Commit Graph

802 Commits

Author SHA1 Message Date
Linus Torvalds
b9085bcbf5 Fairly small update, but there are some interesting new features.
Common: Optional support for adding a small amount of polling on each HLT
 instruction executed in the guest (or equivalent for other architectures).
 This can improve latency up to 50% on some scenarios (e.g. O_DSYNC writes
 or TCP_RR netperf tests).  This also has to be enabled manually for now,
 but the plan is to auto-tune this in the future.
 
 ARM/ARM64: the highlights are support for GICv3 emulation and dirty page
 tracking
 
 s390: several optimizations and bugfixes.  Also a first: a feature
 exposed by KVM (UUID and long guest name in /proc/sysinfo) before
 it is available in IBM's hypervisor! :)
 
 MIPS: Bugfixes.
 
 x86: Support for PML (page modification logging, a new feature in
 Broadwell Xeons that speeds up dirty page tracking), nested virtualization
 improvements (nested APICv---a nice optimization), usual round of emulation
 fixes.  There is also a new option to reduce latency of the TSC deadline
 timer in the guest; this needs to be tuned manually.
 
 Some commits are common between this pull and Catalin's; I see you
 have already included his tree.
 
 ARM has other conflicts where functions are added in the same place
 by 3.19-rc and 3.20 patches.  These are not large though, and entirely
 within KVM.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.22 (GNU/Linux)
 
 iQEcBAABAgAGBQJU28rkAAoJEL/70l94x66DXqQH/1TDOfJIjW7P2kb0Sw7Fy1wi
 cEX1KO/VFxAqc8R0E/0Wb55CXyPjQJM6xBXuFr5cUDaIjQ8ULSktL4pEwXyyv/s5
 DBDkN65mriry2w5VuEaRLVcuX9Wy+tqLQXWNkEySfyb4uhZChWWHvKEcgw5SqCyg
 NlpeHurYESIoNyov3jWqvBjr4OmaQENyv7t2c6q5ErIgG02V+iCux5QGbphM2IC9
 LFtPKxoqhfeB2xFxTOIt8HJiXrZNwflsTejIlCl/NSEiDVLLxxHCxK2tWK/tUXMn
 JfLD9ytXBWtNMwInvtFm4fPmDouv2VDyR0xnK2db+/axsJZnbxqjGu1um4Dqbak=
 =7gdx
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM update from Paolo Bonzini:
 "Fairly small update, but there are some interesting new features.

  Common:
     Optional support for adding a small amount of polling on each HLT
     instruction executed in the guest (or equivalent for other
     architectures).  This can improve latency up to 50% on some
     scenarios (e.g. O_DSYNC writes or TCP_RR netperf tests).  This
     also has to be enabled manually for now, but the plan is to
     auto-tune this in the future.

  ARM/ARM64:
     The highlights are support for GICv3 emulation and dirty page
     tracking

  s390:
     Several optimizations and bugfixes.  Also a first: a feature
     exposed by KVM (UUID and long guest name in /proc/sysinfo) before
     it is available in IBM's hypervisor! :)

  MIPS:
     Bugfixes.

  x86:
     Support for PML (page modification logging, a new feature in
     Broadwell Xeons that speeds up dirty page tracking), nested
     virtualization improvements (nested APICv---a nice optimization),
     usual round of emulation fixes.

     There is also a new option to reduce latency of the TSC deadline
     timer in the guest; this needs to be tuned manually.

     Some commits are common between this pull and Catalin's; I see you
     have already included his tree.

  Powerpc:
     Nothing yet.

     The KVM/PPC changes will come in through the PPC maintainers,
     because I haven't received them yet and I might end up being
     offline for some part of next week"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (130 commits)
  KVM: ia64: drop kvm.h from installed user headers
  KVM: x86: fix build with !CONFIG_SMP
  KVM: x86: emulate: correct page fault error code for NoWrite instructions
  KVM: Disable compat ioctl for s390
  KVM: s390: add cpu model support
  KVM: s390: use facilities and cpu_id per KVM
  KVM: s390/CPACF: Choose crypto control block format
  s390/kernel: Update /proc/sysinfo file with Extended Name and UUID
  KVM: s390: reenable LPP facility
  KVM: s390: floating irqs: fix user triggerable endless loop
  kvm: add halt_poll_ns module parameter
  kvm: remove KVM_MMIO_SIZE
  KVM: MIPS: Don't leak FPU/DSP to guest
  KVM: MIPS: Disable HTW while in guest
  KVM: nVMX: Enable nested posted interrupt processing
  KVM: nVMX: Enable nested virtual interrupt delivery
  KVM: nVMX: Enable nested apic register virtualization
  KVM: nVMX: Make nested control MSRs per-cpu
  KVM: nVMX: Enable nested virtualize x2apic mode
  KVM: nVMX: Prepare for using hardware MSR bitmap
  ...
2015-02-13 09:55:09 -08:00
Andrea Arcangeli
0664e57ff0 mm: gup: kvm use get_user_pages_unlocked
Use the more generic get_user_pages_unlocked which has the additional
benefit of passing FAULT_FLAG_ALLOW_RETRY at the very first page fault
(which allows the first page fault in an unmapped area to be always able
to block indefinitely by being allowed to release the mmap_sem).

Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: Andres Lagar-Cavilla <andreslc@google.com>
Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Peter Feiner <pfeiner@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-11 17:06:05 -08:00
Christian Borntraeger
de8e5d7440 KVM: Disable compat ioctl for s390
We never had a 31bit QEMU/kuli running. We would need to review several
ioctls to check if this creates holes, bugs or whatever to make it work.
Lets just disable compat support for KVM on s390.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-09 12:44:14 +01:00
Paolo Bonzini
f781951299 kvm: add halt_poll_ns module parameter
This patch introduces a new module parameter for the KVM module; when it
is present, KVM attempts a bit of polling on every HLT before scheduling
itself out via kvm_vcpu_block.

This parameter helps a lot for latency-bound workloads---in particular
I tested it with O_DSYNC writes with a battery-backed disk in the host.
In this case, writes are fast (because the data doesn't have to go all
the way to the platters) but they cannot be merged by either the host or
the guest.  KVM's performance here is usually around 30% of bare metal,
or 50% if you use cache=directsync or cache=writethrough (these
parameters avoid that the guest sends pointless flush requests, and
at the same time they are not slow because of the battery-backed cache).
The bad performance happens because on every halt the host CPU decides
to halt itself too.  When the interrupt comes, the vCPU thread is then
migrated to a new physical CPU, and in general the latency is horrible
because the vCPU thread has to be scheduled back in.

With this patch performance reaches 60-65% of bare metal and, more
important, 99% of what you get if you use idle=poll in the guest.  This
means that the tunable gets rid of this particular bottleneck, and more
work can be done to improve performance in the kernel or QEMU.

Of course there is some price to pay; every time an otherwise idle vCPUs
is interrupted by an interrupt, it will poll unnecessarily and thus
impose a little load on the host.  The above results were obtained with
a mostly random value of the parameter (500000), and the load was around
1.5-2.5% CPU usage on one of the host's core for each idle guest vCPU.

The patch also adds a new stat, /sys/kernel/debug/kvm/halt_successful_poll,
that can be used to tune the parameter.  It counts how many HLT
instructions received an interrupt during the polling period; each
successful poll avoids that Linux schedules the VCPU thread out and back
in, and may also avoid a likely trip to C1 and back for the physical CPU.

While the VM is idle, a Linux 4 VCPU VM halts around 10 times per second.
Of these halts, almost all are failed polls.  During the benchmark,
instead, basically all halts end within the polling period, except a more
or less constant stream of 50 per second coming from vCPUs that are not
running the benchmark.  The wasted time is thus very low.  Things may
be slightly different for Windows VMs, which have a ~10 ms timer tick.

The effect is also visible on Marcelo's recently-introduced latency
test for the TSC deadline timer.  Though of course a non-RT kernel has
awful latency bounds, the latency of the timer is around 8000-10000 clock
cycles compared to 20000-120000 without setting halt_poll_ns.  For the TSC
deadline timer, thus, the effect is both a smaller average latency and
a smaller variance.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-06 13:08:37 +01:00
Kai Huang
3b0f1d01e5 KVM: Rename kvm_arch_mmu_write_protect_pt_masked to be more generic for log dirty
We don't have to write protect guest memory for dirty logging if architecture
supports hardware dirty logging, such as PML on VMX, so rename it to be more
generic.

Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-01-29 15:30:38 +01:00
Tiejun Chen
b0165f1b41 kvm: update_memslots: clean flags for invalid memslots
Indeed, any invalid memslots should be new->npages = 0,
new->base_gfn = 0 and new->flags = 0 at the same time.

Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-01-27 21:31:44 +01:00
Christoffer Dall
4b99058995 KVM: Remove unused config symbol
The dirty patch logging series introduced both
HAVE_KVM_ARCH_DIRTY_LOG_PROTECT and KVM_GENERIC_DIRTYLOG_READ_PROTECT
config symbols, but only KVM_GENERIC_DIRTYLOG_READ_PROTECT is used.
Just remove the unused one.

(The config symbol was renamed during the development of the patch
series and the old name just creeped in by accident.()

