ioeventfd is a mechanism to register PIO/MMIO regions to trigger an eventfd
signal when written to by a guest. Host userspace can register any
arbitrary IO address with a corresponding eventfd and then pass the eventfd
to a specific end-point of interest for handling.
Normal IO requires a blocking round-trip since the operation may cause
side-effects in the emulated model or may return data to the caller.
Therefore, an IO in KVM traps from the guest to the host, causes a VMX/SVM
"heavy-weight" exit back to userspace, and is ultimately serviced by qemu's
device model synchronously before returning control back to the vcpu.
However, there is a subclass of IO which acts purely as a trigger for
other IO (such as to kick off an out-of-band DMA request, etc). For these
patterns, the synchronous call is particularly expensive since we really
only want to simply get our notification transmitted asychronously and
return as quickly as possible. All the sychronous infrastructure to ensure
proper data-dependencies are met in the normal IO case are just unecessary
overhead for signalling. This adds additional computational load on the
system, as well as latency to the signalling path.
Therefore, we provide a mechanism for registration of an in-kernel trigger
point that allows the VCPU to only require a very brief, lightweight
exit just long enough to signal an eventfd. This also means that any
clients compatible with the eventfd interface (which includes userspace
and kernelspace equally well) can now register to be notified. The end
result should be a more flexible and higher performance notification API
for the backend KVM hypervisor and perhipheral components.
To test this theory, we built a test-harness called "doorbell". This
module has a function called "doorbell_ring()" which simply increments a
counter for each time the doorbell is signaled. It supports signalling
from either an eventfd, or an ioctl().
We then wired up two paths to the doorbell: One via QEMU via a registered
io region and through the doorbell ioctl(). The other is direct via
ioeventfd.
You can download this test harness here:
ftp://ftp.novell.com/dev/ghaskins/doorbell.tar.bz2
The measured results are as follows:
qemu-mmio: 110000 iops, 9.09us rtt
ioeventfd-mmio: 200100 iops, 5.00us rtt
ioeventfd-pio: 367300 iops, 2.72us rtt
I didn't measure qemu-pio, because I have to figure out how to register a
PIO region with qemu's device model, and I got lazy. However, for now we
can extrapolate based on the data from the NULLIO runs of +2.56us for MMIO,
and -350ns for HC, we get:
qemu-pio: 153139 iops, 6.53us rtt
ioeventfd-hc: 412585 iops, 2.37us rtt
these are just for fun, for now, until I can gather more data.
Here is a graph for your convenience:
http://developer.novell.com/wiki/images/7/76/Iofd-chart.png
The conclusion to draw is that we save about 4us by skipping the userspace
hop.
--------------------
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Today kvm_io_bus_regsiter_dev() returns void and will internally BUG_ON
if it fails. We want to create dynamic MMIO/PIO entries driven from
userspace later in the series, so we need to enhance the code to be more
robust with the following changes:
1) Add a return value to the registration function
2) Fix up all the callsites to check the return code, handle any
failures, and percolate the error up to the caller.
3) Add an unregister function that collapses holes in the array
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This changes bus accesses to use high-level kvm_io_bus_read/kvm_io_bus_write
functions. in_range now becomes unused so it is removed from device ops in
favor of read/write callbacks performing range checks internally.
This allows aliasing (mostly for in-kernel virtio), as well as better error
handling by making it possible to pass errors up to userspace.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use slots_lock to protect device list on the bus. slots_lock is already
taken for read everywhere, so we only need to take it for write when
registering devices. This is in preparation to removing in_range and
kvm->lock around it.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Correct missing locking in a few places in x86's vm_ioctl handling path.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Since
commit 854b5338196b1175706e99d63be43a4f8d8ab607
Author: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
KVM: s390: streamline memslot handling
s390 uses the values of the memslot instead of doing everything in the arch
ioctl handler of the KVM_SET_USER_MEMORY_REGION. Unfortunately we missed to
set the userspace_addr of our memslot due to our s390 ifdef in
__kvm_set_memory_region.
Old s390 userspace launchers did not notice, since they started the guest at
userspace address 0.
Because of CONFIG_DEFAULT_MMAP_MIN_ADDR we now put the guest at 1M userspace,
which does not work. This patch makes sure that new.userspace_addr is set
on s390.
This fix should go in quickly. Nevertheless, looking at the code we should
clean up that ifdef in the long term. Any kernel janitors?
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This allows use of the powerful ftrace infrastructure.
