Re: [PATCH v19 2/8] numa: Extend CLI to provide memory latency and bandwidth information

[Markus Armbruster <armbru@redhat.com>]

Tao Xu <tao3.xu@intel.com> writes:

> From: Liu Jingqi <jingqi.liu@intel.com>
>
> Add -numa hmat-lb option to provide System Locality Latency and
> Bandwidth Information. These memory attributes help to build
> System Locality Latency and Bandwidth Information Structure(s)
> in ACPI Heterogeneous Memory Attribute Table (HMAT). Before using
> hmat-lb option, enable HMAT with -machine hmat=on.
>
> Signed-off-by: Liu Jingqi <jingqi.liu@intel.com>
> Signed-off-by: Tao Xu <tao3.xu@intel.com>
> ---
[...]
> diff --git a/qapi/machine.json b/qapi/machine.json
> index 27d0e37534..cf9851fcd1 100644
> --- a/qapi/machine.json
> +++ b/qapi/machine.json
> @@ -426,10 +426,12 @@
>  #
>  # @cpu: property based CPU(s) to node mapping (Since: 2.10)
>  #
> +# @hmat-lb: memory latency and bandwidth information (Since: 5.0)
> +#
>  # Since: 2.1
>  ##
>  { 'enum': 'NumaOptionsType',
> -  'data': [ 'node', 'dist', 'cpu' ] }
> +  'data': [ 'node', 'dist', 'cpu', 'hmat-lb' ] }
>  
>  ##
>  # @NumaOptions:
> @@ -444,7 +446,8 @@
>    'data': {
>      'node': 'NumaNodeOptions',
>      'dist': 'NumaDistOptions',
> -    'cpu': 'NumaCpuOptions' }}
> +    'cpu': 'NumaCpuOptions',
> +    'hmat-lb': 'NumaHmatLBOptions' }}
>  
>  ##
>  # @NumaNodeOptions:
> @@ -557,6 +560,92 @@
>     'base': 'CpuInstanceProperties',
>     'data' : {} }
>  
> +##
> +# @HmatLBMemoryHierarchy:
> +#
> +# The memory hierarchy in the System Locality Latency and Bandwidth
> +# Information Structure of HMAT (Heterogeneous Memory Attribute Table)
> +#
> +# For more information about @HmatLBMemoryHierarchy see chapter

@HmatLBMemoryHierarchy, see

> +# 5.2.27.4: Table 5-146: Field "Flags" of ACPI 6.3 spec.
> +#
> +# @memory: the structure represents the memory performance
> +#
> +# @first-level: first level of memory side cache
> +#
> +# @second-level: second level of memory side cache
> +#
> +# @third-level: third level of memory side cache
> +#
> +# Since: 5.0
> +##
> +{ 'enum': 'HmatLBMemoryHierarchy',
> +  'data': [ 'memory', 'first-level', 'second-level', 'third-level' ] }
> +
> +##
> +# @HmatLBDataType:
> +#
> +# Data type in the System Locality Latency and Bandwidth
> +# Information Structure of HMAT (Heterogeneous Memory Attribute Table)
> +#
> +# For more information about @HmatLBDataType see chapter

@HmatLBDataType, see

> +# 5.2.27.4: Table 5-146:  Field "Data Type" of ACPI 6.3 spec.
> +#
> +# @access-latency: access latency (nanoseconds)
> +#
> +# @read-latency: read latency (nanoseconds)
> +#
> +# @write-latency: write latency (nanoseconds)
> +#
> +# @access-bandwidth: access bandwidth (Bytes per second)
> +#
> +# @read-bandwidth: read bandwidth (Bytes per second)
> +#
> +# @write-bandwidth: write bandwidth (Bytes per second)
> +#
> +# Since: 5.0
> +##
> +{ 'enum': 'HmatLBDataType',
> +  'data': [ 'access-latency', 'read-latency', 'write-latency',
> +            'access-bandwidth', 'read-bandwidth', 'write-bandwidth' ] }
> +
> +##
> +# @NumaHmatLBOptions:
> +#
> +# Set the system locality latency and bandwidth information
> +# between Initiator and Target proximity Domains.
> +#
> +# For more information about @NumaHmatLBOptions see chapter

