return;
|
|
|
|
|
2014-03-11 18:11:18 +00:00
|
|
|
clr_cr_intercept(svm, INTERCEPT_CR8_WRITE);
|
|
|
|
|
2009-04-21 14:45:08 +00:00
|
|
|
if (irr == -1)
|
2008-04-16 14:51:18 +00:00
|
|
|
return;
|
|
|
|
|
2009-04-21 14:45:08 +00:00
|
|
|
if (tpr >= irr)
|
2010-12-03 09:50:51 +00:00
|
|
|
set_cr_intercept(svm, INTERCEPT_CR8_WRITE);
|
2009-04-21 14:45:08 +00:00
|
|
|
}
|
2008-04-16 14:51:18 +00:00
|
|
|
|
2013-01-25 02:18:50 +00:00
|
|
|
static void svm_set_virtual_x2apic_mode(struct kvm_vcpu *vcpu, bool set)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2013-01-25 02:18:51 +00:00
|
|
|
static int svm_vm_has_apicv(struct kvm *kvm)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_hwapic_isr_update(struct kvm *kvm, int isr)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2013-04-11 11:25:15 +00:00
|
|
|
static void svm_sync_pir_to_irr(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2009-04-21 14:45:08 +00:00
|
|
|
static int svm_nmi_allowed(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
struct vmcb *vmcb = svm->vmcb;
|
2010-04-22 10:33:07 +00:00
|
|
|
int ret;
|
|
|
|
ret = !(vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) &&
|
|
|
|
!(svm->vcpu.arch.hflags & HF_NMI_MASK);
|
|
|
|
ret = ret && gif_set(svm) && nested_svm_nmi(svm);
|
|
|
|
|
|
|
|
return ret;
|
2008-04-16 14:51:18 +00:00
|
|
|
}
|
|
|
|
|
2009-11-12 00:04:25 +00:00
|
|
|
static bool svm_get_nmi_mask(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
return !!(svm->vcpu.arch.hflags & HF_NMI_MASK);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
if (masked) {
|
|
|
|
svm->vcpu.arch.hflags |= HF_NMI_MASK;
|
2010-11-30 17:04:00 +00:00
|
|
|
set_intercept(svm, INTERCEPT_IRET);
|
2009-11-12 00:04:25 +00:00
|
|
|
} else {
|
|
|
|
svm->vcpu.arch.hflags &= ~HF_NMI_MASK;
|
2010-11-30 17:04:00 +00:00
|
|
|
clr_intercept(svm, INTERCEPT_IRET);
|
2009-11-12 00:04:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-03-23 10:12:11 +00:00
|
|
|
static int svm_interrupt_allowed(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
struct vmcb *vmcb = svm->vmcb;
|
2009-09-16 13:24:15 +00:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (!gif_set(svm) ||
|
|
|
|
(vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK))
|
|
|
|
return 0;
|
|
|
|
|
2010-08-02 12:30:20 +00:00
|
|
|
ret = !!(kvm_get_rflags(vcpu) & X86_EFLAGS_IF);
|
2009-09-16 13:24:15 +00:00
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu))
|
2009-09-16 13:24:15 +00:00
|
|
|
return ret && !(svm->vcpu.arch.hflags & HF_VINTR_MASK);
|
|
|
|
|
|
|
|
return ret;
|
2009-03-23 10:12:11 +00:00
|
|
|
}
|
|
|
|
|
2014-03-07 19:03:15 +00:00
|
|
|
static void enable_irq_window(struct kvm_vcpu *vcpu)
|
[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 10:21:36 +00:00
|
|
|
{
|
2009-06-15 13:21:25 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* In case GIF=0 we can't rely on the CPU to tell us when GIF becomes
|
|
|
|
* 1, because that's a separate STGI/VMRUN intercept. The next time we
|
|
|
|
* get that intercept, this function will be called again though and
|
|
|
|
* we'll get the vintr intercept.
|
|
|
|
*/
|
2010-02-19 15:23:01 +00:00
|
|
|
if (gif_set(svm) && nested_svm_intr(svm)) {
|
2009-06-15 13:21:25 +00:00
|
|
|
svm_set_vintr(svm);
|
|
|
|
svm_inject_irq(svm, 0x0);
|
|
|
|
}
|
2007-07-06 09:20:49 +00:00
|
|
|
}
|
|
|
|
|
2014-03-07 19:03:15 +00:00
|
|
|
static void enable_nmi_window(struct kvm_vcpu *vcpu)
|
2007-01-06 00:36:24 +00:00
|
|
|
{
|
2007-09-10 15:10:54 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2007-01-06 00:36:24 +00:00
|
|
|
|
2009-05-11 10:35:52 +00:00
|
|
|
if ((svm->vcpu.arch.hflags & (HF_NMI_MASK | HF_IRET_MASK))
|
|
|
|
== HF_NMI_MASK)
|
2014-03-07 19:03:15 +00:00
|
|
|
return; /* IRET will cause a vm exit */
|
2009-05-11 10:35:52 +00:00
|
|
|
|
2010-02-24 17:59:10 +00:00
|
|
|
/*
|
|
|
|
* Something prevents NMI from been injected. Single step over possible
|
|
|
|
* problem (IRET or exception injection or interrupt shadow)
|
|
|
|
*/
|
2009-10-18 11:24:54 +00:00
|
|
|
svm->nmi_singlestep = true;
|
2009-05-11 10:35:52 +00:00
|
|
|
svm->vmcb->save.rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF);
|
2012-09-21 03:42:55 +00:00
|
|
|
update_db_bp_intercept(vcpu);
|
2007-01-06 00:36:24 +00:00
|
|
|
}
|
|
|
|
|
2007-10-24 22:29:55 +00:00
|
|
|
static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-06-07 16:18:30 +00:00
|
|
|
static void svm_flush_tlb(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
2010-12-03 14:25:16 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
if (static_cpu_has(X86_FEATURE_FLUSHBYASID))
|
|
|
|
svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ASID;
|
|
|
|
else
|
|
|
|
svm->asid_generation--;
|
2007-06-07 16:18:30 +00:00
|
|
|
}
|
|
|
|
|
2007-09-10 15:10:54 +00:00
|
|
|
static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2008-04-16 14:51:17 +00:00
|
|
|
static inline void sync_cr8_to_lapic(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu) && (vcpu->arch.hflags & HF_VINTR_MASK))
|
2010-02-19 15:23:06 +00:00
|
|
|
return;
|
|
|
|
|
2010-12-03 09:50:51 +00:00
|
|
|
if (!is_cr_intercept(svm, INTERCEPT_CR8_WRITE)) {
|
2008-04-16 14:51:17 +00:00
|
|
|
int cr8 = svm->vmcb->control.int_ctl & V_TPR_MASK;
|
2009-04-21 14:45:05 +00:00
|
|
|
kvm_set_cr8(vcpu, cr8);
|
2008-04-16 14:51:17 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-04-16 14:51:15 +00:00
|
|
|
static inline void sync_lapic_to_cr8(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
u64 cr8;
|
|
|
|
|
2010-11-29 16:51:48 +00:00
|
|
|
if (is_guest_mode(vcpu) && (vcpu->arch.hflags & HF_VINTR_MASK))
|
2010-02-19 15:23:06 +00:00
|
|
|
return;
|
|
|
|
|
2008-04-16 14:51:15 +00:00
|
|
|
cr8 = kvm_get_cr8(vcpu);
|
|
|
|
svm->vmcb->control.int_ctl &= ~V_TPR_MASK;
|
|
|
|
svm->vmcb->control.int_ctl |= cr8 & V_TPR_MASK;
|
|
|
|
}
|
|
|
|
|
2009-04-23 14:14:37 +00:00
|
|
|
static void svm_complete_interrupts(struct vcpu_svm *svm)
|
|
|
|
{
|
|
|
|
u8 vector;
|
|
|
|
int type;
|
|
|
|
u32 exitintinfo = svm->vmcb->control.exit_int_info;
|
2010-02-23 16:47:56 +00:00
|
|
|
unsigned int3_injected = svm->int3_injected;
|
|
|
|
|
|
|
|
svm->int3_injected = 0;
|
2009-04-23 14:14:37 +00:00
|
|
|
|
KVM: SVM: check for progress after IRET interception
When we enable an NMI window, we ask for an IRET intercept, since
the IRET re-enables NMIs. However, the IRET intercept happens before
the instruction executes, while the NMI window architecturally opens
afterwards.
