@@ -56,7 +56,7 @@ static int n_mem_addr_cells, n_mem_size_cells;staticintform1_affinity;#define MAX_DISTANCE_REF_POINTS 4-staticintdistance_ref_points_depth;+staticintmax_domain_index;staticconst__be32*distance_ref_points;staticintdistance_lookup_table[MAX_NUMNODES][MAX_DISTANCE_REF_POINTS];
@@ -169,7 +169,7 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)inti,index;-for(i=0;i<distance_ref_points_depth;i++){+for(i=0;i<max_domain_index;i++){index=be32_to_cpu(distance_ref_points[i]);if(cpu1_assoc[index]==cpu2_assoc[index])break;
@@ -193,7 +193,7 @@ int __node_distance(int a, int b)if(!form1_affinity)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);-for(i=0;i<distance_ref_points_depth;i++){+for(i=0;i<max_domain_index;i++){if(distance_lookup_table[a][i]==distance_lookup_table[b][i])break;
The associativity details of the newly added resourced are collected from
the hypervisor via "ibm,configure-connector" rtas call. Update the numa
distance details of the newly added numa node after the above call. In
later patch we will remove updating NUMA distance when we are looking
for node id from associativity array.
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 41 +++++++++++++++++++
arch/powerpc/platforms/pseries/hotplug-cpu.c | 2 +
.../platforms/pseries/hotplug-memory.c | 2 +
arch/powerpc/platforms/pseries/pseries.h | 1 +
4 files changed, 46 insertions(+)
Instead of updating NUMA distance every time we lookup a node id
from the associativity property, add helpers that can be used
during boot which does this only once. Also remove the distance
update from node id lookup helpers.
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 133 +++++++++++++++++++++++++++--------------
1 file changed, 87 insertions(+), 46 deletions(-)
@@ -208,22 +208,6 @@ int __node_distance(int a, int b)}EXPORT_SYMBOL(__node_distance);-staticvoidinitialize_distance_lookup_table(intnid,-const__be32*associativity)-{-inti;--if(affinity_form!=FORM1_AFFINITY)-return;--for(i=0;i<max_domain_index;i++){-const__be32*entry;--entry=&associativity[be32_to_cpu(distance_ref_points[i])-1];-distance_lookup_table[nid][i]=of_read_number(entry,1);-}-}-/**Returnsnidintherange[0..nr_node_ids],or-1ifnousefulNUMA*infoisfound.
@@ -241,15 +225,6 @@ static int associativity_to_nid(const __be32 *associativity)/* POWER4 LPAR uses 0xffff as invalid node */if(nid==0xffff||nid>=nr_node_ids)nid=NUMA_NO_NODE;--if(nid>0&&-of_read_number(associativity,1)>=max_domain_index){-/*-*Skipthelengthfieldandsendstartofassociativityarray-*/-initialize_distance_lookup_table(nid,associativity+1);-}-out:returnnid;}
This helper is only used with the dispatch trace log collection.
A later patch will add Form2 affinity support and this change helps
in keeping that simpler. Also add a comment explaining we don't expect
the code to be called with FORM0
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
@@ -182,6 +182,14 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)returndist;}+intcpu_distance(__be32*cpu1_assoc,__be32*cpu2_assoc)+{+/* We should not get called with FORM0 */+VM_WARN_ON(affinity_form==FORM0_AFFINITY);++return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+}+/* must hold reference to node during call */staticconst__be32*of_get_associativity(structdevice_node*dev){
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.+Form 0 is the older format and is now considered deprecated.++Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.+A value of 1 indicates the usage of Form 1 associativity.++Form 0+-----+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).++Form 1+-----+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity+device tree properties are used to determine the NUMA distance between resource groups/domains. ++The âibm,associativityâ property contains one or more lists of numbers (domainID)+representing the resourceâs platform grouping domains.++The âibm,associativity-reference-pointsâ property contains one or more list of numbers+(domain index) that represents the 1 based ordinal in the associativity lists of the most+significant boundary, with subsequent entries indicating progressively less significant boundaries.++Linux kernel uses the domain id of the most significant boundary (aka primary domain)+as the NUMA node id. Linux kernel computes NUMA distance between two domains by+recursively comparing if they belong to the same higher-level domains. For mismatch+at every higher level of the resource group, the kernel doubles the NUMA distance between+the comparing domains.++Form 2+-------+Form 2 associativity format adds separate device tree properties representing NUMA node distance+thereby making the node distance computation flexible. Form 2 also allows flexible primary+domain numbering. With numa distance computation now detached from the index value of+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the+same domain index representing resource groups of different performance/latency characteristics.++Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the+"ibm,architecture-vec-5" property.++"ibm,numa-lookup-index-table" property contains one or more list numbers representing+the domainIDs present in the system. The offset of the domainID in this property is considered+the domainID index.++prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by+N domainID encoded as with encode-int++For ex:+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.++"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA+distance between resource groups/domains present in the system.++prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.++For ex:+ibm,numa-lookup-index-table = {3, 0, 8, 40}+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}++| 0 8 40+--|------------+ |+0 | 10 20 80+ |+8 | 20 10 160+ |+40| 80 160 10++With Form2 "ibm,associativity" for resources is listed as below:++"ibm,associativity" property for resources in node 0, 8 and 40+{ 4, 6, 7, 0, 0}+{ 4, 6, 9, 8, 8}+{ 4, 6, 7, 0, 40}++With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }++With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes+the kernel should use when using persistent memory devices. Persistent memory devices+can also be used as regular memory using DAX KMEM driver and primary domainID indicates+the numa node number OS should use when using these devices as regular memory. Secondary+domainID is the numa node number that should be used when using this device as+persistent memory. In the later case, we are interested in the locality of the+device to an established numa node. In the above example, if the last row represents a+persistent memory device/resource, NUMA node number 40 will be used when using the device+as regular memory and NUMA node number 0 will be the device numa node when using it as+a persistent memory device.++Each resource (drcIndex) now also supports additional optional device tree properties.+These properties are marked optional because the platform can choose not to export+them and provide the system topology details using the earlier defined device tree+properties alone. The optional device tree properties are used when adding new resources+(DLPAR) and when the platform didn't provide the topology details of the domain which+contains the newly added resource during boot.++"ibm,numa-lookup-index" property contains a number representing the domainID index to be used+when building the NUMA distance of the numa node to which this resource belongs. The domain id+of the new resource can be obtained from the existing "ibm,associativity" property. This+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.+The value is 1 based array index value.++prop-encoded-array: An integer encoded as with encode-int specifying the domainID index++"ibm,numa-distance" property contains one or more list of numbers presenting the NUMA distance+from this resource domain to other resources.++prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.++For ex:+ibm,associativity = { 4, 5, 6, 7, 50}+ibm,numa-lookup-index = { 4 }+ibm,numa-distance = {8, 16, 32, 8, 1, 16, 32, 8, 1}++resulting in a new toplogy as below.+| 0 8 40 50+--|------------------+ |+0 | 10 20 80 160+ |+8 | 20 10 160 320+ |+40| 80 160 10 80+ |+50| 160 320 80 10
