Add some basic examples for the definition of cpu topology
in s390x.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-15-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
The query-s390x-cpu-polarization qmp command returns the current
CPU polarization of the machine.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-14-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
When the guest asks to change the polarization this change
is forwarded to the upper layer using QAPI.
The upper layer is supposed to take according decisions concerning
CPU provisioning.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-13-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
S390x provides two more topology attributes, entitlement and dedication.
Let's add these CPU attributes to the QAPI command query-cpu-fast.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-11-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
The modification of the CPU attributes are done through a monitor
command.
It allows to move the core inside the topology tree to optimize
the cache usage in the case the host's hypervisor previously
moved the CPU.
The same command allows to modify the CPU attributes modifiers
like polarization entitlement and the dedicated attribute to notify
the guest if the host admin modified scheduling or dedication of a vCPU.
With this knowledge the guest has the possibility to optimize the
usage of the vCPUs.
The command has a feature unstable for the moment.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231016183925.2384704-10-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
On interception of STSI(15.1.x) the System Information Block
(SYSIB) is built from the list of pre-ordered topology entries.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-5-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
S390 adds two new SMP levels, drawers and books to the CPU
topology.
S390 CPUs have specific topology features like dedication and
entitlement. These indicate to the guest information on host
vCPU scheduling and help the guest make better scheduling decisions.
Add the new levels to the relevant QAPI structs.
Add all the supported topology levels, dedication and entitlement
as properties to S390 CPUs.
Create machine-common.json so we can later include it in
machine-target.json also.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-3-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Clarify roles of different architectures.
Also change things a bit in anticipation of additional members being
added.
Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231016183925.2384704-2-nsg@linux.ibm.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
[thuth: Updated some comments according to suggestions from Markus]
Signed-off-by: Thomas Huth <thuth@redhat.com>
Commit 57df0dff1a (qapi: Extend -compat to set policy for unstable
interfaces) neglected to update the "Limitation" paragraph to mention
feature 'unstable' in addition to feature 'deprecated'. Do that now.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231009110449.4015601-1-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20231013104736.31722-2-quintela@redhat.com>
Migration bandwidth is a very important value to live migration. It's
because it's one of the major factors that we'll make decision on when to
switchover to destination in a precopy process.
This value is currently estimated by QEMU during the whole live migration
process by monitoring how fast we were sending the data. This can be the
most accurate bandwidth if in the ideal world, where we're always feeding
unlimited data to the migration channel, and then it'll be limited to the
bandwidth that is available.
However in reality it may be very different, e.g., over a 10Gbps network we
can see query-migrate showing migration bandwidth of only a few tens of
MB/s just because there are plenty of other things the migration thread
might be doing. For example, the migration thread can be busy scanning
zero pages, or it can be fetching dirty bitmap from other external dirty
sources (like vhost or KVM). It means we may not be pushing data as much
as possible to migration channel, so the bandwidth estimated from "how many
data we sent in the channel" can be dramatically inaccurate sometimes.
With that, the decision to switchover will be affected, by assuming that we
may not be able to switchover at all with such a low bandwidth, but in
reality we can.
The migration may not even converge at all with the downtime specified,
with that wrong estimation of bandwidth, keeping iterations forever with a
low estimation of bandwidth.
The issue is QEMU itself may not be able to avoid those uncertainties on
measuing the real "available migration bandwidth". At least not something
I can think of so far.
One way to fix this is when the user is fully aware of the available
bandwidth, then we can allow the user to help providing an accurate value.
For example, if the user has a dedicated channel of 10Gbps for migration
for this specific VM, the user can specify this bandwidth so QEMU can
always do the calculation based on this fact, trusting the user as long as
specified. It may not be the exact bandwidth when switching over (in which
case qemu will push migration data as fast as possible), but much better
than QEMU trying to wildly guess, especially when very wrong.
A new parameter "avail-switchover-bandwidth" is introduced just for this.
