These macros expand into error class enumeration constant, comma,
string.  Unclean.  Has been that way since commit 13f59ae8.

The error class is always ERROR_CLASS_GENERIC_ERROR since the previous
commit.

Clean up as follows:

* Prepend every use of a QERR_ macro by ERROR_CLASS_GENERIC_ERROR, and
  delete it from the QERR_ macro.  No change after preprocessing.

* Rewrite error_set(ERROR_CLASS_GENERIC_ERROR, ...) into
  error_setg(...).  Again, no change after preprocessing.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NLuiz Capitulino <lcapitulino@redhat.com>

In postcopy we need the loadvm_handlers to be used in a couple
of different instances of the loadvm loop/routine, and thus
it can't be local any more.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NDavid Gibson <david@gibson.dropbear.id.au>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

There are currently lots of pieces of incoming migration state scattered
around, and postcopy is adding more, and it seems better to try and keep
it together.

allocate MIS in process_incoming_migration_co
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NAmit Shah <amit.shah@redhat.com>
Reviewed-by: NDavid Gibson <david@gibson.dropbear.id.au>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Split qemu_savevm_state_begin to:
  qemu_savevm_state_header   That writes the initial file header.
  qemu_savevm_state_begin    That sets up devices and does the first
                             device pass.

Used later in postcopy.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NAmit Shah <amit.shah@redhat.com>
Reviewed-by: NDavid Gibson <david@gibson.dropbear.id.au>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

We assign the MIGRATION_STATUS_SETUP status in two places.  Just in
succession.  Just remove the second one.
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Add the qmp commands to tune and query the parameters used in live
migration.
Signed-off-by: NLiang Li <liang.z.li@intel.com>
Signed-off-by: NYang Zhang <yang.z.zhang@intel.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Put the three parameters related to multiple thread (de)compression
into an int array, and use an enum type to index the parameter.
Signed-off-by: NLiang Li <liang.z.li@intel.com>
Signed-off-by: NYang Zhang <yang.z.zhang@intel.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

The multiple compression threads can be turned on/off through
qmp and hmp interface before doing live migration.
Signed-off-by: NLiang Li <liang.z.li@intel.com>
Signed-off-by: NYang Zhang <yang.z.zhang@intel.com>
Reviewed-by: NDr.David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Add the code to create and destroy the multiple threads those will be
used to do data decompression. Left some functions empty just to keep
clearness, and the code will be added later.
Signed-off-by: NLiang Li <liang.z.li@intel.com>
Signed-off-by: NYang Zhang <yang.z.zhang@intel.com>
Reviewed-by: NDr.David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Add the code to create and destroy the multiple threads those will
be used to do data compression. Left some functions empty to keep
clearness, and the code will be added later.
Signed-off-by: NLiang Li <liang.z.li@intel.com>
Signed-off-by: NYang Zhang <yang.z.zhang@intel.com>
Reviewed-by: NDr.David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

'cancelling' status was introduced by commit 51cf4c1a, mainly to avoid a
possible start of a new migration process while the previous one still exists.
But we didn't expose this status to user, instead we returned the 'active' state.

Here, we expose it to the user (such as libvirt), 'cancelling' status only
occurs for a short window before the migration aborts, so for users,
if they cancel a migration process, it will observe 'cancelling' status
occasionally.

Testing revealed that with older libvirt (anything 1.2.13 or less) will
print an odd error message if the state is seen, but that the migration
is still properly cancelled. Newer libvirt will be patched to recognize
the new state without the odd error message.
Signed-off-by: Nzhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Cc: libvir-list@redhat.com
Signed-off-by: NJuan Quintela <quintela@redhat.com>

The original 'status' is an open-coded 'str' type, convert it to use an
enum type.
This conversion is backwards compatible, better documented and
more convenient for future extensibility.

In addition, Fix a typo for qapi-schema.json (just remove the typo) :
s/'completed'. 'comppleted' (since 1.2)/'completed' (since 1.2)
Signed-off-by: Nzhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Rename all macro MIG_STATE_* to  MIGRATION_STATUS_* except "MIG_STATE_ERROR",
we rename it to "MIGRATION_STATUS_FAILED" which will match the migration status
string 'failed'.
Signed-off-by: Nzhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

migrate_rdma_pin_all() and qsb_clone() are completely unused and thus
can be deleted.
Signed-off-by: NThomas Huth <thuth@linux.vnet.ibm.com>
Cc: Juan Quintela <quintela@redhat.com>
Cc: Amit Shah <amit.shah@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Create a separate error for the case where migrate_incoming is
used after a succesful migrate_incoming.

Reword the error in the case where '-incoming defer' is missing
to omit the command name so it's right for both hmp and qmp.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NAmit Shah <amit.shah@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

qerror_report_err() is a transitional interface to help with
converting existing monitor commands to QMP.  It should not be used
elsewhere.  Replace by error_report_err() in
process_incoming_migration_co().
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Add migrate_incoming/migrate-incoming to start an incoming
migration.

Once a qemu has been started with
    -incoming defer

the migration can be started by issuing:
    migrate_incoming uri
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

-incoming defer causes qemu to wait for an incoming migration
to be specified later.  The monitor can be used to set migration
capabilities that may affect the incoming connection process.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Force shutdown on migration socket on cancel to cause the cancel
to complete even if the socket is blocked on a dead network.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NAmit Shah <amit.shah@redhat.com>
Signed-off-by: NAmit Shah <amit.shah@redhat.com>

The migration code now occupies a fair chunk of the top level .c
files, it seems time to give it it's own directory.