@@ -178,7 +204,6 @@ static int __cpu_form1_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)break;dist++;}-returndist;}
@@ -186,8 +211,9 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc){/* We should not get called with FORM0 */VM_WARN_ON(affinity_form==FORM0_AFFINITY);--return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+if(affinity_form==FORM1_AFFINITY)+return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+return__cpu_form2_distance(cpu1_assoc,cpu2_assoc);}/* must hold reference to node during call */
@@ -201,7 +227,9 @@ int __node_distance(int a, int b)inti;intdistance=LOCAL_DISTANCE;-if(affinity_form==FORM0_AFFINITY)+if(affinity_form==FORM2_AFFINITY)+returnnuma_distance_table[a][b];+elseif(affinity_form==FORM0_AFFINITY)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);for(i=0;i<max_domain_index;i++){
@@ -303,15 +331,116 @@ static void initialize_form1_numa_distance(struct device_node *node)/**Usedtoupdatedistanceinformationw.r.tnewlyaddednode.+*ibm,numa-lookup-index->4+*ibm,numa-distance->{5,20,40,60,80,10}*/voidupdate_numa_distance(structdevice_node*node){+inti,nid,other_nid,other_nid_index=0;+const__be32*numa_indexp;+const__u8*numa_distancep;+intnuma_index,max_numa_index,numa_distance;+if(affinity_form==FORM0_AFFINITY)return;elseif(affinity_form==FORM1_AFFINITY){initialize_form1_numa_distance(node);return;}+/* FORM2 affinity */++nid=of_node_to_nid_single(node);+if(nid==NUMA_NO_NODE)+return;++/* Already initialized */+if(numa_distance_table[nid][nid]!=-1)+return;+/*+*updatenodedistanceifnotalreadypopulated.+*/+numa_distancep=of_get_property(node,"ibm,numa-distance",NULL);+if(!numa_distancep)+return;++numa_indexp=of_get_property(node,"ibm,numa-lookup-index",NULL);+if(!numa_indexp)+return;++numa_index=of_read_number(numa_indexp,1);+/*+*updatethenuma_id_index_table.Devicetreelookatindextableas+*1basedarrayindexing.+*/+numa_id_index_table[numa_index-1]=nid;++max_numa_index=of_read_number((const__be32*)numa_distancep,1);+VM_WARN_ON(max_numa_index!=2*numa_index);+/* Skip the size which is encoded int */+numa_distancep+=sizeof(__be32);++/*+*Firstfillthedistanceinformationfromothernodetothisnode.+*/+other_nid_index=0;+for(i=0;i<numa_index;i++){+numa_distance=numa_distancep[i];+other_nid=numa_id_index_table[other_nid_index++];+numa_distance_table[other_nid][nid]=numa_distance;+}++other_nid_index=0;+for(;i<max_numa_index;i++){+numa_distance=numa_distancep[i];+other_nid=numa_id_index_table[other_nid_index++];+numa_distance_table[nid][other_nid]=numa_distance;+}+}++/*+*ibm,numa-lookup-index-table={N,domainid1,domainid2,.....domainidN}+*ibm,numa-distance-table={N,1,2,4,5,1,6,....Nelements}+*/+staticvoidinitialize_form2_numa_distance_lookup_table(structdevice_node*root)+{+const__u8*numa_dist_table;+const__be32*numa_lookup_index;+intnuma_dist_table_length;+intmax_numa_index,distance_index;+inti,curr_row=0,curr_column=0;++numa_lookup_index=of_get_property(root,"ibm,numa-lookup-index-table",NULL);+max_numa_index=of_read_number(&numa_lookup_index[0],1);++/* first element of the array is the size and is encode-int */+numa_dist_table=of_get_property(root,"ibm,numa-distance-table",NULL);+numa_dist_table_length=of_read_number((const__be32*)&numa_dist_table[0],1);+/* Skip the size which is encoded int */+numa_dist_table+=sizeof(__be32);++pr_debug(" numa_dist_table_len = %d, numa_dist_indexes_len = %d \n",+numa_dist_table_length,max_numa_index);++for(i=0;i<max_numa_index;i++)+/* +1 skip the max_numa_index in the property */+numa_id_index_table[i]=of_read_number(&numa_lookup_index[i+1],1);+++VM_WARN_ON(numa_dist_table_length!=max_numa_index*max_numa_index);++for(distance_index=0;distance_index<numa_dist_table_length;distance_index++){+intnodeA=numa_id_index_table[curr_row];+intnodeB=numa_id_index_table[curr_column++];++numa_distance_table[nodeA][nodeB]=numa_dist_table[distance_index];++pr_debug("dist[%d][%d]=%d ",nodeA,nodeB,numa_distance_table[nodeA][nodeB]);+if(curr_column>=max_numa_index){+curr_row++;+/* reset the column */+curr_column=0;+}+}}staticint__initfind_primary_domain_index(void)
@@ -324,6 +453,9 @@ static int __init find_primary_domain_index(void)*/if(firmware_has_feature(FW_FEATURE_OPAL)){affinity_form=FORM1_AFFINITY;+}elseif(firmware_has_feature(FW_FEATURE_FORM2_AFFINITY)){+dbg("Using form 2 affinity\n");+affinity_form=FORM2_AFFINITY;}elseif(firmware_has_feature(FW_FEATURE_FORM1_AFFINITY)){dbg("Using form 1 affinity\n");affinity_form=FORM1_AFFINITY;
@@ -368,8 +500,17 @@ static int __init find_primary_domain_index(void)index=of_read_number(&distance_ref_points[1],1);}else{+/*+*BothFORM1andFORM2affinityfindtheprimarydomaindetails+*atthesameoffset.+*/index=of_read_number(distance_ref_points,1);}+/*+*IfitisFORM2alsoinitializethedistancetablehere.+*/+if(affinity_form==FORM2_AFFINITY)+initialize_form2_numa_distance_lookup_table(root);/**Warnandcapifthehardwaresupportsmorethan
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);+*secondary=of_read_number(&associativity[secondary_index],1);+}+if(*primary==0xffff||*primary>=nr_node_ids)+*primary=NUMA_NO_NODE;++if(*secondary==0xffff||*secondary>=nr_node_ids)+*secondary=NUMA_NO_NODE;+return0;+}+/* Returns the nid associated with the given device tree node,*or-1ifnotfound.*/
@@ -1001,9 +1001,6 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)ndr_desc.res,p->dn);gotoerr;}-if(target_nid!=online_nid)-dev_info(dev,"Region registered with target node %d and online node %d",-target_nid,online_nid);mutex_lock(&papr_ndr_lock);list_add_tail(&p->region_list,&papr_nd_regions);
@@ -1096,7 +1093,7 @@ static int papr_scm_probe(struct platform_device *pdev)structpapr_scm_priv*p;constchar*uuid_str;u64uuid[2];-intrc;+intrc,numa_node;/* check we have all the required DT properties */if(of_property_read_u32(dn,"ibm,my-drc-index",&drc_index)){
@@ -1119,11 +1116,20 @@ static int papr_scm_probe(struct platform_device *pdev)return-ENODEV;}-p=kzalloc(sizeof(*p),GFP_KERNEL);if(!p)return-ENOMEM;+if(get_primary_and_secondary_domain(dn,&p->target_node,&numa_node)){+dev_err(&pdev->dev,"%pOF: missing NUMA attributes!\n",dn);+rc=-ENODEV;+gotoerr;+}+p->numa_node=numa_map_to_online_node(numa_node);+if(numa_node!=p->numa_node)+dev_info(&pdev->dev,"Region registered with online node %d and device tree node %d",+p->numa_node,numa_node);+/* Initialize the dimm mutex */mutex_init(&p->health_mutex);
@@ -190,7 +193,7 @@ int __node_distance(int a, int b)inti;intdistance=LOCAL_DISTANCE;-if(!form1_affinity)+if(affinity_form==FORM0_AFFINITY)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);for(i=0;i<max_domain_index;i++){
with 'device-node=1' instead of 'persistent=nodeid=1' in the nvdimm parameter
up here.