To compensate for this mismatch, we only open the NMI window in the
following exit, assuming that the IRET has by then executed; however,
this assumption is not always correct; we may exit due to a host interrupt
or page fault, without having executed the instruction.
Fix by checking for forward progress by recording and comparing the IRET's
rip. This is somewhat of a hack, since an unchaging rip does not mean that
no forward progress has been made, but is the simplest fix for now.
Signed-off-by: Avi Kivity <avi@redhat.com>
2011-02-03 13:29:52 +00:00
|
|
|
/*
|
|
|
|
* If we've made progress since setting HF_IRET_MASK, we've
|
|
|
|
* executed an IRET and can allow NMI injection.
|
|
|
|
*/
|
|
|
|
if ((svm->vcpu.arch.hflags & HF_IRET_MASK)
|
|
|
|
&& kvm_rip_read(&svm->vcpu) != svm->nmi_iret_rip) {
|
2009-05-11 10:35:52 +00:00
|
|
|
svm->vcpu.arch.hflags &= ~(HF_NMI_MASK | HF_IRET_MASK);
|
2010-07-27 09:30:24 +00:00
|
|
|
kvm_make_request(KVM_REQ_EVENT, &svm->vcpu);
|
|
|
|
}
|
2009-05-11 10:35:52 +00:00
|
|
|
|
2009-04-23 14:14:37 +00:00
|
|
|
svm->vcpu.arch.nmi_injected = false;
|
|
|
|
kvm_clear_exception_queue(&svm->vcpu);
|
|
|
|
kvm_clear_interrupt_queue(&svm->vcpu);
|
|
|
|
|
|
|
|
if (!(exitintinfo & SVM_EXITINTINFO_VALID))
|
|
|
|
return;
|
|
|
|
|
2010-07-27 09:30:24 +00:00
|
|
|
kvm_make_request(KVM_REQ_EVENT, &svm->vcpu);
|
|
|
|
|
2009-04-23 14:14:37 +00:00
|
|
|
vector = exitintinfo & SVM_EXITINTINFO_VEC_MASK;
|
|
|
|
type = exitintinfo & SVM_EXITINTINFO_TYPE_MASK;
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
case SVM_EXITINTINFO_TYPE_NMI:
|
|
|
|
svm->vcpu.arch.nmi_injected = true;
|
|
|
|
break;
|
|
|
|
case SVM_EXITINTINFO_TYPE_EXEPT:
|
2010-02-23 16:47:56 +00:00
|
|
|
/*
|
|
|
|
* In case of software exceptions, do not reinject the vector,
|
|
|
|
* but re-execute the instruction instead. Rewind RIP first
|
|
|
|
* if we emulated INT3 before.
|
|
|
|
*/
|
|
|
|
if (kvm_exception_is_soft(vector)) {
|
|
|
|
if (vector == BP_VECTOR && int3_injected &&
|
|
|
|
kvm_is_linear_rip(&svm->vcpu, svm->int3_rip))
|
|
|
|
kvm_rip_write(&svm->vcpu,
|
|
|
|
kvm_rip_read(&svm->vcpu) -
|
|
|
|
int3_injected);
|
2009-04-23 14:14:37 +00:00
|
|
|
break;
|
2010-02-23 16:47:56 +00:00
|
|
|
}
|
2009-04-23 14:14:37 +00:00
|
|
|
if (exitintinfo & SVM_EXITINTINFO_VALID_ERR) {
|
|
|
|
u32 err = svm->vmcb->control.exit_int_info_err;
|
2010-04-22 10:33:13 +00:00
|
|
|
kvm_requeue_exception_e(&svm->vcpu, vector, err);
|
2009-04-23 14:14:37 +00:00
|
|
|
|
|
|
|
} else
|
2010-04-22 10:33:13 +00:00
|
|
|
kvm_requeue_exception(&svm->vcpu, vector);
|
2009-04-23 14:14:37 +00:00
|
|
|
break;
|
|
|
|
case SVM_EXITINTINFO_TYPE_INTR:
|
2009-05-11 10:35:50 +00:00
|
|
|
kvm_queue_interrupt(&svm->vcpu, vector, false);
|
2009-04-23 14:14:37 +00:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-07-20 12:06:17 +00:00
|
|
|
static void svm_cancel_injection(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
struct vmcb_control_area *control = &svm->vmcb->control;
|
|
|
|
|
|
|
|
control->exit_int_info = control->event_inj;
|
|
|
|
control->exit_int_info_err = control->event_inj_err;
|
|
|
|
control->event_inj = 0;
|
|
|
|
svm_complete_interrupts(svm);
|
|
|
|
}
|
|
|
|
|
2009-08-24 08:10:17 +00:00
|
|
|
static void svm_vcpu_run(struct kvm_vcpu *vcpu)
|
[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 10:21:36 +00:00
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
2007-06-07 16:18:30 +00:00
|
|
|
|
2010-04-22 10:33:08 +00:00
|
|
|
svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX];
|
|
|
|
svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP];
|
|
|
|
svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP];
|
|
|
|
|
2009-10-09 14:08:26 +00:00
|
|
|
/*
|
|
|
|
* A vmexit emulation is required before the vcpu can be executed
|
|
|
|
* again.
