Using the previous patches, we're now able to timestamp the SETUP
state. Once we have this time, let the user know about it in the
schema.
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

As described in the previous patch, until now, the MIG_STATE_SETUP
state was not really a 'formal' state. It has been used as a 'zero' state
(what we're calling 'NONE' here) and QEMU has been unconditionally transitioning
into this state when the QMP migration command was called. Instead we want to
introduce MIG_STATE_NONE, which is our starting state in the state machine, and
then immediately transition into the MIG_STATE_SETUP state when the QMP migrate
command is issued.

In order to do this, we must delay the transition into MIG_STATE_ACTIVE until
later in the migration_thread(). This is done to be able to timestamp the amount of
time spent in the SETUP state for proper accounting to the user during
an RDMA migration.

Furthermore, the management software, until now, has never been aware of the
existence of the SETUP state whatsoever. This must change, because, timing of this
state implies that the state actually exists.

These two patches cannot be separated because the 'query_migrate' QMP
switch statement needs to know how to handle this new state transition.
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Tested-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

This patch is in preparation for the next ones: Until now the MIG_STATE_SETUP
state was not really a 'formal' state. It has been used as a 'zero' state
and QEMU has been unconditionally transitioning into this state when
the QMP migrate command was called. In preparation for timing this state,
we have to make this state a a 'real' state which actually gets transitioned
from later in the migration_thread() from SETUP => ACTIVE, rather than just
automatically dropping into this state at the beginninig of the migration.

This means that the state transition function (migration_finish_set_state())
needs to be capable of transitioning from valid states _other_ than just
MIG_STATE_ACTIVE.

The function is in fact already capable of doing that, but was not allowing the
old state to be a parameter specified as an input.

This patch fixes that and only makes the transition if the current state
matches the old state that the caller intended to transition from.
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Tested-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

This takes advantages of the previous patches:

1. use the new QEMUFileOps hook 'save_page'

2. call out to the right accessor methods to invoke
   the iteration hooks defined in QEMUFileOps
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Code that does need to be visible is kept
well contained inside this file and this is the only
new additional file to the entire patch.

This file includes the entire protocol and interfaces
required to perform RDMA migration.

Also, the configure and Makefile modifications to link
this file are included.

Full documentation is in docs/rdma.txt
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

This gives RDMA shared access to madvise() on the destination side
when an entire chunk is found to be zero.
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NChegu Vinod <chegu_vinod@hp.com>
Tested-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

We were not checking for a valid 'bytes_sent' pointer before accessing it.
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

As requested, the protocol now includes memory unpinning support.
This has been implemented in a non-optimized manner, in such a way
that one could devise an LRU or other workload-specific information
on top of the basic mechanism to influence the way unpinning happens
during runtime.

The feature is not yet user-facing, and is thus can only be enabled
at compile-time.
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Add some simple test cases for the new sextract32
and sextract64 functions.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NRichard Henderson <rth@twiddle.net>
Message-id: 1372419632-5521-3-git-send-email-peter.maydell@linaro.org
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

A common operation in instruction decoding is to take a field
from an instruction that represents a signed integer in some
arbitrary number of bits, and sign extend it into a C signed
integer type for manipulation. Provide new functions sextract32()
and sextract64() which perform this operation; they are like
the existing extract32() and extract64() except that the field
is sign-extended into the returned result.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NRichard Henderson <rth@twiddle.net>
Message-id: 1372419632-5521-2-git-send-email-peter.maydell@linaro.org
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

if mr->size == 0, then

int128_get64(int128_sub(mr->size, int128_make64(1))) => assert(!a.hi)

Also, use int128_one().
Signed-off-by: NAlex Williamson <alex.williamson@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Message-id: 20130719184124.15864.20803.stgit@bling.home
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Since 964c6fa1, these files are unused.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Message-id: 1374449966-12926-1-git-send-email-rth@twiddle.net
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This is in fact very simply: When the input in grabbed, everything
should be exclusively passed to the guest - except it has our magic
CTRL-ALT modifier set. Then let GTK filter out those accels that are in
use. When checking the modifier state, we just need to filter out NUM
and CAPS lock.

