The default cpu model on s390x does not provide zPCI, which is
not yet wired up on tcg. Moreover, virtio-ccw is the standard
on s390x.
Using virtio-scsi will implicitly pick the right device, so just
switch to that for simplicity.
Signed-off-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The default cpu model on s390x does not provide zPCI, which is
not yet wired up on tcg. Moreover, virtio-ccw is the standard
on s390x, so use the -ccw instead of the -pci versions of virtio
devices on s390x.
Signed-off-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The default cpu model on s390x does not provide zPCI, which is
not yet wired up on tcg. Moreover, virtio-ccw is the standard
on s390x, so use the -ccw instead of the -pci versions of virtio
devices on s390x.
Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: QingFeng Hao <haoqf@linux.vnet.ibm.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stefan Hajnoczi [Fri, 8 Sep 2017 08:39:41 +0000 (09:39 +0100)]
docs: add qemu-block-drivers(7) man page
Block driver documentation is available in qemu-doc.html. It would be
convenient to have documentation for formats, protocols, and filter
drivers in a man page.
Extract the relevant part of qemu-doc.html into a new file called
docs/qemu-block-drivers.texi. This file can also be built as a
stand-alone document (man, html, etc).
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
People get surprised when, after "qemu-img create -f raw /dev/sdX", they
still see qcow2 with "qemu-img info", if previously the bdev had a qcow2
header. While this is natural because raw doesn't need to write any
magic bytes during creation, hdev_create is free to clear out the first
sector to make sure the stale qcow2 header doesn't cause such confusion.
Signed-off-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Fam Zheng [Fri, 4 Aug 2017 14:36:58 +0000 (22:36 +0800)]
qemu-img: Clarify about relative backing file options
It's not too surprising when a user specifies the backing file relative
to the current working directory instead of the top layer image. This
causes error when they differ. Though the error message has enough
information to infer the fact about the misunderstanding, it is better
if we document this explicitly, so that users don't have to learn from
mistakes.
Signed-off-by: Fam Zheng <famz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Jeff Cody <jcody@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
However, two test cases (172 and 186) overwrite QEMU_OPTIONS and neglect
to manually set '-machine accel=qtest'. Add the missing option for 172.
186 probably only copied the code from 172, it doesn't actually need to
overwrite QEMU_OPTIONS, so remove that in 186.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Tested-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Peter Maydell [Mon, 25 Sep 2017 19:31:24 +0000 (20:31 +0100)]
Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging
slirp updates
# gpg: Signature made Sun 24 Sep 2017 19:07:51 BST
# gpg: using RSA key 0x9E511E01C737F075
# gpg: Good signature from "Samuel Thibault <samuel.thibault@aquilenet.fr>"
# gpg: aka "Samuel Thibault <sthibault@debian.org>"
# gpg: aka "Samuel Thibault <samuel.thibault@gnu.org>"
# gpg: aka "Samuel Thibault <samuel.thibault@inria.fr>"
# gpg: aka "Samuel Thibault <samuel.thibault@labri.fr>"
# gpg: aka "Samuel Thibault <samuel.thibault@ens-lyon.org>"
# gpg: aka "Samuel Thibault <samuel.thibault@u-bordeaux.fr>"
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 900C B024 B679 31D4 0F82 304B D017 8C76 7D06 9EE6
# Subkey fingerprint: 9A37 3D36 64A8 DC62 DA0A 34FD 9E51 1E01 C737 F075
* remotes/thibault/tags/samuel-thibault:
slirp: Add a special case for the NULL socket
slirp: Fix intermittent send queue hangs on a socket
slirp: Add explanation for hostfwd parsing failure
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Kevin Cernekee [Wed, 20 Sep 2017 20:42:05 +0000 (13:42 -0700)]
slirp: Add a special case for the NULL socket
NULL sockets are used for NDP, BOOTP, and other critical operations.
If the topmost mbuf in a NULL session is blocked pending resolution,
it may cause problems if it blocks other packets with a NULL socket.
So do not add mbufs with a NULL socket field to the same session.
