The PCI spec recommends devices use additional alignment for MSI-X
data structures to allow software to map them to separate processor
pages. One advantage of doing this is that we can emulate those data
structures without a significant performance impact to the operation
of the device. Some devices fail to implement that suggestion and
assigned device performance suffers.
One such case of this is a Mellanox MT27500 series, ConnectX-3 VF,
where the MSI-X vector table and PBA are aligned on separate 4K
pages. If PBA emulation is enabled, performance suffers. It's not
clear how much value we get from PBA emulation, but the solution here
is to only lazily enable the emulated PBA when a masked MSI-X vector
fires. We then attempt to more aggresively disable the PBA memory
region any time a vector is unmasked. The expectation is then that
a typical VM will run entirely with PBA emulation disabled, and only
when used is that emulation re-enabled.
Reported-by: Shyam Kaushik <shyam.kaushik@gmail.com>
Tested-by: Shyam Kaushik <shyam.kaushik@gmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Commit c8ee0a4 introduced new events containing PRIx64 constants without
including the % prefix in the preceding string. This results in a compile
error during build if --enable-trace-backends is passed to configure.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 1450566522-6003-1-git-send-email-mark.cave-ayland@ilande.co.uk
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Some functions was moved from block.c to block/io.c, so the trace-events file should reflect that change.
Signed-off-by: Qinghua Jin <qhjin_dev@163.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Add a QIOChannel subclass that is capable of performing I/O
to/from a separate process, via a pair of pipes. The command
can be used for unidirectional or bi-directional I/O.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Add a QIOChannel subclass that can run the websocket protocol over
the top of another QIOChannel instance. This initial implementation
is only capable of acting as a websockets server. There is no support
for acting as a websockets client yet.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Add a QIOChannel subclass that can run the TLS protocol over
the top of another QIOChannel instance. The object provides a
simplified API to perform the handshake when starting the TLS
session. The layering of TLS over the underlying channel does
not have to be setup immediately. It is possible to take an
existing QIOChannel that has done some handshake and then swap
in the QIOChannelTLS layer. This allows for use with protocols
which start TLS right away, and those which start plain text
and then negotiate TLS.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Add a QIOChannel subclass that is capable of operating on things
that are files, such as plain files, pipes, character/block
devices, but notably not sockets.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Implement a QIOChannel subclass that supports sockets I/O.
The implementation is able to manage a single socket file
descriptor, whether a TCP/UNIX listener, TCP/UNIX connection,
or a UDP datagram. It provides APIs which can listen and
connect either asynchronously or synchronously. Since there
is no asynchronous DNS lookup API available, it uses the
QIOTask helper for spawning a background thread to ensure
non-blocking operation.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
A number of I/O operations need to be performed asynchronously
to avoid blocking the main loop. The caller of such APIs need
to provide a callback to be invoked on completion/error and
need access to the error, if any. The small QIOTask provides
a simple framework for dealing with such probes. The API
docs inline provide an outline of how this is to be used.
Some functions don't have the ability to run asynchronously
(eg getaddrinfo always blocks), so to facilitate their use,
the task class provides a mechanism to run a blocking
function in a thread, while triggering the completion
callback in the main event loop thread. This easily allows
any synchronous function to be made asynchronous, albeit
at the cost of spawning a thread.
In this series, the QIOTask class will be used for things like
the TLS handshake, the websockets handshake and TCP connect()
progress.
The concept of QIOTask is inspired by the GAsyncResult
interface / GTask class in the GIO libraries. The min
version requirements on glib don't allow those to be
used from QEMU, so QIOTask provides a facsimilie which
can be easily switched to GTask in the future if the
min version is increased.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
"Unimplemented" messages go to stderr, everything else goes to tracepoints
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Introduce fw_cfg_data_read(), a generic read method which works
on all access widths (1 through 8 bytes, inclusive), and can be
used during both IOPort and MMIO read accesses.
To maintain legibility, only fw_cfg_data_mem_read() (the MMIO
data read method) is replaced by this patch. The new method
essentially unwinds the fw_cfg_data_mem_read() + fw_cfg_read()
combo, but without unnecessarily repeating all the validity
checks performed by the latter on each byte being read.
This patch also modifies the trace_fw_cfg_read prototype to
accept a 64-bit value argument, allowing it to work properly
with the new read method, but also remain backward compatible
with existing call sites.
