| Age | Commit message (Collapse) | Author |
|---|
|
Having large virtio-mem devices that only expose little memory to a VM
is currently a problem: we map the whole sparse memory region into the
guest using a single memslot, resulting in one gigantic memslot in KVM.
KVM allocates metadata for the whole memslot, which can result in quite
some memory waste.
Assuming we have a 1 TiB virtio-mem device and only expose little (e.g.,
1 GiB) memory, we would create a single 1 TiB memslot and KVM has to
allocate metadata for that 1 TiB memslot: on x86, this implies allocating
a significant amount of memory for metadata:
(1) RMAP: 8 bytes per 4 KiB, 8 bytes per 2 MiB, 8 bytes per 1 GiB
-> For 1 TiB: 2147483648 + 4194304 + 8192 = ~ 2 GiB (0.2 %)
With the TDP MMU (cat /sys/module/kvm/parameters/tdp_mmu) this gets
allocated lazily when required for nested VMs
(2) gfn_track: 2 bytes per 4 KiB
-> For 1 TiB: 536870912 = ~512 MiB (0.05 %)
(3) lpage_info: 4 bytes per 2 MiB, 4 bytes per 1 GiB
-> For 1 TiB: 2097152 + 4096 = ~2 MiB (0.0002 %)
(4) 2x dirty bitmaps for tracking: 2x 1 bit per 4 KiB page
-> For 1 TiB: 536870912 = 64 MiB (0.006 %)
So we primarily care about (1) and (2). The bad thing is, that the
memory consumption *doubles* once SMM is enabled, because we create the
memslot once for !SMM and once for SMM.
Having a 1 TiB memslot without the TDP MMU consumes around:
* With SMM: 5 GiB
* Without SMM: 2.5 GiB
Having a 1 TiB memslot with the TDP MMU consumes around:
* With SMM: 1 GiB
* Without SMM: 512 MiB
... and that's really something we want to optimize, to be able to just
start a VM with small boot memory (e.g., 4 GiB) and a virtio-mem device
that can grow very large (e.g., 1 TiB).
Consequently, using multiple memslots and only mapping the memslots we
really need can significantly reduce memory waste and speed up
memslot-related operations. Let's expose the sparse RAM memory region using
multiple memslots, mapping only the memslots we currently need into our
device memory region container.
The feature can be enabled using "dynamic-memslots=on" and requires
"unplugged-inaccessible=on", which is nowadays the default.
Once enabled, we'll auto-detect the number of memslots to use based on the
memslot limit provided by the core. We'll use at most 1 memslot per
gigabyte. Note that our global limit of memslots accross all memory devices
is currently set to 256: even with multiple large virtio-mem devices,
we'd still have a sane limit on the number of memslots used.
The default is to not dynamically map memslot for now
("dynamic-memslots=off"). The optimization must be enabled manually,
because some vhost setups (e.g., hotplug of vhost-user devices) might be
problematic until we support more memslots especially in vhost-user backends.
Note that "dynamic-memslots=on" is just a hint that multiple memslots
*may* be used for internal optimizations, not that multiple memslots
*must* be used. The actual number of memslots that are used is an
internal detail: for example, once memslot metadata is no longer an
issue, we could simply stop optimizing for that. Migration source and
destination can differ on the setting of "dynamic-memslots".
Message-ID: <20230926185738.277351-17-david@redhat.com>
Reviewed-by: Maciej S. Szmigiero <maciej.szmigiero@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
|
|
In many cases, blindly unplugging a virtio-mem device is problematic. We
can only safely remove a device once:
* The guest is not expecting to be able to read unplugged memory
(unplugged-inaccessible == on)
* The virtio-mem device does not have memory plugged (size == 0)
* The virtio-mem device does not have outstanding requests to the VM to
plug memory (requested-size == 0)
So let's add a callback to the virtio-mem device class to check for that.
We'll wire-up virtio-mem-pci next.
Message-ID: <20230711153445.514112-7-david@redhat.com>
Tested-by: Mario Casquero <mcasquer@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
|
|
The bitmap and the size are immutable while migration is active: see
virtio_mem_is_busy(). We can migrate this information early, before
migrating any actual RAM content. Further, all information we need for
sanity checks is immutable as well.
Having this information in place early will, for example, allow for
properly preallocating memory before touching these memory locations
during RAM migration: this way, we can make sure that all memory was
actually preallocated and that any user errors (e.g., insufficient
hugetlb pages) can be handled gracefully.
In contrast, usable_region_size and requested_size can theoretically
still be modified on the source while the VM is running. Keep migrating
these properties the usual, late, way.
Use a new device property to keep behavior of compat machines
unmodified.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>S
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
|
|
With VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE, we signal the VM that reading
unplugged memory is not supported. We have to fail feature negotiation
in case the guest does not support VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE.
First, VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE is required to properly handle
memory backends (or architectures) without support for the shared zeropage
in the hypervisor cleanly. Without the shared zeropage, even reading an
unpopulated virtual memory location can populate real memory and
consequently consume memory in the hypervisor. We have a guaranteed shared
zeropage only on MAP_PRIVATE anonymous memory.
