Currently the only vNVDIMM backend can guarantee the guest write
persistence is device DAX on Linux, because no host-side kernel cache
is involved in the guest access to it. The approach to detect whether
the backend is device DAX needs to access sysfs, which may not work
with SELinux.
Instead, we add the 'unarmed' option to device 'nvdimm', so that users
or management utils, which have enough knowledge about the backend,
can control the unarmed flag in guest ACPI NFIT via this option. The
guest Linux NVDIMM driver, for example, will mark the corresponding
vNVDIMM device read-only if the unarmed flag in guest NFIT is set.
The default value of 'unarmed' option is 'off' in order to keep the
backwards compatibility.
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Message-Id: <20171211072806.2812-4-haozhong.zhang@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Add a new query-memory-size-summary command which provides the
following memory information in bytes:
* base-memory - size of "base" memory specified with command line option -m.
* plugged-memory - amount of memory that was hot-plugged.
If target does not have CONFIG_MEM_HOTPLUG enabled, no
value is reported.
Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis@profitbricks.com>
Signed-off-by: Mohammed Gamal <mohammed.gamal@profitbricks.com>
Signed-off-by: Eduardo Otubo <eduardo.otubo@profitbricks.com>
Signed-off-by: Vadim Galitsyn <vadim.galitsyn@profitbricks.com>
Reviewed-by: Eugene Crosser <evgenii.cherkashin@profitbricks.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Eric Blake <eblake@redhat.com>
Cc: qemu-devel@nongnu.org
Message-Id: <20170829153022.27004-3-vadim.galitsyn@profitbricks.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Fixup comments as per Igor's review
Added 'of' from Vadim's reply
QEMU currently crashes when trying to use a 'pc-dimm' on the pseries
machine without specifying its 'memdev' property. This happens because
pc_dimm_get_memory_region() does not check whether the 'memdev' property
has properly been set by the user. Looking closer at this function, it's
also obvious that it is using &error_abort to call another function - and
this is bad in a function that is used in the hot-plugging calling chain
since this can also cause QEMU to exit unexpectedly.
So let's fix these issues in a proper way now: Add a "Error **errp"
parameter to pc_dimm_get_memory_region() which we use in case the 'memdev'
property has not been set by the user, and which we can use instead of
the &error_abort, and change the callers of get_memory_region() to make
use of this "errp" parameter for proper error checking.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Rename it to nvdimm_plug()
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
as they use completely different way to handle hotplug event
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
as there is a global lock to protect vm-exit handlers and
QMP/monitor, this lock can be dropped
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
The buffer is used to save the FIT info for all the presented nvdimm
devices which is updated after the nvdimm device is plugged or
unplugged. In the later patch, it will be used to construct NVDIMM
ACPI _FIT method which reflects the presented nvdimm devices after
nvdimm hotplug
As FIT buffer can not completely mapped into guest address space,
OSPM will exit to QEMU multiple times, however, there is the race
condition - FIT may be changed during these multiple exits, so that
some rules are introduced:
1) the user should hold the @lock to access the buffer and
2) mark @dirty whenever the buffer is updated.
@dirty is cleared for the first time OSPM gets fit buffer, if
dirty is detected in the later access, OSPM will restart the
access
As fit should be updated after nvdimm device is successfully realized
so that a new hotplug callback, post_hotplug, is introduced
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
For each NVDIMM present or intended to be supported by platform,
platform firmware also exposes an ACPI Namespace Device under
the root device
So it builds nvdimm devices for all slots to support vNVDIMM hotplug
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduce a parameter, 'label-size', which is the size of nvdimm label
data area which is reserved at the end of backend memory. It is required
at least 128k
Two callbacks, read_label_data() and write_label_data(), are used to
operate the label area
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This callback returns the MemoryRegion that is the memory of dimm should
be kept during live migration
nvdimm device is different with pc-dimm as its memory includes not only
the MemoryRegion directly mapping to guest's address space but also the
memory used as label data
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
nvdimm needs to check if the backend memory is large enough to contain
label data and init its memory region when the device is realized, so
introduce realize callback which is called after common dimm has been
realize
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Curretly, the memory region of backed memory is all directly
mapped to guest's address space, however, it will be not true
for nvdimm device if we introduce nvdimm label which only can
be indirectly accessed by ACPI DSM method
Also it improves the comments a bit to reflect this fact
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
'table' argument in bios_linker_add_foo() commands is
a data blob of one of files also passed to the same API.
So instead of passing blob in every API call, add and keep
file name association with related blob at bios_linker_loader_alloc()
time.
And find blob by name looking up allocated file entries
inside of bios_linker_add_foo() commands.
It will:
- make API less confusing,
- enforce calling bios_linker_loader_alloc() before
calling any bios_linker_add_foo()
- make sure that blob is the correct one, i.e.
associated with the right file name
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Patch just changes type of of linker variables to
a structure, there aren't any functional changes.