Note: Filtering based on hard-coded modifiers breaks overriding
accelerators. Needs to be fixed at a later point.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

arm-devs queue

# gpg: Signature made Mon 22 Jul 2013 06:38:52 AM CDT using RSA key ID 14360CDE
# gpg: Can't check signature: public key not found

# By Peter Maydell (8) and Soren Brinkmann (2)
# Via Peter Maydell
* pmaydell/tags/pull-arm-devs-20130722:
  hw/arm: Use 'load_ramdisk()' for loading ramdisks w/ U-Boot header
  hw/loader: Support ramdisk with u-boot header
  vexpress: Add virtio-mmio transports
  vexpress: Make VEDBoardInfo extend arm_boot_info
  arm/boot: Allow boards to modify the FDT blob
  virtio: Implement MMIO based virtio transport
  virtio: Support transports which can specify the vring alignment
  virtio: Add support for guest setting of queue size
  arm/boot: Use qemu_devtree_setprop_sized_cells()
  device_tree: Add qemu_devtree_setprop_sized_cells() utility functions

Message-id: 1374493427-3254-1-git-send-email-peter.maydell@linaro.org
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Paolo Bonzini (2) and others
# Via Paolo Bonzini
* bonzini/iommu-for-anthony:
  exec: fix incorrect assumptions in memory_access_size
  memory: Return -1 again on reads from unsigned regions
  memory: actually set the owner
  exec.c: Pass correct pointer type to qemu_ram_ptr_length

Message-id: 1374264478-23913-1-git-send-email-pbonzini@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Michael Tokarev (2) and others
# Via Michael Tokarev
* mjt/trivial-patches:
  doc: monitor multiplexing rewording
  block/m25p80: Update Micron entries
  Fix command example in qemu.sasl
  slirp: remove mbuf(m_hdr,m_dat) indirection
  linux-user: declare sys_futex to have 6 arguments

Message-id: 1374225073-12959-1-git-send-email-mjt@msgid.tls.msk.ru
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Peter Lieven (5) and others
# Via Stefan Hajnoczi
* stefanha/block:
  block/raw: add .bdrv_get_info
  block: fix bdrv_read_unthrottled()
  cpus: Let vm_stop[_force_state]() always flush block devices
  block-migration: efficiently encode zero blocks
  block/raw: add bdrv_co_write_zeroes
  block: add bdrv_write_zeroes()
  block: fix vvfat error path for enable_write_target
  QEMUBH: make AioContext's bh re-entrant
  dataplane: sync virtio.c and vring.c virtqueue state
  gluster: Add discard support for GlusterFS block driver.
  gluster: Use pkg-config to configure GlusterFS block driver

Message-id: 1374223132-29107-1-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Richard Henderson
# Via Richard Henderson
* rth/axp-next:
  pc-bios: Update palcode-clipper
  target-alpha: Move alarm to vm_clock

Message-id: 1374161033-11449-1-git-send-email-rth@twiddle.net
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Markus Armbruster
# Via Stefan Hajnoczi
* stefanha/tracing:
  trace-events: Fix up source file comments
  trace-events: Drop unused events
  milkymist-minimac2: Fix minimac2_read/_write tracepoints
  slavio_misc: Fix slavio_led_mem_readw/_writew tracepoints
  cleanup-trace-events.pl: New

Message-id: 1374119369-26496-1-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

The load_ramdisk function is used to load ramdisk featuring a U-Boot
header.
Signed-off-by: NSoren Brinkmann <soren.brinkmann@xilinx.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373323202-17083-3-git-send-email-soren.brinkmann@xilinx.com
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Introduce 'load_ramdisk()' which can load "normal" ramdisks and ramdisks
with a u-boot header.
To enable this and leverage synergies 'load_uimage()' is refactored to
accomodate this additional use case.
Signed-off-by: NSoren Brinkmann <soren.brinkmann@xilinx.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373323202-17083-2-git-send-email-soren.brinkmann@xilinx.com
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Add some virtio-mmio transports to the vexpress board model,
together with a modify_dtb hook which adds them to the device
tree so that the kernel will probe for them. We put them
in a reserved area of the address map.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373977512-28932-9-git-send-email-peter.maydell@linaro.org

Make the VEDBoardInfo struct extend arm_boot_info; this will
allow us to get at the VEDBoardInfo information inside callbacks
from arm/boot code.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373977512-28932-8-git-send-email-peter.maydell@linaro.org

Add a callback hook in arm_boot_info to allow board models to
modify the device tree blob if they need to. (The major expected
use case is to add virtio-mmio nodes for virtio-mmio transports
that exist in QEMU but not in the hardware.)
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373977512-28932-7-git-send-email-peter.maydell@linaro.org

Add support for the generic MMIO based virtio transport.

