Commit Graph

35 Commits

Author SHA1 Message Date
Xiao Guangrong 34ab9e9743 migration: detect compression and decompression errors
Currently the page being compressed is allowed to be updated by
the VM on the source QEMU, correspondingly the destination QEMU
just ignores the decompression error. However, we completely miss
the chance to catch real errors, then the VM is corrupted silently

To make the migration more robuster, we copy the page to a buffer
first to avoid it being written by VM, then detect and handle the
errors of both compression and decompression errors properly

Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@tencent.com>
Message-Id: <20180330075128.26919-5-xiaoguangrong@tencent.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-04-25 18:04:08 +01:00
Xiao Guangrong dcaf446ebd migration: stop compression to allocate and free memory frequently
Current code uses compress2() to compress memory which manages memory
internally, that causes huge memory is allocated and freed very
frequently

More worse, frequently returning memory to kernel will flush TLBs
and trigger invalidation callbacks on mmu-notification which
interacts with KVM MMU, that dramatically reduce the performance
of VM

So, we maintain the memory by ourselves and reuse it for each
compression

Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Jiang Biao <jiang.biao2@zte.com.cn>
Signed-off-by: Xiao Guangrong <xiaoguangrong@tencent.com>
Message-Id: <20180330075128.26919-3-xiaoguangrong@tencent.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-04-25 18:04:06 +01:00
Lidong Chen e8a0f2f9a1 migration: Fix rate limiting issue on RDMA migration
RDMA migration implement save_page function for QEMUFile, but
ram_control_save_page do not increase bytes_xfer. So when doing
RDMA migration, it will use whole bandwidth.

Signed-off-by: Lidong Chen <lidongchen@tencent.com>
Message-Id: <1520692378-1835-1-git-send-email-lidongchen@tencent.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-03-23 16:37:15 +00:00
Vladimir Sementsov-Ogievskiy f0d64cb729 migration/qemu-file: add qemu_put_counted_string()
Add function opposite to qemu_get_counted_string.
qemu_put_counted_string puts one-byte length of the string (string
should not be longer than 255 characters), and then it puts the string,
without last zero byte.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20180313180320.339796-9-vsementsov@virtuozzo.com
2018-03-13 17:05:55 -04:00
Juan Quintela 1adc1ceef7 migration: Remove unneeded includes
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
2017-06-14 11:10:19 +02:00
Juan Quintela 6666c96aac migration: Move migration.h to migration/
Nothing uses it outside of migration.h

Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
2017-06-13 11:00:45 +02:00
Juan Quintela 08a0aee15c migration: Split qemu-file.h
Split the file into public and internal interfaces.  I have to rename
the external one because we can't have two include files with the same
name in the same directory.  Build system gets confused.  The only
exported functions are the ones that handle basic types.

Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2017-06-01 18:49:22 +02:00
Pavel Butsykin 53f09a1076 add 'release-ram' migrate capability
This feature frees the migrated memory on the source during postcopy-ram
migration. In the second step of postcopy-ram migration when the source vm
is put on pause we can free unnecessary memory. It will allow, in particular,
to start relaxing the memory stress on the source host in a load-balancing
scenario.

Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com>
Message-Id: <20170203152321.19739-3-pbutsykin@virtuozzo.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
   Manually merged in Pavel's 'migration: madvise error_report fixup!'
2017-02-13 17:27:13 +00:00
Liang Li 787d134fb1 migration: fix live migration failure with compression
Because of commit 11808bb0c4, which remove some condition checks
of 'f->ops->writev_buffer', 'qemu_put_qemu_file' should be enhanced
to clear the 'f_src->iovcnt', or 'f_src->iovcnt' may exceed the
MAX_IOV_SIZE which will break live migration. This should be fixed.

Signed-off-by: Liang Li <liang.z.li@intel.com>
Reported-by: Jinshi Zhang <jinshi.c.zhang@intel.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <1470702146-24399-1-git-send-email-liang.z.li@intel.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-08-11 16:59:53 +05:30
Liang Li b3be28969b qemu-file: Fix qemu_put_compression_data flaw
Current qemu_put_compression_data can only work with no writable
QEMUFile, and can't work with the writable QEMUFile. But it does
not provide any measure to prevent users from using it with a
writable QEMUFile.

We should fix this flaw to make it works with writable QEMUFile.