Reported-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-23 10:52:03 +01:00
Andre Przywara
4fa96afd94 arm/arm64: KVM: force alignment of VGIC dist/CPU/redist addresses
Although the GIC architecture requires us to map the MMIO regions
only at page aligned addresses, we currently do not enforce this from
the kernel side.
Restrict any vGICv2 regions to be 4K aligned and any GICv3 regions
to be 64K aligned. Document this requirement.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:33 +01:00
Andre Przywara
ac3d373564 arm/arm64: KVM: allow userland to request a virtual GICv3
With all of the GICv3 code in place now we allow userland to ask the
kernel for using a virtual GICv3 in the guest.
Also we provide the necessary support for guests setting the memory
addresses for the virtual distributor and redistributors.
This requires some userland code to make use of that feature and
explicitly ask for a virtual GICv3.
Document that KVM_CREATE_IRQCHIP only works for GICv2, but is
considered legacy and using KVM_CREATE_DEVICE is preferred.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:33 +01:00
Andre Przywara
b5d84ff600 arm/arm64: KVM: enable kernel side of GICv3 emulation
With all the necessary GICv3 emulation code in place, we can now
connect the code to the GICv3 backend in the kernel.
The LR register handling is different depending on the emulated GIC
model, so provide different implementations for each.
Also allow non-v2-compatible GICv3 implementations (which don't
provide MMIO regions for the virtual CPU interface in the DT), but
restrict those hosts to support GICv3 guests only.
If the device tree provides a GICv2 compatible GICV resource entry,
but that one is faulty, just disable the GICv2 emulation and let the
user use at least the GICv3 emulation for guests.
To provide proper support for the legacy KVM_CREATE_IRQCHIP ioctl,
note virtual GICv2 compatibility in struct vgic_params and use it
on creating a VGICv2.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:32 +01:00
Andre Przywara
6d52f35af1 arm64: KVM: add SGI generation register emulation
While the generation of a (virtual) inter-processor interrupt (SGI)
on a GICv2 works by writing to a MMIO register, GICv3 uses the system
register ICC_SGI1R_EL1 to trigger them.
Add a trap handler function that calls the new SGI register handler
in the GICv3 code. As ICC_SRE_EL1.SRE at this point is still always 0,
this will not trap yet, but will only be used later when all the data
structures have been initialized properly.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:32 +01:00
Andre Przywara
a0675c25d6 arm/arm64: KVM: add virtual GICv3 distributor emulation
With everything separated and prepared, we implement a model of a
GICv3 distributor and redistributors by using the existing framework
to provide handler functions for each register group.

Currently we limit the emulation to a model enforcing a single
security state, with SRE==1 (forcing system register access) and
ARE==1 (allowing more than 8 VCPUs).

We share some of the functions provided for GICv2 emulation, but take
the different ways of addressing (v)CPUs into account.
Save and restore is currently not implemented.

Similar to the split-off of the GICv2 specific code, the new emulation
code goes into a new file (vgic-v3-emul.c).

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:31 +01:00
Andre Przywara
9fedf14677 arm/arm64: KVM: add opaque private pointer to MMIO data
For a GICv2 there is always only one (v)CPU involved: the one that
does the access. On a GICv3 the access to a CPU redistributor is
memory-mapped, but not banked, so the (v)CPU affected is determined by
looking at the MMIO address region being accessed.
To allow passing the affected CPU into the accessors later, extend
struct kvm_exit_mmio to add an opaque private pointer parameter.
The current GICv2 emulation just does not use it.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:30 +01:00
Andre Przywara
1d916229e3 arm/arm64: KVM: split GICv2 specific emulation code from vgic.c
vgic.c is currently a mixture of generic vGIC emulation code and
functions specific to emulating a GICv2. To ease the addition of
GICv3, split off strictly v2 specific parts into a new file
vgic-v2-emul.c.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>

-------
As the diff isn't always obvious here (and to aid eventual rebases),
here is a list of high-level changes done to the code:
* added new file to respective arm/arm64 Makefiles
* moved GICv2 specific functions to vgic-v2-emul.c:
  - handle_mmio_misc()
  - handle_mmio_set_enable_reg()
  - handle_mmio_clear_enable_reg()
  - handle_mmio_set_pending_reg()
  - handle_mmio_clear_pending_reg()
  - handle_mmio_priority_reg()
  - vgic_get_target_reg()
  - vgic_set_target_reg()
  - handle_mmio_target_reg()
  - handle_mmio_cfg_reg()
  - handle_mmio_sgi_reg()
  - vgic_v2_unqueue_sgi()
  - read_set_clear_sgi_pend_reg()
  - write_set_clear_sgi_pend_reg()
  - handle_mmio_sgi_set()
  - handle_mmio_sgi_clear()
  - vgic_v2_handle_mmio()
  - vgic_get_sgi_sources()
  - vgic_dispatch_sgi()
  - vgic_v2_queue_sgi()
  - vgic_v2_map_resources()
  - vgic_v2_init()
  - vgic_v2_add_sgi_source()
  - vgic_v2_init_model()
  - vgic_v2_init_emulation()
  - handle_cpu_mmio_misc()
  - handle_mmio_abpr()
  - handle_cpu_mmio_ident()
  - vgic_attr_regs_access()
  - vgic_create() (renamed to vgic_v2_create())
  - vgic_destroy() (renamed to vgic_v2_destroy())
  - vgic_has_attr() (renamed to vgic_v2_has_attr())
  - vgic_set_attr() (renamed to vgic_v2_set_attr())
  - vgic_get_attr() (renamed to vgic_v2_get_attr())
  - struct kvm_mmio_range vgic_dist_ranges[]
  - struct kvm_mmio_range vgic_cpu_ranges[]
  - struct kvm_device_ops kvm_arm_vgic_v2_ops {}

Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:30 +01:00
Andre Przywara
832158125d arm/arm64: KVM: add vgic.h header file
vgic.c is currently a mixture of generic vGIC emulation code and
functions specific to emulating a GICv2. To ease the addition of
GICv3 later, we create new header file vgic.h, which holds constants
and prototypes of commonly used functions.
Rename some identifiers to avoid name space clutter.
I removed the long-standing comment about using the kvm_io_bus API
to tackle the GIC register ranges, as it wouldn't be a win for us
anymore.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>

-------
As the diff isn't always obvious here (and to aid eventual rebases),
here is a list of high-level changes done to the code:
* moved definitions and prototypes from vgic.c to vgic.h:
  - VGIC_ADDR_UNDEF
  - ACCESS_{READ,WRITE}_*
  - vgic_init()
  - vgic_update_state()
  - vgic_kick_vcpus()
  - vgic_get_vmcr()
  - vgic_set_vmcr()
  - struct mmio_range {} (renamed to struct kvm_mmio_range)
* removed static keyword and exported prototype in vgic.h:
  - vgic_bitmap_get_reg()
  - vgic_bitmap_set_irq_val()
  - vgic_bitmap_get_shared_map()
  - vgic_bytemap_get_reg()
  - vgic_dist_irq_set_pending()
  - vgic_dist_irq_clear_pending()
  - vgic_cpu_irq_clear()
  - vgic_reg_access()
  - handle_mmio_raz_wi()
  - vgic_handle_enable_reg()
  - vgic_handle_set_pending_reg()
  - vgic_handle_clear_pending_reg()
  - vgic_handle_cfg_reg()
  - vgic_unqueue_irqs()
  - find_matching_range() (renamed to vgic_find_range)
  - vgic_handle_mmio_range()
  - vgic_update_state()
  - vgic_get_vmcr()
  - vgic_set_vmcr()
  - vgic_queue_irq()
  - vgic_kick_vcpus()
  - vgic_init()
  - vgic_v2_init_emulation()
  - vgic_has_attr_regs()
  - vgic_set_common_attr()
  - vgic_get_common_attr()
  - vgic_destroy()
  - vgic_create()
* moved functions to vgic.h (static inline):
  - mmio_data_read()
  - mmio_data_write()
  - is_in_range()

Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:30 +01:00
Andre Przywara
b60da146c1 arm/arm64: KVM: refactor/wrap vgic_set/get_attr()
vgic_set_attr() and vgic_get_attr() contain both code specific for
the emulated GIC as well as code for the userland facing, generic
part of the GIC.
Split the guest GIC facing code of from the generic part to allow
easier splitting later.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:29 +01:00
Andre Przywara
d97f683d0f arm/arm64: KVM: refactor MMIO accessors
The MMIO accessors for GICD_I[CS]ENABLER, GICD_I[CS]PENDR and
GICD_ICFGR behave very similar for GICv2 and GICv3, although the way
the affected VCPU is determined differs.
Since we need them to access the registers from three different
places in the future, we factor out a generic, backend-facing
implementation and use small wrappers in the current GICv2 emulation.
This will ease adding GICv3 accessors later.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:29 +01:00
Andre Przywara
2f5fa41a7a arm/arm64: KVM: make the value of ICC_SRE_EL1 a per-VM variable
ICC_SRE_EL1 is a system register allowing msr/mrs accesses to the
GIC CPU interface for EL1 (guests). Currently we force it to 0, but
for proper GICv3 support we have to allow guests to use it (depending
on their selected virtual GIC model).
So add ICC_SRE_EL1 to the list of saved/restored registers on a
world switch, but actually disallow a guest to change it by only
restoring a fixed, once-initialized value.
This value depends on the GIC model userland has chosen for a guest.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:28 +01:00
Andre Przywara
3caa2d8c3b arm/arm64: KVM: make the maximum number of vCPUs a per-VM value
Currently the maximum number of vCPUs supported is a global value
limited by the used GIC model. GICv3 will lift this limit, but we
still need to observe it for guests using GICv2.
So the maximum number of vCPUs is per-VM value, depending on the
GIC model the guest uses.
Store and check the value in struct kvm_arch, but keep it down to
8 for now.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:28 +01:00
Andre Przywara
4ce7ebdfc6 arm/arm64: KVM: dont rely on a valid GICH base address
To check whether the vGIC was already initialized, we currently check
the GICH base address for not being NULL. Since with GICv3 we may
get along without this address, lets use the irqchip_in_kernel()
function to detect an already initialized vGIC.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:27 +01:00
Andre Przywara
ea2f83a7de arm/arm64: KVM: move kvm_register_device_ops() into vGIC probing
Currently we unconditionally register the GICv2 emulation device
during the host's KVM initialization. Since with GICv3 support we
may end up with only v2 or only v3 or both supported, we move the
registration into the GIC probing function, where we will later know
which combination is valid.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:27 +01:00
Andre Przywara
b26e5fdac4 arm/arm64: KVM: introduce per-VM ops
Currently we only have one virtual GIC model supported, so all guests
use the same emulation code. With the addition of another model we
end up with different guests using potentially different vGIC models,
so we have to split up some functions to be per VM.
Introduce a vgic_vm_ops struct to hold function pointers for those
functions that are different and provide the necessary code to
initialize them.
Also split up the vgic_init() function to separate out VGIC model
specific functionality into a separate function, which will later be
different for a GICv3 model.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:26 +01:00
Andre Przywara
05bc8aafe6 arm/arm64: KVM: wrap 64 bit MMIO accesses with two 32 bit ones
Some GICv3 registers can and will be accessed as 64 bit registers.
Currently the register handling code can only deal with 32 bit
accesses, so we do two consecutive calls to cover this.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:26 +01:00
Andre Przywara
96415257a1 arm/arm64: KVM: refactor vgic_handle_mmio() function
Currently we only need to deal with one MMIO region for the GIC
emulation (the GICv2 distributor), but we soon need to extend this.
Refactor the existing code to allow easier addition of different
ranges without code duplication.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:25 +01:00
Andre Przywara
59892136c4 arm/arm64: KVM: pass down user space provided GIC type into vGIC code
With the introduction of a second emulated GIC model we need to let
userspace specify the GIC model to use for each VM. Pass the
userspace provided value down into the vGIC code and store it there
to differentiate later.

Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-20 18:25:25 +01:00
Mario Smarduch
ba0513b5b8 KVM: Add generic support for dirty page logging
kvm_get_dirty_log() provides generic handling of dirty bitmap, currently reused
by several architectures. Building on that we intrdoduce
kvm_get_dirty_log_protect() adding write protection to mark these pages dirty
for future write access, before next KVM_GET_DIRTY_LOG ioctl call from user
space.

Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Mario Smarduch <m.smarduch@samsung.com>
2015-01-16 14:40:14 +01:00
Mario Smarduch
a6d5101661 KVM: Add architecture-defined TLB flush support
Allow architectures to override the generic kvm_flush_remote_tlbs()
function via HAVE_KVM_ARCH_TLB_FLUSH_ALL. ARMv7 will need this to
provide its own TLB flush interface.

Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Mario Smarduch <m.smarduch@samsung.com>
2015-01-16 14:40:14 +01:00
Eric Auger
065c003482 KVM: arm/arm64: vgic: add init entry to VGIC KVM device
Since the advent of VGIC dynamic initialization, this latter is
initialized quite late on the first vcpu run or "on-demand", when
injecting an IRQ or when the guest sets its registers.

This initialization could be initiated explicitly much earlier
by the users-space, as soon as it has provided the requested
dimensioning parameters.

This patch adds a new entry to the VGIC KVM device that allows
the user to manually request the VGIC init:
- a new KVM_DEV_ARM_VGIC_GRP_CTRL group is introduced.
- Its first attribute is KVM_DEV_ARM_VGIC_CTRL_INIT

The rationale behind introducing a group is to be able to add other
controls later on, if needed.

Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-11 14:12:15 +01:00
Eric Auger
66b030e48a KVM: arm/arm64: vgic: vgic_init returns -ENODEV when no online vcpu
To be more explicit on vgic initialization failure, -ENODEV is
returned by vgic_init when no online vcpus can be found at init.

Signed-off-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2015-01-11 14:12:15 +01:00
Wincy Van
ff651cb613 KVM: nVMX: Add nested msr load/restore algorithm
Several hypervisors need MSR auto load/restore feature.
We read MSRs from VM-entry MSR load area which specified by L1,
and load them via kvm_set_msr in the nested entry.
When nested exit occurs, we get MSRs via kvm_get_msr, writing
them to L1`s MSR store area. After this, we read MSRs from VM-exit
MSR load area, and load them via kvm_set_msr.

Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-01-08 22:45:14 +01:00
Richard Cochran
2eebdde652 timecounter: keep track of accumulated fractional nanoseconds
The current timecounter implementation will drop a variable amount
of resolution, depending on the magnitude of the time delta. In
other words, reading the clock too often or too close to a time
stamp conversion will introduce errors into the time values. This
patch fixes the issue by introducing a fractional nanosecond field
that accumulates the low order bits.

Reported-by: Janusz Użycki <j.uzycki@elproma.com.pl>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-30 18:29:27 -05:00
Paolo Bonzini
dbaff30940 kvm: warn on more invariant breakage
Modifying a non-existent slot is not allowed.  Also check that the
first loop doesn't move a deleted slot beyond the used part of
the mslots array.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-28 10:01:25 +01:00
Paolo Bonzini
efbeec7098 kvm: fix sorting of memslots with base_gfn == 0
Before commit 0e60b0799f (kvm: change memslot sorting rule from size
to GFN, 2014-12-01), the memslots' sorting key was npages, meaning
that a valid memslot couldn't have its sorting key equal to zero.
On the other hand, a valid memslot can have base_gfn == 0, and invalid
memslots are identified by base_gfn == npages == 0.

Because of this, commit 0e60b0799f broke the invariant that invalid
memslots are at the end of the mslots array.  When a memslot with
base_gfn == 0 was created, any invalid memslot before it were left
in place.

This can be fixed by changing the insertion to use a ">=" comparison
instead of "<=", but some care is needed to avoid breaking the case
of deleting a memslot; see the comment in update_memslots.

