The only caller now knows exactly which RAMBlock to free, so it's not
necessary to do the lookup.
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1456813104-25902-6-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
All references to mr->ram_addr are replaced by
memory_region_get_ram_addr(mr) (except for a few assertions that are
replaced with mr->ram_block).
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1456813104-25902-5-git-send-email-famz@redhat.com>
Acked-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Previously we return RAMBlock.offset; now return the pointer to the
whole structure.
ram_block_add returns void now, error is completely passed with errp.
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1456813104-25902-2-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
these two functions consume too much cpu overhead to
find the RAMBlock by ram address.
After this patch, we can pass the RAMBlock pointer
to them so that they don't need to find the RAMBlock
anymore most of the time. We can get better performance
in address translation processing.
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Message-Id: <1455935721-8804-3-git-send-email-arei.gonglei@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Each RAM memory region has a unique corresponding RAMBlock.
In the current realization, the memory region only stored
the ram_addr which means the offset of RAM address space,
We need to qurey the global ram.list to find the ram block
by ram_addr if we want to get the ram block, which is very
expensive.
Now, we store the RAMBlock pointer into memory region
structure. So, if we know the mr, we can easily get the
RAMBlock.
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Message-Id: <1456130097-4208-2-git-send-email-arei.gonglei@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fixes all over the place.
virtio dataplane migration support.
Old q35 machine types removed.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJWzuKeAAoJECgfDbjSjVRpGzIH/1Tz6CoEq1rowiyVJ9B80oQU
gDI2YWnJDSwJllmAF0rmoPRBQR8op3ZETZiCAcADHoZ7kdBNWGbyQeaDrrEPH7Q/
rCDVt8Q3g80vs89aWKG0nQ16J2MW5TbkuiQw7pjQSdc9AbUdWpUqSiWnpZ+sPAql
6DuVpjQ4/rN2alucXoa1Sir8KDDV7kBuY8U6/KoY890qzh842dv2523qvuCza9yR
KX8Imj3oQAFjFSv5t1aOD3yYvWFd73EsReHPLGb1JtsVr/6wjs0sFUyA3JicBgnT
+kWoSObWikfDY69HnqTkJpkun6woMM3zW5h2SkUBf9QP3yqLfGIp9uSriNN84Ak=
=KXyh
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
vhost, virtio, pci, pc
Fixes all over the place.
virtio dataplane migration support.
Old q35 machine types removed.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Thu 25 Feb 2016 11:16:46 GMT using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg: aka "Michael S. Tsirkin <mst@redhat.com>"
* remotes/mst/tags/for_upstream: (21 commits)
q35: No need to check gigabyte_align
q35: Remove unused q35-acpi-dsdt.aml file
ich9: Remove enable_tco arguments from init functions
machine: Remove no_tco field
q35: Remove old machine versions
tests/vhost-user-bridge: fix build on 32 bit systems
vring: remove
virtio-scsi: do not use vring in dataplane
virtio-blk: do not use vring in dataplane
virtio-blk: fix "disabled data plane" mode
virtio: export vring_notify as virtio_should_notify
virtio: add AioContext-specific function for host notifiers
vring: make vring_enable_notification return void
block-migration: acquire AioContext as necessary
pci core: function pci_bus_init() cleanup
pci core: function pci_host_bus_register() cleanup
balloon: Use only 'pc-dimm' type dimm for ballooning
virtio-balloon: rewrite get_current_ram_size()
move get_current_ram_size to virtio-balloon.c
vhost-user: don't merge regions with different fds
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
NB: If this commit breaks compilation for your out-of-tree
patchseries or fork, then you need to make sure you add
#include "qemu/osdep.h" to any new .c files that you have.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
get_current_ram_size() is used only in virtio-balloon.c
This patch moves it into virtio-balloon and make it static, to allow
some balloon-specific tuning.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The last two arguments to these functions are the last and first bit to
check relative to the base. The code was using incorrectly the first
bit and the number of bits. Fix this in cpu_physical_memory_get_dirty
and cpu_physical_memory_all_dirty. This requires a few changes in the
iteration; change the code in cpu_physical_memory_set_dirty_range to
match.