I've not touched:
   arch_init.c - that's mostly RAM migration but has a few random other
                 bits
   savevm.c    - because it's built target specific

This is purely a code move; no code has changed.
   - it fails checkpatch because of old violations, it feels safer
     to keep this as purely a move and fix those at some mythical future
     date.

The xbzrle and vmstate tests are now only run for softmmu builds
since they require files in the migrate/ directory which is only built
for softmmu.
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: NAmit Shah <amit.shah@redhat.com>

When migrated using libvirt with "--copy-storage-all", at the end of
migration there is race between NBD mirroring task trying to do flush
and migration completion, both end up invalidating cache. Since qcow2
driver does not handle this situation very well, random crashes happen.

This disables the BDRV_O_INCOMING flag for the block device being migrated
once the cache has been invalidated.
Signed-off-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>

--

fixed parens by hand
Signed-off-by: NJuan Quintela <quintela@redhat.com>

if a saved vm has unknown flags in the memory data qemu
currently simply ignores this flag and continues which
yields in an unpredictable result.

This patch catches all unknown flags and aborts the
loading of the vm. Additionally error reports are thrown
if the migration aborts abnormally.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

The existing timeout is 30ms which on 100MB/s (1Gbit) gives us
3MB/s rate maximum. If we put some load on the guest, it is easy to
get page dirtying rate too big so live migration will never complete.
In the case of libvirt that means that the guest will be stopped
anyway after a timeout specified in the "virsh migrate" command and
this normally generates even bigger delay.

This changes max_downtime to 300ms which seems to be more
reasonable value.
Signed-off-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

currently the value of the throughput field contains whatever
was the last calculated throughput shortly before the migration
finished.

This patch updates the post migration contents of the field to
the average throughput.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

expose xbzrle cache miss rate
Signed-off-by: NChenLiang <chenliang88@huawei.com>
Signed-off-by: NGonglei <arei.gonglei@huawei.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

expose the count that logs the times of updating the dirty bitmap to
end user.
Signed-off-by: NChenLiang <chenliang88@huawei.com>
Signed-off-by: NGonglei <arei.gonglei@huawei.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

QEMU will assert if you attempt to start an outgoing migration on
a QEMU that's sitting waiting for an incoming migration (started
with -incoming), so disallow it with a proper error.

(This is a fix for https://bugzilla.redhat.com/show_bug.cgi?id=1086987 )
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

This replaces DPRINTF macro with tracepoints.

This moves some messages from migration.c to savevm.c.

This adds tracepoint to signal about fileds failed to migrate.
Signed-off-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAmit Shah <amit.shah@redhat.com>

If it returns an error, the migrated VM will not be started, but qemu
exits with an error message.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NBenoit Canet <benoit@irqsave.net>

If enabled, set the thread name at creation (on GNU systems with
  pthread_set_np)
Fix up all the callers with a thread name
Signed-off-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>

As far as we can tell, all known bugs have been fixed:

1. Parallel migrations are working
2. IPv6 migration is working
3. virt-test is working

I'm not comfortable sending the revised libvirt patch
until this is accepted or review suggestions are addressed,
(including pin-all support. It does not make sense to
remove experimental for one thing and not the other. That's
too many trips through the libvirt community).
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>

Signed-off-by: NOrit Wasserman <owasserm@redhat.com>
Reviewed-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

When qemu do live migration with xbzrle, qemu malloc decoded_buf
at destination end but free it at source end. It will crash qemu
by double free error in some scenarios. Splitting the XBZRLE structure
for clear logic distinguishing src/dst side.
Signed-off-by: NChenLiang <chenliang88@huawei.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NOrit Wasserman <owasserm@redhat.com>
Signed-off-by: NGongLei <arei.gonglei@huawei.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Signed-off-by: NOrit Wasserman <owasserm@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

The migration thread appears to want to allow writeout to occur at full
speed rather than being rate limited during completion of state saving,
but sets the limit to INT_MAX when xfer_limit is INT64_MAX. This causes
problems if there's more than 2GB of state left to save at this point. It
probably ought to just be INT64_MAX instead.
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Introduce MIG_STATE_CANCELLING state to avoid starting a new migration task while the previous one still exist.
Signed-off-by: NZeng Junliang <zengjunliang@huawei.com>
Signed-off-by: NZhang Haoyu <haoyu.zhang@huawei.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

Avoid a bogus COMPLETED->CANCELLED transition.
There is a period of time from the timing of setting COMPLETED state to that of migration thread exits, so during which it's problematic in COMPLETED->CANCELLED transition.
Signed-off-by: NZeng Junliang <zengjunliang@huawei.com>
Signed-off-by: NZhang Haoyu <haoyu.zhang@huawei.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NJuan Quintela <quintela@redhat.com>

If a user or QMP client enter a bad syntax for the migrate
command in QMP/HMP, then the migrate command will never succeed
from that point on.

For example, if you enter:

(qemu) migrate tcp;0:4444
migrate: Parameter 'uri' expects a valid migration protocol

Then the migrate command will always fail from now on:

(qemu) migrate tcp:0:4444
migrate: There's a migration process in progress

The problem is that qmp_migrate() sets the migration status to
MIG_STATE_SETUP and doesn't reset it on syntax error. This bug
was introduced by commit 29ae8a41.
Reviewed-by: NMichael R. Hines <mrhines@us.ibm.com>
Signed-off-by: NLuiz Capitulino <lcapitulino@redhat.com>

Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>
Reviewed-by: NStefan Weil <sw@weilnetz.de>

The printf args are uint64_t and with -Werr QEMU doesn't compile with
migration debugging turned on unless this is fixed.  Fix it.
Signed-off-by: NChristoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: NJuan Quintela <quintela@redhat.com>