Aneesh Kumar K.V (8):
powerpc/pseries: rename min_common_depth to primary_domain_index
powerpc/pseries: rename distance_ref_points_depth to max_domain_index
powerpc/pseries: Rename TYPE1_AFFINITY to FORM1_AFFINITY
powerpc/pseries: Consolidate DLPAR NUMA distance update
powerpc/pseries: Consolidate NUMA distance update during boot
powerpc/pseries: Add a helper for form1 cpu distance
powerpc/pseries: Add support for FORM2 associativity
powerpc/papr_scm: Use FORM2 associativity details
Series:
Tested-by: Daniel Henrique Barboza <redacted>
@@ -56,7 +56,7 @@ static int n_mem_addr_cells, n_mem_size_cells;staticintform1_affinity;#define MAX_DISTANCE_REF_POINTS 4-staticintdistance_ref_points_depth;+staticintmax_domain_index;staticconst__be32*distance_ref_points;staticintdistance_lookup_table[MAX_NUMNODES][MAX_DISTANCE_REF_POINTS];
@@ -169,7 +169,7 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)inti,index;-for(i=0;i<distance_ref_points_depth;i++){+for(i=0;i<max_domain_index;i++){index=be32_to_cpu(distance_ref_points[i]);if(cpu1_assoc[index]==cpu2_assoc[index])break;
@@ -193,7 +193,7 @@ int __node_distance(int a, int b)if(!form1_affinity)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);-for(i=0;i<distance_ref_points_depth;i++){+for(i=0;i<max_domain_index;i++){if(distance_lookup_table[a][i]==distance_lookup_table[b][i])break;
@@ -240,7 +240,7 @@ static int associativity_to_nid(const __be32 *associativity)nid=NUMA_NO_NODE;if(nid>0&&-of_read_number(associativity,1)>=distance_ref_points_depth){+of_read_number(associativity,1)>=max_domain_index){/**Skipthelengthfieldandsendstartofassociativityarray*/
@@ -310,14 +310,14 @@ static int __init find_primary_domain_index(void)*/distance_ref_points=of_get_property(root,"ibm,associativity-reference-points",-&distance_ref_points_depth);+&max_domain_index);if(!distance_ref_points){dbg("NUMA: ibm,associativity-reference-points not found.\n");gotoerr;}-distance_ref_points_depth/=sizeof(int);+max_domain_index/=sizeof(int);if(firmware_has_feature(FW_FEATURE_OPAL)||firmware_has_feature(FW_FEATURE_TYPE1_AFFINITY)){
@@ -328,7 +328,7 @@ static int __init find_primary_domain_index(void)if(form1_affinity){index=of_read_number(distance_ref_points,1);}else{-if(distance_ref_points_depth<2){+if(max_domain_index<2){printk(KERN_WARNING"NUMA: ""short ibm,associativity-reference-points\n");gotoerr;
@@ -341,10 +341,10 @@ static int __init find_primary_domain_index(void)*Warnandcapifthehardwaresupportsmorethan*MAX_DISTANCE_REF_POINTSdomains.*/-if(distance_ref_points_depth>MAX_DISTANCE_REF_POINTS){+if(max_domain_index>MAX_DISTANCE_REF_POINTS){printk(KERN_WARNING"NUMA: distance array capped at ""%d entries\n",MAX_DISTANCE_REF_POINTS);-distance_ref_points_depth=MAX_DISTANCE_REF_POINTS;+max_domain_index=MAX_DISTANCE_REF_POINTS;}of_node_put(root);
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 03:57:10
On Mon, Jun 14, 2021 at 10:09:59PM +0530, Aneesh Kumar K.V wrote:
quoted hunk
The associativity details of the newly added resourced are collected from
the hypervisor via "ibm,configure-connector" rtas call. Update the numa
distance details of the newly added numa node after the above call. In
later patch we will remove updating NUMA distance when we are looking
for node id from associativity array.
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 41 +++++++++++++++++++
arch/powerpc/platforms/pseries/hotplug-cpu.c | 2 +
.../platforms/pseries/hotplug-memory.c | 2 +
arch/powerpc/platforms/pseries/pseries.h | 1 +
4 files changed, 46 insertions(+)
@@ -287,6 +287,47 @@ int of_node_to_nid(struct device_node *device)}EXPORT_SYMBOL(of_node_to_nid);+staticvoid__initialize_form1_numa_distance(const__be32*associativity)+{+inti,nid;++if(of_read_number(associativity,1)>=primary_domain_index){+nid=of_read_number(&associativity[primary_domain_index],1);++for(i=0;i<max_domain_index;i++){+const__be32*entry;++entry=&associativity[be32_to_cpu(distance_ref_points[i])];+distance_lookup_table[nid][i]=of_read_number(entry,1);+}+}+}
This logic is almost identicaly to initialize_distance_lookup_table()
- it would be good if they could be consolidated, so it's clear that
coldplugged and hotplugged nodes are parsing the NUMA information in
the same way.
quoted hunk
+
+static void initialize_form1_numa_distance(struct device_node *node)
+{
+ const __be32 *associativity;
+
+ associativity = of_get_associativity(node);
+ if (!associativity)
+ return;
+
+ __initialize_form1_numa_distance(associativity);
+ return;
+}
+
+/*
+ * Used to update distance information w.r.t newly added node.
