Hyper-V Dynamic Memory Protocol driver (hv-balloon)
cover Hyper-V Dynamic Memory Protocol driver (hv-balloon)
From: "Maciej S. Szmigiero" <[email protected]>
This series adds a Hyper-V Dynamic Memory Protocol driver (hv-balloon) and its protocol definitions.
Also included is a driver providing backing devices for memory hot-add protocols ("haprots").
A haprot device works like a virtual DIMM stick: it allows inserting extra RAM into the guest at run time.
The main differences from the ACPI-based PC DIMM hotplug are:
* Notifying the guest about the new memory range is not done via ACPI but via a protocol handler that registers with the haprot framework. This means that the ACPI DIMM slot limit does not apply.
* A protocol handler can prevent removal of a haprot device when it is still in use by setting its "busy" field.
* A protocol handler can also register an "unplug" callback so it gets notified when an user decides to remove the haprot device.
This way the protocol handler can inform the guest about this fact and / or do its own cleanup.
The hv-balloon driver is like virtio-balloon on steroids: it allows both changing the guest memory allocation via ballooning and inserting extra RAM into it by adding haprot virtual DIMM sticks.
One of advantages of these over ACPI-based PC DIMM hotplug is that such memory can be hotplugged in much smaller granularity because the ACPI DIMM slot limit does not apply.
In contrast with ACPI DIMM hotplug where one can only request to unplug a whole DIMM stick this driver allows removing memory from guest in single page (4k) units via ballooning.
Then, once the guest has released the whole memory backed by a haprot virtual DIMM stick such device is marked "unused" and can be removed from the VM, if one wants so.
A "HV_BALLOON_HAPROT_UNUSED" QMP event is emitted in this case so the software controlling QEMU knows that this operation is now possible.
The haprot devices are also marked unused after a VM reboot (with a corresponding "HV_BALLOON_HAPROT_UNUSED" QMP event).
They are automatically reinserted (if still present) after the guest reconnects to this protocol (a "HV_BALLOON_HAPROT_INUSE" QMP event is then emitted).
For performance reasons, the guest-released memory is tracked in few range trees, as a series of (start, count) ranges.
Each time a new page range is inserted into such tree its neighbors are checked as candidates for possible merging with it.
Besides performance reasons, the Dynamic Memory protocol itself uses page ranges as the data structure in its messages, so relevant pages need to be merged into such ranges anyway.
One has to be careful when tracking the guest-released pages, since the guest can maliciously report returning pages outside its current address space, which later clash with the address range of newly added memory. Similarly, the guest can report freeing the same page twice.
The above design results in much better ballooning performance than when using virtio-balloon with the same guest: 230 GB / minute with this driver versus 70 GB / minute with virtio-balloon.
During a ballooning operation most of time is spent waiting for the guest to come up with newly freed page ranges, processing the received ranges on the host side (in QEMU / KVM) is nearly instantaneous.
The unballoon operation is also pretty much instantaneous:
thanks to the merging of the ballooned out page ranges 200 GB of memory can be returned to the guest in about 1 second.
With virtio-balloon this operation takes about 2.5 minutes.
These tests were done against a Windows Server 2019 guest running on a Xeon E5-2699, after dirtying the whole memory inside guest before each balloon operation.
Using a range tree instead of a bitmap to track the removed memory also means that the solution scales well with the guest size: even a 1 TB range takes just few bytes of memory.
The required GTree operations are available at
https://gitlab.gnome.org/maciejsszmigiero/glib/-/tree/gtree-add-iterators
Since they are not yet in the upstream Glib a check for them was added to "configure" script, together with new "--enable-hv-balloon" and "--disable-hv-balloon" arguments.
If these GTree operations are missing in the system Glib this driver will be skipped during QEMU build.
An optional "status-report=on" device parameter requests memory status events from the guest (typically sent every second), which allow the host to learn both the guest memory available and the guest memory in use counts.
They are emitted externally as "HV_BALLOON_STATUS_REPORT" QMP events.
The driver is named hv-balloon since the Linux kernel client driver for the Dynamic Memory Protocol is named as such and to follow the naming pattern established by the virtio-balloon driver.
The whole protocol runs over Hyper-V VMBus that has its implementation recently merged in.
The driver was tested against Windows Server 2012 R2, Windows Server 2016 and Windows Server 2016 guests and obeys the guest alignment requirements reported to the host via DM_CAPABILITIES_REPORT message. Extensive event tracing is available under 'hv_balloon_*' prefix.
Example usage:
* Add "-device vmbus-bridge,id=vmbus-bridge -device hv-balloon,id=hvb" to the QEMU command line and set "maxmem" value to something large, like 1T.
* Use QEMU monitor commands to add a haprot virtual DIMM stick, together with its memory backend:
object_add memory-backend-ram,id=mem1,size=200G
device_add mem-haprot,id=ha1,memdev=mem1
The first command is actually the same as for ACPI-based DIMM hotplug.
* Use the ballooning interface monitor commands to force the guest to give out as much memory as possible:
balloon 1
The ballooning interface monitor commands can also be used to resize the guest up and down appropriately.
* One can check the current guest size by issuing a "info balloon" command. This is useful to know what is happening, since large ballooning or unballooning operations take some time to complete.
* Once the guest releases the whole memory backed by a haprot device (or is restarted) a "HV_BALLOON_HAPROT_UNUSED" QMP event will be generated.
The haprot device then can be removed, together with its memory backend:
device_del ha1
object_del mem1
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.
* Allow new haport devices to reuse the same address range as the ones that were previously deleted via device_del monitor command without having to restart the VM.
* Add vmstate / live migration support to the hv-balloon driver.
* Use haprot device to also add memory via virtio interface (this requires defining a new operation in virtio-balloon protocol and appropriate support from the client virtio-balloon driver in the Linux kernel).
Kconfig.host | 3 + configure | 35 + hw/hyperv/Kconfig | 5 + hw/hyperv/hv-balloon.c | 2172 ++++++++++++++++++++++++++++++ hw/hyperv/meson.build | 1 + hw/hyperv/trace-events | 17 + hw/i386/Kconfig | 2 + hw/i386/pc.c | 18 +- hw/mem/Kconfig | 4 + hw/mem/haprot.c | 263 ++++ hw/mem/meson.build | 1 + include/hw/hyperv/dynmem-proto.h | 425 ++++++ include/hw/mem/haprot.h | 72 + meson.build | 2 + qapi/misc.json | 74 + 15 files changed, 3093 insertions(+), 1 deletion(-) create mode 100644 hw/hyperv/hv-balloon.c create mode 100644 hw/mem/haprot.c create mode 100644 include/hw/hyperv/dynmem-proto.h create mode 100644 include/hw/mem/haprot.h
PATCH 1/3 haprot: add a device for memory hot-add protocols
From: "Maciej S. Szmigiero" <[email protected]>
This device works like a virtual DIMM stick: it allows inserting extra RAM into the guest at run time.
The main differences from the ACPI-based PC DIMM hotplug are:
* Notifying the guest about the new memory range is not done via ACPI but via a protocol handler that registers with the haprot framework. This means that the ACPI DIMM slot limit does not apply.
* A protocol handler can prevent removal of a haprot device when it is still in use by setting its "busy" field.
* A protocol handler can also register an "unplug" callback so it gets notified when an user decides to remove the haprot device.
This way the protocol handler can inform the guest about this fact and / or do its own cleanup.
--- hw/i386/Kconfig | 2 + hw/i386/pc.c | 18 ++- hw/mem/Kconfig | 4 + hw/mem/haprot.c | 263 ++++++++++++++++++++++++++++++++++++++++ hw/mem/meson.build | 1 + include/hw/mem/haprot.h | 72 +++++++++++ 6 files changed, 359 insertions(+), 1 deletion(-) create mode 100644 hw/mem/haprot.c create mode 100644 include/hw/mem/haprot.h
hw/i386/Kconfig +2 -0
@@ -58,6 +58,7 @@ config I440FX imply E1000_PCI imply VMPORT imply VMMOUSE + imply HAPROT select PC_PCI select PC_ACPI select ACPI_SMBUS
@@ -85,6 +86,7 @@ config Q35 imply E1000E_PCI_EXPRESS imply VMPORT imply VMMOUSE + imply HAPROT select PC_PCI select PC_ACPI select PCI_EXPRESS_Q35
hw/i386/pc.c +17 -1
@@ -77,6 +77,7 @@#include "hw/acpi/cpu_hotplug.h" #include "hw/boards.h" #include "acpi-build.h" +#include "hw/mem/haprot.h" #include "hw/mem/pc-dimm.h" #include "hw/mem/nvdimm.h" #include "qapi/error.h"
@@ -1416,6 +1417,18 @@ static void pc_virtio_md_pci_unplug_request(HotplugHandler *hotplug_dev,error_setg (errp,"virtio based memory devices cannot be unplugged." ); } +static void pc_haprot_unplug_request (DeviceState *dev,Error **errp) +{ +HAProtDevice *haprot =HAPROT (dev); + +if (haprot->busy) { +error_setg (errp,"the memory is still busy, cannot unplug" ); +return ; + } + +object_unparent (OBJECT (dev)); +} +static void pc_virtio_md_pci_unplug (HotplugHandler *hotplug_dev,DeviceState *dev,Error **errp) {
@@ -1458,6 +1471,8 @@ static void pc_machine_device_unplug_request_cb(HotplugHandler *hotplug_dev, }else if (object_dynamic_cast (OBJECT (dev),TYPE_VIRTIO_PMEM_PCI ) ||object_dynamic_cast (OBJECT (dev),TYPE_VIRTIO_MEM_PCI )) {pc_virtio_md_pci_unplug_request (hotplug_dev, dev, errp); + }else if (object_dynamic_cast (OBJECT (dev),TYPE_HAPROT )) { +pc_haprot_unplug_request (dev, errp); }else {error_setg (errp,"acpi: device unplug request for not supported device" " type: %s" ,object_get_typename (OBJECT (dev)));
@@ -1486,7 +1501,8 @@ static HotplugHandler *pc_get_hotplug_handler(MachineState *machine,if (object_dynamic_cast (OBJECT (dev),TYPE_PC_DIMM ) ||object_dynamic_cast (OBJECT (dev),TYPE_CPU ) ||object_dynamic_cast (OBJECT (dev),TYPE_VIRTIO_PMEM_PCI ) || -object_dynamic_cast (OBJECT (dev),TYPE_VIRTIO_MEM_PCI )) { +object_dynamic_cast (OBJECT (dev),TYPE_VIRTIO_MEM_PCI ) || +object_dynamic_cast (OBJECT (dev),TYPE_HAPROT )) {return HOTPLUG_HANDLER (machine); }
hw/mem/Kconfig +4 -0
@@ -10,3 +10,7 @@ config NVDIMM default y depends on (PC || PSERIES || ARM_VIRT) select MEM_DEVICE + +config HAPROT + bool + select MEM_DEVICE
hw/mem/haprot.c new file +263 -0
263 lines not on this page · raw message
hw/mem/meson.build +1 -0
@@ -3,5 +3,6 @@ mem_ss.add(files('memory-device.c')) mem_ss.add(when: 'CONFIG_DIMM', if_true: files('pc-dimm.c')) mem_ss.add(when: 'CONFIG_NPCM7XX', if_true: files('npcm7xx_mc.c')) mem_ss.add(when: 'CONFIG_NVDIMM', if_true: files('nvdimm.c')) +mem_ss.add(when: 'CONFIG_HAPROT', if_true: files('haprot.c')) softmmu_ss.add_all(when: 'CONFIG_MEM_DEVICE', if_true: mem_ss)
include/hw/mem/haprot.h new file +72 -0
72 lines not on this page · raw message
PATCH 2/3 Add Hyper-V Dynamic Memory Protocol definitions
In reply to Maciej S. SzmigieroFrom: "Maciej S. Szmigiero" <[email protected]>
This commit adds Hyper-V Dynamic Memory Protocol definitions, taken from hv_balloon Linux kernel driver, adapted to the QEMU coding style and definitions.