Thanks to Tiejun Chen for posting an initial patch for this bug.

Reported-by: Jamie Heilman <jamie@audible.transient.net>
Reported-by: Andy Lutomirski <luto@amacapital.net>
Tested-by: Jamie Heilman <jamie@audible.transient.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-28 10:01:17 +01:00
Linus Torvalds
66dcff86ba 3.19 changes for KVM:
- spring cleaning: removed support for IA64, and for hardware-assisted
 virtualization on the PPC970
 - ARM, PPC, s390 all had only small fixes
 
 For x86:
 - small performance improvements (though only on weird guests)
 - usual round of hardware-compliancy fixes from Nadav
 - APICv fixes
 - XSAVES support for hosts and guests.  XSAVES hosts were broken because
 the (non-KVM) XSAVES patches inadvertently changed the KVM userspace
 ABI whenever XSAVES was enabled; hence, this part is going to stable.
 Guest support is just a matter of exposing the feature and CPUID leaves
 support.
 
 Right now KVM is broken for PPC BookE in your tree (doesn't compile).
 I'll reply to the pull request with a patch, please apply it either
 before the pull request or in the merge commit, in order to preserve
 bisectability somewhat.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.22 (GNU/Linux)
 
 iQEcBAABAgAGBQJUkpg+AAoJEL/70l94x66DUmoH/jzXYkptSW9NGgm79KqxGJlD
 lzLnLBkitVvx++Mz5YBhdJEhKKLUlCtifFT1zPJQ/pthQhIRSaaAwZyNGgUs5w5x
 yMGKHiPQFyZRbmQtZhCInW0BftJoYHHciO3nUfHCZnp34My9MP2D55W7/z+fYFfQ
 DuqBSE9ThyZJtZ4zh8NRA9fCOeuqwVYRyoBs820Wbsh4cpIBoIK63Dg7k+CLE+ZV
 MZa/mRL6bAfsn9W5bnOUAgHJ3SPznnWbO3/g0aV+roL/5pffblprJx9lKNR08xUM
 6hDFLop2gDehDJesDkY/o8Ckp1hEouvfsVpSShry4vcgtn0hgh2O5/6Orbmj6vE=
 =Zwq1
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM update from Paolo Bonzini:
 "3.19 changes for KVM:

   - spring cleaning: removed support for IA64, and for hardware-
     assisted virtualization on the PPC970

   - ARM, PPC, s390 all had only small fixes

  For x86:
   - small performance improvements (though only on weird guests)
   - usual round of hardware-compliancy fixes from Nadav
   - APICv fixes
   - XSAVES support for hosts and guests.  XSAVES hosts were broken
     because the (non-KVM) XSAVES patches inadvertently changed the KVM
     userspace ABI whenever XSAVES was enabled; hence, this part is
     going to stable.  Guest support is just a matter of exposing the
     feature and CPUID leaves support"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (179 commits)
  KVM: move APIC types to arch/x86/
  KVM: PPC: Book3S: Enable in-kernel XICS emulation by default
  KVM: PPC: Book3S HV: Improve H_CONFER implementation
  KVM: PPC: Book3S HV: Fix endianness of instruction obtained from HEIR register
  KVM: PPC: Book3S HV: Remove code for PPC970 processors
  KVM: PPC: Book3S HV: Tracepoints for KVM HV guest interactions
  KVM: PPC: Book3S HV: Simplify locking around stolen time calculations
  arch: powerpc: kvm: book3s_paired_singles.c: Remove unused function
  arch: powerpc: kvm: book3s_pr.c: Remove unused function
  arch: powerpc: kvm: book3s.c: Remove some unused functions
  arch: powerpc: kvm: book3s_32_mmu.c: Remove unused function
  KVM: PPC: Book3S HV: Check wait conditions before sleeping in kvmppc_vcore_blocked
  KVM: PPC: Book3S HV: ptes are big endian
  KVM: PPC: Book3S HV: Fix inaccuracies in ICP emulation for H_IPI
  KVM: PPC: Book3S HV: Fix KSM memory corruption
  KVM: PPC: Book3S HV: Fix an issue where guest is paused on receiving HMI
  KVM: PPC: Book3S HV: Fix computation of tlbie operand
  KVM: PPC: Book3S HV: Add missing HPTE unlock
  KVM: PPC: BookE: Improve irq inject tracepoint
  arm/arm64: KVM: Require in-kernel vgic for the arch timers
  ...
2014-12-18 16:05:28 -08:00
Paolo Bonzini
333bce5aac Second round of changes for KVM for arm/arm64 for v3.19; fixes reboot
problems, clarifies VCPU init, and fixes a regression concerning the
 VGIC init flow.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJUjsVhAAoJEEtpOizt6ddy5rIH/1V/YVwhprC55YqdHelU9Qu2
 Muzsx+7F71NxC7xgMGFqPD1YrPR+hxvoPhy+ADOBlvcqlolrkDnV9I+8e3geaYNc
 nZ/yEnoGTtbAggiS1smx7usBv34Z88Sd5txNjmj1cmHBy+VOWlyidWMkGBTsfBRe
 mVc61BDUfyC47udgRHXhwS80sbHLJHElmADisFOVmQNBYwwiHiTdx0hMBMnHcC3Y
 /3T0tKxHdeTISnmA+J+n7TcChtTIM4xqC6kwf3rw3b7XX8gdtTKylDHX2GLAg646
 RdebAG2twmGpIc6SxXZbo38f3oY9OFo1Le5xZGa6iUjD56VDw/e4wg4iA2juo0Y=
 =J2Ut
 -----END PGP SIGNATURE-----