+ */
+void update_numa_distance(struct device_node *node)
+{
+ if (affinity_form == FORM0_AFFINITY)
+ return;
+ else if (affinity_form == FORM1_AFFINITY) {
+ initialize_form1_numa_distance(node);
+ return;
+ }
+}
+
static int __init find_primary_domain_index(void)
{
int index;
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 03:57:33
On Mon, Jun 14, 2021 at 10:09:56PM +0530, Aneesh Kumar K.V wrote:
No functional change in this patch.
I think this needs a rationale as to why 'primary_domain_index' is a
better name than 'min_common_depth'. The meaning isn't obvious to me
from either name.
@@ -1259,7 +1259,7 @@ int cpu_to_coregroup_id(int cpu)gotoout;index=of_read_number(associativity,1);-if(index>min_common_depth+1)+if(index>primary_domain_index+1)returnof_read_number(&associativity[index-1],1);out:
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 03:57:56
On Mon, Jun 14, 2021 at 10:09:58PM +0530, Aneesh Kumar K.V wrote:
Also make related code cleanup that will allow adding FORM2_AFFINITY in
later patches. No functional change in this patch.
Signed-off-by: Aneesh Kumar K.V <redacted>
@@ -190,7 +193,7 @@ int __node_distance(int a, int b)inti;intdistance=LOCAL_DISTANCE;-if(!form1_affinity)+if(affinity_form==FORM0_AFFINITY)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);for(i=0;i<max_domain_index;i++){
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource
You describe form 2 below as well, which contradicts this.
+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.
+Form 0 is the older format and is now considered deprecated.
+
+Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
+A value of 1 indicates the usage of Form 1 associativity.
+
+Form 0
+-----
+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).
+
+Form 1
+-----
+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity
+device tree properties are used to determine the NUMA distance between resource groups/domains.
+
+The “ibm,associativity” property contains one or more lists of numbers (domainID)
+representing the resource’s platform grouping domains.
+
+The “ibm,associativity-reference-points” property contains one or more list of numbers
+(domain index) that represents the 1 based ordinal in the associativity lists of the most
+significant boundary, with subsequent entries indicating progressively less significant boundaries.
+
+Linux kernel uses the domain id of the most significant boundary (aka primary domain)
I thought we used the *least* significant boundary (the smallest
grouping, not the largest). That is, the last index, not the first.
Actually... come to think of it, I'm not even sure how to interpret
"most significant". Does that mean a change in grouping at that "most
significant" level results in the largest perfomance difference?
+as the NUMA node id. Linux kernel computes NUMA distance between two domains by
+recursively comparing if they belong to the same higher-level domains. For mismatch
+at every higher level of the resource group, the kernel doubles the NUMA distance between
+the comparing domains.
+
+Form 2
+-------
+Form 2 associativity format adds separate device tree properties representing NUMA node distance
+thereby making the node distance computation flexible. Form 2 also allows flexible primary
+domain numbering. With numa distance computation now detached from the index value of
+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the
+same domain index representing resource groups of different
performance/latency characteristics.
The meaning of "domain index" is not clear to me here.
+
+Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the
+"ibm,architecture-vec-5" property.
+
+"ibm,numa-lookup-index-table" property contains one or more list numbers representing
+the domainIDs present in the system. The offset of the domainID in this property is considered
+the domainID index.
You haven't really introduced the term "domainID". Is "domainID
index" the same as "domain index" above? It's not clear to me.
The distinction between "domain index" and "primary domain id" is also
not clear to me.
+prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by
+N domainID encoded as with encode-int
+
+For ex:
+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.
Above you say "Form 2 allows a large number of primary domain ids at
the same domain index", but this encoding doesn't appear to permit
that.
+"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA
+distance between resource groups/domains present in the system.
+
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,numa-lookup-index-table = {3, 0, 8, 40}
+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}
+
+ | 0 8 40
+--|------------
+ |
+0 | 10 20 80
+ |
+8 | 20 10 160
+ |
+40| 80 160 10
What's the reason for multiplying the values by 10 in the expanded
table version?
+
+With Form2 "ibm,associativity" for resources is listed as below:
+
+"ibm,associativity" property for resources in node 0, 8 and 40
+{ 4, 6, 7, 0, 0}
+{ 4, 6, 9, 8, 8}
+{ 4, 6, 7, 0, 40}
+
+With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }
+
+With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
What the heck is the secondary domainID?
+the kernel should use when using persistent memory devices. Persistent memory devices
+can also be used as regular memory using DAX KMEM driver and primary domainID indicates
+the numa node number OS should use when using these devices as regular memory. Secondary
+domainID is the numa node number that should be used when using this device as
+persistent memory. In the later case, we are interested in the locality of the
+device to an established numa node. In the above example, if the last row represents a
+persistent memory device/resource, NUMA node number 40 will be used when using the device
+as regular memory and NUMA node number 0 will be the device numa node when using it as
+a persistent memory device.
+
+Each resource (drcIndex) now also supports additional optional device tree properties.
+These properties are marked optional because the platform can choose not to export
+them and provide the system topology details using the earlier defined device tree
+properties alone. The optional device tree properties are used when adding new resources
+(DLPAR) and when the platform didn't provide the topology details of the domain which
+contains the newly added resource during boot.
+
+"ibm,numa-lookup-index" property contains a number representing the domainID index to be used
+when building the NUMA distance of the numa node to which this resource belongs. The domain id
+of the new resource can be obtained from the existing "ibm,associativity" property. This
+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.
+The value is 1 based array index value.
Am I correct in thinking that if we have an entirely form2 world, we'd
only need this and the ibm,associativity properties could be dropped?
+
+prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
+
+"ibm,numa-distance" property contains one or more list of numbers presenting the NUMA distance
+from this resource domain to other resources.
IIUC this is about extending the global distance table with
information for a new node. Is that correct?
The global distance table allows for the possibility of asymmetric
distances between nodes, but this does not. Is that intentional?
quoted hunk
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,associativity = { 4, 5, 6, 7, 50}
+ibm,numa-lookup-index = { 4 }
+ibm,numa-distance = {8, 16, 32, 8, 1, 16, 32, 8, 1}
+
+resulting in a new toplogy as below.
