update_transition_efer() masks out some efer bits when deciding whether
to switch the msr during guest entry; for example, NX is emulated using the
mmu so we don't need to disable it, and LMA/LME are handled by the hardware.
However, with shared msrs, the comparison is made against a stale value;
at the time of the guest switch we may be running with another guest's efer.
Fix by deferring the mask/compare to the actual point of guest entry.
Noted by Marcelo.
Signed-off-by: Avi Kivity <avi@redhat.com>
While we are never normally passed an instruction that exceeds 15 bytes,
smp games can cause us to attempt to interpret one, which will cause
large latencies in non-preempt hosts.
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
This new IOCTL exports all yet user-invisible states related to
exceptions, interrupts, and NMIs. Together with appropriate user space
changes, this fixes sporadic problems of vmsave/restore, live migration
and system reset.
[avi: future-proof abi by adding a flags field]
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The various syscall-related MSRs are fairly expensive to switch. Currently
we switch them on every vcpu preemption, which is far too often:
- if we're switching to a kernel thread (idle task, threaded interrupt,
kernel-mode virtio server (vhost-net), for example) and back, then
there's no need to switch those MSRs since kernel threasd won't
be exiting to userspace.
- if we're switching to another guest running an identical OS, most likely
those MSRs will have the same value, so there's little point in reloading
them.
- if we're running the same OS on the guest and host, the MSRs will have
identical values and reloading is unnecessary.
This patch uses the new user return notifiers to implement last-minute
switching, and checks the msr values to avoid unnecessary reloading.
Signed-off-by: Avi Kivity <avi@redhat.com>
When we migrate a kvm guest that uses pvclock between two hosts, we may
suffer a large skew. This is because there can be significant differences
between the monotonic clock of the hosts involved. When a new host with
a much larger monotonic time starts running the guest, the view of time
will be significantly impacted.
Situation is much worse when we do the opposite, and migrate to a host with
a smaller monotonic clock.
This proposed ioctl will allow userspace to inform us what is the monotonic
clock value in the source host, so we can keep the time skew short, and
more importantly, never goes backwards. Userspace may also need to trigger
the current data, since from the first migration onwards, it won't be
reflected by a simple call to clock_gettime() anymore.
[marcelo: future-proof abi with a flags field]
[jan: fix KVM_GET_CLOCK by clearing flags field instead of checking it]
Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Commit 705c5323 opened the doors of hell by unconditionally injecting
single-step flags as long as guest_debug signaled this. This doesn't
work when the guest branches into some interrupt or exception handler
and triggers a vmexit with flag reloading.
Fix it by saving cs:rip when user space requests single-stepping and
restricting the trace flag injection to this guest code position.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Support for Xen PV-on-HVM guests can be implemented almost entirely in
userspace, except for handling one annoying MSR that maps a Xen
hypercall blob into guest address space.
A generic mechanism to delegate MSR writes to userspace seems overkill
and risks encouraging similar MSR abuse in the future. Thus this patch
adds special support for the Xen HVM MSR.
I implemented a new ioctl, KVM_XEN_HVM_CONFIG, that lets userspace tell
KVM which MSR the guest will write to, as well as the starting address
and size of the hypercall blobs (one each for 32-bit and 64-bit) that
userspace has loaded from files. When the guest writes to the MSR, KVM
copies one page of the blob from userspace to the guest.
I've tested this patch with a hacked-up version of Gerd's userspace
code, booting a number of guests (CentOS 5.3 i386 and x86_64, and
FreeBSD 8.0-RC1 amd64) and exercising PV network and block devices.
[jan: fix i386 build warning]
[avi: future proof abi with a flags field]
Signed-off-by: Ed Swierk <eswierk@aristanetworks.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
New AMD processors (Family 0x10 models 8+) support the Pause
Filter Feature. This feature creates a new field in the VMCB
called Pause Filter Count. If Pause Filter Count is greater
than 0 and intercepting PAUSEs is enabled, the processor will
increment an internal counter when a PAUSE instruction occurs
instead of intercepting. When the internal counter reaches the
Pause Filter Count value, a PAUSE intercept will occur.
This feature can be used to detect contended spinlocks,
especially when the lock holding VCPU is not scheduled.