|
|
|
|
*/
|
|
|
|
if (unlikely(svm->nested.exit_required))
|
|
|
|
return;
|
|
|
|
|
2007-07-30 10:07:08 +00:00
|
|
|
pre_svm_run(svm);
|
[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 10:21:36 +00:00
|
|
|
|
2008-04-16 14:51:15 +00:00
|
|
|
sync_lapic_to_cr8(vcpu);
|
|
|
|
|
2009-08-07 09:49:45 +00:00
|
|
|
svm->vmcb->save.cr2 = vcpu->arch.cr2;
|
[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 10:21:36 +00:00
|
|
|
|
2007-09-10 15:10:54 +00:00
|
|
|
clgi();
|
|
|
|
|
|
|
|
local_irq_enable();
|
2006-12-22 09:05:20 +00:00
|
|
|
|
[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 10:21:36 +00:00
|
|
|
asm volatile (
|
2012-09-16 12:10:59 +00:00
|
|
|
"push %%" _ASM_BP "; \n\t"
|
|
|
|
"mov %c[rbx](%[svm]), %%" _ASM_BX " \n\t"
|
|
|
|
"mov %c[rcx](%[svm]), %%" _ASM_CX " \n\t"
|
|
|
|
"mov %c[rdx](%[svm]), %%" _ASM_DX " \n\t"
|
|
|
|
"mov %c[rsi](%[svm]), %%" _ASM_SI " \n\t"
|
|
|
|
"mov %c[rdi](%[svm]), %%" _ASM_DI " \n\t"
|
|
|
|
"mov %c[rbp](%[svm]), %%" _ASM_BP " \n\t"
|
2006-12-13 08:33:45 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
2007-07-27 07:16:56 +00:00
|
|
|
"mov %c[r8](%[svm]), %%r8 \n\t"
|
|
|
|
"mov %c[r9](%[svm]), %%r9 \n\t"
|
|
|
|
"mov %c[r10](%[svm]), %%r10 \n\t"
|
|
|
|
"mov %c[r11](%[svm]), %%r11 \n\t"
|
|
|
|
"mov %c[r12](%[svm]), %%r12 \n\t"
|
|
|
|
"mov %c[r13](%[svm]), %%r13 \n\t"
|
|
|
|
"mov %c[r14](%[svm]), %%r14 \n\t"
|
|
|
|
"mov %c[r15](%[svm]), %%r15 \n\t"
|
[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 10:21:36 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Enter guest mode */
|
2012-09-16 12:10:59 +00:00
|
|
|
"push %%" _ASM_AX " \n\t"
|
|
|
|
"mov %c[vmcb](%[svm]), %%" _ASM_AX " \n\t"
|
2008-05-13 10:23:38 +00:00
|
|
|
__ex(SVM_VMLOAD) "\n\t"
|
|
|
|
__ex(SVM_VMRUN) "\n\t"
|
|
|
|
__ex(SVM_VMSAVE) "\n\t"
|
2012-09-16 12:10:59 +00:00
|
|
|
"pop %%" _ASM_AX " \n\t"
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
/* Save guest registers, load host registers */
|
2012-09-16 12:10:59 +00:00
|
|
|
"mov %%" _ASM_BX ", %c[rbx](%[svm]) \n\t"
|
|
|
|
"mov %%" _ASM_CX ", %c[rcx](%[svm]) \n\t"
|
|
|
|
"mov %%" _ASM_DX ", %c[rdx](%[svm]) \n\t"
|
|
|
|
"mov %%" _ASM_SI ", %c[rsi](%[svm]) \n\t"
|
|
|
|
"mov %%" _ASM_DI ", %c[rdi](%[svm]) \n\t"
|
|
|
|
"mov %%" _ASM_BP ", %c[rbp](%[svm]) \n\t"
|
2006-12-13 08:33:45 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
2007-07-27 07:16:56 +00:00
|
|
|
"mov %%r8, %c[r8](%[svm]) \n\t"
|
|
|
|
"mov %%r9, %c[r9](%[svm]) \n\t"
|
|
|
|
"mov %%r10, %c[r10](%[svm]) \n\t"
|
|
|
|
"mov %%r11, %c[r11](%[svm]) \n\t"
|
|
|
|
"mov %%r12, %c[r12](%[svm]) \n\t"
|
|
|
|
"mov %%r13, %c[r13](%[svm]) \n\t"
|
|
|
|
"mov %%r14, %c[r14](%[svm]) \n\t"
|
|
|
|
"mov %%r15, %c[r15](%[svm]) \n\t"
|
[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 10:21:36 +00:00
|
|
|
#endif
|
2012-09-16 12:10:59 +00:00
|
|
|
"pop %%" _ASM_BP
|
[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 10:21:36 +00:00
|
|
|
:
|
2007-07-27 07:16:56 +00:00
|
|
|
: [svm]"a"(svm),
|
[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 10:21:36 +00:00
|
|
|
[vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)),
|
2007-12-13 15:50:52 +00:00
|
|
|
[rbx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBX])),
|
|
|
|
[rcx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RCX])),
|
|
|
|
[rdx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDX])),
|
|
|
|
[rsi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RSI])),
|
|
|
|
[rdi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDI])),
|
|
|
|
[rbp]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBP]))
|
2006-12-13 08:33:45 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
2007-12-13 15:50:52 +00:00
|
|
|
, [r8]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R8])),
|
|
|
|
[r9]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R9])),
|
|
|
|
[r10]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R10])),
|
|
|
|
[r11]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R11])),
|
|
|
|
[r12]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R12])),
|
|
|
|
[r13]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R13])),
|
|
|
|
[r14]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R14])),
|
|
|
|
[r15]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R15]))
|
[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 10:21:36 +00:00
|
|
|
#endif
|
KVM: SVM: Let gcc to choose which registers to save (x86_64)
This patch lets GCC to determine which registers to save when we
switch to/from a VCPU in the case of AMD x86_64.
* Original code saves following registers:
rbx, rcx, rdx, rsi, rdi, rbp,
r8, r9, r10, r11, r12, r13, r14, r15
* Patched code:
- informs GCC that we modify following registers
using the clobber description:
rbx, rcx, rdx, rsi, rdi
r8, r9, r10, r11, r12, r13, r14, r15
- rbp is saved (pop/push) because GCC seems to ignore its use in the clobber
description.
Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2007-10-25 12:18:53 +00:00
|
|
|
: "cc", "memory"
|
|
|
|
#ifdef CONFIG_X86_64
|
2012-09-16 12:10:59 +00:00
|
|
|
, "rbx", "rcx", "rdx", "rsi", "rdi"
|
KVM: SVM: Let gcc to choose which registers to save (x86_64)
This patch lets GCC to determine which registers to save when we
switch to/from a VCPU in the case of AMD x86_64.
* Original code saves following registers:
rbx, rcx, rdx, rsi, rdi, rbp,
r8, r9, r10, r11, r12, r13, r14, r15
* Patched code:
- informs GCC that we modify following registers
using the clobber description:
rbx, rcx, rdx, rsi, rdi
r8, r9, r10, r11, r12, r13, r14, r15
- rbp is saved (pop/push) because GCC seems to ignore its use in the clobber
description.
Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2007-10-25 12:18:53 +00:00
|
|
|
, "r8", "r9", "r10", "r11" , "r12", "r13", "r14", "r15"
|
2012-09-16 12:10:59 +00:00
|
|
|
#else
|
|
|
|
, "ebx", "ecx", "edx", "esi", "edi"
|
KVM: SVM: Let gcc to choose which registers to save (x86_64)
This patch lets GCC to determine which registers to save when we
switch to/from a VCPU in the case of AMD x86_64.
* Original code saves following registers:
rbx, rcx, rdx, rsi, rdi, rbp,
r8, r9, r10, r11, r12, r13, r14, r15
* Patched code:
- informs GCC that we modify following registers
using the clobber description:
rbx, rcx, rdx, rsi, rdi
r8, r9, r10, r11, r12, r13, r14, r15
- rbp is saved (pop/push) because GCC seems to ignore its use in the clobber
description.
Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2007-10-25 12:18:53 +00:00
|
|
|
#endif
|
|
|
|
);
|
[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 10:21:36 +00:00
|
|
|
|
2010-10-21 10:20:34 +00:00
|
|
|
#ifdef CONFIG_X86_64
|
|
|
|
wrmsrl(MSR_GS_BASE, svm->host.gs_base);
|
|
|
|
#else
|
2010-10-21 10:20:33 +00:00
|
|
|
loadsegment(fs, svm->host.fs);
|
2011-03-08 14:09:51 +00:00
|
|
|
#ifndef CONFIG_X86_32_LAZY_GS
|
|
|
|
loadsegment(gs, svm->host.gs);
|
|
|
|
#endif
|
2010-10-19 14:46:55 +00:00
|
|
|
#endif
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
reload_tss(vcpu);
|
|
|
|
|
2007-11-07 15:14:18 +00:00
|
|
|
local_irq_disable();
|
|
|
|
|
2010-10-21 10:20:31 +00:00
|
|
|
vcpu->arch.cr2 = svm->vmcb->save.cr2;
|
|
|
|
vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
|
|
|
|
vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
|
|
|
|
vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip;
|
|
|
|
|
2011-09-12 08:52:24 +00:00
|
|
|
trace_kvm_exit(svm->vmcb->control.exit_code, vcpu, KVM_ISA_SVM);
|
|
|
|
|
2011-01-14 15:45:02 +00:00
|
|
|
if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI))
|
|
|
|
kvm_before_handle_nmi(&svm->vcpu);
|
|
|
|
|
|
|
|
stgi();
|
|
|
|
|
|
|
|
/* Any pending NMI will happen here */
|
|
|
|
|
|
|
|
if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI))
|
|
|
|
kvm_after_handle_nmi(&svm->vcpu);
|
|
|
|
|
2008-04-16 14:51:17 +00:00
|
|
|
sync_cr8_to_lapic(vcpu);
|
|
|
|
|
2007-07-27 12:13:10 +00:00
|
|
|
svm->next_rip = 0;
|
2009-04-23 14:14:37 +00:00
|
|
|
|
2010-12-03 14:25:16 +00:00
|
|
|
svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
|
|
|
|
|
2010-10-14 09:22:52 +00:00
|
|
|
/* if exit due to PF check for async PF */
|
|
|
|
if (svm->vmcb->control.exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR)
|
|
|
|
svm->apf_reason = kvm_read_and_reset_pf_reason();
|
|
|
|
|
2009-05-31 19:58:47 +00:00
|
|
|
if (npt_enabled) {
|
|
|
|
vcpu->arch.regs_avail &= ~(1 << VCPU_EXREG_PDPTR);
|
|
|
|
vcpu->arch.regs_dirty &= ~(1 << VCPU_EXREG_PDPTR);
|
|
|
|
}
|
2010-05-17 12:43:34 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We need to handle MC intercepts here before the vcpu has a chance to
|
|
|
|
* change the physical cpu
|
|
|
|
*/
|
|
|
|
if (unlikely(svm->vmcb->control.exit_code ==
|
|
|
|
SVM_EXIT_EXCP_BASE + MC_VECTOR))
|
|
|
|
svm_handle_mce(svm);
|
2010-12-03 12:15:21 +00:00
|
|
|
|
|
|
|
mark_all_clean(svm->vmcb);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void svm_set_cr3(struct kvm_vcpu *vcpu, unsigned long root)
|
|
|
|
{
|
2007-07-27 12:13:10 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->save.cr3 = root;
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_CR);
|
2010-12-03 14:25:15 +00:00
|
|
|
svm_flush_tlb(vcpu);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
2010-09-10 15:30:41 +00:00
|
|
|
static void set_tdp_cr3(struct kvm_vcpu *vcpu, unsigned long root)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
|
|
|
svm->vmcb->control.nested_cr3 = root;
|
2010-12-03 10:45:53 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_NPT);