+ | 0 8 40 50
+--|------------------
+ |
+0 | 10 20 80 160
+ |
+8 | 20 10 160 320
+ |
+40| 80 160 10 80
+ |
+50| 160 320 80 10
@@ -178,7 +204,6 @@ static int __cpu_form1_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)break;dist++;}-returndist;}
@@ -186,8 +211,9 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc){/* We should not get called with FORM0 */VM_WARN_ON(affinity_form==FORM0_AFFINITY);--return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+if(affinity_form==FORM1_AFFINITY)+return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+return__cpu_form2_distance(cpu1_assoc,cpu2_assoc);}/* must hold reference to node during call */
@@ -201,7 +227,9 @@ int __node_distance(int a, int b)inti;intdistance=LOCAL_DISTANCE;-if(affinity_form==FORM0_AFFINITY)+if(affinity_form==FORM2_AFFINITY)+returnnuma_distance_table[a][b];+elseif(affinity_form==FORM0_AFFINITY)return((a==b)?LOCAL_DISTANCE:REMOTE_DISTANCE);for(i=0;i<max_domain_index;i++){
@@ -303,15 +331,116 @@ static void initialize_form1_numa_distance(struct device_node *node)/**Usedtoupdatedistanceinformationw.r.tnewlyaddednode.+*ibm,numa-lookup-index->4+*ibm,numa-distance->{5,20,40,60,80,10}*/voidupdate_numa_distance(structdevice_node*node){+inti,nid,other_nid,other_nid_index=0;+const__be32*numa_indexp;+const__u8*numa_distancep;+intnuma_index,max_numa_index,numa_distance;+if(affinity_form==FORM0_AFFINITY)return;elseif(affinity_form==FORM1_AFFINITY){initialize_form1_numa_distance(node);return;}+/* FORM2 affinity */++nid=of_node_to_nid_single(node);+if(nid==NUMA_NO_NODE)+return;++/* Already initialized */+if(numa_distance_table[nid][nid]!=-1)+return;+/*+*updatenodedistanceifnotalreadypopulated.+*/+numa_distancep=of_get_property(node,"ibm,numa-distance",NULL);+if(!numa_distancep)+return;++numa_indexp=of_get_property(node,"ibm,numa-lookup-index",NULL);+if(!numa_indexp)+return;++numa_index=of_read_number(numa_indexp,1);+/*+*updatethenuma_id_index_table.Devicetreelookatindextableas+*1basedarrayindexing.+*/+numa_id_index_table[numa_index-1]=nid;++max_numa_index=of_read_number((const__be32*)numa_distancep,1);+VM_WARN_ON(max_numa_index!=2*numa_index);+/* Skip the size which is encoded int */+numa_distancep+=sizeof(__be32);++/*+*Firstfillthedistanceinformationfromothernodetothisnode.+*/+other_nid_index=0;+for(i=0;i<numa_index;i++){+numa_distance=numa_distancep[i];+other_nid=numa_id_index_table[other_nid_index++];+numa_distance_table[other_nid][nid]=numa_distance;+}++other_nid_index=0;+for(;i<max_numa_index;i++){+numa_distance=numa_distancep[i];+other_nid=numa_id_index_table[other_nid_index++];+numa_distance_table[nid][other_nid]=numa_distance;+}+}++/*+*ibm,numa-lookup-index-table={N,domainid1,domainid2,.....domainidN}+*ibm,numa-distance-table={N,1,2,4,5,1,6,....Nelements}+*/+staticvoidinitialize_form2_numa_distance_lookup_table(structdevice_node*root)+{+const__u8*numa_dist_table;+const__be32*numa_lookup_index;+intnuma_dist_table_length;+intmax_numa_index,distance_index;+inti,curr_row=0,curr_column=0;++numa_lookup_index=of_get_property(root,"ibm,numa-lookup-index-table",NULL);+max_numa_index=of_read_number(&numa_lookup_index[0],1);++/* first element of the array is the size and is encode-int */+numa_dist_table=of_get_property(root,"ibm,numa-distance-table",NULL);+numa_dist_table_length=of_read_number((const__be32*)&numa_dist_table[0],1);+/* Skip the size which is encoded int */+numa_dist_table+=sizeof(__be32);++pr_debug(" numa_dist_table_len = %d, numa_dist_indexes_len = %d \n",+numa_dist_table_length,max_numa_index);++for(i=0;i<max_numa_index;i++)+/* +1 skip the max_numa_index in the property */+numa_id_index_table[i]=of_read_number(&numa_lookup_index[i+1],1);+++VM_WARN_ON(numa_dist_table_length!=max_numa_index*max_numa_index);++for(distance_index=0;distance_index<numa_dist_table_length;distance_index++){+intnodeA=numa_id_index_table[curr_row];+intnodeB=numa_id_index_table[curr_column++];++numa_distance_table[nodeA][nodeB]=numa_dist_table[distance_index];++pr_debug("dist[%d][%d]=%d ",nodeA,nodeB,numa_distance_table[nodeA][nodeB]);+if(curr_column>=max_numa_index){+curr_row++;+/* reset the column */+curr_column=0;+}+}}staticint__initfind_primary_domain_index(void)
@@ -324,6 +453,9 @@ static int __init find_primary_domain_index(void)*/if(firmware_has_feature(FW_FEATURE_OPAL)){affinity_form=FORM1_AFFINITY;+}elseif(firmware_has_feature(FW_FEATURE_FORM2_AFFINITY)){+dbg("Using form 2 affinity\n");+affinity_form=FORM2_AFFINITY;}elseif(firmware_has_feature(FW_FEATURE_FORM1_AFFINITY)){dbg("Using form 1 affinity\n");affinity_form=FORM1_AFFINITY;
@@ -368,8 +500,17 @@ static int __init find_primary_domain_index(void)index=of_read_number(&distance_ref_points[1],1);}else{+/*+*BothFORM1andFORM2affinityfindtheprimarydomaindetails+*atthesameoffset.+*/index=of_read_number(distance_ref_points,1);}+/*+*IfitisFORM2alsoinitializethedistancetablehere.+*/+if(affinity_form==FORM2_AFFINITY)+initialize_form2_numa_distance_lookup_table(root);/**Warnandcapifthehardwaresupportsmorethan
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 03:58:43
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted hunk
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -1001,9 +1001,6 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)ndr_desc.res,p->dn);gotoerr;}-if(target_nid!=online_nid)-dev_info(dev,"Region registered with target node %d and online node %d",-target_nid,online_nid);mutex_lock(&papr_ndr_lock);list_add_tail(&p->region_list,&papr_nd_regions);
@@ -1096,7 +1093,7 @@ static int papr_scm_probe(struct platform_device *pdev)structpapr_scm_priv*p;constchar*uuid_str;u64uuid[2];-intrc;+intrc,numa_node;/* check we have all the required DT properties */if(of_property_read_u32(dn,"ibm,my-drc-index",&drc_index)){
@@ -1119,11 +1116,20 @@ static int papr_scm_probe(struct platform_device *pdev)return-ENODEV;}-p=kzalloc(sizeof(*p),GFP_KERNEL);if(!p)return-ENOMEM;+if(get_primary_and_secondary_domain(dn,&p->target_node,&numa_node)){+dev_err(&pdev->dev,"%pOF: missing NUMA attributes!\n",dn);+rc=-ENODEV;+gotoerr;+}+p->numa_node=numa_map_to_online_node(numa_node);+if(numa_node!=p->numa_node)+dev_info(&pdev->dev,"Region registered with online node %d and device tree node %d",+p->numa_node,numa_node);+/* Initialize the dimm mutex */mutex_init(&p->health_mutex);
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 03:59:19
On Mon, Jun 14, 2021 at 10:10:01PM +0530, Aneesh Kumar K.V wrote:
This helper is only used with the dispatch trace log collection.