See Documentation/trace/ for usage information.
[avi, stephen: various build fixes]
[sheng: fix control register breakage]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Disable usage of 2M pages if VMX_EPT_2MB_PAGE_BIT (bit 16) is clear
in MSR_IA32_VMX_EPT_VPID_CAP and EPT is enabled.
[avi: s/largepages_disabled/largepages_enabled/ to avoid negative logic]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Archs are free to use vcpu_id as they see fit. For x86 it is used as
vcpu's apic id. New ioctl is added to configure boot vcpu id that was
assumed to be 0 till now.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Protect irq injection/acking data structures with a separate irq_lock
mutex. This fixes the following deadlock:
CPU A CPU B
kvm_vm_ioctl_deassign_dev_irq()
mutex_lock(&kvm->lock); worker_thread()
-> kvm_deassign_irq() -> kvm_assigned_dev_interrupt_work_handler()
-> deassign_host_irq() mutex_lock(&kvm->lock);
-> cancel_work_sync() [blocked]
[gleb: fix ia64 path]
Reported-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Introduce irq_lock, and use to protect ioapic data structures.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We modernize the io_device code so that we use container_of() instead of
dev->private, and move the vtable to a separate ops structure
(theoretically allows better caching for multiple instances of the same
ops structure)
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
Disable interrupt at interrupt handler and enable it when guest ack is for
the level triggered interrupt, to prevent reinjected interrupt. MSI/MSI-X don't
need it.
One possible problem is multiply same vector interrupt injected between irq
handler and scheduled work handler would be merged as one for MSI/MSI-X.
But AFAIK, the drivers handle it well.
The patch fixed the oplin card performance issue(MSI-X performance is half of
MSI/INTx).
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
KVM provides a complete virtual system environment for guests, including
support for injecting interrupts modeled after the real exception/interrupt
facilities present on the native platform (such as the IDT on x86).
Virtual interrupts can come from a variety of sources (emulated devices,
pass-through devices, etc) but all must be injected to the guest via
the KVM infrastructure. This patch adds a new mechanism to inject a specific
interrupt to a guest using a decoupled eventfd mechnanism: Any legal signal
on the irqfd (using eventfd semantics from either userspace or kernel) will
translate into an injected interrupt in the guest at the next available
interrupt window.
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
make_all_cpus_request contains a race condition which can
trigger false request completed status, as follows:
CPU0 CPU1
if (test_and_set_bit(req,&vcpu->requests))
.... if (test_and_set_bit(req,&vcpu->requests))
.. return
proceed to smp_call_function_many(wait=1)
Use a spinlock to serialize concurrent CPUs.
Cc: stable@kernel.org
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
When slot is already allocated and being asked to be tracked we need
to break the large pages.
This code flush the mmu when someone ask a slot to start dirty bit
tracking.
Cc: stable@kernel.org
Signed-off-by: Izik Eidus <ieidus@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If userspace specifies a memory slot that is larger than 8 petabytes, it
could overflow the largepages variable.
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
If a slots guest physical address and host virtual address unequal (mod
large page size), then we would erronously try to back guest large pages
with host large pages. Detect this misalignment and diable large page
support for the trouble slot.
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_handle_hva relies on mmu_lock protection to safely access
the memslot structures.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_assigned_dev_ack_irq is vulnerable to a race condition with the
interrupt handler function. It does:
if (dev->host_irq_disabled) {
enable_irq(dev->host_irq);
dev->host_irq_disabled = false;
}
If an interrupt triggers before the host->dev_irq_disabled assignment,
it will disable the interrupt and set dev->host_irq_disabled to true.
On return to kvm_assigned_dev_ack_irq, dev->host_irq_disabled is set to
false, and the next kvm_assigned_dev_ack_irq call will fail to reenable
it.
Other than that, having the interrupt handler and work handlers run in
parallel sounds like asking for trouble (could not spot any obvious
problem, but better not have to, its fragile).
CC: sheng.yang@intel.com
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Intel TXT(Trusted Execution Technology) required VMX off for all cpu to work
when system shutdown.
CC: Joseph Cihula <joseph.cihula@intel.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_vcpu_block() unhalts vpu on an interrupt/timer without checking
if interrupt window is actually opened.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently timer events are processed before entering guest mode. Move it
to main vcpu event loop since timer events should be processed even while
vcpu is halted. Timer may cause interrupt/nmi to be injected and only then
vcpu will be unhalted.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
free_mmu_pages() should only undo what alloc_mmu_pages() does.