--- include/hw/hyperv/dynmem-proto.h | 425 +++++++++++++++++++++++++++++++ 1 file changed, 425 insertions(+) create mode 100644 include/hw/hyperv/dynmem-proto.h
include/hw/hyperv/dynmem-proto.h new file +425 -0
@@ -0,0 +1,425 @@ +#ifndef HW_HYPERV_DYNMEM_PROTO_H +#define HW_HYPERV_DYNMEM_PROTO_H + +/* +* Hyper-V Dynamic Memory Protocol definitions +* +* Copyright (C) 2020 Oracle and/or its affiliates. +* +* Author: Maciej S. Szmigiero <[email protected]> +* +* Based on drivers/hv/hv_balloon.c from Linux kernel: +* Copyright (c) 2012, Microsoft Corporation. +* +* Author: K. Y. Srinivasan <[email protected]> +* +* This work is licensed under the terms of the GNU GPL, version 2. +* See the COPYING file in the top-level directory. +*/ + +/* +* Protocol versions. The low word is the minor version, the high word the major +* version. +* +* History: +* Initial version 1.0 +* Changed to 0.1 on 2009/03/25 +* Changes to 0.2 on 2009/05/14 +* Changes to 0.3 on 2009/12/03 +* Changed to 1.0 on 2011/04/05 +* Changed to 2.0 on 2019/12/10 +*/ + +#define DYNMEM_MAKE_VERSION (Major, Minor) ((uint32_t)(((Major) << 16) | (Minor))) +#define DYNMEM_MAJOR_VERSION (Version) ((uint32_t)(Version) >> 16) +#define DYNMEM_MINOR_VERSION (Version) ((uint32_t)(Version) & 0xff) + +enum { +DYNMEM_PROTOCOL_VERSION_1 =DYNMEM_MAKE_VERSION (0 ,3 ), +DYNMEM_PROTOCOL_VERSION_2 =DYNMEM_MAKE_VERSION (1 ,0 ), +DYNMEM_PROTOCOL_VERSION_3 =DYNMEM_MAKE_VERSION (2 ,0 ), + +DYNMEM_PROTOCOL_VERSION_WIN7 =DYNMEM_PROTOCOL_VERSION_1 , +DYNMEM_PROTOCOL_VERSION_WIN8 =DYNMEM_PROTOCOL_VERSION_2 , +DYNMEM_PROTOCOL_VERSION_WIN10 =DYNMEM_PROTOCOL_VERSION_3 , + +DYNMEM_PROTOCOL_VERSION_CURRENT =DYNMEM_PROTOCOL_VERSION_WIN10 +}; + + + +/* +* Message Types +*/ + +enum dm_message_type { +/* +* Version 0.3 +*/ +DM_ERROR =0 , +DM_VERSION_REQUEST =1 , +DM_VERSION_RESPONSE =2 , +DM_CAPABILITIES_REPORT =3 , +DM_CAPABILITIES_RESPONSE =4 , +DM_STATUS_REPORT =5 , +DM_BALLOON_REQUEST =6 , +DM_BALLOON_RESPONSE =7 , +DM_UNBALLOON_REQUEST =8 , +DM_UNBALLOON_RESPONSE =9 , +DM_MEM_HOT_ADD_REQUEST =10 , +DM_MEM_HOT_ADD_RESPONSE =11 , +DM_VERSION_03_MAX =11 , +/* +* Version 1.0. +*/ +DM_INFO_MESSAGE =12 , +DM_VERSION_1_MAX =12 , + +/* +* Version 2.0 +*/ +DM_MEM_HOT_REMOVE_REQUEST =13 , +DM_MEM_HOT_REMOVE_RESPONSE =14 +}; + + +/* +* Structures defining the dynamic memory management +* protocol. +*/ + +union dm_version { +struct { +uint16_t minor_version; +uint16_t major_version; + }; +uint32_t version; +}QEMU_PACKED ; + + +union dm_caps { +struct { +uint64_t balloon:1 ; +uint64_t hot_add:1 ; +/* +* To support guests that may have alignment +* limitations on hot-add, the guest can specify +* its alignment requirements; a value of n +* represents an alignment of 2^n in mega bytes. +*/ +uint64_t hot_add_alignment:4 ; +uint64_t hot_remove:1 ; +uint64_t reservedz:57 ; + } cap_bits; +uint64_t caps; +}QEMU_PACKED ; + +union dm_mem_page_range { +struct { +/* +* The PFN number of the first page in the range. +* 40 bits is the architectural limit of a PFN +* number for AMD64. +*/ +uint64_t start_page:40 ; +/* +* The number of pages in the range. +*/ +uint64_t page_cnt:24 ; + } finfo; +uint64_t page_range; +}QEMU_PACKED ; + + + +/* +* The header for all dynamic memory messages: +* +* type: Type of the message. +* size: Size of the message in bytes; including the header. +* trans_id: The guest is responsible for manufacturing this ID. +*/ + +struct dm_header { +uint16_t type; +uint16_t size; +uint32_t trans_id; +}QEMU_PACKED ; + +/* +* A generic message format for dynamic memory. +* Specific message formats are defined later in the file. +*/ + +struct dm_message { +struct dm_header hdr; +uint8_t data[];/* enclosed message */ +}QEMU_PACKED ; + + +/* +* Specific message types supporting the dynamic memory protocol. +*/ + +/* +* Version negotiation message. Sent from the guest to the host. +* The guest is free to try different versions until the host +* accepts the version. +* +* dm_version: The protocol version requested. +* is_last_attempt: If TRUE, this is the last version guest will request. +* reservedz: Reserved field, set to zero. +*/ + +struct dm_version_request { +struct dm_header hdr; +union dm_version version; +uint32_t is_last_attempt:1 ; +uint32_t reservedz:31 ; +}QEMU_PACKED ; + +/* +* Version response message; Host to Guest and indicates +* if the host has accepted the version sent by the guest. +* +* is_accepted: If TRUE, host has accepted the version and the guest +* should proceed to the next stage of the protocol. FALSE indicates that +* guest should re-try with a different version. +* +* reservedz: Reserved field, set to zero. +*/ + +struct dm_version_response { +struct dm_header hdr; +uint64_t is_accepted:1 ; +uint64_t reservedz:63 ; +}QEMU_PACKED ; + +/* +* Message reporting capabilities. This is sent from the guest to the +* host. +*/ + +struct dm_capabilities { +struct dm_header hdr; +union dm_caps caps; +uint64_t min_page_cnt; +uint64_t max_page_number; +}QEMU_PACKED ; + +/* +* Response to the capabilities message. This is sent from the host to the +* guest. This message notifies if the host has accepted the guest's +* capabilities. If the host has not accepted, the guest must shutdown +* the service. +* +* is_accepted: Indicates if the host has accepted guest's capabilities. +* reservedz: Must be 0. +*/ + +struct dm_capabilities_resp_msg { +struct dm_header hdr; +uint64_t is_accepted:1 ; +uint64_t hot_remove:1 ; +uint64_t suppress_pressure_reports:1 ; +uint64_t reservedz:61 ; +}QEMU_PACKED ; + +/* +* This message is used to report memory pressure from the guest. +* This message is not part of any transaction and there is no +* response to this message. +* +* num_avail: Available memory in pages. +* num_committed: Committed memory in pages. +* page_file_size: The accumulated size of all page files +* in the system in pages. +* zero_free: The nunber of zero and free pages. +* page_file_writes: The writes to the page file in pages. +* io_diff: An indicator of file cache efficiency or page file activity, +* calculated as File Cache Page Fault Count - Page Read Count. +* This value is in pages. +* +* Some of these metrics are Windows specific and fortunately +* the algorithm on the host side that computes the guest memory +* pressure only uses num_committed value. +*/ + +struct dm_status { +struct dm_header hdr; +uint64_t num_avail; +uint64_t num_committed; +uint64_t page_file_size; +uint64_t zero_free; +uint32_t page_file_writes; +uint32_t io_diff; +}QEMU_PACKED ; + + +/* +* Message to ask the guest to allocate memory - balloon up message. +* This message is sent from the host to the guest. The guest may not be +* able to allocate as much memory as requested. +* +* num_pages: number of pages to allocate. +*/ + +struct dm_balloon { +struct dm_header hdr; +uint32_t num_pages; +uint32_t reservedz; +}QEMU_PACKED ; + + +/* +* Balloon response message; this message is sent from the guest +* to the host in response to the balloon message. +* +* reservedz: Reserved; must be set to zero. +* more_pages: If FALSE, this is the last message of the transaction. +* if TRUE there will atleast one more message from the guest. +* +* range_count: The number of ranges in the range array. +* +* range_array: An array of page ranges returned to the host. +* +*/ + +struct dm_balloon_response { +struct dm_header hdr; +uint32_t reservedz; +uint32_t more_pages:1 ; +uint32_t range_count:31 ; +union dm_mem_page_range range_array[]; +}QEMU_PACKED ; + +/* +* Un-balloon message; this message is sent from the host +* to the guest to give guest more memory. +* +* more_pages: If FALSE, this is the last message of the transaction. +* if TRUE there will atleast one more message from the guest. +* +* reservedz: Reserved; must be set to zero. +* +* range_count: The number of ranges in the range array. +* +* range_array: An array of page ranges returned to the host. +* +*/ + +struct dm_unballoon_request { +struct dm_header hdr; +uint32_t more_pages:1 ; +uint32_t reservedz:31 ; +uint32_t range_count; +union dm_mem_page_range range_array[]; +}QEMU_PACKED ; + +/* +* Un-balloon response message; this message is sent from the guest +* to the host in response to an unballoon request. +* +*/ + +struct dm_unballoon_response { +struct dm_header hdr; +}QEMU_PACKED ; + + +/* +* Hot add request message. Message sent from the host to the guest. +* +* mem_range: Memory range to hot add. +* +*/ + +struct dm_hot_add { +struct dm_header hdr; +union dm_mem_page_range range; +}QEMU_PACKED ; + +/* +* Hot add response message. +* This message is sent by the guest to report the status of a hot add request. +* If page_count is less than the requested page count, then the host should +* assume all further hot add requests will fail, since this indicates that +* the guest has hit an upper physical memory barrier. +* +* Hot adds may also fail due to low resources; in this case, the guest must +* not complete this message until the hot add can succeed, and the host must +* not send a new hot add request until the response is sent. +* If VSC fails to hot add memory DYNMEM_NUMBER_OF_UNSUCCESSFUL_HOTADD_ATTEMPTS +* times it fails the request. +* +* +* page_count: number of pages that were successfully hot added. +* +* result: result of the operation 1: success, 0: failure. +* +*/ + +struct dm_hot_add_response { +struct dm_header hdr; +uint32_t page_count; +uint32_t result; +}QEMU_PACKED ; + +struct dm_hot_remove { +struct dm_header hdr; +uint32_t virtual_node; +uint32_t page_count; +uint32_t qos_flags; +uint32_t reservedZ; +}QEMU_PACKED ; + +struct dm_hot_remove_response { +struct dm_header hdr; +uint32_t result; +uint32_t range_count; +uint64_t more_pages:1 ; +uint64_t reservedz:63 ; +union dm_mem_page_range range_array[]; +}QEMU_PACKED ; + +#define DM_REMOVE_QOS_LARGE (1 << 0) +#define DM_REMOVE_QOS_LOCAL (1 << 1) +#define DM_REMOVE_QOS_MASK (0x3) + +/* +* Types of information sent from host to the guest. +*/ + +enum dm_info_type { +INFO_TYPE_MAX_PAGE_CNT =0 , +MAX_INFO_TYPE +}; + + +/* +* Header for the information message. +*/ + +struct dm_info_header { +enum dm_info_type type; +uint32_t data_size; +uint8_t data[]; +}QEMU_PACKED ; + +/* +* This message is sent from the host to the guest to pass +* some relevant information (win8 addition). +* +* reserved: no used. +* info_size: size of the information blob. +* info: information blob. +*/ + +struct dm_info_msg { +struct dm_header hdr; +uint32_t reserved; +uint32_t info_size; +uint8_t info[]; +}; + +#endif
PATCH 3/3 Add a Hyper-V Dynamic Memory Protocol driver (hv-balloon)
In reply to Maciej S. SzmigieroFrom: "Maciej S. Szmigiero" <[email protected]>
This driver is like virtio-balloon on steroids: it allows both changing the guest memory allocation via ballooning and inserting extra RAM into it by adding haprot virtual DIMM sticks.
One of advantages of these over ACPI-based PC DIMM hotplug is that such memory can be hotplugged in much smaller granularity because the ACPI DIMM slot limit does not apply.
In contrast with ACPI DIMM hotplug where one can only request to unplug a whole DIMM stick this driver allows removing memory from guest in single page (4k) units via ballooning.
Then, once the guest has released the whole memory backed by a haprot virtual DIMM stick such device is marked "unused" and can be removed from the VM, if one wants so.
A "HV_BALLOON_HAPROT_UNUSED" QMP event is emitted in this case so the software controlling QEMU knows that this operation is now possible.
The haprot devices are also marked unused after a VM reboot (with a corresponding "HV_BALLOON_HAPROT_UNUSED" QMP event).
They are automatically reinserted (if still present) after the guest reconnects to this protocol (a "HV_BALLOON_HAPROT_INUSE" QMP event is then emitted).
For performance reasons, the guest-released memory is tracked in few range trees, as a series of (start, count) ranges.
Each time a new page range is inserted into such tree its neighbors are checked as candidates for possible merging with it.
Besides performance reasons, the Dynamic Memory protocol itself uses page ranges as the data structure in its messages, so relevant pages need to be merged into such ranges anyway.
One has to be careful when tracking the guest-released pages, since the guest can maliciously report returning pages outside its current address space, which later clash with the address range of newly added memory. Similarly, the guest can report freeing the same page twice.
The above design results in much better ballooning performance than when using virtio-balloon with the same guest: 230 GB / minute with this driver versus 70 GB / minute with virtio-balloon.
During a ballooning operation most of time is spent waiting for the guest to come up with newly freed page ranges, processing the received ranges on the host side (in QEMU / KVM) is nearly instantaneous.
The unballoon operation is also pretty much instantaneous:
thanks to the merging of the ballooned out page ranges 200 GB of memory can be returned to the guest in about 1 second.
With virtio-balloon this operation takes about 2.5 minutes.
These tests were done against a Windows Server 2019 guest running on a Xeon E5-2699, after dirtying the whole memory inside guest before each balloon operation.
Using a range tree instead of a bitmap to track the removed memory also means that the solution scales well with the guest size: even a 1 TB range takes just few bytes of memory.
Since the required GTree operations aren't present in every Glib version a check for them was added to "configure" script, together with new "--enable-hv-balloon" and "--disable-hv-balloon" arguments. If these GTree operations are missing in the system Glib this driver will be skipped during QEMU build.
An optional "status-report=on" device parameter requests memory status events from the guest (typically sent every second), which allow the host to learn both the guest memory available and the guest memory in use counts.
They are emitted externally as "HV_BALLOON_STATUS_REPORT" QMP events.
The driver is named hv-balloon since the Linux kernel client driver for the Dynamic Memory Protocol is named as such and to follow the naming pattern established by the virtio-balloon driver.
The whole protocol runs over Hyper-V VMBus that has its implementation recently merged in.
The driver was tested against Windows Server 2012 R2, Windows Server 2016 and Windows Server 2016 guests and obeys the guest alignment requirements reported to the host via DM_CAPABILITIES_REPORT message.