A later patch will add Form2 affinity support and this change helps
in keeping that simpler. Also add a comment explaining we don't expect
the code to be called with FORM0
Signed-off-by: Aneesh Kumar K.V <redacted>
@@ -182,6 +182,14 @@ int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)returndist;}+intcpu_distance(__be32*cpu1_assoc,__be32*cpu2_assoc)+{+/* We should not get called with FORM0 */+VM_WARN_ON(affinity_form==FORM0_AFFINITY);++return__cpu_form1_distance(cpu1_assoc,cpu2_assoc);+}+/* must hold reference to node during call */staticconst__be32*of_get_associativity(structdevice_node*dev){
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource
You describe form 2 below as well, which contradicts this.
Fixed as below.
PAPR interface currently supports different ways of communicating these resource
grouping details to the OS. These are referred to as Form 0, Form 1 and Form2
associativity grouping. Form 0 is the older format and is now considered deprecated.
Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
A value of 1 indicates the usage of Form 1 associativity. For Form 2 associativity
bit 2 of byte 5 in the "ibm,architecture-vec-5" property is used.
quoted
+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.
+Form 0 is the older format and is now considered deprecated.
+
+Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
+A value of 1 indicates the usage of Form 1 associativity.
+
+Form 0
+-----
+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).
+
+Form 1
+-----
+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity
+device tree properties are used to determine the NUMA distance between resource groups/domains.
+
+The “ibm,associativity” property contains one or more lists of numbers (domainID)
+representing the resource’s platform grouping domains.
+
+The “ibm,associativity-reference-points” property contains one or more list of numbers
+(domain index) that represents the 1 based ordinal in the associativity lists of the most
+significant boundary, with subsequent entries indicating progressively less significant boundaries.
+
+Linux kernel uses the domain id of the most significant boundary (aka primary domain)
I thought we used the *least* significant boundary (the smallest
grouping, not the largest). That is, the last index, not the first.
Actually... come to think of it, I'm not even sure how to interpret
"most significant". Does that mean a change in grouping at that "most
significant" level results in the largest perfomance difference?
PAPR defines "most significant" as below
When the “ibm,architecture-vec-5” property byte 5 bit 0 has the value of one, the “ibm,associativ-
ity-reference-points” property indicates boundaries between associativity domains presented by the
“ibm,associativity” property containing “near” and “far” resources. The
first such boundary in the list represents the 1 based ordinal in the
associativity lists of the most significant boundary, with subsequent
entries indicating progressively less significant boundaries
I would interpret it as the boundary where we start defining NUMA nodes.
quoted
+as the NUMA node id. Linux kernel computes NUMA distance between two domains by
+recursively comparing if they belong to the same higher-level domains. For mismatch
+at every higher level of the resource group, the kernel doubles the NUMA distance between
+the comparing domains.
+
+Form 2
+-------
+Form 2 associativity format adds separate device tree properties representing NUMA node distance
+thereby making the node distance computation flexible. Form 2 also allows flexible primary
+domain numbering. With numa distance computation now detached from the index value of
+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the
+same domain index representing resource groups of different
performance/latency characteristics.
The meaning of "domain index" is not clear to me here.
Sorry for the confusion there. domain index is the index where domainID
is appearing. W.r.t "ibm,associativity" we have
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains. If we can look at
an example property.
{ 4, 6, 7, 0, 0}
{ 4, 6, 7, 0, 40}
With Form 1 both NUMA node 0 and 40 will appear at the same distance.
They both are at domain index 4. With Form 2 we can represent them with
different NUMA distance values.
quoted
+
+Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the
+"ibm,architecture-vec-5" property.
+
+"ibm,numa-lookup-index-table" property contains one or more list numbers representing
+the domainIDs present in the system. The offset of the domainID in this property is considered
+the domainID index.
You haven't really introduced the term "domainID". Is "domainID
index" the same as "domain index" above? It's not clear to me.
The earlier part of the documented said
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains.
I will update domain index to domainID index.
The distinction between "domain index" and "primary domain id" is also
not clear to me.
primary domain id is the domainID appearing in the primary domainID
index. Linux kenrel also use that as the NUMA node number.
Primary domainID index is defined by ibm,associativity-reference-points
and we consider that as the most significant resource group boundary.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
quoted
+prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by
+N domainID encoded as with encode-int
+
+For ex:
+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.
Above you say "Form 2 allows a large number of primary domain ids at
the same domain index", but this encoding doesn't appear to permit
that.
I didn't follow that question.
quoted
+"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA
+distance between resource groups/domains present in the system.
+
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,numa-lookup-index-table = {3, 0, 8, 40}
+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}
+
+ | 0 8 40
+--|------------
+ |
+0 | 10 20 80
+ |
+8 | 20 10 160
+ |
+40| 80 160 10
What's the reason for multiplying the values by 10 in the expanded
table version?
That was me missing a document update. Since we used 8 bits to encode
distance at some point we were looking at a SCALE factor. But later
realized other architectures also restrict distance to 8 bits. I will
update ibm,numa-distance-table in the document.
quoted
+
+With Form2 "ibm,associativity" for resources is listed as below:
+
+"ibm,associativity" property for resources in node 0, 8 and 40
+{ 4, 6, 7, 0, 0}
+{ 4, 6, 9, 8, 8}
+{ 4, 6, 7, 0, 40}
+
+With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }
+
+With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
What the heck is the secondary domainID
domainID appearing the secondary domainID index.
ibm,associativity-reference-points gives an indication of different
hierachy of resource grouping as below.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
quoted
+the kernel should use when using persistent memory devices. Persistent memory devices
+can also be used as regular memory using DAX KMEM driver and primary domainID indicates
+the numa node number OS should use when using these devices as regular memory. Secondary
+domainID is the numa node number that should be used when using this device as
+persistent memory. In the later case, we are interested in the locality of the
+device to an established numa node. In the above example, if the last row represents a
+persistent memory device/resource, NUMA node number 40 will be used when using the device
+as regular memory and NUMA node number 0 will be the device numa node when using it as
+a persistent memory device.
+
+Each resource (drcIndex) now also supports additional optional device tree properties.
+These properties are marked optional because the platform can choose not to export
+them and provide the system topology details using the earlier defined device tree
+properties alone. The optional device tree properties are used when adding new resources
+(DLPAR) and when the platform didn't provide the topology details of the domain which
+contains the newly added resource during boot.