So when the user specified this parameter, instead of trusting the
estimated value from QEMU itself (based on the QEMUFile send speed), it
trusts the user more by using this value to decide when to switchover,
assuming that we'll have such bandwidth available then.
Note that specifying this value will not throttle the bandwidth for
switchover yet, so QEMU will always use the full bandwidth possible for
sending switchover data, assuming that should always be the most important
way to use the network at that time.
This can resolve issues like "unconvergence migration" which is caused by
hilarious low "migration bandwidth" detected for whatever reason.
Reported-by: Zhiyi Guo <zhguo@redhat.com>
Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20231010221922.40638-1-peterx@redhat.com>
Display it as long as being set, irrelevant of FAILED status. E.g., it may
also be applicable to PAUSED stage of postcopy, to provide hint on what has
gone wrong.
The error_mutex seems to be overlooked when referencing the error, add it
to be very safe.
This will change QAPI behavior by showing up error message outside !FAILED
status, but it's intended and doesn't expect to break anyone.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2018404
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20231004220240.167175-2-peterx@redhat.com>
Currently query-dirty-rate uses QEMU_CLOCK_REALTIME as
the source for start-time field. This translates to
clock_gettime(CLOCK_MONOTONIC), i.e. number of seconds
since host boot. This is not very useful. The only
reasonable use case of start-time I can imagine is to
check whether previously completed measurements are
too old or not. But this makes sense only if start-time
is reported as host wall-clock time.
This patch replaces source of start-time from
QEMU_CLOCK_REALTIME to QEMU_CLOCK_HOST.
Signed-off-by: Andrei Gudkov <gudkov.andrei@huawei.com>
Reviewed-by: Hyman Huang <yong.huang@smartx.com>
Message-Id: <399861531e3b24a1ecea2ba453fb2c3d129fb03a.1693905328.git.gudkov.andrei@huawei.com>
Signed-off-by: Hyman Huang <yong.huang@smartx.com>
Right now "qemu-img map" reports compressed blocks as containing data
but having no host offset. This is not very informative. Instead,
let's add another boolean field named "compressed" in case JSON output
mode is specified. This is achieved by utilizing new allocation status
flag BDRV_BLOCK_COMPRESSED for bdrv_block_status().
Also update the expected qemu-iotests outputs to contain the new field.
Signed-off-by: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
Message-ID: <20230907210226.953821-3-andrey.drobyshev@virtuozzo.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
"Host Memory Backends" and "Memory devices" queue ("mem"):
- Support and document VM templating with R/O files using a new "rom"
parameter for memory-backend-file
- Some cleanups and fixes around NVDIMMs and R/O file handling for guest
RAM
- Optimize ioeventfd updates by skipping address spaces that are not
applicable
-----BEGIN PGP SIGNATURE-----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=iraT
-----END PGP SIGNATURE-----
Merge tag 'mem-2023-09-19' of https://github.com/davidhildenbrand/qemu into staging
Hi,
"Host Memory Backends" and "Memory devices" queue ("mem"):
- Support and document VM templating with R/O files using a new "rom"
parameter for memory-backend-file
- Some cleanups and fixes around NVDIMMs and R/O file handling for guest
RAM
- Optimize ioeventfd updates by skipping address spaces that are not
applicable
# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEG9nKrXNcTDpGDfzKTd4Q9wD/g1oFAmUJdykRHGRhdmlkQHJl
# ZGhhdC5jb20ACgkQTd4Q9wD/g1pf2w//akOUoYMuamySGjXtKLVyMKZkjIys+Ama
# k2C0xzsWAHBP572ezwHi8uxf5j9kzAjsw6GxDZ7FAamD9MhiohkEvkecloBx6f/c
# q3fVHblBNkG7v2urtf4+6PJtJvhzOST2SFXfWeYhO/vaA04AYCDgexv82JN3gA6B
# OS8WyOX62b8wILPSY2GLZ8IqpE9XnOYZwzVBn6YB1yo7ZkYEfXO6cA8nykNuNcOE
# vppqDo7uVIX6317FWj8ygxmzFfOaj0WT2MT2XFzEIDfg8BInQN8HC4mTn0hcVKMa
# N1y+eZH733CQKT+uNBRZ5YOeljOi4d6gEEyvkkA/L7e5D3Qg9hIdvHb4uryCFSWX
# Vt07OP1XLBwCZFobOC6sg+2gtTZJxxYK89e6ZzEd0454S24w5bnEteRAaCGOP0XL
# ww9xYULqhtZs55UC4rvZHJwdUAk1fIY4VqynwkeQXegvz6BxedNeEkJiiEU0Tizx
# N2VpsxAJ7H/LLSFeZoCRESo4azrH6U4n7S/eS1tkCniFqibfe2yIQCDoJVfb42ec
# gfg/vThCrDwHkIHzkMmoV8NndA7Q7SIkyMfYeEEBeZMeg8JzYll4DJEw/jQCacxh
# KRUa+AZvGlTJUq0mkvyOVfLki+iaehoIUuY1yvMrmdWijPO8n3YybmP9Ljhr8VdR
# 9MSYZe+I2v8=
# =iraT
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 19 Sep 2023 06:25:45 EDT
# gpg: using RSA key 1BD9CAAD735C4C3A460DFCCA4DDE10F700FF835A
# gpg: issuer "david@redhat.com"
# gpg: Good signature from "David Hildenbrand <david@redhat.com>" [unknown]
# gpg: aka "David Hildenbrand <davidhildenbrand@gmail.com>" [full]
# gpg: aka "David Hildenbrand <hildenbr@in.tum.de>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 1BD9 CAAD 735C 4C3A 460D FCCA 4DDE 10F7 00FF 835A
* tag 'mem-2023-09-19' of https://github.com/davidhildenbrand/qemu:
memory: avoid updating ioeventfds for some address_space
machine: Improve error message when using default RAM backend id
softmmu/physmem: Hint that "readonly=on,rom=off" exists when opening file R/W for private mapping fails
docs: Start documenting VM templating
docs: Don't mention "-mem-path" in multi-process.rst
softmmu/physmem: Never return directories from file_ram_open()
softmmu/physmem: Fail creation of new files in file_ram_open() with readonly=true
softmmu/physmem: Bail out early in ram_block_discard_range() with readonly files
softmmu/physmem: Remap with proper protection in qemu_ram_remap()
backends/hostmem-file: Add "rom" property to support VM templating with R/O files
softmmu/physmem: Distinguish between file access mode and mmap protection
nvdimm: Reject writing label data to ROM instead of crashing QEMU
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
For now, "share=off,readonly=on" would always result in us opening the
file R/O and mmap'ing the opened file MAP_PRIVATE R/O -- effectively
turning it into ROM.
Especially for VM templating, "share=off" is a common use case. However,
that use case is impossible with files that lack write permissions,
because "share=off,readonly=on" will not give us writable RAM.
The sole user of ROM via memory-backend-file are R/O NVDIMMs, but as we
have users (Kata Containers) that rely on the existing behavior --
malicious VMs should not be able to consume COW memory for R/O NVDIMMs --
we cannot change the semantics of "share=off,readonly=on"
So let's add a new "rom" property with on/off/auto values. "auto" is
the default and what most people will use: for historical reasons, to not
change the old semantics, it defaults to the value of the "readonly"
property.
For VM templating, one can now use:
-object memory-backend-file,share=off,readonly=on,rom=off,...
But we'll disallow:
-object memory-backend-file,share=on,readonly=on,rom=off,...
because we would otherwise get an error when trying to mmap the R/O file
shared and writable. An explicit error message is cleaner.
We will also disallow for now:
-object memory-backend-file,share=off,readonly=off,rom=on,...