Free mmu pages from the generic VM destruction function, kvm_destroy_vm().
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
After discussion with Marcelo, we decided to rework device assignment framework
together. The old problems are kernel logic is unnecessary complex. So Marcelo
suggest to split it into a more elegant way:
1. Split host IRQ assign and guest IRQ assign. And userspace determine the
combination. Also discard msi2intx parameter, userspace can specific
KVM_DEV_IRQ_HOST_MSI | KVM_DEV_IRQ_GUEST_INTX in assigned_irq->flags to
enable MSI to INTx convertion.
2. Split assign IRQ and deassign IRQ. Import two new ioctls:
KVM_ASSIGN_DEV_IRQ and KVM_DEASSIGN_DEV_IRQ.
This patch also fixed the reversed _IOR vs _IOW in definition(by deprecated the
old interface).
[avi: replace homemade bitcount() by hweight_long()]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch finally enable MSI-X.
What we need for MSI-X:
1. Intercept one page in MMIO region of device. So that we can get guest desired
MSI-X table and set up the real one. Now this have been done by guest, and
transfer to kernel using ioctl KVM_SET_MSIX_NR and KVM_SET_MSIX_ENTRY.
2. Information for incoming interrupt. Now one device can have more than one
interrupt, and they are all handled by one workqueue structure. So we need to
identify them. The previous patch enable gsi_msg_pending_bitmap get this done.
3. Mapping from host IRQ to guest gsi as well as guest gsi to real MSI/MSI-X
message address/data. We used same entry number for the host and guest here, so
that it's easy to find the correlated guest gsi.
What we lack for now:
1. The PCI spec said nothing can existed with MSI-X table in the same page of
MMIO region, except pending bits. The patch ignore pending bits as the first
step (so they are always 0 - no pending).
2. The PCI spec allowed to change MSI-X table dynamically. That means, the OS
can enable MSI-X, then mask one MSI-X entry, modify it, and unmask it. The patch
didn't support this, and Linux also don't work in this way.
3. The patch didn't implement MSI-X mask all and mask single entry. I would
implement the former in driver/pci/msi.c later. And for single entry, userspace
should have reposibility to handle it.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We have to handle more than one interrupt with one handler for MSI-X. Avi
suggested to use a flag to indicate the pending. So here is it.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Introduce KVM_SET_MSIX_NR and KVM_SET_MSIX_ENTRY two ioctls.
This two ioctls are used by userspace to specific guest device MSI-X entry
number and correlate MSI-X entry with GSI during the initialization stage.
MSI-X should be well initialzed before enabling.
Don't support change MSI-X entry number for now.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Under CONFIG_MAXSMP, cpus_hardware_enabled is allocated from the heap and
not statically initialized. This causes a crash on reboot when kvm thinks
vmx is enabled on random nonexistent cpus and accesses nonexistent percpu
lists.
Fix by explicitly clearing the variable.
Cc: stable@kernel.org
Reported-and-tested-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
When checking for overlapping slots on registration of a new one, kvm
currently also considers zero-length (ie. deleted) slots and rejects
requests incorrectly. This finally denies user space from joining slots.
Fix the check by skipping deleted slots and advertise this via a
KVM_CAP_JOIN_MEMORY_REGIONS_WORKS.
Cc: stable@kernel.org
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The large page initialization code concludes there are two large pages spanned
by a slot covering 1 (small) page starting at gfn 1. This is incorrect, and
also results in incorrect write_count initialization in some cases (base = 1,
npages = 513 for example).
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
only need to set assigned_dev_id for deassignment, use
match->flags to judge and deassign it.
Acked-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The function kvm_is_mmio_pfn is called before put_page is called on a
page by KVM. This is a problem when when this function is called on some
struct page which is part of a compund page. It does not test the
reserved flag of the compound page but of the struct page within the
compount page. This is a problem when KVM works with hugepages allocated
at boot time. These pages have the reserved bit set in all tail pages.
Only the flag in the compount head is cleared. KVM would not put such a
page which results in a memory leak.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Merge MSI userspace interface with IRQ routing table. Notice the API have been
changed, and using IRQ routing table would be the only interface kvm-userspace
supported.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently KVM has a static routing from GSI numbers to interrupts (namely,
0-15 are mapped 1:1 to both PIC and IOAPIC, and 16:23 are mapped 1:1 to
the IOAPIC). This is insufficient for several reasons:
- HPET requires non 1:1 mapping for the timer interrupt
- MSIs need a new method to assign interrupt numbers and dispatch them
- ACPI APIC mode needs to be able to reassign the PCI LINK interrupts to the
ioapics
This patch implements an interrupt routing table (as a linked list, but this
can be easily changed) and a userspace interface to replace the table. The
routing table is initialized according to the current hardwired mapping.