+
+"ibm,numa-lookup-index" property contains a number representing the domainID index to be used
+when building the NUMA distance of the numa node to which this resource belongs. The domain id
+of the new resource can be obtained from the existing "ibm,associativity" property. This
+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.
+The value is 1 based array index value.
Am I correct in thinking that if we have an entirely form2 world, we'd
only need this and the ibm,associativity properties could be dropped?
Not really. ibm,numa-lookup-index-table was added to have a concise
representation of numa distance via ibm,numa-distance-table.
For ex: With domainID 0, 4, 5 we could do a 5x5 matrix to represent the
numa distance. Instead ibm,numa-lookup-index-table allows us to present
the same in a 3x3 matrix distance[index0][index1] is the distance
between NUMA node 0 and 4 and distance[index0][index2] is the distance
between NUMA node 0 and 5
quoted
+
+prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
+
+"ibm,numa-distance" property contains one or more list of numbers presenting the NUMA distance
+from this resource domain to other resources.
IIUC this is about extending the global distance table with
information for a new node. Is that correct?
correct.
The global distance table allows for the possibility of asymmetric
distances between nodes, but this does not. Is that intentional?
This also does, For example with 3 nodes currently present and 4 node
getting added ibm,numa-distance have 8 elements enabling us to have
distance[Node0][Node50] being different from distance[Node50][Node0]
as shown below.
quoted
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,associativity = { 4, 5, 6, 7, 50}
+ibm,numa-lookup-index = { 4 }
+ibm,numa-distance = {8, 16, 32, 8, 1, 16, 32, 8, 1}
+
+resulting in a new toplogy as below.
+ | 0 8 40 50
+--|------------------
+ |
+0 | 10 20 80 160
+ |
+8 | 20 10 160 320
+ |
+40| 80 160 10 80
+ |
+50| 160 320 80 10
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices. Persistent memory devices
can also be used as regular memory using DAX KMEM driver and primary domainID indicates
the numa node number OS should use when using these devices as regular memory. Secondary
domainID is the numa node number that should be used when using this device as
persistent memory. In the later case, we are interested in the locality of the
device to an established numa node. In the above example, if the last row represents a
persistent memory device/resource, NUMA node number 40 will be used when using the device
as regular memory and NUMA node number 0 will be the device numa node when using it as
a persistent memory device.
quoted
+ *secondary = of_read_number(&associativity[secondary_index], 1);
+ }
+ if (*primary == 0xffff || *primary >= nr_node_ids)
+ *primary = NUMA_NO_NODE;
+
+ if (*secondary == 0xffff || *secondary >= nr_node_ids)
+ *secondary = NUMA_NO_NODE;
+ return 0;
+}
+
/* Returns the nid associated with the given device tree node,
* or -1 if not found.
*/
@@ -1001,9 +1001,6 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)ndr_desc.res,p->dn);gotoerr;}-if(target_nid!=online_nid)-dev_info(dev,"Region registered with target node %d and online node %d",-target_nid,online_nid);mutex_lock(&papr_ndr_lock);list_add_tail(&p->region_list,&papr_nd_regions);
@@ -1096,7 +1093,7 @@ static int papr_scm_probe(struct platform_device *pdev)structpapr_scm_priv*p;constchar*uuid_str;u64uuid[2];-intrc;+intrc,numa_node;/* check we have all the required DT properties */if(of_property_read_u32(dn,"ibm,my-drc-index",&drc_index)){
@@ -1119,11 +1116,20 @@ static int papr_scm_probe(struct platform_device *pdev)return-ENODEV;}-p=kzalloc(sizeof(*p),GFP_KERNEL);if(!p)return-ENOMEM;+if(get_primary_and_secondary_domain(dn,&p->target_node,&numa_node)){+dev_err(&pdev->dev,"%pOF: missing NUMA attributes!\n",dn);+rc=-ENODEV;+gotoerr;+}+p->numa_node=numa_map_to_online_node(numa_node);+if(numa_node!=p->numa_node)+dev_info(&pdev->dev,"Region registered with online node %d and device tree node %d",+p->numa_node,numa_node);+/* Initialize the dimm mutex */mutex_init(&p->health_mutex);
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource
You describe form 2 below as well, which contradicts this.
Fixed as below.
PAPR interface currently supports different ways of communicating these resource
grouping details to the OS. These are referred to as Form 0, Form 1 and Form2
associativity grouping. Form 0 is the older format and is now considered deprecated.
Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
A value of 1 indicates the usage of Form 1 associativity. For Form 2 associativity
bit 2 of byte 5 in the "ibm,architecture-vec-5" property is used.
LGTM.
quoted
quoted
+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.
+Form 0 is the older format and is now considered deprecated.
+
+Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
+A value of 1 indicates the usage of Form 1 associativity.
+
+Form 0
+-----
+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).
+
+Form 1
+-----
+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity
+device tree properties are used to determine the NUMA distance between resource groups/domains.
+
+The “ibm,associativity” property contains one or more lists of numbers (domainID)
+representing the resource’s platform grouping domains.
+
+The “ibm,associativity-reference-points” property contains one or more list of numbers
+(domain index) that represents the 1 based ordinal in the associativity lists of the most
+significant boundary, with subsequent entries indicating progressively less significant boundaries.
+
+Linux kernel uses the domain id of the most significant boundary (aka primary domain)
I thought we used the *least* significant boundary (the smallest
grouping, not the largest). That is, the last index, not the first.
Actually... come to think of it, I'm not even sure how to interpret
"most significant". Does that mean a change in grouping at that "most
significant" level results in the largest perfomance difference?
PAPR defines "most significant" as below
When the “ibm,architecture-vec-5” property byte 5 bit 0 has the value of one, the “ibm,associativ-
ity-reference-points” property indicates boundaries between associativity domains presented by the
“ibm,associativity” property containing “near” and “far” resources. The
first such boundary in the list represents the 1 based ordinal in the
associativity lists of the most significant boundary, with subsequent
entries indicating progressively less significant boundaries
No... that's not a definition. Like your draft PAPR uses the term
while entirely failing to define it. From what I can tell about how
it is used the "most significant" boundary corresponds to what Linux
simply thinks of as the node id. But intuitively, I'd think of that
as the "least significant" boundary, since that's basically the
smallest granularity at which we care about NUMA distances.
I would interpret it as the boundary where we start defining NUMA
nodes.
That isn't any clearer to me.
quoted
quoted
+as the NUMA node id. Linux kernel computes NUMA distance between two domains by
+recursively comparing if they belong to the same higher-level domains. For mismatch
+at every higher level of the resource group, the kernel doubles the NUMA distance between
+the comparing domains.
+
+Form 2
+-------
+Form 2 associativity format adds separate device tree properties representing NUMA node distance
+thereby making the node distance computation flexible. Form 2 also allows flexible primary
+domain numbering. With numa distance computation now detached from the index value of
+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the
+same domain index representing resource groups of different
performance/latency characteristics.