Merge tag 'kvm-arm-for-3.19-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD

Second round of changes for KVM for arm/arm64 for v3.19; fixes reboot
problems, clarifies VCPU init, and fixes a regression concerning the
VGIC init flow.

Conflicts:
	arch/ia64/kvm/kvm-ia64.c [deleted in HEAD and modified in kvmarm]
2014-12-15 13:06:40 +01:00
Christoffer Dall
05971120fc arm/arm64: KVM: Require in-kernel vgic for the arch timers
It is curently possible to run a VM with architected timers support
without creating an in-kernel VGIC, which will result in interrupts from
the virtual timer going nowhere.

To address this issue, move the architected timers initialization to the
time when we run a VCPU for the first time, and then only initialize
(and enable) the architected timers if we have a properly created and
initialized in-kernel VGIC.

When injecting interrupts from the virtual timer to the vgic, the
current setup should ensure that this never calls an on-demand init of
the VGIC, which is the only call path that could return an error from
kvm_vgic_inject_irq(), so capture the return value and raise a warning
if there's an error there.

We also change the kvm_timer_init() function from returning an int to be
a void function, since the function always succeeds.

Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2014-12-15 11:50:42 +01:00
Christoffer Dall
ca7d9c829d arm/arm64: KVM: Initialize the vgic on-demand when injecting IRQs
Userspace assumes that it can wire up IRQ injections after having
created all VCPUs and after having created the VGIC, but potentially
before starting the first VCPU.  This can currently lead to lost IRQs
because the state of that IRQ injection is not stored anywhere and we
don't return an error to userspace.

We haven't seen this problem manifest itself yet, presumably because
guests reset the devices on boot, but this could cause issues with
migration and other non-standard startup configurations.

Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2014-12-15 11:36:21 +01:00
Christoffer Dall
1f57be2895 arm/arm64: KVM: Add (new) vgic_initialized macro
Some code paths will need to check to see if the internal state of the
vgic has been initialized (such as when creating new VCPUs), so
introduce such a macro that checks the nr_cpus field which is set when
the vgic has been initialized.

Also set nr_cpus = 0 in kvm_vgic_destroy, because the error path in
vgic_init() will call this function, and code should never errornously
assume the vgic to be properly initialized after an error.

Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2014-12-13 14:17:10 +01:00
Christoffer Dall
c52edf5f8c arm/arm64: KVM: Rename vgic_initialized to vgic_ready
The vgic_initialized() macro currently returns the state of the
vgic->ready flag, which indicates if the vgic is ready to be used when
running a VM, not specifically if its internal state has been
initialized.

Rename the macro accordingly in preparation for a more nuanced
initialization flow.

Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2014-12-13 14:17:05 +01:00
Peter Maydell
6d3cfbe21b arm/arm64: KVM: vgic: move reset initialization into vgic_init_maps()
VGIC initialization currently happens in three phases:
 (1) kvm_vgic_create() (triggered by userspace GIC creation)
 (2) vgic_init_maps() (triggered by userspace GIC register read/write
     requests, or from kvm_vgic_init() if not already run)
 (3) kvm_vgic_init() (triggered by first VM run)

We were doing initialization of some state to correspond with the
state of a freshly-reset GIC in kvm_vgic_init(); this is too late,
since it will overwrite changes made by userspace using the
register access APIs before the VM is run. Move this initialization
earlier, into the vgic_init_maps() phase.

This fixes a bug where QEMU could successfully restore a saved
VM state snapshot into a VM that had already been run, but could
not restore it "from cold" using the -loadvm command line option
(the symptoms being that the restored VM would run but interrupts
were ignored).

Finally rename vgic_init_maps to vgic_init and renamed kvm_vgic_init to
kvm_vgic_map_resources.

  [ This patch is originally written by Peter Maydell, but I have
    modified it somewhat heavily, renaming various bits and moving code
    around.  If something is broken, I am to be blamed. - Christoffer ]

Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
2014-12-13 14:15:52 +01:00
Christian Borntraeger
7a72f7a140 KVM: track pid for VCPU only on KVM_RUN ioctl
We currently track the pid of the task that runs the VCPU in vcpu_load.
If a yield to that VCPU is triggered while the PID of the wrong thread
is active, the wrong thread might receive a yield, but this will most
likely not help the executing thread at all.  Instead, if we only track
the pid on the KVM_RUN ioctl, there are two possibilities:

1) the thread that did a non-KVM_RUN ioctl is holding a mutex that
the VCPU thread is waiting for.  In this case, the VCPU thread is not
runnable, but we also do not do a wrong yield.