Signed-off-by: Avi Kivity <avi@redhat.com>
Allow clients to request notifications when the guest masks or unmasks a
particular irq line. This complements irq ack notifications, as the guest
will not ack an irq line that is masked.
Currently implemented for the ioapic only.
Signed-off-by: Avi Kivity <avi@redhat.com>
MSI is always enabled by default for msi2intx=1. But if msi2intx=0, we
have to disable MSI if guest require to do so.
The patch also discard unnecessary msi2intx judgment if guest want to update
MSI state.
Notice KVM_DEV_IRQ_ASSIGN_MSI_ACTION is a mask which should cover all MSI
related operations, though we only got one for now.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This rips out the support for KVM_DEBUG_GUEST and introduces a new IOCTL
instead: KVM_SET_GUEST_DEBUG. The IOCTL payload consists of a generic
part, controlling the "main switch" and the single-step feature. The
arch specific part adds an x86 interface for intercepting both types of
debug exceptions separately and re-injecting them when the host was not
interested. Moveover, the foundation for guest debugging via debug
registers is layed.
To signal breakpoint events properly back to userland, an arch-specific
data block is now returned along KVM_EXIT_DEBUG. For x86, the arch block
contains the PC, the debug exception, and relevant debug registers to
tell debug events properly apart.
The availability of this new interface is signaled by
KVM_CAP_SET_GUEST_DEBUG. Empty stubs for not yet supported archs are
provided.
Note that both SVM and VTX are supported, but only the latter was tested
yet. Based on the experience with all those VTX corner case, I would be
fairly surprised if SVM will work out of the box.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm->slots_lock is outer to kvm->lock, so take slots_lock
in kvm_vm_ioctl_assign_device() before taking kvm->lock,
rather than taking it in kvm_iommu_map_memslots().
Cc: stable@kernel.org
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In the past, kvm_get_kvm() and kvm_put_kvm() was called in assigned device irq
handler and interrupt_work, in order to prevent cancel_work_sync() in
kvm_free_assigned_irq got a illegal state when waiting for interrupt_work done.
But it's tricky and still got two problems:
1. A bug ignored two conditions that cancel_work_sync() would return true result
in a additional kvm_put_kvm().
2. If interrupt type is MSI, we would got a window between cancel_work_sync()
and free_irq(), which interrupt would be injected again...
This patch discard the reference count used for irq handler and interrupt_work,
and ensure the legal state by moving the free function at the very beginning of
kvm_destroy_vm(). And the patch fix the second bug by disable irq before
cancel_work_sync(), which may result in nested disable of irq but OK for we are
going to free it.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_arch_sync_events is introduced to quiet down all other events may happen
contemporary with VM destroy process, like IRQ handler and work struct for
assigned device.
For kvm_arch_sync_events is called at the very beginning of kvm_destroy_vm(), so
the state of KVM here is legal and can provide a environment to quiet down other
events.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The destructor for huge pages uses the backing inode for adjusting
hugetlbfs accounting.
Hugepage mappings are destroyed by exit_mmap, after
mmu_notifier_release, so there are no notifications through
unmap_hugepage_range at this point.
The hugetlbfs inode can be freed with pages backed by it referenced
by the shadow. When the shadow releases its reference, the huge page
destructor will access a now freed inode.
Implement the release operation for kvm mmu notifiers to release page
refs before the hugetlbfs inode is gone.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In kvm_iommu_unmap_memslots(), assigned_dev_head is already empty.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Support device deassignment, it can be used in device hotplug.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
intel iommu APIs are updated, use the new APIs.
In addition, change kvm_iommu_map_guest() to just create the domain, let kvm_iommu_assign_device() assign device.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
If an assigned device shares a guest irq with an emulated
device then we currently interpret an ack generated by the
emulated device as originating from the assigned device
leading to e.g. "Unbalanced enable for IRQ 4347" from the
enable_irq() in kvm_assigned_dev_ack_irq().
The fix is fairly simple - don't enable the physical device
irq unless it was previously disabled.