Signed-off-by: Liang Li <liang.z.li@intel.com>
Suggested-by: Juan Quintela <quintela@redhat.com>
Message-Id: <1462433579-13691-5-git-send-email-liang.z.li@intel.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-06-17 18:24:18 +05:30
Daniel P. Berrange 12992c16d9 migration: remove qemu_get_fd method from QEMUFile
Now that there is a set_blocking callback in QEMUFileOps,
and all users needing non-blocking support have been
converted to QIOChannel, there is no longer any codepath
requiring the qemu_get_fd() method for QEMUFile. Remove it
to avoid further code being introduced with an expectation
of direct file handle access.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1461751518-12128-29-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:32:21 +05:30
Daniel P. Berrange 11808bb0c4 migration: remove support for non-iovec based write handlers
All the remaining QEMUFile implementations provide an iovec
based write handler, so the put_buffer callback can be removed
to simplify the code.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1461751518-12128-28-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:32:18 +05:30
Daniel P. Berrange a24939f279 migration: move definition of struct QEMUFile back into qemu-file.c
Now that the memory buffer based QEMUFile impl is gone, there
is no need for any backend to be accessing internals of the
QEMUFile struct, so it can be moved back into qemu-file.c

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1461751518-12128-24-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:32:05 +05:30
Daniel P. Berrange 06ad513532 migration: introduce set_blocking function in QEMUFileOps
Remove the assumption that every QEMUFile implementation has
a file descriptor available by introducing a new function
in QEMUFileOps to change the blocking state of a QEMUFile.

If not set, it will fallback to the original code using
the get_fd method.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <1461751518-12128-7-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:31:19 +05:30
Daniel P. Berrange 0436e09f96 migration: split migration hooks out of QEMUFileOps
The QEMUFileOps struct contains the I/O subsystem callbacks
and the migration stage hooks. Split the hooks out into a
separate QEMUFileHooks struct to make it easier to refactor
the I/O side of QEMUFile without affecting the hooks.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <1461751518-12128-6-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:31:16 +05:30
Daniel P. Berrange baf51e7739 migration: ensure qemu_fflush() always writes full data amount
The QEMUFile writev_buffer / put_buffer functions are expected
to write out the full set of requested data, blocking until
complete. The qemu_fflush() caller does not expect to deal with
partial writes. Clarify the function comments and add a sanity
check to the code to catch mistaken implementations.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <1461751518-12128-5-git-send-email-berrange@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26 11:31:14 +05:30
Peter Maydell 1393a48526 migration: Clean up includes
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.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1453832250-766-2-git-send-email-peter.maydell@linaro.org
2016-01-29 15:07:22 +00:00
Dr. David Alan Gilbert adc468e9b9 Return path: Open a return path on QEMUFile for sockets
Postcopy needs a method to send messages from the destination back to
the source, this is the 'return path'.

Wire it up for 'socket' QEMUFile's.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-11-10 14:51:49 +01:00
Dr. David Alan Gilbert a800cd5c38 Add wrapper for setting blocking status on a QEMUFile
Add a wrapper to change the blocking status on a QEMUFile
rather than having to use qemu_set_block(qemu_get_fd(f));
it seems best to avoid exposing the fd since not all QEMUFile's
really have one.  With this wrapper we could move the implementation
down to be different on different transports.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-11-10 14:51:48 +01:00
Dr. David Alan Gilbert 9504fb510c Add qemu_get_buffer_in_place to avoid copies some of the time
qemu_get_buffer always copies the data it reads to a users buffer,
however in many cases the file buffer inside qemu_file could be given
back to the caller, avoiding the copy.  This isn't always possible
depending on the size and alignment of the data.

Thus 'qemu_get_buffer_in_place' either copies the data to a supplied
buffer or updates a pointer to the internal buffer if convenient.

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>
2015-11-10 14:51:48 +01:00
Daniel P. Berrange 10817bf09d coroutine: move into libqemuutil.a library
The coroutine files are currently referenced by the block-obj-y
variable. The coroutine functionality though is already used by
more than just the block code. eg migration code uses coroutine
yield. In the future the I/O channel code will also use the
coroutine yield functionality. Since the coroutine code is nicely
self-contained it can be easily built as part of the libqemuutil.a
library, making it widely available.

The headers are also moved into include/qemu, instead of the
include/block directory, since they are now part of the util
codebase, and the impl was never in the block/ directory
either.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2015-10-20 14:59:04 +01:00
Markus Armbruster 97f3ad3551 migration: Use g_new() & friends where that makes obvious sense
g_new(T, n) is neater than g_malloc(sizeof(T) * n).  It's also safer,
for two reasons.  One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.