|
2010-09-10 15:30:41 +00:00
|
|
|
|
|
|
|
/* Also sync guest cr3 here in case we live migrate */
|
2010-12-05 15:30:00 +00:00
|
|
|
svm->vmcb->save.cr3 = kvm_read_cr3(vcpu);
|
2010-12-03 10:45:54 +00:00
|
|
|
mark_dirty(svm->vmcb, VMCB_CR);
|
2010-09-10 15:30:41 +00:00
|
|
|
|
2010-12-03 14:25:15 +00:00
|
|
|
svm_flush_tlb(vcpu);
|
2010-09-10 15:30:41 +00:00
|
|
|
}
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
static int is_disabled(void)
|
|
|
|
{
|
2007-06-22 09:29:50 +00:00
|
|
|
u64 vm_cr;
|
|
|
|
|
|
|
|
rdmsrl(MSR_VM_CR, vm_cr);
|
|
|
|
if (vm_cr & (1 << SVM_VM_CR_SVM_DISABLE))
|
|
|
|
return 1;
|
|
|
|
|
[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 10:21:36 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2007-02-19 12:37:47 +00:00
|
|
|
static void
|
|
|
|
svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Patch in the VMMCALL instruction:
|
|
|
|
*/
|
|
|
|
hypercall[0] = 0x0f;
|
|
|
|
hypercall[1] = 0x01;
|
|
|
|
hypercall[2] = 0xd9;
|
|
|
|
}
|
|
|
|
|
2007-07-31 11:23:01 +00:00
|
|
|
static void svm_check_processor_compat(void *rtn)
|
|
|
|
{
|
|
|
|
*(int *)rtn = 0;
|
|
|
|
}
|
|
|
|
|
2007-12-26 11:57:04 +00:00
|
|
|
static bool svm_cpu_has_accelerated_tpr(void)
|
|
|
|
{
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2009-04-27 12:35:42 +00:00
|
|
|
static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
|
2008-10-09 08:01:57 +00:00
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-12-18 08:48:46 +00:00
|
|
|
static void svm_cpuid_update(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2010-04-22 10:33:11 +00:00
|
|
|
static void svm_set_supported_cpuid(u32 func, struct kvm_cpuid_entry2 *entry)
|
|
|
|
{
|
2010-04-22 10:33:12 +00:00
|
|
|
switch (func) {
|
2010-09-10 15:31:06 +00:00
|
|
|
case 0x80000001:
|
|
|
|
if (nested)
|
|
|
|
entry->ecx |= (1 << 2); /* Set SVM bit */
|
|
|
|
break;
|
2010-04-22 10:33:12 +00:00
|
|
|
case 0x8000000A:
|
|
|
|
entry->eax = 1; /* SVM revision 1 */
|
|
|
|
entry->ebx = 8; /* Lets support 8 ASIDs in case we add proper
|
|
|
|
ASID emulation to nested SVM */
|
|
|
|
entry->ecx = 0; /* Reserved */
|
2010-07-27 16:14:21 +00:00
|
|
|
entry->edx = 0; /* Per default do not support any
|
|
|
|
additional features */
|
|
|
|
|
|
|
|
/* Support next_rip if host supports it */
|
2010-11-09 14:15:42 +00:00
|
|
|
if (boot_cpu_has(X86_FEATURE_NRIPS))
|
2010-07-27 16:14:21 +00:00
|
|
|
entry->edx |= SVM_FEATURE_NRIP;
|
2010-04-22 10:33:12 +00:00
|
|
|
|
2010-09-10 15:31:05 +00:00
|
|
|
/* Support NPT for the guest if enabled */
|
|
|
|
if (npt_enabled)
|
|
|
|
entry->edx |= SVM_FEATURE_NPT;
|
|
|
|
|
2010-04-22 10:33:12 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-04-22 10:33:11 +00:00
|
|
|
}
|
|
|
|
|
2010-01-05 11:02:27 +00:00
|
|
|
static int svm_get_lpage_level(void)
|
2009-07-27 14:30:48 +00:00
|
|
|
{
|
2010-01-05 11:02:27 +00:00
|
|
|
return PT_PDPE_LEVEL;
|
2009-07-27 14:30:48 +00:00
|
|
|
}
|
|
|
|
|
2009-12-18 08:48:47 +00:00
|
|
|
static bool svm_rdtscp_supported(void)
|
|
|
|
{
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2012-07-02 01:18:48 +00:00
|
|
|
static bool svm_invpcid_supported(void)
|
|
|
|
{
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2014-03-05 22:19:52 +00:00
|
|
|
static bool svm_mpx_supported(void)
|
|
|
|
{
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2010-06-30 04:25:15 +00:00
|
|
|
static bool svm_has_wbinvd_exit(void)
|
|
|
|
{
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2009-12-30 10:40:26 +00:00
|
|
|
static void svm_fpu_deactivate(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
|
2010-11-30 17:03:59 +00:00
|
|
|
set_exception_intercept(svm, NM_VECTOR);
|
2010-02-19 15:23:08 +00:00
|
|
|
update_cr0_intercept(svm);
|
2009-12-30 10:40:26 +00:00
|
|
|
}
|
|
|
|
|
2011-04-04 10:39:33 +00:00
|
|
|
#define PRE_EX(exit) { .exit_code = (exit), \
|
2011-04-21 09:35:41 +00:00
|
|
|
.stage = X86_ICPT_PRE_EXCEPT, }
|
2011-04-04 10:39:28 +00:00
|
|
|
#define POST_EX(exit) { .exit_code = (exit), \
|
2011-04-21 09:35:41 +00:00
|
|
|
.stage = X86_ICPT_POST_EXCEPT, }
|
2011-04-04 10:39:32 +00:00
|
|
|
#define POST_MEM(exit) { .exit_code = (exit), \
|
2011-04-21 09:35:41 +00:00
|
|
|
.stage = X86_ICPT_POST_MEMACCESS, }
|
2011-04-04 10:39:28 +00:00
|
|
|
|
2012-08-29 23:30:20 +00:00
|
|
|
static const struct __x86_intercept {
|
2011-04-04 10:39:28 +00:00
|
|
|
u32 exit_code;
|
|
|
|
enum x86_intercept_stage stage;
|
|
|
|
} x86_intercept_map[] = {
|
|
|
|
[x86_intercept_cr_read] = POST_EX(SVM_EXIT_READ_CR0),
|
|
|