Of course, this can still lead to a situation where a
non-assigned device ACK can cause the physical device irq to
be reenabled before the device was serviced. However, being
level sensitive, the interrupt will merely be regenerated.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This changes cpus_hardware_enabled from a cpumask_t to a cpumask_var_t:
equivalent for CONFIG_CPUMASKS_OFFSTACK=n, otherwise dynamically allocated.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
We're getting rid on on-stack cpumasks for large NR_CPUS.
1) Use cpumask_var_t/alloc_cpumask_var.
2) smp_call_function_mask -> smp_call_function_many
3) cpus_clear, cpus_empty, cpu_set -> cpumask_clear, cpumask_empty,
cpumask_set_cpu.
This actually generates slightly smaller code than the old one with
CONFIG_CPUMASKS_OFFSTACK=n. (gcc knows that cpus cannot be NULL in
that case, where cpumask_var_t is cpumask_t[1]).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
Avi said:
> Wow, code duplication from Rusty. Things must be bad.
Something about glass houses comes to mind. But instead, a patch.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@redhat.com>
There is a race between a "close of the file descriptors" and module
unload in the kvm module.
You can easily trigger this problem by applying this debug patch:
>--- kvm.orig/virt/kvm/kvm_main.c
>+++ kvm/virt/kvm/kvm_main.c
>@@ -648,10 +648,14 @@ void kvm_free_physmem(struct kvm *kvm)
> kvm_free_physmem_slot(&kvm->memslots[i], NULL);
> }
>
>+#include <linux/delay.h>
> static void kvm_destroy_vm(struct kvm *kvm)
> {
> struct mm_struct *mm = kvm->mm;
>
>+ printk("off1\n");
>+ msleep(5000);
>+ printk("off2\n");
> spin_lock(&kvm_lock);
> list_del(&kvm->vm_list);
> spin_unlock(&kvm_lock);
and killing the userspace, followed by an rmmod.
The problem is that kvm_destroy_vm can run while the module count
is 0. That means, you can remove the module while kvm_destroy_vm
is running. But kvm_destroy_vm is part of the module text. This
causes a kerneloops. The race exists without the msleep but is much
harder to trigger.
This patch requires the fix for anon_inodes (anon_inodes: use fops->owner
for module refcount).
With this patch, we can set the owner of all anonymous KVM inodes file
operations. The VFS will then control the KVM module refcount as long as there
is an open file. kvm_destroy_vm will be called by the release function of the
last closed file - before the VFS drops the module refcount.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Right now, KVM does not remove a slot when we do a
register ioctl for size 0 (would be the expected behaviour).
Instead, we only mark it as empty, but keep all bitmaps
and allocated data structures present. It completely
nullifies our chances of reusing that same slot again
for mapping a different piece of memory.
In this patch, we destroy rmaps, and vfree() the
pointers that used to hold the dirty bitmap, rmap
and lpage_info structures.
Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Split out the logic corresponding to undoing assign_irq() and
clean it up a bit.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Set assigned_dev->irq_source_id to -1 so that we can avoid freeing
a source ID which we never allocated.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Impact: make global function static
virt/kvm/kvm_main.c:85:6: warning: symbol 'kvm_rebooting' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Avi Kivity <avi@redhat.com>
Now we use MSI as default one, and translate MSI to INTx when guest need
INTx rather than MSI. For legacy device, we provide support for non-sharing
host IRQ.
Provide a parameter msi2intx for this method. The value is true by default in
x86 architecture.
We can't guarantee this mode can work on every device, but for most of us
tested, it works. If your device encounter some trouble with this mode, you can
try set msi2intx modules parameter to 0. If the device is OK with msi2intx=0,
then please report it to KVM mailing list or me. We may prepare a blacklist for
the device that can't work in this mode.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We enable guest MSI and host MSI support in this patch. The userspace want to
enable MSI should set KVM_DEV_IRQ_ASSIGN_ENABLE_MSI in the assigned_irq's flag.
Function would return -ENOTTY if can't enable MSI, userspace shouldn't set MSI
Enable bit when KVM_ASSIGN_IRQ return -ENOTTY with
KVM_DEV_IRQ_ASSIGN_ENABLE_MSI.
Userspace can tell the support of MSI device from #ifdef KVM_CAP_DEVICE_MSI.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The function is used to dispatch MSI to lapic according to MSI message
address and message data.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Separate guest irq type and host irq type, for we can support guest using INTx
with host using MSI (but not opposite combination).