-object memory-backend-file,share=on,readonly=off,rom=on,...
It's not harmful, but also not really required for now.
Alternatives that were abandoned:
* Make "unarmed=on" for the NVDIMM set the memory region container
readonly. We would still see a change of ROM->RAM and possibly run
into memslot limits with vhost-user. Further, there might be use cases
for "unarmed=on" that should still allow writing to that memory
(temporary files, system RAM, ...).
* Add a new "readonly=on/off/auto" parameter for NVDIMMs. Similar issues
as with "unarmed=on".
* Make "readonly" consume "on/off/file" instead of being a 'bool' type.
This would slightly changes the behavior of the "readonly" parameter:
values like true/false (as accepted by a 'bool'type) would no longer be
accepted.
Message-ID: <20230906120503.359863-4-david@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
AF_XDP is a network socket family that allows communication directly
with the network device driver in the kernel, bypassing most or all
of the kernel networking stack. In the essence, the technology is
pretty similar to netmap. But, unlike netmap, AF_XDP is Linux-native
and works with any network interfaces without driver modifications.
Unlike vhost-based backends (kernel, user, vdpa), AF_XDP doesn't
require access to character devices or unix sockets. Only access to
the network interface itself is necessary.
This patch implements a network backend that communicates with the
kernel by creating an AF_XDP socket. A chunk of userspace memory
is shared between QEMU and the host kernel. 4 ring buffers (Tx, Rx,
Fill and Completion) are placed in that memory along with a pool of
memory buffers for the packet data. Data transmission is done by
allocating one of the buffers, copying packet data into it and
placing the pointer into Tx ring. After transmission, device will
return the buffer via Completion ring. On Rx, device will take
a buffer form a pre-populated Fill ring, write the packet data into
it and place the buffer into Rx ring.
AF_XDP network backend takes on the communication with the host
kernel and the network interface and forwards packets to/from the
peer device in QEMU.
Usage example:
-device virtio-net-pci,netdev=guest1,mac=00:16:35:AF:AA:5C
-netdev af-xdp,ifname=ens6f1np1,id=guest1,mode=native,queues=1
XDP program bridges the socket with a network interface. It can be
attached to the interface in 2 different modes:
1. skb - this mode should work for any interface and doesn't require
driver support. With a caveat of lower performance.
2. native - this does require support from the driver and allows to
bypass skb allocation in the kernel and potentially use
zero-copy while getting packets in/out userspace.
By default, QEMU will try to use native mode and fall back to skb.
Mode can be forced via 'mode' option. To force 'copy' even in native
mode, use 'force-copy=on' option. This might be useful if there is
some issue with the driver.
Option 'queues=N' allows to specify how many device queues should
be open. Note that all the queues that are not open are still
functional and can receive traffic, but it will not be delivered to
QEMU. So, the number of device queues should generally match the
QEMU configuration, unless the device is shared with something
else and the traffic re-direction to appropriate queues is correctly
configured on a device level (e.g. with ethtool -N).
'start-queue=M' option can be used to specify from which queue id
QEMU should start configuring 'N' queues. It might also be necessary
to use this option with certain NICs, e.g. MLX5 NICs. See the docs
for examples.
In a general case QEMU will need CAP_NET_ADMIN and CAP_SYS_ADMIN
or CAP_BPF capabilities in order to load default XSK/XDP programs to
the network interface and configure BPF maps. It is possible, however,
to run with no capabilities. For that to work, an external process
with enough capabilities will need to pre-load default XSK program,
create AF_XDP sockets and pass their file descriptors to QEMU process
on startup via 'sock-fds' option. Network backend will need to be
configured with 'inhibit=on' to avoid loading of the program.
QEMU will need 32 MB of locked memory (RLIMIT_MEMLOCK) per queue
or CAP_IPC_LOCK.
There are few performance challenges with the current network backends.