|
[x86_intercept_cr_write] = POST_EX(SVM_EXIT_WRITE_CR0),
|
|
|
|
[x86_intercept_clts] = POST_EX(SVM_EXIT_WRITE_CR0),
|
|
|
|
[x86_intercept_lmsw] = POST_EX(SVM_EXIT_WRITE_CR0),
|
|
|
|
[x86_intercept_smsw] = POST_EX(SVM_EXIT_READ_CR0),
|
2011-04-04 10:39:29 +00:00
|
|
|
[x86_intercept_dr_read] = POST_EX(SVM_EXIT_READ_DR0),
|
|
|
|
[x86_intercept_dr_write] = POST_EX(SVM_EXIT_WRITE_DR0),
|
2011-04-04 10:39:30 +00:00
|
|
|
[x86_intercept_sldt] = POST_EX(SVM_EXIT_LDTR_READ),
|
|
|
|
[x86_intercept_str] = POST_EX(SVM_EXIT_TR_READ),
|
|
|
|
[x86_intercept_lldt] = POST_EX(SVM_EXIT_LDTR_WRITE),
|
|
|
|
[x86_intercept_ltr] = POST_EX(SVM_EXIT_TR_WRITE),
|
|
|
|
[x86_intercept_sgdt] = POST_EX(SVM_EXIT_GDTR_READ),
|
|
|
|
[x86_intercept_sidt] = POST_EX(SVM_EXIT_IDTR_READ),
|
|
|
|
[x86_intercept_lgdt] = POST_EX(SVM_EXIT_GDTR_WRITE),
|
|
|
|
[x86_intercept_lidt] = POST_EX(SVM_EXIT_IDTR_WRITE),
|
2011-04-04 10:39:31 +00:00
|
|
|
[x86_intercept_vmrun] = POST_EX(SVM_EXIT_VMRUN),
|
|
|
|
[x86_intercept_vmmcall] = POST_EX(SVM_EXIT_VMMCALL),
|
|
|
|
[x86_intercept_vmload] = POST_EX(SVM_EXIT_VMLOAD),
|
|
|
|
[x86_intercept_vmsave] = POST_EX(SVM_EXIT_VMSAVE),
|
|
|
|
[x86_intercept_stgi] = POST_EX(SVM_EXIT_STGI),
|
|
|
|
[x86_intercept_clgi] = POST_EX(SVM_EXIT_CLGI),
|
|
|
|
[x86_intercept_skinit] = POST_EX(SVM_EXIT_SKINIT),
|
|
|
|
[x86_intercept_invlpga] = POST_EX(SVM_EXIT_INVLPGA),
|
2011-04-04 10:39:32 +00:00
|
|
|
[x86_intercept_rdtscp] = POST_EX(SVM_EXIT_RDTSCP),
|
|
|
|
[x86_intercept_monitor] = POST_MEM(SVM_EXIT_MONITOR),
|
|
|
|
[x86_intercept_mwait] = POST_EX(SVM_EXIT_MWAIT),
|
2011-04-04 10:39:33 +00:00
|
|
|
[x86_intercept_invlpg] = POST_EX(SVM_EXIT_INVLPG),
|
|
|
|
[x86_intercept_invd] = POST_EX(SVM_EXIT_INVD),
|
|
|
|
[x86_intercept_wbinvd] = POST_EX(SVM_EXIT_WBINVD),
|
|
|
|
[x86_intercept_wrmsr] = POST_EX(SVM_EXIT_MSR),
|
|
|
|
[x86_intercept_rdtsc] = POST_EX(SVM_EXIT_RDTSC),
|
|
|
|
[x86_intercept_rdmsr] = POST_EX(SVM_EXIT_MSR),
|
|
|
|
[x86_intercept_rdpmc] = POST_EX(SVM_EXIT_RDPMC),
|
|
|
|
[x86_intercept_cpuid] = PRE_EX(SVM_EXIT_CPUID),
|
|
|
|
[x86_intercept_rsm] = PRE_EX(SVM_EXIT_RSM),
|
2011-04-04 10:39:34 +00:00
|
|
|
[x86_intercept_pause] = PRE_EX(SVM_EXIT_PAUSE),
|
|
|
|
[x86_intercept_pushf] = PRE_EX(SVM_EXIT_PUSHF),
|
|
|
|
[x86_intercept_popf] = PRE_EX(SVM_EXIT_POPF),
|
|
|
|
[x86_intercept_intn] = PRE_EX(SVM_EXIT_SWINT),
|
|
|
|
[x86_intercept_iret] = PRE_EX(SVM_EXIT_IRET),
|
|
|
|
[x86_intercept_icebp] = PRE_EX(SVM_EXIT_ICEBP),
|
|
|
|
[x86_intercept_hlt] = POST_EX(SVM_EXIT_HLT),
|
2011-04-04 10:39:35 +00:00
|
|
|
[x86_intercept_in] = POST_EX(SVM_EXIT_IOIO),
|
|
|
|
[x86_intercept_ins] = POST_EX(SVM_EXIT_IOIO),
|
|
|
|
[x86_intercept_out] = POST_EX(SVM_EXIT_IOIO),
|
|
|
|
[x86_intercept_outs] = POST_EX(SVM_EXIT_IOIO),
|
2011-04-04 10:39:28 +00:00
|
|
|
};
|
|
|
|
|
2011-04-04 10:39:33 +00:00
|
|
|
#undef PRE_EX
|
2011-04-04 10:39:28 +00:00
|
|
|
#undef POST_EX
|
2011-04-04 10:39:32 +00:00
|
|
|
#undef POST_MEM
|
2011-04-04 10:39:28 +00:00
|
|
|
|
2011-04-04 10:39:27 +00:00
|
|
|
static int svm_check_intercept(struct kvm_vcpu *vcpu,
|
|
|
|
struct x86_instruction_info *info,
|
|
|
|
enum x86_intercept_stage stage)
|
|
|
|
{
|
2011-04-04 10:39:28 +00:00
|
|
|
struct vcpu_svm *svm = to_svm(vcpu);
|
|
|
|
int vmexit, ret = X86EMUL_CONTINUE;
|
|
|
|
struct __x86_intercept icpt_info;
|
|
|
|
struct vmcb *vmcb = svm->vmcb;
|
|
|
|
|
|
|
|
if (info->intercept >= ARRAY_SIZE(x86_intercept_map))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
icpt_info = x86_intercept_map[info->intercept];
|
|
|
|
|
2011-04-21 09:35:41 +00:00
|
|
|
if (stage != icpt_info.stage)
|
2011-04-04 10:39:28 +00:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
switch (icpt_info.exit_code) {
|
|
|
|
case SVM_EXIT_READ_CR0:
|
|
|
|
if (info->intercept == x86_intercept_cr_read)
|
|
|
|
icpt_info.exit_code += info->modrm_reg;
|
|
|
|
break;
|
|
|
|
case SVM_EXIT_WRITE_CR0: {
|
|
|
|
unsigned long cr0, val;
|
|
|
|
u64 intercept;
|
|
|
|
|
|
|
|
if (info->intercept == x86_intercept_cr_write)
|
|
|
|
icpt_info.exit_code += info->modrm_reg;
|
|
|
|
|
2014-06-29 19:55:53 +00:00
|
|
|
if (icpt_info.exit_code != SVM_EXIT_WRITE_CR0 ||
|
|
|
|
info->intercept == x86_intercept_clts)
|
2011-04-04 10:39:28 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
intercept = svm->nested.intercept;
|
|
|
|
|
|
|
|
if (!(intercept & (1ULL << INTERCEPT_SELECTIVE_CR0)))
|
|
|
|
break;
|
|
|
|
|
|
|
|
cr0 = vcpu->arch.cr0 & ~SVM_CR0_SELECTIVE_MASK;
|
|
|
|
val = info->src_val & ~SVM_CR0_SELECTIVE_MASK;
|
|
|
|
|
|
|
|