Fixes: 5b82b70
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Tested-by: Leon Alrae <leon.alrae@imgtec.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-id: 1455113505-11237-1-git-send-email-pbonzini@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Although accesses to ram_list.dirty_memory[] use atomics so multiple
threads can safely dirty the bitmap, the data structure is not fully
thread-safe yet.
This patch handles the RAM hotplug case where ram_list.dirty_memory[] is
grown. ram_list.dirty_memory[] is change from a regular bitmap to an
RCU array of pointers to fixed-size bitmap blocks. Threads can continue
accessing bitmap blocks while the array is being extended. See the
comments in the code for an in-depth explanation of struct
DirtyMemoryBlocks.
I have tested that live migration with virtio-blk dataplane works.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <1453728801-5398-2-git-send-email-stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This condition is true in the common case, so we can cut out the body of
the function. In addition, this makes it easier for the compiler to do
at least partial inlining, even if it decides that fully inlining the
function is unreasonable.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Split host_from_stream_offset() into two parts:
One is to get ram block, which the block idstr may be get from migration
stream, the other is to get hva (host) address from block and the offset.
Besides, we will do the check working in a new helper offset_in_ramblock().
Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Message-Id: <1452829066-9764-2-git-send-email-zhang.zhanghailiang@huawei.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Split the bits that require it to exec/log.h.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-id: 1452174932-28657-8-git-send-email-den@openvz.org
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
This will either create a new AS or return a pointer to an
already existing equivalent one, if we have already created
an AS for the specified root memory region.
The motivation is to reuse address spaces as much as possible.
It's going to be quite common that bus masters out in device land
have pointers to the same memory region for their mastering yet
each will need to create its own address space. Let the memory
API implement sharing for them.
Aside from the perf optimisations, this should reduce the amount
of redundant output on info mtree as well.
Thee returned value will be malloced, but the malloc will be
automatically freed when the AS runs out of refs.
Signed-off-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
[PMM: dropped check for NULL root as unused; added doc-comment;
squashed Peter C's reference-counting patch into this one;
don't compare name string when deciding if we can share ASes;
read as->malloced before the unref of as->root to avoid possible
read-after-free if as->root was the owner of as]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Add a function to return the AddressSpace for a CPU based on
its numerical index. (Callers outside exec.c don't have access
to the CPUAddressSpace struct so can't just fish it out of the
CPUState struct directly.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Pass the MemTxAttrs for the memory access to iotlb_to_region(); this
allows it to determine the correct AddressSpace to use for the lookup.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
When looking up the MemoryRegionSection for the new TLB entry in
tlb_set_page_with_attrs(), use cpu_asidx_from_attrs() to determine
the correct address space index for the lookup, and pass it into
address_space_translate_for_iotlb().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Add documentation comments for tlb_set_page_with_attrs()
and tlb_set_page().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Allow multiple calls to cpu_address_space_init(); each
call adds an entry to the cpu->ases array at the specified
index. It is up to the target-specific CPU code to actually use
these extra address spaces.
Since this multiple AddressSpace support won't work with
KVM, add an assertion to avoid confusing failures.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Rather than setting cpu->as unconditionally in cpu_exec_init
(and then having target-i386 override this later), don't set
it until the first call to cpu_address_space_init.
This requires us to initialise the address space for
both TCG and KVM (KVM doesn't need the AS listener but
it does require cpu->as to be set).
For target CPUs which don't set up any address spaces (currently
everything except i386), add the default address_space_memory
in qemu_init_vcpu().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Acked-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
If virtio-net driver allocates memory in ivshmem shared memory,
vhost-net will work correctly, but vhost-user will not work because
a fd of shared memory will not be sent to vhost-user backend.
This patch fixes ivshmem to store file descriptor of shared memory.
It will be used when vhost-user negotiates vhost-user backend.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
memcpy can take a large amount of time for small reads and writes.