First is that they do not support IO threads. This means that data
path is handled by the main thread in QEMU and may slow down other
work or may be slowed down by some other work. This also means that
taking advantage of multi-queue is generally not possible today.
Another thing is that data path is going through the device emulation
code, which is not really optimized for performance. The fastest
"frontend" device is virtio-net. But it's not optimized for heavy
traffic either, because it expects such use-cases to be handled via
some implementation of vhost (user, kernel, vdpa). In practice, we
have virtio notifications and rcu lock/unlock on a per-packet basis
and not very efficient accesses to the guest memory. Communication
channels between backend and frontend devices do not allow passing
more than one packet at a time as well.
Some of these challenges can be avoided in the future by adding better
batching into device emulation or by implementing vhost-af-xdp variant.
There are also a few kernel limitations. AF_XDP sockets do not
support any kinds of checksum or segmentation offloading. Buffers
are limited to a page size (4K), i.e. MTU is limited. Multi-buffer
support implementation for AF_XDP is in progress, but not ready yet.
Also, transmission in all non-zero-copy modes is synchronous, i.e.
done in a syscall. That doesn't allow high packet rates on virtual
interfaces.
However, keeping in mind all of these challenges, current implementation
of the AF_XDP backend shows a decent performance while running on top
of a physical NIC with zero-copy support.
Test setup:
2 VMs running on 2 physical hosts connected via ConnectX6-Dx card.
Network backend is configured to open the NIC directly in native mode.
The driver supports zero-copy. NIC is configured to use 1 queue.
Inside a VM - iperf3 for basic TCP performance testing and dpdk-testpmd
for PPS testing.
iperf3 result:
TCP stream : 19.1 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 3.4 Mpps
Rx only : 2.0 Mpps
L2 FWD Loopback : 1.5 Mpps
In skb mode the same setup shows much lower performance, similar to
the setup where pair of physical NICs is replaced with veth pair:
iperf3 result:
TCP stream : 9 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 1.2 Mpps
Rx only : 1.0 Mpps
L2 FWD Loopback : 0.7 Mpps
Results in skb mode or over the veth are close to results of a tap
backend with vhost=on and disabled segmentation offloading bridged
with a NIC.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> (docker/lcitool)
Signed-off-by: Jason Wang <jasowang@redhat.com>
In commit 6f974c843c ("gtk: overwrite the console.c char driver"), I
shared the VC console parse handler with GTK. And later on in commit
d8aec9d9 ("display: add -display spice-app launching a Spice client"),
I also used it to handle spice-app VC.
This is not necessary, the VC console options (width/height/cols/rows)
are specific, and unused by tty-level GTK/Spice VC.
This is not a breaking change, as those options are still being parsed
by QAPI ChardevVC. Adjust the documentation about it.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230830093843.3531473-44-marcandre.lureau@redhat.com>
Reformat the dirty-limit migration doc comments to conform
to current conventions as commit a937b6aa73 (qapi: Reformat
doc comments to conform to current conventions).
Signed-off-by: Hyman Huang(黄勇) <yong.huang@smartx.com>
Message-ID: <169073570563.19893.2928364761104733482-1@git.sr.ht>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Whitespace tidied up]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Since commit a937b6aa73 (qapi: Reformat doc comments to conform to
current conventions), a number of comments not conforming to the
current formatting conventions were added. No problem, just sweep
the entire documentation once more.
To check the generated documentation does not change, I compared the
generated HTML before and after this commit with "wdiff -3". Finds no
differences. Comparing with diff is not useful, as the reflown
paragraphs are visible there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-7-armbru@redhat.com>
trace-event-set-state's explanation of how events are selected is
under "Features". Doesn't belong there. Simply delete it, as it
feels redundant with documentation of member @name.
trace-event-get-state's explanation is under "Returns". Tolerable,
but similarly redundant. Delete it, too.