Converting linker to a structure will allow to extend
it functionality in follow up patch adding sanity blob
checks.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
correct comment and remove an unused macro. commit adcb4ee6
already correct its type
Signed-off-by: Cao jin <caoj.fnst@cn.fujitsu.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Emulate dsm method after IO VM-exit
Currently, we only introduce the framework and no function is actually
supported
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
32 bits IO port starting from 0x0a18 in guest is reserved for NVDIMM
ACPI emulation. The table, NVDIMM_DSM_MEM_FILE, will be patched into
NVDIMM ACPI binary code
OSPM uses this port to tell QEMU the final address of the DSM memory
and notify QEMU to emulate the DSM method
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The 'base' field of MemoryHotplugState is ram_addr_t, which indicates that
it exists in the abstract address space of RAM regions.
However, the actual usage of this field indicates that it is a concrete
physical address (it's passed as an offset to memory_region_add_subgregion
for example).
So, correct its type to 'hwaddr'.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Eduardo Habkost <ehabkost@redhat.com>
NFIT is defined in ACPI 6.0: 5.2.25 NVDIMM Firmware Interface Table (NFIT)
Currently, we only support PMEM mode. Each device has 3 structures:
- SPA structure, defines the PMEM region info
- MEM DEV structure, it has the @handle which is used to associate specified
ACPI NVDIMM device we will introduce in later patch.
Also we can happily ignored the memory device's interleave, the real
nvdimm hardware access is hidden behind host
- DCR structure, it defines vendor ID used to associate specified vendor
nvdimm driver. Since we only implement PMEM mode this time, Command
window and Data window are not needed
The NVDIMM functionality is controlled by the parameter, 'nvdimm', which
is introduced for the machine, there is a example to enable it:
-machine pc,nvdimm -m 8G,maxmem=100G,slots=100 -object \
memory-backend-file,id=mem1,share,mem-path=/tmp/nvdimm1,size=10G -device \
nvdimm,memdev=mem1,id=nv1
It is disabled on default
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduce "nvdimm" device which is based on pc-dimm device type
Currently, nothing is specific for nvdimm but hotplug is disabled
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
setting gap to TRUE will make sparse DIMM
address auto allocation, leaving gaps between
a new DIMM address and preceeding existing DIMM.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
pc_dimm_plug() has code that will be needed for memory plug handlers
in other archs too. Extract code from pc_dimm_plug() into a generic
routine pc_dimm_memory_plug() that resides in pc-dimm.c. Also
correspondingly refactor re-usable unplug code into pc_dimm_memory_unplug().
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Tested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Move hotplug_memory_base and hotplug_memory fields of PCMachineState
into a separate structure so that the same can be made use of from
other architectures supporing memory hotplug.
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Tested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Now that pc_existing_dimms_capacity() is an API, include Error pointer
as an argument and modify the caller appropriately.
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Move pc_existing_dimms_capacity() to pc-dimm.c since it would be needed
by PowerPC memory hotplug code too.
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
When running in KVM mode, kvm_set_phys_mem() will silently
fail if registered MemoryRegion address/size is not page
aligned. Causing memory hotplug failure in guest.
Mapping non aligned MemoryRegion in TCG mode 'works', but
sane guest OS still expects page aligned memory module
and fails to initialize it if it's not aligned.
So do not allow non aligned (i.e. valid) address/size
values for DIMM to avoid either KVM failure or guest
issues caused by it.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
... allowing to get state of present memory devices.
Currently implemented only for PCDIMMDevice.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
- if slot property is not specified on -device/device_add command,
treat default value as request for assigning PCDIMMDevice to
the first free slot.
- if slot is provided with -device/device_add command, attempt to
use it or fail command if it's already occupied.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
- if 'addr' property is not specified on -device/device_add command,
treat the default value as request for assigning PCDIMMDevice to
the first free memory region.
- if 'addr' is provided with -device/device_add command, attempt to
use it or fail command if it's already occupied or falls inside
of an existing PCDIMMDevice memory region.
Note:
GCompareFunc(a, b) used by g_slist_insert_sorted() returns 'gint',
however it might be too small to fit difference between
2 addresses. So use 128bit to calculate the difference and normalize
result to -1/0/1 return values.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Andrey Korolyov <andrey@xdel.ru>
MST: commit log tweaks
Each hotplug-able memory slot is a PCDIMMDevice.
A hot-add operation for a memory device:
- creates a new PCDIMMDevice and makes hotplug controller to map it into
guest address space
Hotplug operations are done through normal device_add commands.
For migration case, all hotplugged memory devices on source should be
specified on target's command line using '-device' option with
properties set to the same values as on source.
To simplify review, patch introduces only PCDIMMDevice QOM skeleton that
will be extended by following patches to implement actual memory hotplug
and related functions.
Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis@profitbricks.com>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