--- Kconfig.host | 3 + configure | 35 + hw/hyperv/Kconfig | 5 + hw/hyperv/hv-balloon.c | 2172 ++++++++++++++++++++++++++++++++++++++++ hw/hyperv/meson.build | 1 + hw/hyperv/trace-events | 17 + meson.build | 2 + qapi/misc.json | 74 ++ 8 files changed, 2309 insertions(+) create mode 100644 hw/hyperv/hv-balloon.c
Kconfig.host +3 -0
@@ -33,3 +33,6 @@ config VIRTFS config PVRDMA bool + +config HV_BALLOON_POSSIBLE + bool
configure +35 -0
@@ -543,6 +543,7 @@ fuzzing="no" rng_none="no" secret_keyring="" libdaxctl="" +hv_balloon="" meson="" ninja="" skip_meson=no
@@ -1628,6 +1629,10 @@ for opt do ;; --disable-libdaxctl) libdaxctl=no ;; + --enable-hv-balloon) hv_balloon=yes + ;; + --disable-hv-balloon) hv_balloon=no + ;; *) echo "ERROR: unknown option $opt" echo "Try '$0 --help' for more information"
@@ -1949,6 +1954,7 @@ disabled with --disable-FEATURE, default is enabled if available: xkbcommon xkbcommon support rng-none dummy RNG, avoid using /dev/(u)random and getrandom() libdaxctl libdaxctl support + hv-balloon hv-balloon driver where supported (requires extended GTree API) NOTE: The object files are built at the place where configure is launched EOF
@@ -6119,6 +6125,32 @@ if test "$libdaxctl" != "no"; then fi fi +########################################## +# check for hv-balloon + +if test "$hv_balloon" != "no"; then + cat > $TMPC << EOF +#include <string.h> +#include <gmodule.h> +int main(void) { + GTree *tree; + + tree = g_tree_new((GCompareFunc)strcmp); + (void)g_tree_node_first(tree); + g_tree_destroy(tree); + return 0; +} +EOF + if compile_prog "$glib_cflags" "$glib_libs" ; then + hv_balloon=yes + else + if test "$hv_balloon" = "yes" ; then + feature_not_found "hv-balloon" "Update Glib" + fi + hv_balloon="no" + fi +fi + ########################################## # check for slirp
@@ -7352,6 +7384,9 @@ fi if test "$sheepdog" = "yes" ; then echo "CONFIG_SHEEPDOG=y" >> $config_host_mak fi +if test "$hv_balloon" = "yes" ; then + echo "CONFIG_HV_BALLOON_POSSIBLE=y" >> $config_host_mak +fi if test "$pty_h" = "yes" ; then echo "HAVE_PTY_H=y" >> $config_host_mak fi
hw/hyperv/Kconfig +5 -0
@@ -11,3 +11,8 @@ config VMBUS bool default y depends on HYPERV + +config HV_BALLOON + bool + default y + depends on HV_BALLOON_POSSIBLE && VMBUS && HAPROT
hw/hyperv/hv-balloon.c new file +2172 -0
2,172 lines not on this page · raw message
hw/hyperv/meson.build +1 -0
@@ -1,3 +1,4 @@ specific_ss.add(when: 'CONFIG_HYPERV', if_true: files('hyperv.c')) specific_ss.add(when: 'CONFIG_HYPERV_TESTDEV', if_true: files('hyperv_testdev.c')) specific_ss.add(when: 'CONFIG_VMBUS', if_true: files('vmbus.c')) +specific_ss.add(when: 'CONFIG_HV_BALLOON', if_true: files('hv-balloon.c'))
hw/hyperv/trace-events +17 -0
@@ -16,3 +16,20 @@ vmbus_gpadl_torndown(uint32_t gpadl_id) "gpadl #%d" vmbus_open_channel(uint32_t chan_id, uint32_t gpadl_id, uint32_t target_vp) "channel #%d gpadl #%d target vp %d" vmbus_channel_open(uint32_t chan_id, uint32_t status) "channel #%d status %d" vmbus_close_channel(uint32_t chan_id) "channel #%d" + +# hv-balloon +hv_balloon_state_change(const char *tostr) "-> %s" +hv_balloon_incoming_version(uint16_t major, uint16_t minor) "incoming proto version %u.%u" +hv_balloon_incoming_caps(uint32_t caps) "incoming caps 0x%x" +hv_balloon_outgoing_unballoon(uint32_t trans_id, uint64_t count, uint64_t start, uint64_t rempages) "posting unballoon %"PRIu32" for %"PRIu64" @ %"PRIu64", remaining %"PRIu64 +hv_balloon_incoming_unballoon(uint32_t trans_id) "incoming unballoon response %"PRIu32 +hv_balloon_outgoing_hot_add(uint32_t trans_id, uint64_t count, uint64_t start) "posting hot add %"PRIu32" for %"PRIu64" @ %"PRIu64 +hv_balloon_incoming_hot_add(uint32_t trans_id, uint32_t result, uint32_t count) "incoming hot add response %"PRIu32", result %"PRIu32", count %"PRIu32 +hv_balloon_outgoing_balloon(uint32_t trans_id, uint64_t count, uint64_t rempages) "posting balloon %"PRIu32" for %"PRIu64", remaining %"PRIu64 +hv_balloon_incoming_balloon(uint32_t trans_id, uint32_t range_count, uint32_t more_pages) "incoming balloon response %"PRIu32", ranges %"PRIu32", more %"PRIu32 +hv_balloon_haprot_range_add(uint64_t count, uint64_t start) "adding haprot range %"PRIu64" @ %"PRIu64 +hv_balloon_haprot_range_remove(uint64_t count, uint64_t start, uint64_t removed_guest_range, uint64_t removed_both_range, uint64_t removed_guest, uint64_t removed_both) "removing haprot range %"PRIu64" @ %"PRIu64" counts (g %"PRIu64", b %"PRIu64"), global counts (g %"PRIu64", b %"PRIu64")" +hv_balloon_remove_response(uint64_t count, uint64_t start, unsigned int both) "processing remove response range %"PRIu64" @ %"PRIu64", both %u" +hv_balloon_remove_response_hole(uint64_t counthole, uint64_t starthole, uint64_t countrange, uint64_t startrange, uint64_t starthpr, unsigned int both) "response range hole %"PRIu64" @ %"PRIu64" from range %"PRIu64" @ %"PRIu64", before hpr start %"PRIu64", both %u" +hv_balloon_remove_response_common(uint64_t countcommon, uint64_t startcommon, uint64_t countrange, uint64_t startrange, uint64_t counthpr, uint64_t starthpr, uint64_t removed, unsigned int both) "response common range %"PRIu64" @ %"PRIu64" from range %"PRIu64" @ %"PRIu64" with hpr %"PRIu64" @ %"PRIu64", removed %"PRIu64", both %u" +hv_balloon_remove_response_remainder(uint64_t count, uint64_t start, unsigned int both) "remove response remaining range %"PRIu64" @ %"PRIu64", both %u"
meson.build +2 -0
@@ -525,6 +525,7 @@ kconfig_external_symbols = [ 'CONFIG_VIRTFS', 'CONFIG_LINUX', 'CONFIG_PVRDMA', + 'CONFIG_HV_BALLOON_POSSIBLE', ] ignored = ['TARGET_XML_FILES', 'TARGET_ABI_DIR', 'TARGET_DIRS']
@@ -1525,6 +1526,7 @@ endif summary_info += {'thread sanitizer': config_host.has_key('CONFIG_TSAN')} summary_info += {'rng-none': config_host.has_key('CONFIG_RNG_NONE')} summary_info += {'Linux keyring': config_host.has_key('CONFIG_SECRET_KEYRING')} +summary_info += {'hv-balloon support': config_host.has_key('CONFIG_HV_BALLOON_POSSIBLE')} summary(summary_info, bool_yn: true) if not supported_cpus.contains(cpu)
qapi/misc.json +74 -0
80 lines not on this page · raw message
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series seems to have some coding style problems. See output below for more information:
Type: series
Message-id: [email protected]
Subject: [PATCH 0/3] Hyper-V Dynamic Memory Protocol driver (hv-balloon)
=== TEST SCRIPT BEGIN ===
#!/bin/bash
git rev-parse base > /dev/null || exit 0
git config --local diff.renamelimit 0
git config --local diff.renames True
git config --local diff.algorithm histogram./scripts/checkpatch.pl --mailback base..
=== TEST SCRIPT END ===
Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384 Switched to a new branch 'test'
180104d Add a Hyper-V Dynamic Memory Protocol driver (hv-balloon) cb9f9d7 Add Hyper-V Dynamic Memory Protocol definitions 3f22e8f haprot: add a device for memory hot-add protocols
=== OUTPUT BEGIN ===
1/3 Checking commit 3f22e8fb4473 (haprot: add a device for memory hot-add protocols)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating? #109:
new file mode 100644
WARNING: line over 80 characters
#212: FILE: hw/mem/haprot.c:99:
+ "Node property value %"PRIu32" exceeds the number of numa nodes %d",
total: 0 errors, 2 warnings, 404 lines checked
Patch 1/3 has style problems, please review. If any of these errors are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
2/3 Checking commit cb9f9d7b2fb4 (Add Hyper-V Dynamic Memory Protocol definitions)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating? #15:
new file mode 100644
WARNING: line over 80 characters
#52: FILE: include/hw/hyperv/dynmem-proto.h:33:
+#define DYNMEM_MAKE_VERSION(Major, Minor) ((uint32_t)(((Major) << 16) | (Minor)))
ERROR: space prohibited after that '&' (ctx:WxW)
#54: FILE: include/hw/hyperv/dynmem-proto.h:35:
+#define DYNMEM_MINOR_VERSION(Version) ((uint32_t)(Version) & 0xff)
^
total: 1 errors, 2 warnings, 425 lines checked
Patch 2/3 has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS.
3/3 Checking commit 180104d55786 (Add a Hyper-V Dynamic Memory Protocol driver (hv-balloon))
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating? #188:
new file mode 100644
WARNING: line over 80 characters
#248: FILE: hw/hyperv/hv-balloon.c:56:
+#define HV_BALLOON_HA_CHUNK_PAGES (HV_BALLOON_HA_CHUNK_SIZE / HV_BALLOON_PAGE_SIZE)
WARNING: line over 80 characters
#1236: FILE: hw/hyperv/hv-balloon.c:1044:
+ (1 << balloon->caps.cap_bits.hot_add_alignment)
WARNING: line over 80 characters
#1618: FILE: hw/hyperv/hv-balloon.c:1426:
+ warn_report("guest reported removed page backed by unsupported page size %zu",
WARNING: line over 80 characters
#1692: FILE: hw/hyperv/hv-balloon.c:1500:
+ warn_report("guest reported removed page %"PRIu64" not found in RAM",
ERROR: line over 90 characters
#1700: FILE: hw/hyperv/hv-balloon.c:1508:
+ warn_report("guest reported removed page %"PRIu64" in a section smaller than page size",
ERROR: line over 90 characters
#1710: FILE: hw/hyperv/hv-balloon.c:1518:
+ warn_report("guest reported removed page %"PRIu64" in a section that is not an ordinary RAM",
ERROR: line over 90 characters
#1749: FILE: hw/hyperv/hv-balloon.c:1557:
+ warn_report("guest reported more pages removed than currently pending (%"PRIu64" vs %"PRIu64")",
WARNING: line over 80 characters
#1863: FILE: hw/hyperv/hv-balloon.c:1671:
+ qapi_event_send_hv_balloon_status_report((uint64_t)msgStatus->num_committed *
ERROR: line over 90 characters
#1918: FILE: hw/hyperv/hv-balloon.c:1726:
+ warn_report("DM_HOT_ADD_RESPONSE page count higher than requested (%"PRIu32" vs %"PRIu64")",
total: 4 errors, 6 warnings, 2369 lines checked
Patch 3/3 has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS.
=== OUTPUT END ===
Test command exited with code: 1
The full log is available at
http://patchew.org/logs/[email protected]/testing.checkpatch/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]
Quoted 1 line · On Sun, 20 Sep 2020 15:25:19 +0200 "Maciej S. Szmigiero"…
From: "Maciej S. Szmigiero" <[email protected]>
From description it sounds like an alternative of virtio-mem, CCing David.
Quoted 168 lines
This series adds a Hyper-V Dynamic Memory Protocol driver (hv-balloon) and its protocol definitions.
Also included is a driver providing backing devices for memory hot-add protocols ("haprots").A haprot device works like a virtual DIMM stick: it allows inserting extra RAM into the guest at run time.
The main differences from the ACPI-based PC DIMM hotplug are:
* Notifying the guest about the new memory range is not done via ACPI but via a protocol handler that registers with the haprot framework. This means that the ACPI DIMM slot limit does not apply.* A protocol handler can prevent removal of a haprot device when it is still in use by setting its "busy" field.
* A protocol handler can also register an "unplug" callback so it gets notified when an user decides to remove the haprot device.
This way the protocol handler can inform the guest about this fact and / or do its own cleanup.The hv-balloon driver is like virtio-balloon on steroids: it allows both changing the guest memory allocation via ballooning and inserting extra RAM into it by adding haprot virtual DIMM sticks.
One of advantages of these over ACPI-based PC DIMM hotplug is that such memory can be hotplugged in much smaller granularity because the ACPI DIMM slot limit does not apply.In contrast with ACPI DIMM hotplug where one can only request to unplug a whole DIMM stick this driver allows removing memory from guest in single page (4k) units via ballooning.
Then, once the guest has released the whole memory backed by a haprot virtual DIMM stick such device is marked "unused" and can be removed from the VM, if one wants so.
A "HV_BALLOON_HAPROT_UNUSED" QMP event is emitted in this case so the software controlling QEMU knows that this operation is now possible.The haprot devices are also marked unused after a VM reboot (with a corresponding "
HV_BALLOON_HAPROT_UNUSED" QMP event).