Cc: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-5-armbru@redhat.com>
The notes section comes out like this:
Notes
Additional arguments depend on the type.
1. For detailed information about this command, please refer to the
‘docs/qdev-device-use.txt’ file.
2. It’s possible to list device properties by running QEMU with the
“-device DEVICE,help” command-line argument, where DEVICE is the
device’s name
The first item isn't numbered. Fix that:
1. Additional arguments depend on the type.
2. For detailed information about this command, please refer to the
‘docs/qdev-device-use.txt’ file.
3. It’s possible to list device properties by running QEMU with the
“-device DEVICE,help” command-line argument, where DEVICE is the
device’s name
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-4-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Examples come out like
Example
set new histograms for all io types with intervals [0, 10), [10,
50), [50, 100), [100, +inf):
The sentence "set new histograms ..." starts with a lower case letter.
Capitalize it. Same for the other examples.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-3-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Documentation for member @bin comes out like
list of io request counts corresponding to histogram intervals.
len("bins") = len("boundaries") + 1 For the example above, "bins"
may be something like [3, 1, 5, 2], and corresponding histogram
looks like:
Note how the equation and the sentence following it run together.
Replace the equation:
list of io request counts corresponding to histogram intervals,
one more element than "boundaries" has. For the example above,
"bins" may be something like [3, 1, 5, 2], and corresponding
histogram looks like:
Cc: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20230720071610.1096458-2-armbru@redhat.com>
[Off by one fixed]
Has return zero for more than 10 years.
Specifically we introduced the field in 1.5.0
commit f1c72795af
Author: Peter Lieven <pl@kamp.de>
Date: Tue Mar 26 10:58:37 2013 +0100
migration: do not sent zero pages in bulk stage
during bulk stage of ram migration if a page is a
zero page do not send it at all.
the memory at the destination reads as zero anyway.
even if there is an madvise with QEMU_MADV_DONTNEED
at the target upon receipt of a zero page I have observed
that the target starts swapping if the memory is overcommitted.
it seems that the pages are dropped asynchronously.
this patch also updates QMP to return the number of
skipped pages in MigrationStats.
but removed its usage in 1.5.3
commit 9ef051e553
Author: Peter Lieven <pl@kamp.de>
Date: Mon Jun 10 12:14:19 2013 +0200
Revert "migration: do not sent zero pages in bulk stage"
Not sending zero pages breaks migration if a page is zero
at the source but not at the destination. This can e.g. happen
if different BIOS versions are used at source and destination.
It has also been reported that migration on pseries is completely
broken with this patch.
This effectively reverts commit f1c72795af.
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20230612193344.3796-2-quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Extend query-migrate to provide throttle time and estimated
ring full time with dirty-limit capability enabled, through which
we can observe if dirty limit take effect during live migration.
Signed-off-by: Hyman Huang(黄勇) <yong.huang@smartx.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-ID: <168733225273.5845.15871826788879741674-8@git.sr.ht>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_dirty_rate_stat_collect
so that period can be configured instead of hardcoded.
dirty-limit capability is kind of like auto-converge
but using dirty limit instead of traditional cpu-throttle
to throttle guest down. To enable this feature, turn on
the dirty-limit capability before live migration using
migrate-set-capabilities, and set the parameters
"x-vcpu-dirty-limit-period", "vcpu-dirty-limit" suitably
to speed up convergence.
Signed-off-by: Hyman Huang(黄勇) <yong.huang@smartx.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <168618975839.6361.17407633874747688653-4@git.sr.ht>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-parameters.
This two parameters are used to help implement the dirty
page rate limit algo of migration.
Signed-off-by: Hyman Huang(黄勇) <yong.huang@smartx.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <168618975839.6361.17407633874747688653-3@git.sr.ht>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is in the range of 1 to 1000ms and used to
make dirtyrate calculation period configurable.