Signed-off-by: Kevin Cernekee <cernekee@chromium.org> Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Kevin Cernekee [Wed, 20 Sep 2017 20:42:04 +0000 (13:42 -0700)]
slirp: Fix intermittent send queue hangs on a socket
if_output() originally sent one mbuf per call and used the slirp->next_m
variable to keep track of where it left off. But nowadays it tries to
send all of the mbufs from the fastq, and one mbuf from each session on
the batchq. The next_m variable is both redundant and harmful: there is
a case[0] involving delayed packets in which next_m ends up pointing
to &slirp->if_batchq when an active session still exists, and this
blocks all traffic for that session until qemu is restarted.
The test case was created to reproduce a problem that was seen on
long-running Chromium OS VM tests[1] which rapidly create and
destroy ssh connections through hostfwd.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
* remotes/bonzini/tags/for-upstream: (32 commits)
chardev: remove context in chr_update_read_handler
chardev: use per-dev context for io_add_watch_poll
chardev: add Chardev.gcontext field
chardev: new qemu_chr_be_update_read_handlers()
scsi: add persistent reservation manager using qemu-pr-helper
scsi: add multipath support to qemu-pr-helper
scsi: build qemu-pr-helper
scsi, file-posix: add support for persistent reservation management
memory: Share special empty FlatView
memory: seek FlatView sharing candidates among children subregions
memory: trace FlatView creation and destruction
memory: Create FlatView directly
memory: Get rid of address_space_init_shareable
memory: Rework "info mtree" to print flat views and dispatch trees
memory: Do not allocate FlatView in address_space_init
memory: Share FlatView's and dispatch trees between address spaces
memory: Move address_space_update_ioeventfds
memory: Alloc dispatch tree where topology is generared
memory: Store physical root MR in FlatView
memory: Rename mem_begin/mem_commit/mem_add helpers
...
Peter Xu [Thu, 21 Sep 2017 06:35:54 +0000 (14:35 +0800)]
chardev: remove context in chr_update_read_handler
We had a per-chardev cache for context, then we don't need this
parameter to be passed in every time when chr_update_read_handler()
called. As long as we are calling chr_update_read_handler() using
qemu_chr_be_update_read_handlers() we'll be fine.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1505975754-21555-5-git-send-email-peterx@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Peter Xu [Thu, 21 Sep 2017 06:35:53 +0000 (14:35 +0800)]
chardev: use per-dev context for io_add_watch_poll
It was only passed in by chr_update_read_handlers(). However when
reconnect, we'll lose that context information. So if a chardev was
running on another context (rather than the default context, the NULL
pointer), it'll switch back to the default context if reconnection
happens. But, it should really stick to the old context.
Convert all the callers of io_add_watch_poll() to use the internally
cached gcontext. Then the context should be able to survive even after
reconnections.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1505975754-21555-4-git-send-email-peterx@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Peter Xu [Thu, 21 Sep 2017 06:35:52 +0000 (14:35 +0800)]
chardev: add Chardev.gcontext field
It caches the gcontext that is used to poll the chardev IO. Before this
patch, we only passed it in via chr_update_read_handlers(). However
that may not be enough if the char backend is disconnected and
reconnected afterward. There are chardev codes that still assumed the
context be NULL (which is the main context). Will fix that up in
following up patches.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1505975754-21555-3-git-send-email-peterx@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 22 Aug 2017 04:50:55 +0000 (06:50 +0200)]
scsi: add multipath support to qemu-pr-helper
Proper support of persistent reservation for multipath devices requires
communication with the multipath daemon, so that the reservation is
registered and applied when a path comes up. The device mapper
utilities provide a library to do so; this patch makes qemu-pr-helper.c
detect multipath devices and, when one is found, delegate the operation
to libmpathpersist.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 22 Aug 2017 04:50:18 +0000 (06:50 +0200)]
scsi: build qemu-pr-helper
Introduce a privileged helper to run persistent reservation commands.