This commit only touches allocations with size arguments of the form
sizeof(T).  Same Coccinelle semantic patch as in commit b45c03f.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1442231491-23352-1-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2015-09-29 11:36:35 +05:30
Dr. David Alan Gilbert 56f3835ff1 migration: qemu-file more size_t'ifying
This time convert the external functions:
  qemu_get_buffer, qemu_peek_buffer
  qemu_put_buffer and qemu_put_buffer_async

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <1439463094-5394-6-git-send-email-dgilbert@redhat.com>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2015-09-29 11:33:02 +05:30
Dr. David Alan Gilbert 632e3a5cd8 Rework ram_control_load_hook to hook during block load
We need the names of RAMBlocks as they're loaded for RDMA,
reuse a slightly modified ram_control_load_hook:
  a) Pass a 'data' parameter to use for the name in the block-reg
     case
  b) Only some hook types now require the presence of a hook function.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-07-07 14:54:48 +02:00
Markus Armbruster d49b683644 qerror: Move #include out of qerror.h
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Luiz Capitulino <lcapitulino@redhat.com>
2015-06-22 18:20:40 +02:00
Dr. David Alan Gilbert 7c1e52ba6f Move copy out of qemu_peek_buffer
qemu_peek_buffer currently copies the data it reads into a buffer,
however a future patch wants access to the buffer without the copy,
hence rework to remove the copy to the layer above.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-06-12 06:54:01 +02:00
Dr. David Alan Gilbert b3af1bc9d2 Add qemu_get_counted_string to read a string prefixed by a count byte
and use it in loadvm_state and ram_load.

Where ever it's used, check the return and error if it failed.

Minor: ram_load was using a 257 byte array for its string, the
       maximum length is 255 bytes + 0 terminator, so fix to 256

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-06-12 06:54:01 +02:00
Liang Li 44f0eadc33 qemu-file: Add compression functions to QEMUFile
qemu_put_compression_data() compress the data and put it to QEMUFile.
qemu_put_qemu_file() put the data in the buffer of source QEMUFile to
destination QEMUFile.

Signed-off-by: Liang Li <liang.z.li@intel.com>
Signed-off-by: Yang Zhang <yang.z.zhang@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-05-06 17:36:57 +02:00
Juan Quintela 6e1dea46b8 ram: make all save_page functions take a uint64_t parameter
It used to be an int, but then we can't pass directly the
bytes_transferred parameter, that would happen later in the series.

Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
2015-03-16 14:32:47 +01:00
Alexander Graf 9722140011 qemu-file: Add fast ftell code path
For ftell we flush the output buffer to ensure that we don't have anything
lingering in our internal buffers. This is a very safe thing to do.

However, with the dynamic size measurement that the dynamic vmstate
description will bring this would turn out quite slow.

Instead, we can fast path this specific measurement and just take the
internal buffers into account when telling the kernel our position.

I'm sure I overlooked some corner cases where this doesn't work, so
instead of tuning the safe, existing version, this patch adds a fast
variant of ftell that gets used by the dynamic vmstate description code
which isn't critical when it fails.

Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
2015-02-05 17:16:14 +01:00
Dr. David Alan Gilbert e1a8c9b67f socket shutdown
Add QEMUFile interface to allow a socket to be 'shut down' - i.e. any
reads/writes will fail (and any blocking read/write will be woken).

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2015-01-16 13:06:17 +05:30
Peter Maydell 90d6a6730b migration/qemu-file.c: Don't shift left into sign bit
Add a cast in qemu_get_be32() to avoid shifting left into the sign
bit of a signed integer (which is undefined behaviour in C).

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2015-01-15 10:44:13 +03:00
Dr. David Alan Gilbert 977184db39 Split the QEMU buffered file code out
The splitting of qemu-file and addition of the buffered file landed
at the same time; so now split the buffered file code out.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2014-12-16 17:47:36 +05:30
Dr. David Alan Gilbert 4f9d090012 Split struct QEMUFile out
Now we've got multiple QEMUFile source files, some of them need
access to things that were defined in qemu-file.c, so create
a -internal header for them.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2014-12-16 17:47:36 +05:30
Dr. David Alan Gilbert 60fe637bf0 Start migrating migration code into a migration directory
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: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
2014-12-16 17:47:36 +05:30