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Marc Marí <markmb@redhat.com>
Signed-off-by: Gabriel Somlo <somlo@cmu.edu>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-id: 1446733972-1602-6-git-send-email-somlo@cmu.edu
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
For now, we use inotify watches to track only a small number of
events, namely, add, delete and modify. Note that for delete, the kernel
already deactivates the watch for us and we just need to
take care of modifying our internal state.
inotify is a linux only mechanism.
Suggested-by: Gerd Hoffman <kraxel@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Message-id: 1448314625-3855-4-git-send-email-bsd@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
To support adding/removal of objects, we will need to update
the object cache hierarchy we have built internally. Convert
to using a Qlist for easier management.
Signed-off-by: Bandan Das <bsd@redhat.com>
Message-id: 1448314625-3855-2-git-send-email-bsd@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
The assertion problem was noticed in 06c3916b35, but it wasn't
completely fixed, because even though the req is not marked as
serialising, it still gets serialised by wait_serialising_requests
against other serialising requests, which could lead to the same
assertion failure.
Fix it by even more explicitly skipping the serialising for this
specific case.
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1448962590-2842-2-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Tweak the end of migration cleanup; we don't want to close stuff down
at the end of the main stream, since the postcopy is still sending pages
on the other thread.
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>
Prior to servicing userfault requests we must ensure we've not got
huge pages in the area that might include non-transferred memory,
since a hugepage could incorrectly mark the whole huge page as present.
We mark the area as non-huge page (nhp) just before we perform
discards; the discard code now tells us to discard any areas
that haven't been sent (as well as any that are redirtied);
any already formed transparent-huge-pages get fragmented
by this discard process if they cotnain any discards.
Transparent huge pages that have been entirely transferred
and don't contain any discards are not broken by this mechanism;
they stay as huge pages.
By starting postcopy after a full precopy pass, many of the pages
then stay as huge pages; this is important for maintaining performance
after the end of the migration.
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>
Wire up more of the handlers for the commands on the destination side,
in particular loadvm_postcopy_handle_run now has enough to start the
guest running.
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>
The loading of a device state (during postcopy) may access guest
memory that's still on the source machine and thus might need
a page fill; split off a separate thread that handles the incoming
page data so that the original incoming migration code can finish
off the device data.
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>
userfaultfd is a Linux syscall that gives an fd that receives a stream
of notifications of accesses to pages registered with it and allows
the program to acknowledge those stalls and tell the accessing
thread to carry on.
We convert the requests from the kernel into messages back to the
source asking for the pages.
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>
In postcopy, the destination guest is running at the same time
as it's receiving pages; as we receive new pages we must put
them into the guests address space atomically to avoid a running
CPU accessing a partially written page.
Use the helpers in postcopy-ram.c to map these pages.
qemu_get_buffer_in_place is used to avoid a copy out of qemu_file
in the case that postcopy is going to do a copy anyway.
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_place_page (etc) provide a way for postcopy to place a page
into guests memory atomically (using the copy ioctl on the ufd).
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>
When transmitting RAM pages, consume pages that have been queued by
MIG_RPCOMM_REQPAGE commands and send them ahead of normal page scanning.
Note:
a) After a queued page the linear walk carries on from after the
unqueued page; there is a reasonable chance that the destination
was about to ask for other closeby pages anyway.
b) We have to be careful of any assumptions that the page walking
code makes, in particular it does some short cuts on its first linear
walk that break as soon as we do a queued page.
c) We have to be careful to not break up host-page size chunks, since
this makes it harder to place the pages on the destination.
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>
On receiving MIG_RPCOMM_REQ_PAGES look up the address and
queue the page.
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>
Add MIG_RP_MSG_REQ_PAGES command on Return path for the postcopy
destination to request a page from the source.
Two versions exist:
MIG_RP_MSG_REQ_PAGES_ID that includes a RAMBlock name and start/len
MIG_RP_MSG_REQ_PAGES that just has start/len for use with the same
RAMBlock as a previous MIG_RP_MSG_REQ_PAGES_ID
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>
The end of migration in postcopy is a bit different since some of
the things normally done at the end of migration have already been
done on the transition to postcopy.
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>
Rework the migration thread to setup and start postcopy.
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>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Soon we'll be in either ACTIVE or POSTCOPY_ACTIVE when we
complete migration, and we need to know which we expect to be
in to change state safely.