Rescheduling another VCPU prevents the VCPU seeking the
lock from wasting its quantum by spinning idly.
Experimental results show that most spinlocks are held
for less than 1000 PAUSE cycles or more than a few
thousand. Default the Pause Filter Counter to 3000 to
detect the contended spinlocks.
Processor support for this feature is indicated by a CPUID
bit.
On a 24 core system running 4 guests each with 16 VCPUs,
this patch improved overall performance of each guest's
32 job kernbench by approximately 3-5% when combined
with a scheduler algorithm thati caused the VCPU to
sleep for a brief period. Further performance improvement
may be possible with a more sophisticated yield algorithm.
Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
New NHM processors will support Pause-Loop Exiting by adding 2 VM-execution
control fields:
PLE_Gap - upper bound on the amount of time between two successive
executions of PAUSE in a loop.
PLE_Window - upper bound on the amount of time a guest is allowed to execute in
a PAUSE loop
If the time, between this execution of PAUSE and previous one, exceeds the
PLE_Gap, processor consider this PAUSE belongs to a new loop.
Otherwise, processor determins the the total execution time of this loop(since
1st PAUSE in this loop), and triggers a VM exit if total time exceeds the
PLE_Window.
* Refer SDM volume 3b section 21.6.13 & 22.1.3.
Pause-Loop Exiting can be used to detect Lock-Holder Preemption, where one VP
is sched-out after hold a spinlock, then other VPs for same lock are sched-in
to waste the CPU time.
Our tests indicate that most spinlocks are held for less than 212 cycles.
Performance tests show that with 2X LP over-commitment we can get +2% perf
improvement for kernel build(Even more perf gain with more LPs).
Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Push TF and RF injection and filtering on guest single-stepping into the
vender get/set_rflags callbacks. This makes the whole mechanism more
robust wrt user space IOCTL order and instruction emulations.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Much of so far vendor-specific code for setting up guest debug can
actually be handled by the generic code. This also fixes a minor deficit
in the SVM part /wrt processing KVM_GUESTDBG_ENABLE.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
X86 CPUs need to have some magic happening to enable the virtualization
extensions on them. This magic can result in unpleasant results for
users, like blocking other VMMs from working (vmx) or using invalid TLB
entries (svm).
Currently KVM activates virtualization when the respective kernel module
is loaded. This blocks us from autoloading KVM modules without breaking
other VMMs.
To circumvent this problem at least a bit, this patch introduces on
demand activation of virtualization. This means, that instead
virtualization is enabled on creation of the first virtual machine
and disabled on destruction of the last one.
So using this, KVM can be easily autoloaded, while keeping other
hypervisors usable.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Maintain back mapping from irqchip/pin to gsi to speedup
interrupt acknowledgment notifications.
[avi: build fix on non-x86/ia64]
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This removes assumptions that max GSIs is smaller than number of pins.
Sharing is tracked on pin level not GSI level.
[avi: no PIC on ia64]
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This at once also gets the alignment specification right for
x86-64.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <4B0FF8F80200007800022708@vpn.id2.novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Having run into the run-(boot-)time check a couple of times lately,
I finally took time to find a build-time check so that one doesn't
need to analyze the register/stack dump and resolve this (through
manual lookup in vmlinux) to the offending construct.
The assembler will emit a message like "Error: value of <num> too
large for field of 1 bytes at <offset>", which while not pointing
out the source location still makes analysis quite a bit easier.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <4B0FF8AA0200007800022703@vpn.id2.novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The semantics the PAT code expect of is_untracked_pat_range() is "is
this range completely contained inside the untracked region." This
means that checkin 8a27138924f64d2f30c1022f909f74480046bc3f was
technically wrong, because the implementation needlessly confusing.
The sane interface is for it to take a semiclosed range like just
about everything else (as evidenced by the sheer number of "- 1"'s
removed by that patch) so change the actual implementation to match.
Reported-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Jack Steiner <steiner@sgi.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <20091119202341.GA4420@sgi.com>
The data that was stored in this table is now available in
dev->archdata.iommu. So this table is not longer necessary.