They are automatically reinserted (if still present) after the guest reconnects to this protocol (a "HV_BALLOON_HAPROT_INUSE" QMP event is then emitted).For performance reasons, the guest-released memory is tracked in few range trees, as a series of (start, count) ranges.
Each time a new page range is inserted into such tree its neighbors are checked as candidates for possible merging with it.Besides performance reasons, the Dynamic Memory protocol itself uses page ranges as the data structure in its messages, so relevant pages need to be merged into such ranges anyway.
One has to be careful when tracking the guest-released pages, since the guest can maliciously report returning pages outside its current address space, which later clash with the address range of newly added memory. Similarly, the guest can report freeing the same page twice.
The above design results in much better ballooning performance than when using virtio-balloon with the same guest: 230 GB / minute with this driver versus 70 GB / minute with virtio-balloon.
During a ballooning operation most of time is spent waiting for the guest to come up with newly freed page ranges, processing the received ranges on the host side (in QEMU / KVM) is nearly instantaneous.
The unballoon operation is also pretty much instantaneous:
thanks to the merging of the ballooned out page ranges 200 GB of memory can be returned to the guest in about 1 second.
With virtio-balloon this operation takes about 2.5 minutes.These tests were done against a Windows Server 2019 guest running on a Xeon E5-2699, after dirtying the whole memory inside guest before each balloon operation.
Using a range tree instead of a bitmap to track the removed memory also means that the solution scales well with the guest size: even a 1 TB range takes just few bytes of memory.
The required GTree operations are available at
https://gitlab.gnome.org/maciejsszmigiero/glib/-/tree/gtree-add-iterators
Since they are not yet in the upstream Glib a check for them was added to "configure" script, together with new "--enable-hv-balloon" and "--disable-hv-balloon" arguments.
If these GTree operations are missing in the system Glib this driver will be skipped during QEMU build.An optional "status-report=on" device parameter requests memory status events from the guest (typically sent every second), which allow the host to learn both the guest memory available and the guest memory in use counts.
They are emitted externally as "HV_BALLOON_STATUS_REPORT" QMP events.The driver is named hv-balloon since the Linux kernel client driver for the Dynamic Memory Protocol is named as such and to follow the naming pattern established by the virtio-balloon driver.
The whole protocol runs over Hyper-V VMBus that has its implementation recently merged in.The driver was tested against Windows Server 2012 R2, Windows Server 2016 and Windows Server 2016 guests and obeys the guest alignment requirements reported to the host via
DM_CAPABILITIES_REPORTmessage. Extensive event tracing is available under 'hv_balloon_*' prefix.Example usage:
* Add "-device vmbus-bridge,id=vmbus-bridge -device hv-balloon,id=hvb" to the QEMU command line and set "maxmem" value to something large, like 1T.* Use QEMU monitor commands to add a haprot virtual DIMM stick, together with its memory backend:
object_addmemory-backend-ram,id=mem1,size=200G
device_addmem-haprot,id=ha1,memdev=mem1
The first command is actually the same as for ACPI-based DIMM hotplug.* Use the ballooning interface monitor commands to force the guest to give out as much memory as possible:
balloon 1
The ballooning interface monitor commands can also be used to resize the guest up and down appropriately.* One can check the current guest size by issuing a "info balloon" command. This is useful to know what is happening, since large ballooning or unballooning operations take some time to complete.
* Once the guest releases the whole memory backed by a haprot device (or is restarted) a "
HV_BALLOON_HAPROT_UNUSED" QMP event will be generated.
The haprot device then can be removed, together with its memory backend:
device_delha1
object_delmem1Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.* Allow new haport devices to reuse the same address range as the ones that were previously deleted via
device_delmonitor command without having to restart the VM.* Add vmstate / live migration support to the hv-balloon driver.
* Use haprot device to also add memory via virtio interface (this requires defining a new operation in virtio-balloon protocol and appropriate support from the client virtio-balloon driver in the Linux kernel).
Kconfig.host | 3 + configure | 35 + hw/hyperv/Kconfig | 5 + hw/hyperv/hv-balloon.c | 2172 ++++++++++++++++++++++++++++++ hw/hyperv/meson.build | 1 + hw/hyperv/trace-events | 17 + hw/i386/Kconfig | 2 + hw/i386/pc.c | 18 +- hw/mem/Kconfig | 4 + hw/mem/haprot.c | 263 ++++ hw/mem/meson.build | 1 + include/hw/hyperv/dynmem-proto.h | 425 ++++++ include/hw/mem/haprot.h | 72 + meson.build | 2 + qapi/misc.json | 74 + 15 files changed, 3093 insertions(+), 1 deletion(-) create mode 100644 hw/hyperv/hv-balloon.c create mode 100644 hw/mem/haprot.c create mode 100644 include/hw/hyperv/dynmem-proto.h create mode 100644 include/hw/mem/haprot.h
Quoted 7 lines · Igor Mammedov
Quoted 1 line · On Sun, 20 Sep 2020 15:25:19 +0200 "Maciej S. Szmigiero"…
From: "Maciej S. Szmigiero" <[email protected]>
From description it sounds like an alternative of virtio-mem, CCing David.
Hah! Was already replying when your cc came in (thanks anyway!).
Not quite an alternative to virtio-mem, more like a (in some factors) optimized alternative to DIMM-based memory hotplug + virtio-balloon.
The core design of virtio-mem avoids basic issues related to memory ballooning. (it was the result of analyzing existing technologies and their shortcomings - including HV dynamic memory).
Signature · 3 lines
Thanks, David / dhildenb
Quoted 23 lines · Maciej S. Szmigiero
From: "Maciej S. Szmigiero" <[email protected]>
This series adds a Hyper-V Dynamic Memory Protocol driver (hv-balloon) and its protocol definitions.
Also included is a driver providing backing devices for memory hot-add protocols ("haprots").A haprot device works like a virtual DIMM stick: it allows inserting extra RAM into the guest at run time.
The main differences from the ACPI-based PC DIMM hotplug are:
* Notifying the guest about the new memory range is not done via ACPI but via a protocol handler that registers with the haprot framework. This means that the ACPI DIMM slot limit does not apply.* A protocol handler can prevent removal of a haprot device when it is still in use by setting its "busy" field.
* A protocol handler can also register an "unplug" callback so it gets notified when an user decides to remove the haprot device.
This way the protocol handler can inform the guest about this fact and / or do its own cleanup.
Quoted 6 lines
The hv-balloon driver is like virtio-balloon on steroids: it allows both changing the guest memory allocation via ballooning and inserting extra RAM into it by adding haprot virtual DIMM sticks.
One of advantages of these over ACPI-based PC DIMM hotplug is that such memory can be hotplugged in much smaller granularity because the ACPI DIMM slot limit does not apply.
Reading further below, it's essentially DIMM-based memory hotplug + virtio-balloon - except the 256MB DIMM limit. But reading below, I don't see how you want to avoid the KVM memory slot limit that's in a similar size (I recall 256*2 due to 2 address spaces). Or avoid VMA limits when wanting to grow a VM big in very tiny steps over time (e.g., adding 64MB at a time).
Quoted 25 lines
In contrast with ACPI DIMM hotplug where one can only request to unplug a whole DIMM stick this driver allows removing memory from guest in single page (4k) units via ballooning.
Then, once the guest has released the whole memory backed by a haprot virtual DIMM stick such device is marked "unused" and can be removed from the VM, if one wants so.
A "HV_BALLOON_HAPROT_UNUSED" QMP event is emitted in this case so the software controlling QEMU knows that this operation is now possible.The haprot devices are also marked unused after a VM reboot (with a corresponding "
HV_BALLOON_HAPROT_UNUSED" QMP event).
They are automatically reinserted (if still present) after the guest reconnects to this protocol (a "HV_BALLOON_HAPROT_INUSE" QMP event is then emitted).For performance reasons, the guest-released memory is tracked in few range trees, as a series of (start, count) ranges.
Each time a new page range is inserted into such tree its neighbors are checked as candidates for possible merging with it.Besides performance reasons, the Dynamic Memory protocol itself uses page ranges as the data structure in its messages, so relevant pages need to be merged into such ranges anyway.
Quoted 8 lines
One has to be careful when tracking the guest-released pages, since the guest can maliciously report returning pages outside its current address space, which later clash with the address range of newly added memory. Similarly, the guest can report freeing the same page twice.
The above design results in much better ballooning performance than when using virtio-balloon with the same guest: 230 GB / minute with this driver versus 70 GB / minute with virtio-balloon.
I assume these numbers apply with Windows guests only. IIRC Linux hv_balloon does not support page migration/compaction, while virtio-balloon does. So you might end up with quite some fragmented memory with hv_balloon in Linux guests - of course, usually only in corner cases.
Quoted 28 lines
During a ballooning operation most of time is spent waiting for the guest to come up with newly freed page ranges, processing the received ranges on the host side (in QEMU / KVM) is nearly instantaneous.
The unballoon operation is also pretty much instantaneous:
thanks to the merging of the ballooned out page ranges 200 GB of memory can be returned to the guest in about 1 second.
With virtio-balloon this operation takes about 2.5 minutes.These tests were done against a Windows Server 2019 guest running on a Xeon E5-2699, after dirtying the whole memory inside guest before each balloon operation.
Using a range tree instead of a bitmap to track the removed memory also means that the solution scales well with the guest size: even a 1 TB range takes just few bytes of memory.
Example usage:
* Add "-device vmbus-bridge,id=vmbus-bridge -device hv-balloon,id=hvb" to the QEMU command line and set "maxmem" value to something large, like 1T.* Use QEMU monitor commands to add a haprot virtual DIMM stick, together with its memory backend:
object_addmemory-backend-ram,id=mem1,size=200G
device_addmem-haprot,id=ha1,memdev=mem1
The first command is actually the same as for ACPI-based DIMM hotplug.
Quoted 3 lines
* Use the ballooning interface monitor commands to force the guest to give out as much memory as possible:
balloon 1
At least under virtio-balloon with Linux, that will pretty sure trigger a guest crash. Is something like that expected to work with Windows guests reasonably well?
Quoted 6 lines
The ballooning interface monitor commands can also be used to resize the guest up and down appropriately.
* One can check the current guest size by issuing a "info balloon" command. This is useful to know what is happening, since large ballooning or unballooning operations take some time to complete.
So, every time you want to add more memory (after the balloon was deflated) to a guest, you have to plug a new mem-haprot device, correct?
So your QEMU user has to be well aware of how to balance "balloon" and "object_add/device_add/object_del_device_del" commands to achieve the desired guest size.
Quoted 7 lines
* Once the guest releases the whole memory backed by a haprot device (or is restarted) a "
HV_BALLOON_HAPROT_UNUSED" QMP event will be generated.
The haprot device then can be removed, together with its memory backend:
device_delha1
object_delmem1
So, you rely on some external entity to properly shrink a guest again (e.g., during reboot).
Quoted 5 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.
Yeah, that makes sense. Only one at a time.
Quoted 10 lines
* Allow new haport devices to reuse the same address range as the ones that were previously deleted via
device_delmonitor command without having to restart the VM.* Add vmstate / live migration support to the hv-balloon driver.
* Use haprot device to also add memory via virtio interface (this requires defining a new operation in virtio-balloon protocol and appropriate support from the client virtio-balloon driver in the Linux kernel).
Most probably not the direction we are going to take. We have virtio-mem for clean, fine-grained, NUMA-aware, paravirtualized memory hot(un)plug now, and we are well aware of various issues with (base-page size based) memory ballooning that are fairly impossible to solve (especially in the context of vfio).
Signature · 3 lines
Thanks, David / dhildenb
Hi David,
Thank you for your comments.
First, I want to underline that this driver targets Windows guests, where ability to modify and adapt the guest memory management code is extremely limited.
While it does work with Linux guests, too, this is definitely not its native environment.
It also has to support rather big guests, up to 1 TB of RAM, so performance-related things are important.
Further answers are bellow.
Quoted 31 lines · David Hildenbrand
Quoted 23 lines · Maciej S. Szmigiero
From: "Maciej S. Szmigiero" <[email protected]>
This series adds a Hyper-V Dynamic Memory Protocol driver (hv-balloon) and its protocol definitions.
Also included is a driver providing backing devices for memory hot-add protocols ("haprots").A haprot device works like a virtual DIMM stick: it allows inserting extra RAM into the guest at run time.
The main differences from the ACPI-based PC DIMM hotplug are:
* Notifying the guest about the new memory range is not done via ACPI but via a protocol handler that registers with the haprot framework. This means that the ACPI DIMM slot limit does not apply.* A protocol handler can prevent removal of a haprot device when it is still in use by setting its "busy" field.
* A protocol handler can also register an "unplug" callback so it gets notified when an user decides to remove the haprot device.
This way the protocol handler can inform the guest about this fact and / or do its own cleanup.Quoted 6 lines
The hv-balloon driver is like virtio-balloon on steroids: it allows both changing the guest memory allocation via ballooning and inserting extra RAM into it by adding haprot virtual DIMM sticks.
One of advantages of these over ACPI-based PC DIMM hotplug is that such memory can be hotplugged in much smaller granularity because the ACPI DIMM slot limit does not apply.
Quoted 4 lines
Reading further below, it's essentially DIMM-based memory hotplug + virtio-balloon - except the 256MB DIMM limit. But reading below, I don't see how you want to avoid the KVM memory slot limit that's in a similar size (I recall 256*2 due to 2 address spaces).
The idea is to use virtual DIMM sticks for hot-adding extra memory at runtime, while using ballooning for runtime adjustment of the guest memory size within the current maximum.
When the guest is rebooted the virtual DIMMs configuration is adjusted by the software controlling QEMU (some are removed and / or some are added) to give the guest the same effective memory size as it had before the reboot.