This patch includes some fixes for bugs spotted by
Ying-Shiuan Pan <yspan@itri.org.tw>.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NKONRAD Frederic <fred.konrad@greensocs.com>
Message-id: 1373977512-28932-6-git-send-email-peter.maydell@linaro.org
[Fred changes: updated to new virtio-bus mechanisms]
Signed-off-by: NKONRAD Frederic <fred.konrad@greensocs.com>
[PMM changes:
 * fixed trivial makefile conflict
 * removed unused int_enable
 * host_features doesn't need migrating
 * reset guest accessible state in the reset function
 * minor style fixes like extra blank lines
 * RAZ/WI if there's no backend
 * made transport size 0x200, in line with kvmtool
 * set has_variable_vring_alignment
]

Support virtio transports which can specify the vring alignment
(ie where the guest communicates this to the host) by providing
a new virtio_queue_set_align() function. (The default alignment
remains as before.)

Transports which wish to make use of this must set the
has_variable_vring_alignment field in their VirtioBusClass
struct to true; they can then change the alignment via
virtio_queue_set_align().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373977512-28932-5-git-send-email-peter.maydell@linaro.org

The MMIO virtio transport spec allows the guest to tell the host how
large the queue size is. Add virtio_queue_set_num() function which
implements this in the QEMU common virtio support code.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1373977512-28932-4-git-send-email-peter.maydell@linaro.org

Replace the opencoded assembly of the reg property array for the
/memory node with a call to qemu_devtree_setprop_sized_cells().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-id: 1373977512-28932-3-git-send-email-peter.maydell@linaro.org

We already have a qemu_devtree_setprop_cells() which sets a dtb
property to an array of cells whose values are specified by varargs.
However for the fairly common case of setting a property to a list
of addresses or of address,size pairs the number of cells used by
each element in the list depends on the parent's #address-cells
and #size-cells properties. To make this easier we provide an analogous
qemu_devtree_setprop_sized_cells() macro which allows the number
of cells used by each element to be specified. This is implemented
using an underlying qemu_devtree_setprop_sized_cells_from_array()
function which takes the values and sizes as an array; this may
also be directly useful for cases where the cell contents are
constructed programmatically.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-id: 1373977512-28932-2-git-send-email-peter.maydell@linaro.org

Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

- Split 32Mb and 256Mb parts into a11 and a13 variants.
- Add the 4K sector flag to the 128Mb parts.  (These entries were taken from
  the Linux kernel list, which is missing the flag.)
- Fill out the table of sizes with entries for 64Mb parts.

Prodded by Peter Crosthwaite.
Signed-off-by: NEd Maste <emaste@freebsd.org>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

sasldblistusers2 doesn't have a '-a' option
Signed-off-by: NCole Robinson <crobinso@redhat.com>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

sys_futex has 6 arguments, and all of these need to be copied. Fix incorrect
declaration in the mips_syscall_args array.

This change fixes the cases where the 5th and 6th arguments have non-zero
value and have importance. An example is a Linux implementation of
pthread_cond_wait() function.
Signed-off-by: NPetar Jovanovic <petar.jovanovic@imgtec.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

Signed-off-by: NPeter Lieven <pl@kamp.de>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Even if the VM is already stopped, we cannot assume that all data has
already been successfully flushed to disk. The flush during the previous
vm_stop() could have failed.

Run bdrv_flush_all() unconditionally so that we get an error each time
if the block device isn't really flushed.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

this patch adds a efficient encoding for zero blocks by
adding a new flag indicating a block is completely zero.

additionally bdrv_write_zeros() is used at the destination
to efficiently write these zeroes. depending on the implementation
this avoids that the destination target gets fully provisioned.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NKevin Wolf <kwolf@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NKevin Wolf <kwolf@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>