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Separate INTx enabling part to a independence function, so that we can add MSI
enabling part easily.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Distinguish common part for device assignment and INTx part, perparing for
refactor later.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Commit 7fd49de9773fdcb7b75e823b21c1c5dc1e218c14 "KVM: ensure that memslot
userspace addresses are page-aligned" broke kernel space allocated memory
slot, for the userspace_addr is invalid.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Bad page translation and silent guest failure ensue if the userspace address is
not page-aligned. I hit this problem using large (host) pages with qemu,
because qemu currently has a hardcoded 4096-byte alignment for guest memory
allocations.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Some areas of kvm x86 mmu are using gfn offset inside a slot without
unaliasing the gfn first. This patch makes sure that the gfn will be
unaliased and add gfn_to_memslot_unaliased() to save the calculating
of the gfn unaliasing in case we have it unaliased already.
Signed-off-by: Izik Eidus <ieidus@redhat.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Ideally, every assigned device should in a clear condition before and after
assignment, so that the former state of device won't affect later work.
Some devices provide a mechanism named Function Level Reset, which is
defined in PCI/PCI-e document. We should execute it before and after device
assignment.
(But sadly, the feature is new, and most device on the market now don't
support it. We are considering using D0/D3hot transmit to emulate it later,
but not that elegant and reliable as FLR itself.)
[Update: Reminded by Xiantao, execute FLR after we ensure that the device can
be assigned to the guest.]
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Also remove unnecessary parameter of unregister irq ack notifier.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Every call of kvm_set_irq() should offer an irq_source_id, which is
allocated by kvm_request_irq_source_id(). Based on irq_source_id, we
identify the irq source and implement logical OR for shared level
interrupts.
The allocated irq_source_id can be freed by kvm_free_irq_source_id().
Currently, we support at most sizeof(unsigned long) different irq sources.
[Amit: - rebase to kvm.git HEAD
- move definition of KVM_USERSPACE_IRQ_SOURCE_ID to common file
- move kvm_request_irq_source_id to the update_irq ioctl]
[Xiantao: - Add kvm/ia64 stuff and make it work for kvm/ia64 guests]
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Add a kvm_ prefix to avoid polluting kernel's name space.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
To share with other archs, this patch moves device assignment
logic to common parts.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Assigned device could DMA to mmio pages, so also need to map mmio pages
into VT-d page table.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently "#include <linux/intel-iommu.h>" is not needed in
virt/kvm/kvm_main.c.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
One of vcpu_setup responsibilities is to do mmu initialization.
However, in case we fail in kvm_arch_vcpu_reset, before we get the
chance to init mmu. OTOH, vcpu_destroy will attempt to destroy mmu,
triggering a bug. Keeping track of whether or not mmu is initialized
would unnecessarily complicate things. Rather, we just make return,
making sure any needed uninitialization is done before we return, in
case we fail.
Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Convert gfn_to_pfn to use get_user_pages_fast, which can do lockless
pagetable lookups on x86. Kernel compilation on 4-way guest is 3.7%
faster on VMX.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_vm_fault is invoked with mmap_sem held in read mode. Since gfn_to_page
will be converted to get_user_pages_fast, which requires this lock NOT
to be held, switch to opencoded get_user_pages.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Based on a patch by: Kay, Allen M <allen.m.kay@intel.com>
This patch enables PCI device assignment based on VT-d support.
When a device is assigned to the guest, the guest memory is pinned and
the mapping is updated in the VT-d IOMMU.
[Amit: Expose KVM_CAP_IOMMU so we can check if an IOMMU is present
and also control enable/disable from userspace]
Signed-off-by: Kay, Allen M <allen.m.kay@intel.com>
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Ben-Ami Yassour <benami@il.ibm.com>
Signed-off-by: Amit Shah <amit.shah@qumranet.com>
Acked-by: Mark Gross <mgross@linux.intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Offline or uninitialized vcpu's can be executed if requested to perform
userspace work.
Follow Avi's suggestion to handle halted vcpu's in the main loop,
simplifying kvm_emulate_halt(). Introduce a new vcpu->requests bit to
indicate events that promote state from halted to running.
Also standardize vcpu wake sites.
Signed-off-by: Marcelo Tosatti <mtosatti <at> redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This is esoteric and only needed to break COW on MAP_SHARED mappings. Since
KVM no longer does these sorts of mappings, breaking COW on them is no longer
necessary.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Userspace may specify memory slots that are backed by mmio pages rather than
normal RAM. In some cases it is not enough to identify these mmio pages
by pfn_valid(). This patch adds checking the PageReserved as well.