So, yes, it will be a problem if the user expands their running guest ~256 times, each time making it even bigger than previously, without rebooting it even once, but this does seem to be an edge use case.
In the future it would be better to automatically turn the current effective guest size into its boot memory size when the VM restarts (the VM will then have no virtual DIMMs inserted after a reboot), but doing this requires quite a few changes to QEMU, that's why it isn't there yet.
The above is basically how Hyper-V hypervisor handles its memory size changes and it seems to be as close to having a transparently resizable guest as reasonably possible.
Quoted 2 lines
Or avoid VMA limits when wanting to grow a VM big in very tiny steps over time (e.g., adding 64MB at a time).
Not sure if you are taking about VMA limits inside the host or the guest.
Quoted 34 lines
Quoted 25 lines
In contrast with ACPI DIMM hotplug where one can only request to unplug a whole DIMM stick this driver allows removing memory from guest in single page (4k) units via ballooning.
Then, once the guest has released the whole memory backed by a haprot virtual DIMM stick such device is marked "unused" and can be removed from the VM, if one wants so.
A "HV_BALLOON_HAPROT_UNUSED" QMP event is emitted in this case so the software controlling QEMU knows that this operation is now possible.The haprot devices are also marked unused after a VM reboot (with a corresponding "
HV_BALLOON_HAPROT_UNUSED" QMP event).
They are automatically reinserted (if still present) after the guest reconnects to this protocol (a "HV_BALLOON_HAPROT_INUSE" QMP event is then emitted).For performance reasons, the guest-released memory is tracked in few range trees, as a series of (start, count) ranges.
Each time a new page range is inserted into such tree its neighbors are checked as candidates for possible merging with it.Besides performance reasons, the Dynamic Memory protocol itself uses page ranges as the data structure in its messages, so relevant pages need to be merged into such ranges anyway.
Quoted 8 lines
One has to be careful when tracking the guest-released pages, since the guest can maliciously report returning pages outside its current address space, which later clash with the address range of newly added memory. Similarly, the guest can report freeing the same page twice.
The above design results in much better ballooning performance than when using virtio-balloon with the same guest: 230 GB / minute with this driver versus 70 GB / minute with virtio-balloon.
Quoted 5 lines
I assume these numbers apply with Windows guests only. IIRC Linux
hv_balloondoes not support page migration/compaction, while virtio-balloon does. So you might end up with quite some fragmented memory withhv_balloonin Linux guests - of course, usually only in corner cases.
As I previously mentioned, this driver targets mainly Windows guests.
And Windows seems to be rather determined to free the requested number of pages: waiting for the guest to reply to a 2GB balloon request sometimes takes 2-3 seconds.
So i guess it does some kind of memory compaction during that request processing time.
Quoted 28 lines
Quoted 28 lines
During a ballooning operation most of time is spent waiting for the guest to come up with newly freed page ranges, processing the received ranges on the host side (in QEMU / KVM) is nearly instantaneous.
The unballoon operation is also pretty much instantaneous:
thanks to the merging of the ballooned out page ranges 200 GB of memory can be returned to the guest in about 1 second.
With virtio-balloon this operation takes about 2.5 minutes.These tests were done against a Windows Server 2019 guest running on a Xeon E5-2699, after dirtying the whole memory inside guest before each balloon operation.
Using a range tree instead of a bitmap to track the removed memory also means that the solution scales well with the guest size: even a 1 TB range takes just few bytes of memory.
Example usage:
* Add "-device vmbus-bridge,id=vmbus-bridge -device hv-balloon,id=hvb" to the QEMU command line and set "maxmem" value to something large, like 1T.* Use QEMU monitor commands to add a haprot virtual DIMM stick, together with its memory backend:
object_addmemory-backend-ram,id=mem1,size=200G
device_addmem-haprot,id=ha1,memdev=mem1
The first command is actually the same as for ACPI-based DIMM hotplug.
Quoted 7 lines
Quoted 3 lines
* Use the ballooning interface monitor commands to force the guest to give out as much memory as possible:
balloon 1At least under virtio-balloon with Linux, that will pretty sure trigger a guest crash. Is something like that expected to work with Windows guests reasonably well?
Windows will generally leave some memory free when processing balloon requests, although the precise amount varies between few hundred MB to values like 1+ GB.
Usually it runs stable even with these few hundred MBs of free memory remaining but I have seen occasional crashes at shutdown time in this case (probably something critical failing to initialize due to the system running out of memory).
While the above command was just a quick example, I personally think it is the guest who should be enforcing a balloon floor since it is the guest that knows its internal memory requirements, not the host.
For this reason the hv_balloon client driver inside the Linux kernel implements its own, rough balloon floor - see compute_balloon_floor().
On the other hand, one can also argue that the user wish should be respected as much as possible.
Quoted 9 lines
Quoted 6 lines
The ballooning interface monitor commands can also be used to resize the guest up and down appropriately.
* One can check the current guest size by issuing a "info balloon" command. This is useful to know what is happening, since large ballooning or unballooning operations take some time to complete.
So, every time you want to add more memory (after the balloon was deflated) to a guest, you have to plug a new mem-haprot device, correct?
Yes.
Quoted 3 lines
So your QEMU user has to be well aware of how to balance "balloon" and "object_add/device_add/object_del_device_del" commands to achieve the desired guest size.
In this case the VM user does not interact directly with the QEMU process.
Rather, the user tells the software controlling QEMU (think: libvirt) their wish how large they want the guest to be and this software then does everything what is necessary to achieve such target and make it persistent across guest reboots.
Quoted 10 lines
Quoted 7 lines
* Once the guest releases the whole memory backed by a haprot device (or is restarted) a "
HV_BALLOON_HAPROT_UNUSED" QMP event will be generated.
The haprot device then can be removed, together with its memory backend:
device_delha1
object_delmem1So, you rely on some external entity to properly shrink a guest again (e.g., during reboot).
Yes.
Quoted 7 lines
Quoted 5 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.Yeah, that makes sense. Only one at a time.
Having only one *active* at a time makes sense, however it ultimately would be nice to be able to have them both inserted into a VM:
one for Windows guests and one for Linux ones.
Even though only one obviously would be active at the same time.
Quoted 11 lines
Quoted 10 lines
* Allow new haport devices to reuse the same address range as the ones that were previously deleted via
device_delmonitor command without having to restart the VM.* Add vmstate / live migration support to the hv-balloon driver.
* Use haprot device to also add memory via virtio interface (this requires defining a new operation in virtio-balloon protocol and appropriate support from the client virtio-balloon driver in the Linux kernel).
Quoted 6 lines
Most probably not the direction we are going to take. We have virtio-mem for clean, fine-grained, NUMA-aware, paravirtualized memory hot(un)plug now, and we are well aware of various issues with (base-page size based) memory ballooning that are fairly impossible to solve (especially in the context of vfio).
Thanks,
Maciej
Quoted 7 lines · Maciej S. Szmigiero
Hi David,
Thank you for your comments.
First, I want to underline that this driver targets Windows guests, where ability to modify and adapt the guest memory management code is extremely limited.
Yeah, I know the pain.
[...]
Quoted 9 lines
The idea is to use virtual DIMM sticks for hot-adding extra memory at runtime, while using ballooning for runtime adjustment of the guest memory size within the current maximum.
When the guest is rebooted the virtual DIMMs configuration is adjusted by the software controlling QEMU (some are removed and / or some are added) to give the guest the same effective memory size as it had before the reboot.
Okay, so while "the ACPI DIMM slot limit does not apply", the KVM memory slot limit (currently) applies, resulting in exactly the same behavior.
The only (conceptual difference) I am able to spot is then a notification to the user on reboot, so the guest memory layout can be adjusted (which I consider very ugly, but it's the same thing when mixing ballooning and DIMMs - which is why it's usually never done).
[...]
Quoted 4 lines
So, yes, it will be a problem if the user expands their running guest ~256 times, each time making it even bigger than previously, without rebooting it even once, but this does seem to be an edge use case.
IIRC, that's exactly what dynamic memory under Windows does in automatic mode, no? Monitor the guests, distribute memory accordingly - usually in smaller steps. But I am no expert on Hyper-V.
Quoted 6 lines
In the future it would be better to automatically turn the current effective guest size into its boot memory size when the VM restarts (the VM will then have no virtual DIMMs inserted after a reboot), but doing this requires quite a few changes to QEMU, that's why it isn't there yet.
Will most probably never happen as reshuffling the layout of your boot memory (especially with NUMA) within QEMU can break live migration in various ways.
If you already notify the user on a reboot, the user can just kill the VM and start it with an adjusted boot memory size. Yeah, that's ugly, but so is the whole "adjust DIMM/balloon configuration during a reboot from outside QEMU".
BTW, how would you handle: Start guest with 10G. Inflate balloon to 5G. Reboot. There are no virtual DIMMs to adjust.
Quoted 4 lines
The above is basically how Hyper-V hypervisor handles its memory size changes and it seems to be as close to having a transparently resizable guest as reasonably possible.
"having a transparently resizable _Windows_ guests right now" :)
Quoted 6 lines
Quoted 4 lines
Or avoid VMA limits when wanting to grow a VM big in very tiny steps over time (e.g., adding 64MB at a time).
Not sure if you are taking about VMA limits inside the host or the guest.
Host. one virtual DIMM corresponds to one VMA. But the KVM memory limit already applies before that, so it doesn't matter.
[...]
Quoted 7 lines
Quoted 5 lines
I assume these numbers apply with Windows guests only. IIRC Linux
hv_balloondoes not support page migration/compaction, while virtio-balloon does. So you might end up with quite some fragmented memory withhv_balloonin Linux guests - of course, usually only in corner cases.As I previously mentioned, this driver targets mainly Windows guests.
... and you cannot enforce that people will only use it with Windows guests :)
[...]
Quoted 4 lines
Windows will generally leave some memory free when processing balloon requests, although the precise amount varies between few hundred MB to values like 1+ GB.
Quoted 8 lines
Usually it runs stable even with these few hundred MBs of free memory remaining but I have seen occasional crashes at shutdown time in this case (probably something critical failing to initialize due to the system running out of memory).
While the above command was just a quick example, I personally think it is the guest who should be enforcing a balloon floor since it is the guest that knows its internal memory requirements, not the host.
Even the guest has no idea about the (future) working set size. That's a known problem.
There are always cases where the calculation is wrong, and if the monitoring process isn't fast enough to react and adjust the guest size, your things will end up baldy in your guest. Just as the reboot case you mentioned, where the VM crashes.
[...]
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.
Quoted 6 lines
Quoted 1 line
Yeah, that makes sense. Only one at a time.
Having only one *active* at a time makes sense, however it ultimately would be nice to be able to have them both inserted into a VM:
one for Windows guests and one for Linux ones.
Even though only one obviously would be active at the same time.
I don't think that's the right way forward - that should be configured when the VM is started.
Personal opinion: I can understand the motivation to implement hypervisor-specific devices to better support closed-source operating systems. But I doubt we want to introduce+support ten different proprietary devices based on proprietary standards doing roughly the same thing just because closed-source operating systems are too lazy to support open standards properly.
Signature · 3 lines
Thanks, David / dhildenb
Quoted 5 lines · David Hildenbrand
Quoted 4 lines · Maciej S. Szmigiero
Hi David,
Thank you for your comments.
(...)
Quoted 10 lines
Quoted 9 lines
The idea is to use virtual DIMM sticks for hot-adding extra memory at runtime, while using ballooning for runtime adjustment of the guest memory size within the current maximum.
When the guest is rebooted the virtual DIMMs configuration is adjusted by the software controlling QEMU (some are removed and / or some are added) to give the guest the same effective memory size as it had before the reboot.
Quoted 7 lines
Okay, so while "the ACPI DIMM slot limit does not apply", the KVM memory slot limit (currently) applies, resulting in exactly the same behavior.
The only (conceptual difference) I am able to spot is then a notification to the user on reboot, so the guest memory layout can be adjusted (which I consider very ugly, but it's the same thing when mixing ballooning and DIMMs - which is why it's usually never done).
If you want to shrink a guest at runtime you'll pretty much have to use ballooning as {ACPI-based PC, virtual} DIMM stick sizes are far too large to make anything but rough adjustments to the guest memory size.
In addition to that with ACPI-based PC DIMM hotplug it is the host that chooses which particular DIMM stick to unplug while having no feedback from the guest how much of each DIMM stick memory range is currently in use and so will have to be copied somewhere else.
I know that this a source of significant hot removal slowdown, especially when a "ripple effect" happens on removal:
1) There are 3 extra DIMMs plugged into the guest: A, B, C.
A and B are nearly empty, but C is nearly full.
2) The host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick C,
3) The guest copies the content of the stick C to the stick B,
4) Once again, the host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick B,
5) The guest now has to copy the same data from the stick B to the stick A, once again.
With virtual DIMM sticks + this driver it is the guest which chooses which particular pages to release, hopefully choosing the already unused ones.
Once the whole memory behind a DIMM stick is released the host knows that it can be unplugged now without any copying.
While it might seem like this will cause a lot of fragmentation in practice Windows seems to try to give out the largest continuous range of pages it is able to find.
One can also see in the hv_balloon client driver from the Linux kernel that this driver tries to do 2 MB allocations for as long as it can before giving out single pages.
The reason why ballooning and DIMMs wasn't being used together previously is probably because virtio-balloon will (at least on Windows) mark the ballooned out pages as in use inside the guest, preventing the removal of the DIMM stick backing them.
In addition to the above, virtio-balloon is also very slow, as the whole protocol operates on single pages only, not on page ranges. There might also be some interference with Windows memory management causing an extra slowdown in comparison to the native Windows DM protocol.
If the KVM slot limit starts to be a problem in practice then we can think what can be done about it.
It's always one obstacle less.
I see that the same KVM slot limit probably applies also for virtio-mem, since it uses memory-backend-ram as its backing memory device, too, right?