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>
'MIGRATION_STATUS_POSTCOPY_ACTIVE' is entered after migrate_start_postcopy
'migration_in_postcopy' is provided for other sections to know if
they're in postcopy.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Blake <eblake@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>
Modify save_live_pending to return separate postcopiable and
non-postcopiable counts.
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>
MIG_CMD_PACKAGED is a migration command that wraps a chunk of migration
stream inside a package whose length can be determined purely by reading
its header. The destination guarantees that the whole MIG_CMD_PACKAGED
is read off the stream prior to parsing the contents.
This is used by postcopy to load device state (from the package)
while leaving the main stream free to receive memory pages.
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>
The state of the postcopy process is managed via a series of messages;
* Add wrappers and handlers for sending/receiving these messages
* Add state variable that track the current state of postcopy
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>
Postcopy needs to have two migration streams loading concurrently;
one from memory (with the device state) and the other from the fd
with the memory transactions.
Split the core of qemu_loadvm_state out so we can use it for both.
Allow the inner loadvm loop to quit and cause the parent loops to
exit as well.
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>
Open a return path, and handle messages that are received upon it.
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>
Add migrate_send_rp_message to send a message from destination to source along the return path.
(It uses a mutex to let it be called from multiple threads)
Add migrate_send_rp_shut to send a 'shut' message to indicate
the destination is finished with the RP.
Add migrate_send_rp_ack to send a 'PONG' message in response to a PING
Use it in the MSG_RP_PING handler
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>
Add two src->dest commands:
* OPEN_RETURN_PATH - To request that the destination open the return path
* PING - Request an acknowledge from the destination
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>
Create QEMU_VM_COMMAND section type for sending commands from
source to destination. These commands are not intended to convey
guest state but to control the migration process.
For use in postcopy.
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>
In postcopy we're going to need to perform the complete phase
for postcopiable devices at a different point, start out by
renaming all of the 'complete's to make the difference obvious.
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>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Peter Lieven <pl@kamp.de>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 1446203414-4013-7-git-send-email-kraxel@redhat.com
These messages are disabled by default; a perfect usecase for tracepoints,
which in fact already exist. Add the missing information to them and
stop using qemu_log_mask.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The function qemu_savevm_state_cancel is called after the migration
in migration_thread, it seems strange to 'cancel' it after completion,
rename it to qemu_savevm_state_cleanup looks better.
Signed-off-by: Liang Li <liang.z.li@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>al3
Reviewed-by: Amit Shah <amit.shah@redhat.com>al3
Signed-off-by: Juan Quintela <quintela@redhat.com>al3
The event throttling state machine is hard to understand. I'm not
sure it's entirely correct. Rewrite it in a more straightforward
manner:
State 1: No event sent recently (less than evconf->rate ns ago)
Invariant: evstate->timer is not pending, evstate->qdict is null
On event: send event, arm timer, goto state 2
State 2: Event sent recently, no additional event being delayed
Invariant: evstate->timer is pending, evstate->qdict is null
On event: store it in evstate->qdict, goto state 3
On timer: goto state 1
State 3: Event sent recently, additional event being delayed
Invariant: evstate->timer is pending, evstate->qdict is non-null
On event: store it in evstate->qdict, goto state 3
On timer: send evstate->qdict, clear evstate->qdict,
arm timer, goto state 2
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1444921716-9511-3-git-send-email-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
These messages are disabled by default; a perfect usecase for tracepoints.
Convert them over.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Replace error_report() and use tracing instead. It's not an error to get
a connection or a disconnection, so silence this and trace it instead.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Thibaut Collet <thibaut.collet@6wind.com>
The malloc vtable is not supported anymore in glib, because it broke
when constructors called g_malloc. Remove tracing of g_malloc,
g_realloc and g_free calls.
Note that, for systemtap users, glib also provides tracepoints
glib.mem_alloc, glib.mem_free, glib.mem_realloc, glib.slice_alloc
and glib.slice_free.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 1442417924-25831-1-git-send-email-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Alex Williamson
Mark Cave-Ayland
Qinghua Jin
Daniel P. Berrange
Paolo Bonzini
Peter Maydell
Gabriel L. Somlo
Bandan Das
Fam Zheng
Dr. David Alan Gilbert
Gerd Hoffmann
Liang Li
Markus Armbruster
Marc-André Lureau