Signed-off-by: Ben-Ami Yassour <benami@il.ibm.com>
Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Synchronize changes to host virtual addresses which are part of
a KVM memory slot to the KVM shadow mmu. This allows pte operations
like swapping, page migration, and madvise() to transparently work
with KVM.
Signed-off-by: Andrea Arcangeli <andrea@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This allows reading memslots with only the mmu_lock hold for mmu
notifiers that runs in atomic context and with mmu_lock held.
Signed-off-by: Andrea Arcangeli <andrea@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch just extends the anon_inode_getfd interface to take an additional
parameter with a flag value. The flag value is passed on to
get_unused_fd_flags in anticipation for a use with the O_CLOEXEC flag.
No actual semantic changes here, the changed callers all pass 0 for now.
[akpm@linux-foundation.org: KVM fix]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
smp_call_function_mask() now complains when called in a preemptible context;
adjust its callers accordingly.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Flush the shadow mmu before removing regions to avoid stale entries.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch #ifdefs the bitmap array for dirty tracking. We don't have dirty
tracking on s390 today, and we'd love to use our storage keys to store the
dirty information for migration. Therefore, we won't need this array at all,
and due to our limited amount of vmalloc space this limits the amount of guests
we can run.
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch adds all needed structures to coalesce MMIOs.
Until an architecture uses it, it is not compiled.
Coalesced MMIO introduces two ioctl() to define where are the MMIO zones that
can be coalesced:
- KVM_REGISTER_COALESCED_MMIO registers a coalesced MMIO zone.
It requests one parameter (struct kvm_coalesced_mmio_zone) which defines
a memory area where MMIOs can be coalesced until the next switch to
user space. The maximum number of MMIO zones is KVM_COALESCED_MMIO_ZONE_MAX.
- KVM_UNREGISTER_COALESCED_MMIO cancels all registered zones inside
the given bounds (bounds are also given by struct kvm_coalesced_mmio_zone).
The userspace client can check kernel coalesced MMIO availability by asking
ioctl(KVM_CHECK_EXTENSION) for the KVM_CAP_COALESCED_MMIO capability.
The ioctl() call to KVM_CAP_COALESCED_MMIO will return 0 if not supported,
or the page offset where will be stored the ring buffer.
The page offset depends on the architecture.
After an ioctl(KVM_RUN), the first page of the KVM memory mapped points to
a kvm_run structure. The offset given by KVM_CAP_COALESCED_MMIO is
an offset to the coalesced MMIO ring expressed in PAGE_SIZE relatively
to the address of the start of th kvm_run structure. The MMIO ring buffer
is defined by the structure kvm_coalesced_mmio_ring.
[akio: fix oops during guest shutdown]
Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Modify member in_range() of structure kvm_io_device to pass length and the type
of the I/O (write or read).
This modification allows to use kvm_io_device with coalesced MMIO.
Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>
KVM turns off hardware virtualization extensions during reboot, in order
to disassociate the memory used by the virtualization extensions from the
processor, and in order to have the system in a consistent state.
Unfortunately virtual machines may still be running while this goes on,
and once virtualization extensions are turned off, any virtulization
instruction will #UD on execution.
Fix by adding an exception handler to virtualization instructions; if we get
an exception during reboot, we simply spin waiting for the reset to complete.
If it's a true exception, BUG() so we can have our stack trace.
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch allows VMAs that contain no backing page to be used for guest
memory. This is useful for assigning mmio regions to a guest.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
kvm_dev_ioctl casts the arg value to void __user *, just to recast it
again to long. This seems unnecessary.
According to objdump the binary code on x86 is unchanged by this patch.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
It's not even passed on to smp_call_function() anymore, since that
was removed. So kill it.
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
It's never used and the comments refer to nonatomic and retry
interchangably. So get rid of it.
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
There's still a race in kvm_vcpu_block(), if a wake_up_interruptible()
call happens before the task state is set to TASK_INTERRUPTIBLE:
CPU0 CPU1
kvm_vcpu_block
add_wait_queue
kvm_cpu_has_interrupt = 0
set interrupt
if (waitqueue_active())
wake_up_interruptible()
kvm_cpu_has_pending_timer
kvm_arch_vcpu_runnable
signal_pending
set_current_state(TASK_INTERRUPTIBLE)
schedule()
Can be fixed by using prepare_to_wait() which sets the task state before
testing for the wait condition.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
a) none of the callers even looks at inode or file returned by anon_inode_getfd()
b) any caller that would try to look at those would be racy, since by the time
it returns we might have raced with close() from another thread and that
file would be pining for fjords.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Use kvm own refcounting instead of playing with ->filp->f_count.