If not, then how do you map a totally new memory range into the guest address space without consuming a KVM memory slot?
If that's somehow possible then maybe the same mechanism can simply be reused for this driver.
Quoted 10 lines
[...]
Quoted 4 lines
So, yes, it will be a problem if the user expands their running guest ~256 times, each time making it even bigger than previously, without rebooting it even once, but this does seem to be an edge use case.
IIRC, that's exactly what dynamic memory under Windows does in automatic mode, no? Monitor the guests, distribute memory accordingly - usually in smaller steps. But I am no expert on Hyper-V.
Yes, they call their automatic mode "Dynamic Memory" in recent Windows versions.
This is a bit confusing because even if you disable this feature the Hyper-V hypervisor will still provide this Dynamic Memory Protocol service and use it to resize the guest on (user) demand. Just it won't do such resize on its own but only when explicitly requested.
Don't know if they internally have any limit that is similar to the KVM memory slot limit, though.
Quoted 10 lines
Quoted 6 lines
In the future it would be better to automatically turn the current effective guest size into its boot memory size when the VM restarts (the VM will then have no virtual DIMMs inserted after a reboot), but doing this requires quite a few changes to QEMU, that's why it isn't there yet.
Will most probably never happen as reshuffling the layout of your boot memory (especially with NUMA) within QEMU can break live migration in various ways.
That's why this functionality is not in the current driver version as it is a bit hard to implement :)
Quoted 7 lines
If you already notify the user on a reboot, the user can just kill the VM and start it with an adjusted boot memory size. Yeah, that's ugly, but so is the whole "adjust DIMM/balloon configuration during a reboot from outside QEMU".
BTW, how would you handle: Start guest with 10G. Inflate balloon to 5G. Reboot. There are no virtual DIMMs to adjust.
You'll typically want to avoid relaunching QEMU as much as possible since things like chardev sockets and a VNC connection will disconnect if the QEMU process exits.
Not to mention that it takes some time for it to actually start again.
However, there is a trade-off here: one can either start the guest with a relatively large boot memory size, but then shrinking the guest means that it will see the whole boot memory size again during reboot, until it is ballooned down again after it has connected to the DM protocol.
Or it can be started with a small boot memory size, but this means that few virtual DIMMs might always be inserted (their size and / or count can be optimized during the next reboot or if they become unused due to ballooning).
Or one can choose some point in between these two scenarios.
I think a virtio-mem user has to choose a similar trade-off between the boot memory size and the size and count of plugged-in virtio-mem devices, right?
Quoted 6 lines
Quoted 4 lines
The above is basically how Hyper-V hypervisor handles its memory size changes and it seems to be as close to having a transparently resizable guest as reasonably possible.
"having a transparently resizable _Windows_ guests right now" :)
Right.
(...)
Quoted 11 lines
Quoted 8 lines
Quoted 6 lines
I assume these numbers apply with Windows guests only. IIRC Linux
hv_balloondoes not support page migration/compaction, while virtio-balloon does. So you might end up with quite some fragmented memory withhv_balloonin Linux guests - of course, usually only in corner cases.As I previously mentioned, this driver targets mainly Windows guests.
... and you cannot enforce that people will only use it with Windows guests :)
If people want to run this driver with Linux or port the hv_balloon client driver from the Linux kernel to, for example, GNU Hurd and run the DM protocol there then they are free to do so.
Just it really isn't this driver target environment.
Quoted 15 lines
[...]
Quoted 4 lines
Windows will generally leave some memory free when processing balloon requests, although the precise amount varies between few hundred MB to values like 1+ GB.
Quoted 8 lines
Usually it runs stable even with these few hundred MBs of free memory remaining but I have seen occasional crashes at shutdown time in this case (probably something critical failing to initialize due to the system running out of memory).
While the above command was just a quick example, I personally think it is the guest who should be enforcing a balloon floor since it is the guest that knows its internal memory requirements, not the host.
Quoted 7 lines
Even the guest has no idea about the (future) working set size. That's a known problem.
There are always cases where the calculation is wrong, and if the monitoring process isn't fast enough to react and adjust the guest size, your things will end up baldy in your guest. Just as the reboot case you mentioned, where the VM crashes.
The actual Hyper-V hypervisor somehow manages not to over-balloon its guests to the point that they run of of memory and crash. So this is definitely doable (with a margin of safety).
However, such heuristics are really an issue for the software controlling QEMU and so are outside the scope of this driver.
By the way, that's why DM guests emit a STATUS message each second with various memory counters (translated into a QMP event by this driver)
- to give its host hints about the guest memory pressure.
Quoted 18 lines
[...]
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.Quoted 6 lines
Quoted 1 line
Yeah, that makes sense. Only one at a time.
Having only one *active* at a time makes sense, however it ultimately would be nice to be able to have them both inserted into a VM:
one for Windows guests and one for Linux ones.
Even though only one obviously would be active at the same time.I don't think that's the right way forward - that should be configured when the VM is started.
Quoted 7 lines
Personal opinion: I can understand the motivation to implement hypervisor-specific devices to better support closed-source operating systems. But I doubt we want to introduce+support ten different proprietary devices based on proprietary standards doing roughly the same thing just because closed-source operating systems are too lazy to support open standards properly.
What do you mean by "ten" proprietary devices?
Is there another balloon protocol driver currently in the tree other than virtio-balloon running over various buses?
People are running Windows guests using QEMU, too.
That's why there are dozen or so Hyper-V enlightenments implemented, even though they duplicate KVM PV stuff or that there is kvmvapic with its Windows guest live-patching.
Not to mention many, many devices like e1000 or VMware vmxnet3 even though virtio-net exists or PIIX IDE even though virtio-{blk,scsi} exist. Or the applesmc driver, which is cleanly designed to help run just one proprietary OS.
Thanks,
Maciej
Quoted 11 lines
Quoted 7 lines
Okay, so while "the ACPI DIMM slot limit does not apply", the KVM memory slot limit (currently) applies, resulting in exactly the same behavior.
The only (conceptual difference) I am able to spot is then a notification to the user on reboot, so the guest memory layout can be adjusted (which I consider very ugly, but it's the same thing when mixing ballooning and DIMMs - which is why it's usually never done).
If you want to shrink a guest at runtime you'll pretty much have to use ballooning as {ACPI-based PC, virtual} DIMM stick sizes are far too large to make anything but rough adjustments to the guest memory size.
Right.
Quoted 5 lines
In addition to that with ACPI-based PC DIMM hotplug it is the host that chooses which particular DIMM stick to unplug while having no feedback from the guest how much of each DIMM stick memory range is currently in use and so will have to be copied somewhere else.
Yeah, these external requests are bad in general. Not only for performance - also because they can fail silently (if there is unmovable data on it).
Quoted 8 lines
I know that this a source of significant hot removal slowdown, especially when a "ripple effect" happens on removal:
1) There are 3 extra DIMMs plugged into the guest: A, B, C.
A and B are nearly empty, but C is nearly full.2) The host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick C,
In theory, the host can simply track inflation requests. In practice, guests tend to sometimes re-access balloon-inflated memory (e.g., simple kexec-style reboot under Linux, kdump on older Linux versions), so it's not completely safe to do.
Quoted 15 lines
3) The guest copies the content of the stick C to the stick B,
4) Once again, the host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick B,
5) The guest now has to copy the same data from the stick B to the stick A, once again.
With virtual DIMM sticks + this driver it is the guest which chooses which particular pages to release, hopefully choosing the already unused ones.
Once the whole memory behind a DIMM stick is released the host knows that it can be unplugged now without any copying.
Quoted 7 lines
While it might seem like this will cause a lot of fragmentation in practice Windows seems to try to give out the largest continuous range of pages it is able to find.
One can also see in the
hv_balloonclient driver from the Linux kernel that this driver tries to do 2 MB allocations for as long as it can before giving out single pages.
Yeah, something similar was proposed for virtio-balloon already (and there is a paper about that - see below). For virito-balloon, we were not yet convinced that stealing most hugepages from the guest is always such a good idea. It at least would have to be configurable, to not mess with existing use cases.
Quoted 5 lines
The reason why ballooning and DIMMs wasn't being used together previously is probably because virtio-balloon will (at least on Windows) mark the ballooned out pages as in use inside the guest, preventing the removal of the DIMM stick backing them.
Same on Linux - but the pages are movable, such they can at least moved around to offline + unplug a DIMM.
Some of the reasons why ballooning + DIMMs are not used as far as I know is:
- Management issues. Using it without some managing instance (plug/unplug DIMM, control balloon) is impossible. Try using it with bare QEMU - basically impossible.
- Memory Hotplug limitations: Maximum DIMM count. Minimum DIMM size that an OS can use (e.g., >= 128MiB under Linux, sometimes even 1GB). The granularity restrictions you mentioned apply.
- Memory Hotplug reliability: It can happen easily that hotplugging a DIMM / onlining it under Linux fails (e.g., minimum DIMM size). "What you think the guests actually has as available memory might be wrong". If you ignore that (and you don't even get notified) and adjust the balloon later, your (Linux) guest might be in trouble. Assume you hotplug a 8GB to a 2G VM and later try to inflate the balloon to 4GB ... so you need reliable monitoring and error handling.
So yeah, I can understand how hv-balloon tries to work around some of these issues.
Quoted 6 lines
In addition to the above, virtio-balloon is also very slow, as the whole protocol operates on single pages only, not on page ranges. There might also be some interference with Windows memory management causing an extra slowdown in comparison to the native Windows DM protocol.
Yes, because I assume Windows doesn't really care too much about optimizing for virtio-balloon. There isn't too much external developers can do about that. See below for hugepage ballooning in virtio-balloon.
Quoted 4 lines
If the KVM slot limit starts to be a problem in practice then we can think what can be done about it.
It's always one obstacle less.
I'm not a friend of leaving the challenging/fundamental problems to be sported out in the future (e.g., resizing initial boot memory, dealing with fundamental limits - like KVM memory slots or VMA). But I get how it's easier to get something running this way :)
Quoted 4 lines
I see that the same KVM slot limit probably applies also for virtio-mem, since it uses memory-backend-ram as its backing memory device, too, right?
Yes, one memory backend per virtio-mem device. You usually have one device per NUMA node.
Quoted 5 lines
If not, then how do you map a totally new memory range into the guest address space without consuming a KVM memory slot?
If that's somehow possible then maybe the same mechanism can simply be reused for this driver.
So, virtio-mem will (in the future, still to be upstreamed by me) use resizeable memory regions / ramblocks / KVM memory slots. The region can grow (e.g., memory hotplug) and shrink (e.g., during reboot, but later also if unplugged memory would allow for it). Nice thing is that migration code fully supports resizeable ramblocks already.
Resizes are triggered by the virtio-mem device, so stuff is completely handled inside QEMU. For hv-balloon, you could grow the region when required (e.g., balloon X, whereby X is > ram size after inflating), and shrink during reboot (or whenever it might be valid to shrink). However, you cannot "rip out" anything in between, you'll have to rely on MADV_DONTNEED until the guests reboots (well, just like basic ballooning) and you can definitely shrink the region.
That's the tradeoff virtio-mem decided to take for now to be able to manage any size changes + migration completely in QEMU, avoiding any coordination with an external entity (e.g., libvirt in your example) when resizing a guest.
Quoted 12 lines
Quoted 11 lines
[...]
Quoted 4 lines
So, yes, it will be a problem if the user expands their running guest ~256 times, each time making it even bigger than previously, without rebooting it even once, but this does seem to be an edge use case.
IIRC, that's exactly what dynamic memory under Windows does in automatic mode, no? Monitor the guests, distribute memory accordingly - usually in smaller steps. But I am no expert on Hyper-V.
Quoted 8 lines
Yes, they call their automatic mode "Dynamic Memory" in recent Windows versions.
This is a bit confusing because even if you disable this feature the Hyper-V hypervisor will still provide this Dynamic Memory Protocol service and use it to resize the guest on (user) demand. Just it won't do such resize on its own but only when explicitly requested.
Interesting, thanks.
Quoted 3 lines
Don't know if they internally have any limit that is similar to the KVM memory slot limit, though.
That would be interesting for me - like which limits do actually apply under Hyper-V.
[...]
Quoted 6 lines
Quoted 6 lines
If you already notify the user on a reboot, the user can just kill the VM and start it with an adjusted boot memory size. Yeah, that's ugly, but so is the whole "adjust DIMM/balloon configuration during a reboot from outside QEMU".
Quoted 7 lines
Quoted 2 lines
BTW, how would you handle: Start guest with 10G. Inflate balloon to 5G. Reboot. There are no virtual DIMMs to adjust.
You'll typically want to avoid relaunching QEMU as much as possible since things like chardev sockets and a VNC connection will disconnect if the QEMU process exits.
Not to mention that it takes some time for it to actually start again.
Exactly my thoughts, that's why I tried to avoid that as well with virtio-mem.
Quoted 5 lines
However, there is a trade-off here: one can either start the guest with a relatively large boot memory size, but then shrinking the guest means that it will see the whole boot memory size again during reboot, until it is ballooned down again after it has connected to the DM protocol.
Yeah, one of the main issues of memory ballooning.
Quoted 11 lines
Or it can be started with a small boot memory size, but this means that few virtual DIMMs might always be inserted (their size and / or count can be optimized during the next reboot or if they become unused due to ballooning).
Or one can choose some point in between these two scenarios.
I think a virtio-mem user has to choose a similar trade-off between the boot memory size and the size and count of plugged-in virtio-mem devices, right?
Partially yes, partially no. It doesn't really care about the second case you mention ("few virtual DIMMs might always be inserted") due to the way it works. And it does noever have to deal with "inflate balloon after/during reboot".
A virtio-mem device manages only its assigned memory, it does not work on random system memory like memory ballooning. So you can never unplug initial memory.