Handle the common case of reading s/g descriptors from memory (there
is no corresponding "write" case that is as common, because writes
often use address_space_st* functions) by inlining the relevant
parts of address_space_read into the caller.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We want to inline the case where there is only one iteration, because
then the compiler can also inline the memcpy. As a start, extract
everything after the first address_space_translate call.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
For the common case of DMA into non-hotplugged RAM, it is unnecessary
but expensive to do object_ref/unref. Add back an owner field to
MemoryRegion, so that these memory regions can skip the reference
counting.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Simplify the code and document the assumption. The only caller
that is not within rcu_read_lock is memory_region_get_ram_ptr.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replace qemu_ram_free_from_ptr() with qemu_ram_free().
The only difference between qemu_ram_free_from_ptr() and
qemu_ram_free() is that g_free_rcu() is used instead of
call_rcu(reclaim_ramblock). We can safely replace it because:
* RAM blocks allocated by qemu_ram_alloc_from_ptr() always have
RAM_PREALLOC set;
* reclaim_ramblock(block) will do nothing except g_free(block)
if RAM_PREALLOC is set at block->flags.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <1446844805-14492-2-git-send-email-ehabkost@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Anthony reported that >4GB guests on Xen with 32bit QEMU broke after
commit 4ed023c ("Round up RAMBlock sizes to host page sizes", 2015-11-05).
In that patch sizes are masked against qemu_host_page_size/mask which
are uintptr_t, and thus 32bit on a 32bit QEMU, even though the ram space
might be bigger than 4GB on Xen.
Since ram_addr_t is not available on user-mode emulation targets, ensure
that we get a sign extension when masking away the low bits of the address.
Remove the ~10 year old scary comment that the type of these variables
is probably wrong, with another equally scary comment. The new comment
however does not have "???" in it, which is arguably an improvement.
For completeness use the alignment macros in linux-user and bsd-user
instead of manually doing an &. linux-user and bsd-user are not affected
by the Xen issue, however.
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reported-by: Anthony PERARD <anthony.perard@citrix.com>
Fixes: 4ed023ce2a
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a function to find a RAMBlock by name; use it in two
of the places that already open code that loop; we've
got another use later in postcopy.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Postcopy sends RAMBlock names and offsets over the wire (since it can't
rely on the order of ramaddr being the same), and it starts out with
HVA fault addresses from the kernel.
qemu_ram_block_from_host translates a HVA into a RAMBlock, an offset
in the RAMBlock and the global ram_addr_t value.
Rewrite qemu_ram_addr_from_host to use qemu_ram_block_from_host.
Provide qemu_ram_get_idstr since its the actual name text sent on the
wire.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
The HOST_PAGE_ALIGN macros don't work until the page size variables
have been set up; later in postcopy I use those macros in the RAM
code, and it can be triggered using -object.
Fix this by initialising page_size_init() earlier - it's currently
initialised inside the accelerators, move it up into vl.c.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
This patch is required for deterministic replay to generate an exception
by trying executing an instruction without changing icount.
It adds new flag to TB for disabling icount while translating it.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Message-Id: <20150917162359.8676.77011.stgit@PASHA-ISP.def.inno>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
For GICv3 ITS implementation we are going to use requester IDs in KVM IRQ
routing code. This patch introduces reusable convenient way to obtain this
ID from the device pointer. The new function is now used in some places,
where the same calculation was used.
MemTxAttrs.stream_id also renamed to requester_id in order to better
reflect semantics of the field.
Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <5814bcb03a297f198e796b13ed9c35059c52f89b.1444916432.git.p.fedin@samsung.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The header is included from basically everywhere, thanks to cpu.h.
It should be moved to the (TCG only) files that actually need it.
As a start, remove non-TCG stuff.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Gather up all the fields currently in CPUState which deal with the CPU's
AddressSpace into a separate CPUAddressSpace struct. This paves the way
for allowing the CPU to know about more than one AddressSpace.
The rearrangement also allows us to make the MemoryListener a directly
embedded object in the CPUAddressSpace (it could not be embedded in
CPUState because 'struct MemoryListener' isn't defined for the user-only
builds). This allows us to resolve the FIXME in tcg_commit() by going
directly from the MemoryListener to the CPUAddressSpace.