if (info->intercept == x86_intercept_lmsw) {
|
|
|
|
cr0 &= 0xfUL;
|
|
|
|
val &= 0xfUL;
|
|
|
|
/* lmsw can't clear PE - catch this here */
|
|
|
|
if (cr0 & X86_CR0_PE)
|
|
|
|
val |= X86_CR0_PE;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (cr0 ^ val)
|
|
|
|
icpt_info.exit_code = SVM_EXIT_CR0_SEL_WRITE;
|
|
|
|
|
|
|
|
break;
|
|
|
|
}
|
2011-04-04 10:39:29 +00:00
|
|
|
case SVM_EXIT_READ_DR0:
|
|
|
|
case SVM_EXIT_WRITE_DR0:
|
|
|
|
icpt_info.exit_code += info->modrm_reg;
|
|
|
|
break;
|
2011-04-04 10:39:33 +00:00
|
|
|
case SVM_EXIT_MSR:
|
|
|
|
if (info->intercept == x86_intercept_wrmsr)
|
|
|
|
vmcb->control.exit_info_1 = 1;
|
|
|
|
else
|
|
|
|
vmcb->control.exit_info_1 = 0;
|
|
|
|
break;
|
2011-04-04 10:39:34 +00:00
|
|
|
case SVM_EXIT_PAUSE:
|
|
|
|
/*
|
|
|
|
* We get this for NOP only, but pause
|
|
|
|
* is rep not, check this here
|
|
|
|
*/
|
|
|
|
if (info->rep_prefix != REPE_PREFIX)
|
|
|
|
goto out;
|
2011-04-04 10:39:35 +00:00
|
|
|
case SVM_EXIT_IOIO: {
|
|
|
|
u64 exit_info;
|
|
|
|
u32 bytes;
|
|
|
|
|
|
|
|
exit_info = (vcpu->arch.regs[VCPU_REGS_RDX] & 0xffff) << 16;
|
|
|
|
|
|
|
|
if (info->intercept == x86_intercept_in ||
|
|
|
|
info->intercept == x86_intercept_ins) {
|
|
|
|
exit_info |= SVM_IOIO_TYPE_MASK;
|
|
|
|
bytes = info->src_bytes;
|
|
|
|
} else {
|
|
|
|
bytes = info->dst_bytes;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (info->intercept == x86_intercept_outs ||
|
|
|
|
info->intercept == x86_intercept_ins)
|
|
|
|
exit_info |= SVM_IOIO_STR_MASK;
|
|
|
|
|
|
|
|
if (info->rep_prefix)
|
|
|
|
exit_info |= SVM_IOIO_REP_MASK;
|
|
|
|
|
|
|
|
bytes = min(bytes, 4u);
|
|
|
|
|
|
|
|
exit_info |= bytes << SVM_IOIO_SIZE_SHIFT;
|
|
|
|
|
|
|
|
exit_info |= (u32)info->ad_bytes << (SVM_IOIO_ASIZE_SHIFT - 1);
|
|
|
|
|
|
|
|
vmcb->control.exit_info_1 = exit_info;
|
|
|
|
vmcb->control.exit_info_2 = info->next_rip;
|
|
|
|
|
|
|
|
break;
|
|
|
|
}
|
2011-04-04 10:39:28 +00:00
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
vmcb->control.next_rip = info->next_rip;
|
|
|
|
vmcb->control.exit_code = icpt_info.exit_code;
|
|
|
|
vmexit = nested_svm_exit_handled(svm);
|
|
|
|
|
|
|
|
ret = (vmexit == NESTED_EXIT_DONE) ? X86EMUL_INTERCEPTED
|
|
|
|
: X86EMUL_CONTINUE;
|
|
|
|
|
|
|
|
out:
|
|
|
|
return ret;
|
2011-04-04 10:39:27 +00:00
|
|
|
}
|
|
|
|
|
2013-04-11 11:25:10 +00:00
|
|
|
static void svm_handle_external_intr(struct kvm_vcpu *vcpu)
|
|
|
|
{
|
|
|
|
local_irq_enable();
|
|
|
|
}
|
|
|
|
|
2007-09-09 12:41:59 +00:00
|
|
|
static struct kvm_x86_ops svm_x86_ops = {
|
[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 10:21:36 +00:00
|
|
|
.cpu_has_kvm_support = has_svm,
|
|
|
|
.disabled_by_bios = is_disabled,
|
|
|
|
.hardware_setup = svm_hardware_setup,
|
|
|
|
.hardware_unsetup = svm_hardware_unsetup,
|
2007-07-31 11:23:01 +00:00
|
|
|
.check_processor_compatibility = svm_check_processor_compat,
|
[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 10:21:36 +00:00
|
|
|
.hardware_enable = svm_hardware_enable,
|
|
|
|
.hardware_disable = svm_hardware_disable,
|
2007-12-26 11:57:04 +00:00
|
|
|
.cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr,
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
.vcpu_create = svm_create_vcpu,
|
|
|
|
.vcpu_free = svm_free_vcpu,
|
2007-09-10 15:10:54 +00:00
|
|
|
.vcpu_reset = svm_vcpu_reset,
|
[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 10:21:36 +00:00
|
|
|
|
2007-09-10 15:10:54 +00:00
|
|
|
.prepare_guest_switch = svm_prepare_guest_switch,
|
[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 10:21:36 +00:00
|
|
|
.vcpu_load = svm_vcpu_load,
|
|
|
|
.vcpu_put = svm_vcpu_put,
|
|
|
|
|
2012-09-21 03:42:55 +00:00
|
|
|
.update_db_bp_intercept = update_db_bp_intercept,
|
[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 10:21:36 +00:00
|
|
|
.get_msr = svm_get_msr,
|
|
|
|
.set_msr = svm_set_msr,
|
|
|
|
.get_segment_base = svm_get_segment_base,
|
|
|
|
.get_segment = svm_get_segment,
|
|
|
|
.set_segment = svm_set_segment,
|
2008-03-24 17:38:34 +00:00
|
|
|
.get_cpl = svm_get_cpl,
|
2007-09-05 15:21:32 +00:00
|
|
|
.get_cs_db_l_bits = kvm_get_cs_db_l_bits,
|
2009-12-29 16:43:06 +00:00
|
|
|
.decache_cr0_guest_bits = svm_decache_cr0_guest_bits,
|
2010-12-05 16:56:11 +00:00
|
|
|
.decache_cr3 = svm_decache_cr3,
|
2007-04-27 06:29:21 +00:00
|
|
|
.decache_cr4_guest_bits = svm_decache_cr4_guest_bits,
|
[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 10:21:36 +00:00
|
|
|
.set_cr0 = svm_set_cr0,
|
|
|
|
.set_cr3 = svm_set_cr3,
|
|
|
|
.set_cr4 = svm_set_cr4,
|
|
|
|