That will allow to get rid of a lot of crap in anon_inode_getfd() and
kill a race in kvm_dev_ioctl_create_vm() (file might have been closed
immediately by another thread, so ->filp might point to already freed
struct file when we get around to setting it).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Avi Kivity <avi@qumranet.com>
So userspace can save/restore the mpstate during migration.
[avi: export the #define constants describing the value]
[christian: add s390 stubs]
[avi: ditto for ia64]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Timers that fire between guest hlt and vcpu_block's add_wait_queue() are
ignored, possibly resulting in hangs.
Also make sure that atomic_inc and waitqueue_active tests happen in the
specified order, otherwise the following race is open:
CPU0 CPU1
if (waitqueue_active(wq))
add_wait_queue()
if (!atomic_read(pit_timer->pending))
schedule()
atomic_inc(pit_timer->pending)
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This interface allows user a space application to read the trace of kvm
related events through relayfs.
Signed-off-by: Feng (Eric) Liu <eric.e.liu@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch introduces a gfn_to_pfn() function and corresponding functions like
kvm_release_pfn_dirty(). Using these new functions, we can modify the x86
MMU to no longer assume that it can always get a struct page for any given gfn.
We don't want to eliminate gfn_to_page() entirely because a number of places
assume they can do gfn_to_page() and then kmap() the results. When we support
IO memory, gfn_to_page() will fail for IO pages although gfn_to_pfn() will
succeed.
This does not implement support for avoiding reference counting for reserved
RAM or for IO memory. However, it should make those things pretty straight
forward.
Since we're only introducing new common symbols, I don't think it will break
the non-x86 architectures but I haven't tested those. I've tested Intel,
AMD, NPT, and hugetlbfs with Windows and Linux guests.
[avi: fix overflow when shifting left pfns by adding casts]
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
the main purpose of adding this functions is the abilaty to release the
spinlock that protect the kvm list while still be able to do operations
on a specific kvm in a safe way.
Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Since the size of kvm_regs is too big to allocate from kernel stack on ia64,
use kzalloc to allocate it.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Create large pages mappings if the guest PTE's are marked as such and
the underlying memory is hugetlbfs backed. If the largepage contains
write-protected pages, a large pte is not used.
Gives a consistent 2% improvement for data copies on ram mounted
filesystem, without NPT/EPT.
Anthony measures a 4% improvement on 4-way kernbench, with NPT.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Mark zapped root pagetables as invalid and ignore such pages during lookup.
This is a problem with the cr3-target feature, where a zapped root table fools
the faulting code into creating a read-only mapping. The result is a lockup
if the instruction can't be emulated.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
With CONFIG_PREEMPT=n, this is needed in order to disable the fault-in
code from sleeping.
Signed-off-by: Andrea Arcangeli <andrea@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch replaces the mmap_sem lock for the memory slots with a new
kvm private lock, it is needed beacuse untill now there were cases where
kvm accesses user memory while holding the mmap semaphore.
Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Sometimes simple attributes might need to return an error, e.g. for
acquiring a mutex interruptibly. In fact we have that situation in
spufs already which is the original user of the simple attributes. This
patch merged the temporarily forked attributes in spufs back into the
main ones and allows to return errors.
[akpm@linux-foundation.org: build fix]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: <stefano.brivio@polimi.it>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Greg KH <greg@kroah.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Convert the synchronization of the shadow handling to a separate mmu_lock
spinlock.
Also guard fetch() by mmap_sem in read-mode to protect against alias
and memslot changes.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
In preparation for a mmu spinlock, add kvm_read_guest_atomic()
and use it in fetch() and prefetch_page().
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Do not hold kvm->lock mutex across the entire pagefault code,
only acquire it in places where it is necessary, such as mmu
hash list, active list, rmap and parent pte handling.
Allow concurrent guest walkers by switching walk_addr() to use
mmap_sem in read-mode.
And get rid of the lockless __gfn_to_page.
[avi: move kvm_mmu_pte_write() locking inside the function]
[avi: add locking for real mode]
[avi: fix cmpxchg locking]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Gregory Haskins