Currently with the "x-vcpu-dirty-limit-period" varies, the
total time of live migration changes, test results show the
optimal value of "x-vcpu-dirty-limit-period" ranges from
500ms to 1000 ms. "x-vcpu-dirty-limit-period" should be made
stable once it proves best value can not be determined with
developer's experiments.
Signed-off-by: Hyman Huang(黄勇) <yong.huang@smartx.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <168618975839.6361.17407633874747688653-2@git.sr.ht>
Signed-off-by: Juan Quintela <quintela@redhat.com>
So all the file is consistent.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230612191604.2219-1-quintela@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Rewrote calc-dirty-rate documentation. Briefly described
different modes of dirty page rate measurement. Added some
examples. Fixed obvious grammar errors.
Signed-off-by: Andrei Gudkov <gudkov.andrei@huawei.com>
Message-Id: <fe7d32a621ebd69ef6974beb2499c0b5dccb9e19.1684854849.git.gudkov.andrei@huawei.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
[Prose tweaked and spacing corrected, as per review]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Migration downtime estimation is calculated based on bandwidth and
remaining migration data. This assumes that loading of migration data in
the destination takes a negligible amount of time and that downtime
depends only on network speed.
While this may be true for RAM, it's not necessarily true for other
migrated devices. For example, loading the data of a VFIO device in the
destination might require from the device to allocate resources, prepare
internal data structures and so on. These operations can take a
significant amount of time which can increase migration downtime.
This patch adds a new capability "switchover ack" that prevents the
source from stopping the VM and completing the migration until an ACK
is received from the destination that it's OK to do so.
This can be used by migrated devices in various ways to reduce downtime.
For example, a device can send initial precopy metadata to pre-allocate
resources in the destination and use this capability to make sure that
the pre-allocation is completed before the source VM is stopped, so it
will have full effect.
This new capability relies on the return path capability to communicate
from the destination back to the source.
The actual implementation of the capability will be added in the
following patches.
Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Tested-by: YangHang Liu <yanghliu@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Make GBM optional for EGL code, and enable the build for win32.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230606115658.677673-13-marcandre.lureau@redhat.com>
TBStats will be introduced to replace CONFIG_PROFILER totally, here
remove all CONFIG_PROFILER related stuffs first.
Signed-off-by: Vanderson M. do Rosario <vandersonmr2@gmail.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Fei Wu <fei2.wu@intel.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230607122411.3394702-2-fei2.wu@intel.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
These events include a copy of the device health information at the
time of the event. Actually using the emulated device health would
require a lot of controls to manipulate that state. Given the aim
of this injection code is to just test the flows when events occur,
inject the contents of the device health state as well.
Future work may add more sophisticate device health emulation
including direct generation of these records when events occur
(such as a temperature threshold being crossed). That does not
reduce the usefulness of this more basic generation of the events.
Acked-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230530133603.16934-8-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Defined in CXL r3.0 8.2.9.2.1.2 DRAM Event Record, this event
provides information related to DRAM devices.
Example injection command in QMP:
{ "execute": "cxl-inject-dram-event",
"arguments": {
"path": "/machine/peripheral/cxl-mem0",
"log": "informational",
"flags": 1,
"dpa": 1000,
"descriptor": 3,
"type": 3,
"transaction-type": 192,
"channel": 3,
"rank": 17,
"nibble-mask": 37421234,
"bank-group": 7,
"bank": 11,
"row": 2,
"column": 77,
"correction-mask": [33, 44, 55,66]
}}
Acked-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230530133603.16934-7-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
To facilitate testing provide a QMP command to inject a general media
event. The event can be added to the log specified.
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230530133603.16934-6-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Inject poison using QMP command cxl-inject-poison to add an entry to the
poison list.
For now, the poison is not returned CXL.mem reads, but only via the
mailbox command Get Poison List. So a normal memory read to an address
that is on the poison list will not yet result in a synchronous exception
(and similar for partial cacheline writes).
That is left for a future patch.