This patch removes the remaining uses of that variable and
removes it from the code.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch cleans up the code to flush device table entries
in the IOMMU. With this chance the driver can get rid of the
iommu_queue_inv_dev_entry() function.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch introduces a list to each protection domain which
keeps all devices associated with the domain. This can be
used later to optimize certain functions and to completly
remove the amd_iommu_pd_table.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds a reference count to each device to count
how often the device was bound to that domain. This is
important for single devices that act as an alias for a
number of others. These devices must stay bound to their
domains until all devices that alias to it are unbound from
the same domain.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch changes IOMMU code to use dev->archdata->iommu to
store information about the alias device and the domain the
device is attached to.
This allows the driver to get rid of the amd_iommu_pd_table
in the future.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch makes device isolation mandatory and removes
support for the amd_iommu=share option. This simplifies the
code in several places.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The non-present cache flag was IOMMU local until now which
doesn't make sense. Make this a global flag so we can remove
the lase user of 'struct iommu' in the map/unmap path.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds code to keep a global list of all protection
domains. This allows to simplify the resume code.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds reference counting for protection domains
per IOMMU. This allows a smarter TLB flushing strategy.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds an index field to struct amd_iommu which can
be used to lookup it up in an array. This index will be used
in struct protection_domain to keep track which protection
domain has devices behind which IOMMU.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch updates the copyright headers in the relevant AMD
IOMMU driver files to match the date of the latest changes.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch moves all function declarations which are only
used inside the driver code to a seperate header file.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Make it readable in the source too, not just in the assembly output.
No change in functionality.
Cc: Brian Gerst <brgerst@gmail.com>
LKML-Reference: <1259176706-5908-1-git-send-email-brgerst@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Zero the input register in the exception handler instead of
using an extra register to pass in a zero value.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
LKML-Reference: <1259176706-5908-1-git-send-email-brgerst@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Mtdblock driver doesn't call flush_dcache_page for pages in request. So,
this causes problems on architectures where the icache doesn't fill from
the dcache or with dcache aliases. The patch fixes this.
The ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE symbol was introduced to avoid
pointless empty cache-thrashing loops on architectures for which
flush_dcache_page() is a no-op. Every architecture was provided with this
flush pages on architectires where ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE is
equal 1 or do nothing otherwise.
See "fix mtd_blkdevs problem with caches on some architectures" discussion
on LKML for more information.
Signed-off-by: Ilya Loginov <isloginov@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Peter Horton <phorton@bitbox.co.uk>
Cc: "Ed L. Cashin" <ecashin@coraid.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Percpu symbols now occupy the same namespace as other global
symbols and as such short global symbols without subsystem
prefix tend to collide with local variables. dr7 percpu
variable used by x86 was hit by this. Rename it to cpu_dr7.
The rename also makes it more consistent with its fellow
cpu_debugreg percpu variable.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <20091125115856.GA17856@elte.hu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
This patch factors out the search for an MMCONFIG region, which was
previously implemented in both mmconfig_32 and mmconfig_64. No functional
change.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This changes pci_mmcfg_region from a table to a list, to make it easier
to add and remove MMCONFIG regions for PCI host bridge hotplug.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
The virtual address is only used for x86_64, but it's so much simpler
to manage it as part of the pci_mmcfg_region that I think it's worth
wasting a pointer per MMCONFIG region on x86_32.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This patch adds a resource and corresponding name to the MMCONFIG
structure. This makes allocation simpler (we can allocate the
resource and name at the same time we allocate the pci_mmcfg_region),
and gives us a way to hang onto the resource after inserting it.
This will be needed so we can release and free it when hot-removing
a host bridge.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This only renames the struct pci_mmcfg_region members; no functional change.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This adds a struct pci_mmcfg_region with a little more information
than the struct acpi_mcfg_allocation used previously. The acpi_mcfg
structure is defined by the spec, so we can't change it.
To begin with, struct pci_mmcfg_region is basically the same as the
ACPI MCFG version, but future patches will add more information.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This factors out the common "bus << 20" expression used when computing the
MMCONFIG address.
Reviewed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Move the find_smp_config() call to before bootmem is initialized.
Use reserve_early() instead of reserve_bootmem() in it.
This simplifies the code, we only need to call find_smp_config()
once and can remove the now unneeded reserve parameter from
x86_init_mpparse::find_smp_config.
We thus also reduce x86's dependency on bootmem allocations.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4B0BB9F2.70907@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