This lets virtual machines send persistent reservations without using
CAP_SYS_RAWIO or out-of-tree patches. The helper uses Unix permissions
and SCM_RIGHTS to restrict access to processes that can access its socket
and prove that they have an open file descriptor for a raw SCSI device.
The next patch will also correct the usage of persistent reservations
with multipath devices.
It would also be possible to support for Linux's IOC_PR_* ioctls in
the future, to support NVMe devices. For now, however, only SCSI is
supported.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Not all scripts using qemu.py configure the Python logging
module, and end up generating a "No handlers could be found for
logger" message instead of actual log messages.
To avoid requiring every script using qemu.py to configure
logging manually, call basicConfig() when creating a QEMUMachine
object. This won't affect scripts that already set up logging,
but will ensure that scripts that don't configure logging keep
working.
Reported-by: Kevin Wolf <kwolf@redhat.com> Fixes: 4738b0a85a0c2031fddc71b51cccebce0c4ba6b1 Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20170921162234.847-1-ehabkost@redhat.com> Reviewed-by: Cleber Rosa <crosa@redhat.com> Acked-by: Lukáš Doktor <ldoktor@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
John Snow [Thu, 21 Sep 2017 01:38:21 +0000 (21:38 -0400)]
ide: fix enum comparison for gcc 4.7
Apparently GCC gets bent over comparing enum values against zero.
Replace the conditional with something less readable.
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Signed-off-by: John Snow <jsnow@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 20170921013821.1673-1-jsnow@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
migration: split ufd_version_check onto receive/request features part
This modification is necessary for userfault fd features which are
required to be requested from userspace.
UFFD_FEATURE_THREAD_ID is a one of such "on demand" feature, which will
be introduced in the next patch.
QEMU have to use separate userfault file descriptor, due to
userfault context has internal state, and after first call of
ioctl UFFD_API it changes its state to UFFD_STATE_RUNNING (in case of
success), but kernel while handling ioctl UFFD_API expects UFFD_STATE_WAIT_API.
So only one ioctl with UFFD_API is possible per ufd.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Alexey Perevalov <a.perevalov@samsung.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
migration: fix hardcoded function name in error report
Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Alexey Perevalov <a.perevalov@samsung.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
migration: pass MigrationIncomingState* into migration check functions
That tiny refactoring is necessary to be able to set
UFFD_FEATURE_THREAD_ID while requesting features, and then
to create downtime context in case when kernel supports it.
Signed-off-by: Alexey Perevalov <a.perevalov@samsung.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
migration: split common postcopy out of ram postcopy
Split common postcopy staff from ram postcopy staff.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Fill postcopy-able pending only if ram postcopy is enabled.
It is necessary because of there will be other postcopy-able states and
when ram postcopy is disabled, it should not spoil common postcopy
related pending.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Now postcopy-able states are recognized by not NULL
save_live_complete_postcopy handler. But when we have several different
postcopy-able states, it is not convenient. Ram postcopy may be
disabled, while some other postcopy enabled, in this case Ram state
should behave as it is not postcopy-able.
This patch add separate has_postcopy handler to specify behaviour of
savevm state.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Wed, 30 Aug 2017 08:32:00 +0000 (16:32 +0800)]
bitmap: provide to_le/from_le helpers
Provide helpers to convert bitmaps to little endian format. It can be
used when we want to send one bitmap via network to some other hosts.
One thing to mention is that, these helpers only solve the problem of
endianess, but it does not solve the problem of different word size on
machines (the bitmaps managing same count of bits may contains different
size when malloced). So we need to take care of the size alignment issue
on the callers for now.
Signed-off-by: Peter Xu <peterx@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Wed, 30 Aug 2017 08:31:59 +0000 (16:31 +0800)]
bitmap: introduce bitmap_count_one()
Count how many bits set in the bitmap.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Juan Quintela [Mon, 17 Jul 2017 10:30:25 +0000 (12:30 +0200)]
migration: Split migration_fd_process_incoming
We need that on later patches.
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Juan Quintela [Thu, 27 Apr 2017 08:48:25 +0000 (10:48 +0200)]
migration: Create x-multifd-page-count parameter
Indicates how many pages we are going to send in each batch to a multifd
thread.