However, you can do something like
-m 4G,maxmem=104G
and define a virtio-mem device with a maximum size of 100G and an initial size of - say 16G. When booting up, the guest will detect the additional 16GB and have effectively 20GB. However, you can only ever shrink back down to 4GB (e.g., reliably during a reboot).
And it might not always be desirable (at least under Linux) to have little boot memory - say the CMA allocator might want to reserve bigger chunks of memory early during boot - if the initial memory size is too small, this can fail easily.
[...]
Quoted 22 lines
Quoted 14 lines
Quoted 5 lines
Windows will generally leave some memory free when processing balloon requests, although the precise amount varies between few hundred MB to values like 1+ GB.
Quoted 8 lines
Usually it runs stable even with these few hundred MBs of free memory remaining but I have seen occasional crashes at shutdown time in this case (probably something critical failing to initialize due to the system running out of memory).
While the above command was just a quick example, I personally think it is the guest who should be enforcing a balloon floor since it is the guest that knows its internal memory requirements, not the host.
Quoted 7 lines
Even the guest has no idea about the (future) working set size. That's a known problem.
There are always cases where the calculation is wrong, and if the monitoring process isn't fast enough to react and adjust the guest size, your things will end up baldy in your guest. Just as the reboot case you mentioned, where the VM crashes.
Quoted 6 lines
The actual Hyper-V hypervisor somehow manages not to over-balloon its guests to the point that they run of of memory and crash. So this is definitely doable (with a margin of safety).
However, such heuristics are really an issue for the software controlling QEMU and so are outside the scope of this driver.
Yeah, just like any heuristic, it can be wrong. I wonder if we could add something similar for virtio-balloon (at least don't immediately deflate until your VM dies ...).
Quoted 4 lines
By the way, that's why DM guests emit a STATUS message each second with various memory counters (translated into a QMP event by this driver)
- to give its host hints about the guest memory pressure.
Right, we have something similar for virtio-balloon, via the stats VQ. IIRC, it's used by auto-ballooning mechanisms implemented in openstack (I'd say similar to dynamic memory, it just won't try to increase the size of a guest using new DIMMs).
Quoted 25 lines
Quoted 17 lines
Quoted 7 lines
Quoted 7 lines
Quoted 6 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.Quoted 6 lines
Quoted 1 line
Yeah, that makes sense. Only one at a time.
Having only one *active* at a time makes sense, however it ultimately would be nice to be able to have them both inserted into a VM:
one for Windows guests and one for Linux ones.
Even though only one obviously would be active at the same time.I don't think that's the right way forward - that should be configured when the VM is started.
Quoted 7 lines
Personal opinion: I can understand the motivation to implement hypervisor-specific devices to better support closed-source operating systems. But I doubt we want to introduce+support ten different proprietary devices based on proprietary standards doing roughly the same thing just because closed-source operating systems are too lazy to support open standards properly.
Quoted 3 lines
What do you mean by "ten" proprietary devices?
Is there another balloon protocol driver currently in the tree other than virtio-balloon running over various buses?
Maybe OSX wants to be next and re-invent the wheel with a proprietary balloon driver for a custom hypervisor. I think you get the idea.
What I'm saying is that I'd much rather want to see Windows improve+extend virtio-balloon and such (virtio-mem), instead of requiring hypervisors to implement undocumented, proprietary devices to get stuff running somewhat smoothly in modern cloud environments. I have the feeling that quite some stuff you mention can simply be "fixed" by extending/improving virtio-baloon under Windows - for example, inflation speed can be improved significantly by inflating in bigger chunks. See
https://dl.acm.org/doi/10.1145/3240302.3240420
In contrast to hv-balloon *we* can extend/improve the interface/stadnard/implementation on both sides (host/guest).
Quoted 6 lines
People are running Windows guests using QEMU, too.
That's why there are dozen or so Hyper-V enlightenments implemented, even though they duplicate KVM PV stuff or that there is kvmvapic with its Windows guest live-patching.
IIRC the Hyper-V enlightenment stuff is properly publicly documented - whereby last time I checked, the hv-balloon is completely undocumented and has to be reverse engineered from the Linux implementation. Please correct me if I'm wrong - I am not able to spot references in your cover letter as well - I'd be interested into that!
Quoted 5 lines
Not to mention many, many devices like e1000 or VMware vmxnet3 even though virtio-net exists or PIIX IDE even though virtio-{blk,scsi} exist. Or the applesmc driver, which is cleanly designed to help run just one proprietary OS.
IIRC we need the devices either to bootstrap - e.g., use e1000 until we can install virtio-net once the guest is up and running, or to support older, unmodified guests. I'd like to stress that what you are proposing is different in that sense. Your Windows VM will work just fine without a hv-balloon device.
Again, just my personal opinion, I don't make any decisions around here :)
Signature · 3 lines
Thanks, David / dhildenb
On 23.09.2020 14:48, David Hildenbrand wrote:
(...)
Quoted 6 lines
Quoted 6 lines
I know that this a source of significant hot removal slowdown, especially when a "ripple effect" happens on removal:
1) There are 3 extra DIMMs plugged into the guest: A, B, C.
A and B are nearly empty, but C is nearly full.
Quoted 7 lines
Quoted 2 lines
2) The host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick C,
In theory, the host can simply track inflation requests. In practice, guests tend to sometimes re-access balloon-inflated memory (e.g., simple kexec-style reboot under Linux, kdump on older Linux versions), so it's not completely safe to do.
You are describing Linux situation here, while this driver targets Windows.
I think the issues you describe (kexec, etc.) are probably fixable once somebody is determined enough to do so.
I mean, either the old kernel needs to transfer information about "forbidden" memory areas to the new kernel or the new kernel needs to query these somehow (probably only if is receives a "do it" flag from the old kernel).
Quoted 23 lines
Quoted 15 lines
3) The guest copies the content of the stick C to the stick B,
4) Once again, the host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick B,
5) The guest now has to copy the same data from the stick B to the stick A, once again.
With virtual DIMM sticks + this driver it is the guest which chooses which particular pages to release, hopefully choosing the already unused ones.
Once the whole memory behind a DIMM stick is released the host knows that it can be unplugged now without any copying.Quoted 7 lines
While it might seem like this will cause a lot of fragmentation in practice Windows seems to try to give out the largest continuous range of pages it is able to find.
One can also see in the
hv_balloonclient driver from the Linux kernel that this driver tries to do 2 MB allocations for as long as it can before giving out single pages.
Quoted 5 lines
Yeah, something similar was proposed for virtio-balloon already (and there is a paper about that - see below). For virito-balloon, we were not yet convinced that stealing most hugepages from the guest is always such a good idea. It at least would have to be configurable, to not mess with existing use cases.
Thanks for the paper.
The hv_balloon Linux client driver does large continuous allocations of ordinary 4k pages, not only transparent hugepages.
Even if the virtio-balloon client driver is changed to do higher-older allocations the protocol itself only supports transporting individual page numbers and not ranges.
Either the virtio-balloon protocol will need to be changed to allow sending page ranges (good) or there will need to be some implicit agreement between the host and client drivers that pages from a continuous range will be sent consecutively.
Then the host driver can reassemble the whole range, ending either when it receives a page outside the range or when a reassembly timeout happens.
But that's just really ugly - realistically the virtio-balloon protocol will need to be changed in such case.
Quoted 7 lines
Quoted 6 lines
The reason why ballooning and DIMMs wasn't being used together previously is probably because virtio-balloon will (at least on Windows) mark the ballooned out pages as in use inside the guest, preventing the removal of the DIMM stick backing them.
Quoted 7 lines
Same on Linux - but the pages are movable, such they can at least moved around to offline + unplug a DIMM.
Some of the reasons why ballooning + DIMMs are not used as far as I know is:
- Management issues. Using it without some managing instance (plug/unplug DIMM, control balloon) is impossible. Try using it with bare QEMU - basically impossible.
We'll it's not impossible, just needs some manual reconfiguration.
But such advanced usage scenarios usually are using QEMU with an external controller anyway.
Quoted 3 lines
- Memory Hotplug limitations: Maximum DIMM count. Minimum DIMM size that an OS can use (e.g., >= 128MiB under Linux, sometimes even 1GB). The granularity restrictions you mentioned apply.
Windows has a memory hotplug granularity of 1 MB.
Quoted 7 lines
- Memory Hotplug reliability: It can happen easily that hotplugging a DIMM / onlining it under Linux fails (e.g., minimum DIMM size). "What you think the guests actually has as available memory might be wrong". If you ignore that (and you don't even get notified) and adjust the balloon later, your (Linux) guest might be in trouble. Assume you hotplug a 8GB to a 2G VM and later try to inflate the balloon to 4GB ... so you need reliable monitoring and error handling.
The DM "hot add" message has a "page_count" field to tell the host how much memory the guest has actually added.
Quoting the DM protocol header file:
Quoted 5 lines
Hot adds may also fail due to low resources; in this case, the guest must not complete this message until the hot add can succeed, and the host must not send a new hot add request until the response is sent.
If VSC fails to hot add memoryDYNMEM_NUMBER_OF_UNSUCCESSFUL_HOTADD_ATTEMPTStimes it fails the request.
The host also knows the current guest memory size from its STATUS messages.
Quoted 10 lines
So yeah, I can understand how hv-balloon tries to work around some of these issues.
Quoted 6 lines
In addition to the above, virtio-balloon is also very slow, as the whole protocol operates on single pages only, not on page ranges. There might also be some interference with Windows memory management causing an extra slowdown in comparison to the native Windows DM protocol.
Quoted 3 lines
Yes, because I assume Windows doesn't really care too much about optimizing for virtio-balloon. There isn't too much external developers can do about that. See below for hugepage ballooning in virtio-balloon.
Exactly, that's why the DM protocol is the best thing we have to offer for Windows guests right now.
Quoted 9 lines
Quoted 4 lines
If the KVM slot limit starts to be a problem in practice then we can think what can be done about it.
It's always one obstacle less.I'm not a friend of leaving the challenging/fundamental problems to be sported out in the future (e.g., resizing initial boot memory, dealing with fundamental limits - like KVM memory slots or VMA). But I get how it's easier to get something running this way :)
Constraints can be removed step-by-step, when they actually start tobite.
It is unrealistic to have a perfect guest resize solution in a single patch series upfront, the issue is just too complex.
Quoted 7 lines
Quoted 4 lines
I see that the same KVM slot limit probably applies also for virtio-mem, since it uses memory-backend-ram as its backing memory device, too, right?
Yes, one memory backend per virtio-mem device. You usually have one device per NUMA node.
So if you want to dynamically manage most of the guest memory these virtio-mem devices + their backends will be very large (circa 1/4 guest size each for a 4-node machine).
That means in practice they won't ever be able to be hot-removed before the VM is rebooted since there will very likely be at least single stuck page in each of these backing devices preventing their removal. (If hot-removal support is ever enabled for virtio-mem, it looks like it is not possible yet even on a VM reboot).
And I can see that removing a single RAM block in virio-mem is done by discarding it via MADV_DONTNEED, just like in ballooning. Only the minimum block size is 2 MB and not 4 KB so all consecutive 512 pages in a block will need to be free in order to discard it.
Quoted 20 lines
Quoted 5 lines
If not, then how do you map a totally new memory range into the guest address space without consuming a KVM memory slot?
If that's somehow possible then maybe the same mechanism can simply be reused for this driver.So, virtio-mem will (in the future, still to be upstreamed by me) use resizeable memory regions / ramblocks / KVM memory slots. The region can grow (e.g., memory hotplug) and shrink (e.g., during reboot, but later also if unplugged memory would allow for it). Nice thing is that migration code fully supports resizeable ramblocks already.
Resizes are triggered by the virtio-mem device, so stuff is completely handled inside QEMU. For hv-balloon, you could grow the region when required (e.g., balloon X, whereby X is > ram size after inflating), and shrink during reboot (or whenever it might be valid to shrink). However, you cannot "rip out" anything in between, you'll have to rely on
MADV_DONTNEEDuntil the guests reboots (well, just like basic ballooning) and you can definitely shrink the region.
Quoted 4 lines
That's the tradeoff virtio-mem decided to take for now to be able to manage any size changes + migration completely in QEMU, avoiding any coordination with an external entity (e.g., libvirt in your example) when resizing a guest.
For hv-balloon (that is, Windows guests), it seems to me that the slot resizing would be a viable solution for growing the guest without hitting the KVM slot limit or requiring a VM reboot.
For shrinking such guests it isn't really necessary since DM ballooning is enough until a reboot happens.
During a reboot, virtual DIMM config optimization can be used for now, while the best, long-term, solution would be to actually do what the Hyper-V hypervisor does in this case: resize the boot memory to match the target guest size.
This way the whole issue of the guest seeing only the boot memory part, not the dynamic part, during its boot will no longer be there.
(...)
Quoted 30 lines
Quoted 11 lines
Or it can be started with a small boot memory size, but this means that few virtual DIMMs might always be inserted (their size and / or count can be optimized during the next reboot or if they become unused due to ballooning).
Or one can choose some point in between these two scenarios.
I think a virtio-mem user has to choose a similar trade-off between the boot memory size and the size and count of plugged-in virtio-mem devices, right?
Partially yes, partially no. It doesn't really care about the second case you mention ("few virtual DIMMs might always be inserted") due to the way it works. And it does noever have to deal with "inflate balloon after/during reboot".
A virtio-mem device manages only its assigned memory, it does not work on random system memory like memory ballooning. So you can never unplug initial memory.
However, you can do something like
-m 4G,maxmem=104G
and define a virtio-mem device with a maximum size of 100G and an initial size of - say 16G. When booting up, the guest will detect the additional 16GB and have effectively 20GB. However, you can only ever shrink back down to 4GB (e.g., reliably during a reboot).