@NumaHmatLBOptions, see

> +# 5.2.27.4: Table 5-146 of ACPI 6.3 spec.
> +#
> +# @initiator: the Initiator Proximity Domain.
> +#
> +# @target: the Target Proximity Domain.
> +#
> +# @hierarchy: the Memory Hierarchy. Indicates the performance
> +#             of memory or side cache.
> +#
> +# @data-type: presents the type of data, access/read/write
> +#             latency or hit latency.
> +#
> +# @latency: the value of latency from @initiator to @target
> +#           proximity domain, the latency unit is "ns(nanosecond)".
> +#
> +# @bandwidth: the value of bandwidth between @initiator and @target
> +#             proximity domain, the bandwidth unit is
> +#             "Bytes per second".
> +#
> +# Since: 5.0
> +##
> +{ 'struct': 'NumaHmatLBOptions',
> +    'data': {
> +    'initiator': 'uint16',
> +    'target': 'uint16',
> +    'hierarchy': 'HmatLBMemoryHierarchy',
> +    'data-type': 'HmatLBDataType',
> +    '*latency': 'uint64',
> +    '*bandwidth': 'size' }}
> +
>  ##
>  # @HostMemPolicy:
>  #
> diff --git a/qemu-options.hx b/qemu-options.hx
> index 63f6b33322..23303fc7d7 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -168,16 +168,19 @@ DEF("numa", HAS_ARG, QEMU_OPTION_numa,
>      "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
>      "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
>      "-numa dist,src=source,dst=destination,val=distance\n"
> -    "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n",
> +    "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
> +    "-numa hmat-lb,initiator=node,target=node,hierarchy=memory|first-level|second-level|third-level,data-type=access-latency|read-latency|write-latency[,latency=lat][,bandwidth=bw]\n",
>      QEMU_ARCH_ALL)
>  STEXI
>  @item -numa node[,mem=@var{size}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}][,initiator=@var{initiator}]
>  @itemx -numa node[,memdev=@var{id}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}][,initiator=@var{initiator}]
>  @itemx -numa dist,src=@var{source},dst=@var{destination},val=@var{distance}
>  @itemx -numa cpu,node-id=@var{node}[,socket-id=@var{x}][,core-id=@var{y}][,thread-id=@var{z}]
> +@itemx -numa hmat-lb,initiator=@var{node},target=@var{node},hierarchy=@var{hierarchy},data-type=@var{tpye}[,latency=@var{lat}][,bandwidth=@var{bw}]
>  @findex -numa
>  Define a NUMA node and assign RAM and VCPUs to it.
>  Set the NUMA distance from a source node to a destination node.
> +Set the ACPI Heterogeneous Memory Attributes for the given nodes.
>  
>  Legacy VCPU assignment uses @samp{cpus} option where
>  @var{firstcpu} and @var{lastcpu} are CPU indexes. Each
> @@ -256,6 +259,49 @@ specified resources, it just assigns existing resources to NUMA
>  nodes. This means that one still has to use the @option{-m},
>  @option{-smp} options to allocate RAM and VCPUs respectively.
>  
> +Use @samp{hmat-lb} to set System Locality Latency and Bandwidth Information
> +between initiator and target NUMA nodes in ACPI Heterogeneous Attribute Memory Table (HMAT).
> +Initiator NUMA node can create memory requests, usually it has one or more processors.
> +Target NUMA node contains addressable memory.
> +
> +In @samp{hmat-lb} option, @var{node} are NUMA node IDs. @var{hierarchy} is the memory
> +hierarchy of the target NUMA node: if @var{hierarchy} is 'memory', the structure
> +represents the memory performance; if @var{hierarchy} is 'first-level|second-level|third-level',
> +this structure represents aggregated performance of memory side caches for each domain.
> +@var{type} of 'data-type' is type of data represented by this structure instance:
> +if 'hierarchy' is 'memory', 'data-type' is 'access|read|write' latency or 'access|read|write'
> +bandwidth of the target memory; if 'hierarchy' is 'first-level|second-level|third-level',
> +'data-type' is 'access|read|write' hit latency or 'access|read|write' hit bandwidth of the
> +target memory side cache.
> +
> +@var{lat} is latency value in nanoseconds. @var{bw} is bandwidth value,
> +the possible value and units are NUM[M|G|T], mean that the bandwidth value are
> +NUM byte per second (or MB/s, GB/s or TB/s depending on used suffix).
> +And if input bandwidth value without any unit, the unit will be byte per second.

This sentence feels redundant to me.

> +Note that if latency or bandwidth value is 0, means the corresponding latency or
> +bandwidth information is not provided.
> +
> +For example, the following options describe 2 NUMA nodes. Node 0 has 2 cpus and
> +a ram, node 1 has only a ram. The processors in node 0 access memory in node
> +0 with access-latency 5 nanoseconds, access-bandwidth is 200 MB/s;
> +The processors in NUMA node 0 access memory in NUMA node 1 with access-latency 10
> +nanoseconds, access-bandwidth is 100 MB/s.
> +@example
> +-machine hmat=on \
> +-m 2G \
> +-object memory-backend-ram,size=1G,id=m0 \
> +-object memory-backend-ram,size=1G,id=m1 \
> +-smp 2 \
> +-numa node,nodeid=0,memdev=m0 \
> +-numa node,nodeid=1,memdev=m1,initiator=0 \
> +-numa cpu,node-id=0,socket-id=0 \
> +-numa cpu,node-id=0,socket-id=1 \
> +-numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
> +-numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
> +-numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
> +-numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M
> +@end example
> +
>  ETEXI
>  
>  DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,