.set_efer = svm_set_efer,
|
|
|
|
.get_idt = svm_get_idt,
|
|
|
|
.set_idt = svm_set_idt,
|
|
|
|
.get_gdt = svm_get_gdt,
|
|
|
|
.set_gdt = svm_set_gdt,
|
2014-01-04 17:47:16 +00:00
|
|
|
.get_dr6 = svm_get_dr6,
|
|
|
|
.set_dr6 = svm_set_dr6,
|
2010-04-13 07:05:23 +00:00
|
|
|
.set_dr7 = svm_set_dr7,
|
2014-02-21 09:32:27 +00:00
|
|
|
.sync_dirty_debug_regs = svm_sync_dirty_debug_regs,
|
2009-05-31 19:58:47 +00:00
|
|
|
.cache_reg = svm_cache_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 10:21:36 +00:00
|
|
|
.get_rflags = svm_get_rflags,
|
|
|
|
.set_rflags = svm_set_rflags,
|
2010-01-21 13:31:47 +00:00
|
|
|
.fpu_activate = svm_fpu_activate,
|
2009-12-30 10:40:26 +00:00
|
|
|
.fpu_deactivate = svm_fpu_deactivate,
|
[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 10:21:36 +00:00
|
|
|
|
|
|
|
.tlb_flush = svm_flush_tlb,
|
|
|
|
|
|
|
|
.run = svm_vcpu_run,
|
2007-09-10 15:10:54 +00:00
|
|
|
.handle_exit = handle_exit,
|
[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 10:21:36 +00:00
|
|
|
.skip_emulated_instruction = skip_emulated_instruction,
|
2009-05-12 20:21:05 +00:00
|
|
|
.set_interrupt_shadow = svm_set_interrupt_shadow,
|
|
|
|
.get_interrupt_shadow = svm_get_interrupt_shadow,
|
2007-02-19 12:37:47 +00:00
|
|
|
.patch_hypercall = svm_patch_hypercall,
|
2007-08-06 13:29:07 +00:00
|
|
|
.set_irq = svm_set_irq,
|
2009-04-21 14:45:08 +00:00
|
|
|
.set_nmi = svm_inject_nmi,
|
2007-11-25 11:41:11 +00:00
|
|
|
.queue_exception = svm_queue_exception,
|
2010-07-20 12:06:17 +00:00
|
|
|
.cancel_injection = svm_cancel_injection,
|
2009-03-23 10:12:11 +00:00
|
|
|
.interrupt_allowed = svm_interrupt_allowed,
|
2009-04-21 14:45:08 +00:00
|
|
|
.nmi_allowed = svm_nmi_allowed,
|
2009-11-12 00:04:25 +00:00
|
|
|
.get_nmi_mask = svm_get_nmi_mask,
|
|
|
|
.set_nmi_mask = svm_set_nmi_mask,
|
2009-04-21 14:45:08 +00:00
|
|
|
.enable_nmi_window = enable_nmi_window,
|
|
|
|
.enable_irq_window = enable_irq_window,
|
|
|
|
.update_cr8_intercept = update_cr8_intercept,
|
2013-01-25 02:18:50 +00:00
|
|
|
.set_virtual_x2apic_mode = svm_set_virtual_x2apic_mode,
|
2013-01-25 02:18:51 +00:00
|
|
|
.vm_has_apicv = svm_vm_has_apicv,
|
|
|
|
.load_eoi_exitmap = svm_load_eoi_exitmap,
|
|
|
|
.hwapic_isr_update = svm_hwapic_isr_update,
|
2013-04-11 11:25:15 +00:00
|
|
|
.sync_pir_to_irr = svm_sync_pir_to_irr,
|
2007-10-24 22:29:55 +00:00
|
|
|
|
|
|
|
.set_tss_addr = svm_set_tss_addr,
|
2008-04-25 02:20:22 +00:00
|
|
|
.get_tdp_level = get_npt_level,
|
2009-04-27 12:35:42 +00:00
|
|
|
.get_mt_mask = svm_get_mt_mask,
|
2009-06-17 12:22:14 +00:00
|
|
|
|
2010-11-18 11:09:54 +00:00
|
|
|
.get_exit_info = svm_get_exit_info,
|
|
|
|
|
2010-01-05 11:02:27 +00:00
|
|
|
.get_lpage_level = svm_get_lpage_level,
|
2009-12-18 08:48:46 +00:00
|
|
|
|
|
|
|
.cpuid_update = svm_cpuid_update,
|
2009-12-18 08:48:47 +00:00
|
|
|
|
|
|
|
.rdtscp_supported = svm_rdtscp_supported,
|
2012-07-02 01:18:48 +00:00
|
|
|
.invpcid_supported = svm_invpcid_supported,
|
2014-03-05 22:19:52 +00:00
|
|
|
.mpx_supported = svm_mpx_supported,
|
2010-04-22 10:33:11 +00:00
|
|
|
|
|
|
|
.set_supported_cpuid = svm_set_supported_cpuid,
|
2010-06-30 04:25:15 +00:00
|
|
|
|
|
|
|
.has_wbinvd_exit = svm_has_wbinvd_exit,
|
2010-08-20 08:07:17 +00:00
|
|
|
|
2011-03-25 08:44:49 +00:00
|
|
|
.set_tsc_khz = svm_set_tsc_khz,
|
2012-11-29 20:42:50 +00:00
|
|
|
.read_tsc_offset = svm_read_tsc_offset,
|
2010-08-20 08:07:17 +00:00
|
|
|
.write_tsc_offset = svm_write_tsc_offset,
|
2010-08-20 08:07:23 +00:00
|
|
|
.adjust_tsc_offset = svm_adjust_tsc_offset,
|
2011-03-25 08:44:50 +00:00
|
|
|
.compute_tsc_offset = svm_compute_tsc_offset,
|
2011-08-02 12:54:20 +00:00
|
|
|
.read_l1_tsc = svm_read_l1_tsc,
|
2010-09-10 15:30:41 +00:00
|
|
|
|
|
|
|
.set_tdp_cr3 = set_tdp_cr3,
|
2011-04-04 10:39:27 +00:00
|
|
|
|
|
|
|
.check_intercept = svm_check_intercept,
|
2013-04-11 11:25:10 +00:00
|
|
|
.handle_external_intr = svm_handle_external_intr,
|
[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 10:21:36 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
static int __init svm_init(void)
|
|
|
|
{
|
2007-11-14 12:39:31 +00:00
|
|
|
return kvm_init(&svm_x86_ops, sizeof(struct vcpu_svm),
|
2010-04-28 12:39:01 +00:00
|
|
|
__alignof__(struct vcpu_svm), THIS_MODULE);
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit svm_exit(void)
|
|
|
|
{
|
2007-11-14 12:39:31 +00:00
|
|
|
kvm_exit();
|
[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 10:21:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
module_init(svm_init)
|
|
|
|
module_exit(svm_exit)
|