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com>
--
Be consistent with defaults and documentation
Use new DEFINE_PROP_*
Rename x-multifd-group to x-multifd-page-count
Juan Quintela [Fri, 15 Jan 2016 07:56:17 +0000 (08:56 +0100)]
migration: Create x-multifd-channels parameter
Indicates the number of channels that we will create. By default we
create 2 channels.
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com>
--
Catch inconsistent defaults (eric).
Improve comment stating that number of threads is the same than number
of sockets
Use new DEFIN_PROP_*
Rename x-multifd-threads to x-multifd-threads
Juan Quintela [Thu, 14 Jan 2016 11:23:00 +0000 (12:23 +0100)]
migration: Add multifd capability
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
--
We pass the ioc instead of the fd. This will allow us to have more
than one channel open. We also make sure that we set the
from_src_file sooner, so we don't need to pass it as a parameter.
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
--
Peter Maydell [Fri, 22 Sep 2017 11:14:27 +0000 (12:14 +0100)]
Merge remote-tracking branch 'remotes/famz/tags/build-and-test-automation-pull-request' into staging
# gpg: Signature made Fri 22 Sep 2017 08:28:38 BST
# gpg: using RSA key 0xCA35624C6A9171C6
# gpg: Good signature from "Fam Zheng <famz@redhat.com>"
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 5003 7CB7 9706 0F76 F021 AD56 CA35 624C 6A91 71C6
* remotes/famz/tags/build-and-test-automation-pull-request: (36 commits)
docker: Drop 'set -e' from run script
docker: Use archive-source.py
tests: Add README for vm tests
MAINTAINERS: Add tests/vm entry
Makefile: Add rules to run vm tests
tests: Add OpenBSD image
tests: Add NetBSD image
tests: Add FreeBSD image
tests: Add ubuntu.i386 image
tests: Add vm test lib
tests: Add a test key pair
scripts: Add archive-source.sh
qemu.py: Add "wait()" method
gitignore: Ignore vm test images
MAINTAINERS: Fix subsystem name for "Build and test automation"
buildsys: Move rdma libs to per object
buildsys: Move brlapi libs to per object
buildsys: Move usb redir cflags/libs to per object
buildsys: Move libusb cflags/libs to per object
buildsys: Move libcacard cflags/libs to per object
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
seccomp: Don't include libseccomp from QEMU header
The only prototype doesn't need anything from the lib header, and not
including it here allows files that include this header, for example
vl.c, to compile without the libseccomp cflags.
The breakage is since c3883e1f93 for environments where `pkg-config
--cflags libseccomp" is non-empty.
Reported-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Fam Zheng <famz@redhat.com> Acked-by: Eduardo Otubo <otubo@redhat.com>
Message-id: 20170920083647.14599-1-famz@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The migration interface for ais was introduced with kernel 4.13
but the capability itself had been active since 4.12. As migration
support is considered necessary lets disable ais in the 2.10
stable version. A proper fix and re-enablement will be done
for qemu 2.11.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20170921140834.14233-2-borntraeger@de.ibm.com> Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <20170907084230.26493-1-famz@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Fam Zheng <famz@redhat.com>
Alex Bennée [Tue, 25 Jul 2017 13:34:23 +0000 (14:34 +0100)]
docker: docker.py make --no-cache skip checksum test
If you invoke with NOCACHE=1 we pass --no-cache in the argv to
docker.py but may still not force a rebuild if the dockerfile checksum
hasn't changed. By testing for its presence we can force builds
without having to manually remove the docker image.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20170725133425.436-5-alex.bennee@linaro.org> Signed-off-by: Fam Zheng <famz@redhat.com>
Paolo Bonzini [Mon, 21 Aug 2017 16:58:56 +0000 (18:58 +0200)]
scsi, file-posix: add support for persistent reservation management
It is a common requirement for virtual machine to send persistent
reservations, but this currently requires either running QEMU with
CAP_SYS_RAWIO, or using out-of-tree patches that let an unprivileged
QEMU bypass Linux's filter on SG_IO commands.