2) the thread that did a non-KVM_RUN ioctl is sleeping, or doing
something that does not block the VCPU thread.  In this case, the
VCPU thread can receive the directed yield correctly.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
CC: Rik van Riel <riel@redhat.com>
CC: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
CC: Michael Mueller <mimu@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:13 +01:00
David Hildenbrand
eed6e79d73 KVM: don't check for PF_VCPU when yielding
kvm_enter_guest() has to be called with preemption disabled and will
set PF_VCPU.  Current code takes PF_VCPU as a hint that the VCPU thread
is running and therefore needs no yield.

However, the check on PF_VCPU is wrong on s390, where preemption has
to stay enabled in order to correctly process page faults.  Thus,
s390 reenables preemption and starts to execute the guest.  The thread
might be scheduled out between kvm_enter_guest() and kvm_exit_guest(),
resulting in PF_VCPU being set but not being run.  When this happens,
the opportunity for directed yield is missed.

However, this check is done already in kvm_vcpu_on_spin before calling
kvm_vcpu_yield_loop:

        if (!ACCESS_ONCE(vcpu->preempted))
                continue;

so the check on PF_VCPU is superfluous in general, and this patch
removes it.

Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:12 +01:00
Igor Mammedov
9c1a5d3878 kvm: optimize GFN to memslot lookup with large slots amount
Current linear search doesn't scale well when
large amount of memslots is used and looked up slot
is not in the beginning memslots array.
Taking in account that memslots don't overlap, it's
possible to switch sorting order of memslots array from
'npages' to 'base_gfn' and use binary search for
memslot lookup by GFN.

As result of switching to binary search lookup times
are reduced with large amount of memslots.

Following is a table of search_memslot() cycles
during WS2008R2 guest boot.

                         boot,          boot + ~10 min
                         mostly same    of using it,
                         slot lookup    randomized lookup
                max      average        average
                cycles   cycles         cycles

13 slots      : 1450       28           30

13 slots      : 1400       30           40
binary search

117 slots     : 13000      30           460

117 slots     : 2000       35           180
binary search

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:11 +01:00
Igor Mammedov
0e60b0799f kvm: change memslot sorting rule from size to GFN
it will allow to use binary search for GFN -> memslot
lookups, reducing lookup cost with large slots amount.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:11 +01:00
Igor Mammedov
7f379cff11 kvm: update_memslots: drop not needed check for the same slot
UP/DOWN shift loops will shift array in needed
direction and stop at place where new slot should
be placed regardless of old slot size.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:09 +01:00
Igor Mammedov
5a38b6e6b4 kvm: update_memslots: drop not needed check for the same number of pages
if number of pages haven't changed sorting algorithm
will do nothing, so there is no need to do extra check
to avoid entering sorting logic.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-12-04 15:29:09 +01:00
Ard Biesheuvel
d3fccc7ef8 kvm: fix kvm_is_mmio_pfn() and rename to kvm_is_reserved_pfn()
This reverts commit 85c8555ff0 ("KVM: check for !is_zero_pfn() in
kvm_is_mmio_pfn()") and renames the function to kvm_is_reserved_pfn.

The problem being addressed by the patch above was that some ARM code
based the memory mapping attributes of a pfn on the return value of
kvm_is_mmio_pfn(), whose name indeed suggests that such pfns should
be mapped as device memory.

However, kvm_is_mmio_pfn() doesn't do quite what it says on the tin,
and the existing non-ARM users were already using it in a way which
suggests that its name should probably have been 'kvm_is_reserved_pfn'
from the beginning, e.g., whether or not to call get_page/put_page on
it etc. This means that returning false for the zero page is a mistake
and the patch above should be reverted.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-11-26 14:40:45 +01:00
Christoffer Dall
6b50f54064 arm/arm64: KVM: vgic: Fix error code in kvm_vgic_create()
If we detect another vCPU is running we just exit and return 0 as if we
succesfully created the VGIC, but the VGIC wouldn't actual be created.

This shouldn't break in-kernel behavior because the kernel will not
observe the failed the attempt to create the VGIC, but userspace could
be rightfully confused.

Cc: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-11-26 14:40:44 +01:00
Shannon Zhao
016ed39c54 arm/arm64: KVM: vgic: kick the specific vcpu instead of iterating through all
When call kvm_vgic_inject_irq to inject interrupt, we can known which
vcpu the interrupt for by the irq_num and the cpuid. So we should just
kick this vcpu to avoid iterating through all.

Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Shannon Zhao <zhaoshenglong@huawei.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
2014-11-26 10:19:37 +00:00
Christoffer Dall
b1e952b4e4 arm/arm64: vgic: Remove unreachable irq_clear_pending
When 'injecting' an edge-triggered interrupt with a falling edge we
shouldn't clear the pending state on the distributor.  In fact, we
don't, because the check in vgic_validate_injection would prevent us
from ever reaching this bit of code.

Remove the unreachable snippet.

Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
2014-11-25 13:57:28 +00:00