This patch extracts the actual update of the cached dispatch pointer
from cpu_reload_memory_map() (which is renamed accordingly to
cpu_reloading_memory_map() as it is only responsible for breaking
cpu-exec.c's RCU critical section now). This lets us keep the definition
of the CPUAddressSpace struct private to exec.c.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <1443709790-25180-4-git-send-email-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
At present, the "average" guestimate of TB size is way too small, leading
to many unused entries in the pre-allocated TB array. For a guest with 1GB
ram, we're currently allocating 256MB for the array.
Survey arm, alpha, aarch64, ppc, sparc, i686, x86_64 guests running on
x86_64 and ppc64 hosts and select a new average. The size of the array
drops to 81MB with no more flushing than before.
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Signed-off-by: Richard Henderson <rth@twiddle.net>
We currently pre-compute an worst case code size for any TB, which
works out to be 122kB. Since the average TB size is near 1kB, this
wastes quite a lot of storage.
Instead, check for overflow in between generating code for each opcode.
The overhead of the check isn't measurable and wastage is minimized.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
It is no longer used, so tidy up everything reached by it.
This includes the gen_opc_* arrays, the search_pc parameter
and the inline gen_intermediate_code_internal functions.
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
We can now restore state without retranslation.
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
The gen_opc_* arrays are already redundant with the data stored in
the insn_start arguments. Transition restore_state_to_opc to use
data from the latter.
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
As it's only caller, this tidies things a bit.
Reviewed-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
When we have guest visible IOMMUs, we allow notifiers to be registered
which will be informed of all changes to IOMMU mappings. This is used by
vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings.
However, unlike with a memory region listener, an iommu notifier won't be
told about any mappings which already exist in the (guest) IOMMU at the
time it is registered. This can cause problems if hotplugging a VFIO
device onto a guest bus which had existing guest IOMMU mappings, but didn't
previously have an VFIO devices (and hence no host IOMMU mappings).
This adds a memory_region_iommu_replay() function to handle this case. It
replays any existing mappings in an IOMMU memory region to a specified
notifier. Because the IOMMU memory region doesn't internally remember the
granularity of the guest IOMMU it has a small hack where the caller must
specify a granularity at which to replay mappings.
If there are finer mappings in the guest IOMMU these will be reported in
the iotlb structures passed to the notifier which it must handle (probably
causing it to flag an error). This isn't new - the VFIO iommu notifier
must already handle notifications about guest IOMMU mappings too short
for it to represent in the host IOMMU.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Move the architecture agnostic function prototypes for exec.c out of
cputlb.h to exec-all.h. This allows hiding of the arch specific
cputlb.h from exec.c which should be getting close to having no
architecture specifics. Prepares support for multi-arch, which will have
a minimal cpu.h that services exec.c but not cputlb.h.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Peter Crosthwaite <crosthwaite.peter@gmail.com>
Message-Id: <b4fe754c58c860315e35d44430c26b1c967ce2c9.1441614289.git.crosthwaite.peter@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Change tlb_set_dirty() to accept a CPU instead of an env pointer. This
allows for removal of another CPUArchState usage from prototypes that
need to be QOMified.
Signed-off-by: Peter Crosthwaite <crosthwaite.peter@gmail.com>
Message-Id: <d2b1dcbe7945112989861d8ba7369449c11cc273.1441614289.git.crosthwaite.peter@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
To prepare for multi-arch, cputlb.c should only have awareness of one
single architecture. This means it should not have access to the full
CPU lists which may be heterogeneous. Instead, push the CPU_LOOP() up
to the one and only caller in exec.c.
Signed-off-by: Peter Crosthwaite <crosthwaite.peter@gmail.com>
Message-Id: <db06dc6c49f8970caaf116d0385f00ee10a56f2f.1441614289.git.crosthwaite.peter@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fam Zheng
Peter Maydell
Gonglei
Vladimir Sementsov-Ogievskiy
Paolo Bonzini
Stefan Hajnoczi
zhanghailiang
Peter Crosthwaite
Tetsuya Mukawa
Eduardo Habkost
Dr. David Alan Gilbert
Pavel Dovgalyuk
Pavel Fedin
Richard Henderson
David Gibson
Peter Crosthwaite