Quoted 5 lines
And it might not always be desirable (at least under Linux) to have little boot memory - say the CMA allocator might want to reserve bigger chunks of memory early during boot - if the initial memory size is too small, this can fail easily.
So the virtio-mem trade-off is between the size of the boot memory and the dynamically managed part(s).
And also the block size, as as far I can see, a single stuck page in a block will prevent it from being discarded.
Of course, we are talking about Linux guests here - Windows guest will see just the boot memory part.
(...)
Quoted 40 lines
Quoted 25 lines
Quoted 17 lines
Quoted 7 lines
Quoted 7 lines
Quoted 6 lines
Future directions:
* Allow sharing the ballooning QEMU interface between hv-balloon and virtio-balloon drivers.
Currently, only one of them can be added to the VM at the same time.Quoted 6 lines
Quoted 1 line
Yeah, that makes sense. Only one at a time.
Having only one *active* at a time makes sense, however it ultimately would be nice to be able to have them both inserted into a VM:
one for Windows guests and one for Linux ones.
Even though only one obviously would be active at the same time.I don't think that's the right way forward - that should be configured when the VM is started.
Quoted 7 lines
Personal opinion: I can understand the motivation to implement hypervisor-specific devices to better support closed-source operating systems. But I doubt we want to introduce+support ten different proprietary devices based on proprietary standards doing roughly the same thing just because closed-source operating systems are too lazy to support open standards properly.
Quoted 3 lines
What do you mean by "ten" proprietary devices?
Is there another balloon protocol driver currently in the tree other than virtio-balloon running over various buses?Maybe OSX wants to be next and re-invent the wheel with a proprietary balloon driver for a custom hypervisor. I think you get the idea.
What I'm saying is that I'd much rather want to see Windows improve+extend virtio-balloon and such (virtio-mem), instead of requiring hypervisors to implement undocumented, proprietary devices to get stuff running somewhat smoothly in modern cloud environments. I have the feeling that quite some stuff you mention can simply be "fixed" by extending/improving virtio-baloon under Windows - for example, inflation speed can be improved significantly by inflating in bigger chunks. See
Quoted 4 lines
https://dl.acm.org/doi/10.1145/3240302.3240420
In contrast to hv-balloon *we* can extend/improve the interface/stadnard/implementation on both sides (host/guest).
Even if we switched virtio-balloon to bigger allocations and made the protocol return page ranges the allocation is still done by simply using an ordinary alloc_pages()-equivalent API.
I don't see any exported Windows kernel API for allocating balloon memory.
The same goes for adding new RAM.
The above alone means that supporting virtio-mem semantics on current Windows versions is likely not possible.
Quoted 8 lines
Quoted 6 lines
People are running Windows guests using QEMU, too.
That's why there are dozen or so Hyper-V enlightenments implemented, even though they duplicate KVM PV stuff or that there is kvmvapic with its Windows guest live-patching.
IIRC the Hyper-V enlightenment stuff is properly publicly documented -
They are only documented from the guest perspective - basically what the guest of a Hyper-V hypervisor can possibly use.
There is nothing in the Hyper-V TLFS which of the functionality it documents is supported or required by any Windows version or how Windows guests actually make use of these features.
Not to mention that the documented interface could say the guest can expect values A or B or C for parameter X, which is technically true, however the actual Hyper-V hypervisor always uses A and that's what Windows will expect.
Quoted 4 lines
whereby last time I checked, the hv-balloon is completely undocumented and has to be reverse engineered from the Linux implementation. Please correct me if I'm wrong - I am not able to spot references in your cover letter as well - I'd be interested into that!
The DM protocol is rather straightforward - the Linux driver contains well-commented definitions of its messages.
For a hot add the host simply provides the start page frame number and the count of pages to add and in response receives the number of pages the guest was actually able to hot add.
For a balloon request the host provides a count of pages it wants the guest to free and the guest responds with page ranges it has managed to release.
The reverse happens for an unballoon request.
The protocol consists of just a few simple messages, well described in the Linux driver.
The VMBus part of the protocol works in the same way as in other VMBus devices.
Quoted 11 lines
Quoted 5 lines
Not to mention many, many devices like e1000 or VMware vmxnet3 even though virtio-net exists or PIIX IDE even though virtio-{blk,scsi} exist. Or the applesmc driver, which is cleanly designed to help run just one proprietary OS.
IIRC we need the devices either to bootstrap - e.g., use e1000 until we can install virtio-net once the guest is up and running, or to support older, unmodified guests. I'd like to stress that what you are proposing is different in that sense. Your Windows VM will work just fine without a hv-balloon device.
There are also "accelerator" devices like kvmvapic or proprietary devices where an older open standard exists, like sii3112 SATA, even though IDE / AHCI can be used to bootstrap the guest.
The sii3112 driver was only added in 2017, although the hardware that it emulates comes from early 21st century.
Quoted 2 lines
Again, just my personal opinion, I don't make any decisions around here :)
Thanks for taking the time to provide your feedback and insight, Maciej
Quoted 11 lines · Maciej S. Szmigiero
On 23.09.2020 14:48, David Hildenbrand wrote:
(...)Quoted 9 lines
Quoted 8 lines
I know that this a source of significant hot removal slowdown, especially when a "ripple effect" happens on removal:
1) There are 3 extra DIMMs plugged into the guest: A, B, C.
A and B are nearly empty, but C is nearly full.2) The host does not know anything which DIMM is empty and which is full, so it requests the guest to unplug the stick C,
Quoted 7 lines
Quoted 4 lines
In theory, the host can simply track inflation requests. In practice, guests tend to sometimes re-access balloon-inflated memory (e.g., simple kexec-style reboot under Linux, kdump on older Linux versions), so it's not completely safe to do.
You are describing Linux situation here, while this driver targets Windows.
Yeah, this stuff is also broken with Linux under Hyper-V IIRC (I remember fixing kdump/makedumpfile to not touch inflated pages).
Quoted 8 lines
I think the issues you describe (kexec, etc.) are probably fixable once somebody is determined enough to do so.
I mean, either the old kernel needs to transfer information about "forbidden" memory areas to the new kernel or the new kernel needs to query these somehow (probably only if is receives a "do it" flag from the old kernel).
One idea is to notify the host like "I'll reuse any page again". But doing that via virtio while your dying isn't always possible. So the new kernel would have to do it - but then virtio-balloon would have to be up and running extremely early during boot, before any possibly-inflated page gets touched - also a head scratcher. Then, what if your new kernel doesn't support virtio-balloon ... not easy.
[...]
Quoted 2 lines
But that's just really ugly - realistically the virtio-balloon protocol will need to be changed in such case.
Yes, and there were RFCs already.
https://lkml.kernel.org/r/[email protected]
[...]
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
If the KVM slot limit starts to be a problem in practice then we can think what can be done about it.
It's always one obstacle less.
Quoted 6 lines
Quoted 4 lines
I'm not a friend of leaving the challenging/fundamental problems to be sported out in the future (e.g., resizing initial boot memory, dealing with fundamental limits - like KVM memory slots or VMA). But I get how it's easier to get something running this way :)
Constraints can be removed step-by-step, when they actually start tobite.
Not if they are fundamental. Try implementing vfio support for memory ballooning, you're going to have a hard time.
Quoted 3 lines
It is unrealistic to have a perfect guest resize solution in a single patch series upfront, the issue is just too complex.
I disagree, but that's a different discussion. :)
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
I see that the same KVM slot limit probably applies also for virtio-mem, since it uses memory-backend-ram as its backing memory device, too, right?
Quoted 6 lines
Quoted 2 lines
Yes, one memory backend per virtio-mem device. You usually have one device per NUMA node.
So if you want to dynamically manage most of the guest memory these virtio-mem devices + their backends will be very large (circa 1/4 guest size each for a 4-node machine).
Right, or even bigger, depending on the setup.
Quoted 6 lines
That means in practice they won't ever be able to be hot-removed before the VM is rebooted since there will very likely be at least single stuck page in each of these backing devices preventing their removal. (If hot-removal support is ever enabled for virtio-mem, it looks like it is not possible yet even on a VM reboot).
Yeah, hot-removing a virtio-mem device is not one of the important use cases (it's completely blocked). If you want to unplug memory, adjust the requested size.
There are plans to support it in the future (for example during reboot), but I barely see a need for it currently (especially once we support resizeable memory backends upstream).
Quoted 5 lines
And I can see that removing a single RAM block in virio-mem is done by discarding it via
MADV_DONTNEED, just like in ballooning. Only the minimum block size is 2 MB and not 4 KB so all consecutive 512 pages in a block will need to be free in order to discard it.
Right, that was a decision to avoid issues known from base-page-size based ballooning (like fragmenting guest memory, big tracking bitmaps, incompatibility with vfio, breaking THP and degrading performance, incompatibility with hugetlbfs ...)
[...]
Quoted 9 lines
During a reboot, virtual DIMM config optimization can be used for now, while the best, long-term, solution would be to actually do what the Hyper-V hypervisor does in this case: resize the boot memory to match the target guest size.
This way the whole issue of the guest seeing only the boot memory part, not the dynamic part, during its boot will no longer be there.
Yeah, I had the same thought back when designing virtio-mem (and looking into similar handling), but decided that it's impossible to get right - at least in NUMA setups (and regarding migration). But I can see that Hyper-V Dynamic Memory doesn't care too much about NUMA at all (and NUMA-aware ballooning has its own set of issues).
[...]
Quoted 5 lines
So the virtio-mem trade-off is between the size of the boot memory and the dynamically managed part(s).
And also the block size, as as far I can see, a single stuck page in a block will prevent it from being discarded.
Yes. It's really something in-between memory ballooning and DIMM-based memory hot(un)plug. The block size will be comparatively large in some setups (esp., with vfio). You're definitely not able to squeeze out the page of your guest - we have virtio-balloon for that if one really wants to do that - not the target use case of virtio-mem.
Quoted 3 lines
Of course, we are talking about Linux guests here - Windows guest will see just the boot memory part.
Yes, until we have support for it.
[...]
Quoted 10 lines
Even if we switched virtio-balloon to bigger allocations and made the protocol return page ranges the allocation is still done by simply using an ordinary alloc_pages()-equivalent API.
I don't see any exported Windows kernel API for allocating balloon memory.The same goes for adding new RAM.
The above alone means that supporting virtio-mem semantics on current Windows versions is likely not possible.
All I can say is, that there are (unofficial?) APIs :) (e.g., the ones used by DM). But yeah, that's what you get with closed-source operating systems - and personally, I think, it shouldn't be us that have to suffer.
Quoted 13 lines
Quoted 9 lines
Quoted 7 lines
People are running Windows guests using QEMU, too.
That's why there are dozen or so Hyper-V enlightenments implemented, even though they duplicate KVM PV stuff or that there is kvmvapic with its Windows guest live-patching.
IIRC the Hyper-V enlightenment stuff is properly publicly documented -
They are only documented from the guest perspective - basically what the guest of a Hyper-V hypervisor can possibly use.
Quoted 9 lines
There is nothing in the Hyper-V TLFS which of the functionality it documents is supported or required by any Windows version or how Windows guests actually make use of these features.
Not to mention that the documented interface could say the guest can expect values A or B or C for parameter X, which is technically true, however the actual Hyper-V hypervisor always uses A and that's what Windows will expect.
Interesting, thanks.
Quoted 17 lines
Quoted 4 lines
whereby last time I checked, the hv-balloon is completely undocumented and has to be reverse engineered from the Linux implementation. Please correct me if I'm wrong - I am not able to spot references in your cover letter as well - I'd be interested into that!
The DM protocol is rather straightforward - the Linux driver contains well-commented definitions of its messages.
For a hot add the host simply provides the start page frame number and the count of pages to add and in response receives the number of pages the guest was actually able to hot add.
For a balloon request the host provides a count of pages it wants the guest to free and the guest responds with page ranges it has managed to release.
The reverse happens for an unballoon request.
Quoted 6 lines
The protocol consists of just a few simple messages, well described in the Linux driver.
The VMBus part of the protocol works in the same way as in other VMBus devices.
Yeah, but then there are things like
https://lkml.kernel.org/r/[email protected]
that left me clueless - it seems like we're missing some things, maybe there is more (I am pretty sure there is more ... :) )? (I do some work on the Linux hv_balloon driver every now of then when working on optimizing the other balloon drivers / virtio-mem).
Quoted 18 lines
Quoted 11 lines
Quoted 5 lines
Not to mention many, many devices like e1000 or VMware vmxnet3 even though virtio-net exists or PIIX IDE even though virtio-{blk,scsi} exist. Or the applesmc driver, which is cleanly designed to help run just one proprietary OS.
IIRC we need the devices either to bootstrap - e.g., use e1000 until we can install virtio-net once the guest is up and running, or to support older, unmodified guests. I'd like to stress that what you are proposing is different in that sense. Your Windows VM will work just fine without a hv-balloon device.
There are also "accelerator" devices like kvmvapic or proprietary devices where an older open standard exists, like sii3112 SATA, even though IDE / AHCI can be used to bootstrap the guest.
The sii3112 driver was only added in 2017, although the hardware that it emulates comes from early 21st century.
Quoted 4 lines
Quoted 2 lines
Again, just my personal opinion, I don't make any decisions around here :)
Thanks for taking the time to provide your feedback and insight,
I'll try to give your series a look. I can definitely say that
1) I dislike that an external entity has to do vDIMM adaptions / ballooning adaptions when rebooting or when wanting to resize a guest.
2) I am not sure ignoring the kvm memory slot limit is a good idea. (or the fundamental issue of resizing boot memory - ever)
Once you have the current approach upstream (vDIMMs, ballooning), there is no easy way to change that later (requires deprecating, etc.).
But we talked about that already.
Signature · 3 lines
Thanks, David / dhildenb