Second, we want VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE to be the default
long-term as even populating the shared zeropage can be problematic: for
example, without THP support (possible) or without support for the shared
huge zeropage with THP (unlikely), the PTE page tables to hold the shared
zeropage entries can consume quite some memory that cannot be reclaimed
easily.
Third, there are other optimizations+features (e.g., protection of
unplugged memory, reducing the total memory slot size and bitmap sizes)
that will require VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE.
We really only support x86 targets with virtio-mem for now (and
Linux similarly only support x86), but that might change soon, so prepare
for different targets already.
Add a new "unplugged-inaccessible" tristate property for x86 targets:
- "off" will keep VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE unset and legacy
guests working.
- "on" will set VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE and stop legacy guests
from using the device.
- "auto" selects the default based on support for the shared zeropage.
Warn in case the property is set to "off" and we don't have support for the
shared zeropage.
For existing compat machines, the property will default to "off", to
not change the behavior but eventually warn about a problematic setup.
Short-term, we'll set the property default to "auto" for new QEMU machines.
Mid-term, we'll set the property default to "on" for new QEMU machines.
Long-term, we'll deprecate the parameter and disallow legacy
guests completely.
The property has to match on the migration source and destination. "auto"
will result in the same VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE setting as long
as the qemu command line (esp. memdev) match -- so "auto" is good enough
for migration purposes and the parameter doesn't have to be migrated
explicitly.
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20211217134039.29670-3-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
|
For scarce memory resources, such as hugetlb, we want to be able to
prealloc such memory resources in order to not crash later on access. On
simple user errors we could otherwise easily run out of memory resources
an crash the VM -- pretty much undesired.
For ordinary memory devices, such as DIMMs, we preallocate memory via the
memory backend for such use cases; however, with virtio-mem we're dealing
with sparse memory backends; preallocating the whole memory backend
destroys the whole purpose of virtio-mem.
Instead, we want to preallocate memory when actually exposing memory to the
VM dynamically, and fail plugging memory gracefully + warn the user in case
preallocation fails.
A common use case for hugetlb will be using "reserve=off,prealloc=off" for
the memory backend and "prealloc=on" for the virtio-mem device. This
way, no huge pages will be reserved for the process, but we can recover
if there are no actual huge pages when plugging memory. Libvirt is
already prepared for this.
Note that preallocation cannot protect from the OOM killer -- which
holds true for any kind of preallocation in QEMU. It's primarily useful
only for scarce memory resources such as hugetlb, or shared file-backed
memory. It's of little use for ordinary anonymous memory that can be
swapped, KSM merged, ... but we won't forbid it.
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20211217134611.31172-9-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
|
Migration code now properly handles RAMBlocks which are indirectly managed
by a RamDiscardManager. No need for manual handling via the free page
optimization interface, let's get rid of it.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
|
|
Let's properly notify when (un)plugging blocks, after discarding memory
and before allowing the guest to consume memory. Handle errors from
notifiers gracefully (e.g., no remaining VFIO mappings) when plugging,
rolling back the change and telling the guest that the VM is busy.
One special case to take care of is replaying all notifications after
restoring the vmstate. The device starts out with all memory discarded,
so after loading the vmstate, we have to notify about all plugged
blocks.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Auger Eric <eric.auger@redhat.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Marek Kedzierski <mkedzier@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210413095531.25603-6-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
|
|
One of the goals of having less boilerplate on QOM declarations
is to avoid human error. Requiring an extra argument that is
never used is an opportunity for mistakes.
Remove the unused argument from OBJECT_DECLARE_TYPE and
OBJECT_DECLARE_SIMPLE_TYPE.
Coccinelle patch used to convert all users of the macros:
@@
declarer name OBJECT_DECLARE_TYPE;
identifier InstanceType, ClassType, lowercase, UPPERCASE;
@@
OBJECT_DECLARE_TYPE(InstanceType, ClassType,
- lowercase,
UPPERCASE);
@@
declarer name OBJECT_DECLARE_SIMPLE_TYPE;
identifier InstanceType, lowercase, UPPERCASE;
@@
OBJECT_DECLARE_SIMPLE_TYPE(InstanceType,
- lowercase,
UPPERCASE);
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Paul Durrant <paul@xen.org>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20200916182519.415636-4-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
|
|
Replace DECLARE_OBJ_CHECKERS with OBJECT_DECLARE_TYPE where the
typedefs can be safely removed.
Generated running:
$ ./scripts/codeconverter/converter.py -i \
--pattern=DeclareObjCheckers $(git grep -l '' -- '*.[ch]')
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20200831210740.126168-16-ehabkost@redhat.com>
Message-Id: <20200831210740.126168-17-ehabkost@redhat.com>
Message-Id: <20200831210740.126168-18-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
|
|
Generated using:
$ ./scripts/codeconverter/converter.py -i \
--pattern=TypeCheckMacro $(git grep -l '' -- '*.[ch]')
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-12-ehabkost@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-13-ehabkost@redhat.com>
Message-Id: <20200831210740.126168-14-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
|
|
Some typedefs and macros are defined after the type check macros.