As an alternative mechanism, the next patches will introduce a
privileged helper to run persistent reservation commands without
expanding QEMU's attack surface unnecessarily.
The helper is invoked through a "pr-manager" QOM object, to which
file-posix.c passes SG_IO requests for PERSISTENT RESERVE OUT and
PERSISTENT RESERVE IN commands. For example:
Multiple pr-manager implementations are conceivable and possible, though
only one is implemented right now. For example, a pr-manager could:
- talk directly to the multipath daemon from a privileged QEMU
(i.e. QEMU links to libmpathpersist); this makes reservation work
properly with multipath, but still requires CAP_SYS_RAWIO
- use the Linux IOC_PR_* ioctls (they require CAP_SYS_ADMIN though)
- more interestingly, implement reservations directly in QEMU
through file system locks or a shared database (e.g. sqlite)
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This shares an cached empty FlatView among address spaces. The empty
FV is used every time when a root MR renders into a FV without memory
sections which happens when MR or its children are not enabled or
zero-sized. The empty_view is not NULL to keep the rest of memory
API intact; it also has a dispatch tree for the same reason.
On POWER8 with 255 CPUs, 255 virtio-net, 40 PCI bridges guest this halves
the amount of FlatView's in use (557 -> 260) and dispatch tables
(~800000 -> ~370000). In an unrelated experiment with 112 non-virtio
devices on x86 ("-M pc"), only 4 FlatViews are alive, and about ~2000
are created at startup.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-16-aik@ozlabs.ru> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 21 Sep 2017 10:28:16 +0000 (12:28 +0200)]
memory: seek FlatView sharing candidates among children subregions
A container can be used instead of an alias to allow switching between
multiple subregions. In this case we cannot directly share the
subregions (since they only belong to a single parent), but if the
subregions are aliases we can in turn walk those.
This is not enough to remove all source of quadratic FlatView creation,
but it enables sharing of the PCI bus master FlatViews (and their
AddressSpaceDispatch structures) across all PCI devices. For 112
virtio-net-pci devices, boot time is reduced from 25 to 10 seconds and
memory consumption from 1.4 to 1 G.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This avoids usual memory_region_transaction_commit() which rebuilds
all FVs.
On POWER8 with 255 CPUs, 255 virtio-net, 40 PCI bridges guest this brings
down the boot time from 25s to 20s and reduces the amount of temporary FVs
allocated during machine constructon (~800000 -> ~640000) and amount of
temporary dispatch trees (~370000 -> ~300000), the total memory footprint
goes down (18G -> 17G).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-18-aik@ozlabs.ru> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
memory: Share FlatView's and dispatch trees between address spaces
This allows sharing flat views between address spaces (AS) when
the same root memory region is used when creating a new address space.
This is done by walking through all ASes and caching one FlatView per
a physical root MR (i.e. not aliased).
This removes search for duplicates from address_space_init_shareable() as
FlatViews are shared elsewhere and keeping as::ref_count correct seems
an unnecessary and useless complication.
This should cause no change and memory use or boot time yet.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-13-aik@ozlabs.ru> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Address spaces get to keep a root MR (alias or not) but FlatView stores
the actual MR as this is going to be used later on to decide whether to
share a particular FlatView or not.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-10-aik@ozlabs.ru> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
memory: Move AddressSpaceDispatch from AddressSpace to FlatView
As we are going to share FlatView's between AddressSpace's,
and AddressSpaceDispatch is a structure to perform quick lookup
in FlatView, this moves ASD to FlatView.
After previosly open coded ASD rendering, we can also remove
as->next_dispatch as the new FlatView pointer is stored
on a stack and set to an AS atomically.
flatview_destroy() is executed under RCU instead of
address_space_dispatch_free() now.
This makes mem_begin/mem_commit to work with ASD and mem_add with FV
as later on mem_add will be taking FV as an argument anyway.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-5-aik@ozlabs.ru> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
projects / qemu-xen.git / log