See CXL rev 3.0, sec 8.2.9.8.4.1 Get Poison list (Opcode 4300h)
Kernel patches to use this interface here:
https://lore.kernel.org/linux-cxl/cover.1665606782.git.alison.schofield@intel.com/
To inject poison using QMP (telnet to the QMP port)
{ "execute": "qmp_capabilities" }
{ "execute": "cxl-inject-poison",
"arguments": {
"path": "/machine/peripheral/cxl-pmem0",
"start": 2048,
"length": 256
}
}
Adjusted to select a device on your machine.
Note that the poison list supported is kept short enough to avoid the
complexity of state machine that is needed to handle the MORE flag.
Reviewed-by: Fan Ni <fan.ni@samsung.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Message-Id: <20230526170010.574-3-Jonathan.Cameron@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Since we *might* have user emulation with softmmu,
use the clearer 'CONFIG_SYSTEM_ONLY' key to check
for system emulation.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230613133347.82210-9-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
migrate_ignore_shared() is an optimization that avoids copying memory
that is visible and can be mapped on the target. However, a
memory-backend-ram or a memory-backend-memfd block with the RAM_SHARED
flag set is not migrated when migrate_ignore_shared() is true. This is
wrong, because the block has no named backing store, and its contents will
be lost. To fix, ignore shared memory iff it is a named file. Define a
new flag RAM_NAMED_FILE to distinguish this case.
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <1686151116-253260-1-git-send-email-steven.sistare@oracle.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
When we for example have a sparse qcow2 image and discard: unmap is enabled,
there can be a lot of fragmentation in the image after some time. Especially on VM's
that do a lot of writes/deletes.
This causes the qcow2 image to grow even over 110% of its virtual size,
because the free gaps in the image get too small to allocate new
continuous clusters. So it allocates new space at the end of the image.
Disabling discard is not an option, as discard is needed to keep the
incremental backup size as low as possible. Without discard, the
incremental backups would become large, as qemu thinks it's just dirty
blocks but it doesn't know the blocks are unneeded.
So we need to avoid fragmentation but also 'empty' the unneeded blocks in
the image to have a small incremental backup.
In addition, we also want to send the discards further down the stack, so
the underlying blocks are still discarded.
Therefor we introduce a new qcow2 option "discard-no-unref".
When setting this option to true, discards will no longer have the qcow2
driver relinquish cluster allocations. Other than that, the request is
handled as normal: All clusters in range are marked as zero, and, if
pass-discard-request is true, it is passed further down the stack.
The only difference is that the now-zero clusters are preallocated
instead of being unallocated.
This will avoid fragmentation on the qcow2 image.
Fixes: https://gitlab.com/qemu-project/qemu/-/issues/1621
Signed-off-by: Jean-Louis Dupond <jean-louis@dupond.be>
Message-Id: <20230605084523.34134-2-jean-louis@dupond.be>
Reviewed-by: Hanna Czenczek <hreitz@redhat.com>
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
It's already confusing that we have two very similar functions for
wrapping the parse of a 64-bit unsigned value, differing mainly on
whether they permit leading '-'. Adjust the signature of parse_uint()
and parse_uint_full() to be like all of qemu_strto*(): put the result
parameter last, use the same types (uint64_t and unsigned long long
have the same width, but are not always the same type), and mark
endptr const (this latter change only affects the rare caller of
parse_uint). Adjust all callers in the tree.
While at it, note that since cutils.c already includes:
QEMU_BUILD_BUG_ON(sizeof(int64_t) != sizeof(long long));
we are guaranteed that the result of parse_uint* cannot exceed
UINT64_MAX (or the build would have failed), so we can drop
pre-existing dead comparisons in opts-visitor.c that were never false.
Reviewed-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230522190441.64278-8-eblake@redhat.com>
[eblake: Drop dead code spotted by Markus]
Signed-off-by: Eric Blake <eblake@redhat.com>