This makes it difficult to automatically replace their
definitions with OBJECT_DECLARE_TYPE.
Patch generated using:
$ ./scripts/codeconverter/converter.py -i \
--pattern=QOMStructTypedefSplit $(git grep -l '' -- '*.[ch]')
which will split "typdef struct { ... } TypedefName"
declarations.
Followed by:
$ ./scripts/codeconverter/converter.py -i --pattern=MoveSymbols \
$(git grep -l '' -- '*.[ch]')
which will:
- move the typedefs and #defines above the type check macros
- add missing #include "qom/object.h" lines if necessary
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-9-ehabkost@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-10-ehabkost@redhat.com>
Message-Id: <20200831210740.126168-11-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
|
|
The content of unplugged memory is undefined and should not be migrated,
ever. Exclude all unplugged memory during precopy using the precopy notifier
infrastructure introduced for free page hinting in virtio-balloon.
Unplugged memory is marked as "not dirty", meaning it won't be
considered for migration.
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200626072248.78761-21-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
|
We want to send qapi events in case the size of a virtio-mem device
changes. This allows upper layers to always know how much memory is
actually currently consumed via a virtio-mem device.
Unfortuantely, we have to report the id of our proxy device. Let's provide
an easy way for our proxy device to register, so it can send the qapi
events. Piggy-backing on the notifier infrastructure (although we'll
only ever have one notifier registered) seems to be an easy way.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200626072248.78761-17-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
|
This is the very basic/initial version of virtio-mem. An introduction to
virtio-mem can be found in the Linux kernel driver [1]. While it can be
used in the current state for hotplug of a smaller amount of memory, it
will heavily benefit from resizeable memory regions in the future.
Each virtio-mem device manages a memory region (provided via a memory
backend). After requested by the hypervisor ("requested-size"), the
guest can try to plug/unplug blocks of memory within that region, in order
to reach the requested size. Initially, and after a reboot, all memory is
unplugged (except in special cases - reboot during postcopy).
The guest may only try to plug/unplug blocks of memory within the usable
region size. The usable region size is a little bigger than the
requested size, to give the device driver some flexibility. The usable
region size will only grow, except on reboots or when all memory is
requested to get unplugged. The guest can never plug more memory than
requested. Unplugged memory will get zapped/discarded, similar to in a
balloon device.
The block size is variable, however, it is always chosen in a way such that
THP splits are avoided (e.g., 2MB). The state of each block
(plugged/unplugged) is tracked in a bitmap.
As virtio-mem devices (e.g., virtio-mem-pci) will be memory devices, we now
expose "VirtioMEMDeviceInfo" via "query-memory-devices".
--------------------------------------------------------------------------
There are two important follow-up items that are in the works:
1. Resizeable memory regions: Use resizeable allocations/RAM blocks to
grow/shrink along with the usable region size. This avoids creating
initially very big VMAs, RAM blocks, and KVM slots.
2. Protection of unplugged memory: Make sure the gust cannot actually
make use of unplugged memory.
Other follow-up items that are in the works:
1. Exclude unplugged memory during migration (via precopy notifier).
2. Handle remapping of memory.
3. Support for other architectures.
--------------------------------------------------------------------------
Example usage (virtio-mem-pci is introduced in follow-up patches):
Start QEMU with two virtio-mem devices (one per NUMA node):
$ qemu-system-x86_64 -m 4G,maxmem=20G \
-smp sockets=2,cores=2 \
-numa node,nodeid=0,cpus=0-1 -numa node,nodeid=1,cpus=2-3 \
[...]
-object memory-backend-ram,id=mem0,size=8G \
-device virtio-mem-pci,id=vm0,memdev=mem0,node=0,requested-size=0M \
-object memory-backend-ram,id=mem1,size=8G \
-device virtio-mem-pci,id=vm1,memdev=mem1,node=1,requested-size=1G
Query the configuration:
(qemu) info memory-devices
Memory device [virtio-mem]: "vm0"
memaddr: 0x140000000
node: 0
requested-size: 0
size: 0
max-size: 8589934592
block-size: 2097152
memdev: /objects/mem0
Memory device [virtio-mem]: "vm1"
memaddr: 0x340000000
node: 1
requested-size: 1073741824
size: 1073741824
max-size: 8589934592
block-size: 2097152
memdev: /objects/mem1
Add some memory to node 0:
(qemu) qom-set vm0 requested-size 500M
Remove some memory from node 1:
(qemu) qom-set vm1 requested-size 200M
Query the configuration again:
(qemu) info memory-devices
Memory device [virtio-mem]: "vm0"
memaddr: 0x140000000
node: 0
requested-size: 524288000
size: 524288000
max-size: 8589934592
block-size: 2097152
memdev: /objects/mem0
Memory device [virtio-mem]: "vm1"
memaddr: 0x340000000
node: 1
requested-size: 209715200
size: 209715200
max-size: 8589934592
block-size: 2097152
memdev: /objects/mem1
[1] https://lkml.kernel.org/r/20200311171422.10484-1-david@redhat.com
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Eric Blake <eblake@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200626072248.78761-11-david@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