The meaning of "domain index" is not clear to me here.
Sorry for the confusion there. domain index is the index where domainID
is appearing. W.r.t "ibm,associativity" we have
Ok, I think I eventually deduced that. We should start out clearly
defining both domainID and index here.
Also.. I think we need to find more distinct terms, because "index" is
being used for both where the ID appears in an associativity array,
and also when an ID appears in the Form2 "lookup-index-table" and the
two usages are totally unconnected.
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains. If we can look at
an example property.
{ 4, 6, 7, 0, 0}
{ 4, 6, 7, 0, 40}
With Form 1 both NUMA node 0 and 40 will appear at the same distance.
They both are at domain index 4. With Form 2 we can represent them with
different NUMA distance values.
Ok. Note that PAPR was never clear about what space domain IDs need
to be unique within: do they need to be (a) globally unique (not true
in practice), (b) unique at their index level or (c) unique only
within their "parent" node at the previous index level.
We should take the opportunity with Form2 to make that clearer.
My understanding is that with Form2 it should be entirely feasible to
built a dt have associativity arrays that are always of length 1. Is
that correct?
quoted
quoted
+
+Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the
+"ibm,architecture-vec-5" property.
+
+"ibm,numa-lookup-index-table" property contains one or more list numbers representing
+the domainIDs present in the system. The offset of the domainID in this property is considered
+the domainID index.
You haven't really introduced the term "domainID". Is "domainID
index" the same as "domain index" above? It's not clear to me.
The earlier part of the documented said
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains.
I will update domain index to domainID index.
quoted
The distinction between "domain index" and "primary domain id" is also
not clear to me.
primary domain id is the domainID appearing in the primary domainID
index. Linux kenrel also use that as the NUMA node number.
nit s/kenrel/kernel/
Primary domainID index is defined by ibm,associativity-reference-points
and we consider that as the most significant resource group boundary.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Ok, explicitly stating that in the doc would help a lot.
quoted
quoted
+prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by
+N domainID encoded as with encode-int
+
+For ex:
+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.
Above you say "Form 2 allows a large number of primary domain ids at
the same domain index", but this encoding doesn't appear to permit
that.
I didn't follow that question.
Ah, that's because I was thinking of index here as the index within
the lookup-index-table, not the index within the associativity
arrays.
quoted
quoted
+"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA
+distance between resource groups/domains present in the system.
+
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,numa-lookup-index-table = {3, 0, 8, 40}
+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}
+
+ | 0 8 40
+--|------------
+ |
+0 | 10 20 80
+ |
+8 | 20 10 160
+ |
+40| 80 160 10
What's the reason for multiplying the values by 10 in the expanded
table version?
That was me missing a document update. Since we used 8 bits to encode
distance at some point we were looking at a SCALE factor. But later
realized other architectures also restrict distance to 8 bits. I will
update ibm,numa-distance-table in the document.
Ok.
quoted
quoted
+
+With Form2 "ibm,associativity" for resources is listed as below:
+
+"ibm,associativity" property for resources in node 0, 8 and 40
+{ 4, 6, 7, 0, 0}
+{ 4, 6, 9, 8, 8}
+{ 4, 6, 7, 0, 40}
+
+With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }
+
+With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
What the heck is the secondary domainID
domainID appearing the secondary domainID index.
I understand that from the clarifications you've made about, but
second domainID index wasn't any more defined in the original draft.
ibm,associativity-reference-points gives an indication of different
hierachy of resource grouping as below.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
quoted
quoted
+the kernel should use when using persistent memory devices. Persistent memory devices
+can also be used as regular memory using DAX KMEM driver and primary domainID indicates
+the numa node number OS should use when using these devices as regular memory. Secondary
+domainID is the numa node number that should be used when using this device as
+persistent memory. In the later case, we are interested in the locality of the
+device to an established numa node. In the above example, if the last row represents a
+persistent memory device/resource, NUMA node number 40 will be used when using the device
+as regular memory and NUMA node number 0 will be the device numa node when using it as
+a persistent memory device.
+
+Each resource (drcIndex) now also supports additional optional device tree properties.
+These properties are marked optional because the platform can choose not to export
+them and provide the system topology details using the earlier defined device tree
+properties alone. The optional device tree properties are used when adding new resources
+(DLPAR) and when the platform didn't provide the topology details of the domain which
+contains the newly added resource during boot.
+
+"ibm,numa-lookup-index" property contains a number representing the domainID index to be used
+when building the NUMA distance of the numa node to which this resource belongs. The domain id
+of the new resource can be obtained from the existing "ibm,associativity" property. This
+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.
+The value is 1 based array index value.
Am I correct in thinking that if we have an entirely form2 world, we'd
only need this and the ibm,associativity properties could be dropped?
Not really. ibm,numa-lookup-index-table was added to have a concise
representation of numa distance via ibm,numa-distance-table.
For ex: With domainID 0, 4, 5 we could do a 5x5 matrix to represent the
numa distance. Instead ibm,numa-lookup-index-table allows us to present
the same in a 3x3 matrix distance[index0][index1] is the distance
between NUMA node 0 and 4 and distance[index0][index2] is the distance
between NUMA node 0 and 5
Right, I get the purpose of it, and I realized I misphrashed my
question. My point is that in a Form2 world, the *only* thing the
associativity array is used for is to deduce its position in
lookup-index-table. Once you have have that for each resource, you
have everything you need, yes?
quoted
quoted
+
+prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
+
+"ibm,numa-distance" property contains one or more list of numbers presenting the NUMA distance
+from this resource domain to other resources.
IIUC this is about extending the global distance table with
information for a new node. Is that correct?
correct.
quoted
The global distance table allows for the possibility of asymmetric
distances between nodes, but this does not. Is that intentional?
This also does, For example with 3 nodes currently present and 4 node
getting added ibm,numa-distance have 8 elements enabling us to have
distance[Node0][Node50] being different from distance[Node50][Node0]
as shown below.
Ok that's not clear from the above. Rather than "one or more lists of
numbers" I think you want to explicitly give two options. Either one
list, which gives symmetric distances, or two which gives distances
to, then distance from.
quoted
quoted
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,associativity = { 4, 5, 6, 7, 50}
+ibm,numa-lookup-index = { 4 }
+ibm,numa-distance = {8, 16, 32, 8, 1, 16, 32, 8, 1}
+
+resulting in a new toplogy as below.
+ | 0 8 40 50
+--|------------------
+ |
+0 | 10 20 80 160
+ |
+8 | 20 10 160 320
+ |
+40| 80 160 10 80
+ |
+50| 160 320 80 10
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-15 06:35:38
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
Persistent memory devices
can also be used as regular memory using DAX KMEM driver and primary domainID indicates
the numa node number OS should use when using these devices as regular memory. Secondary
domainID is the numa node number that should be used when using this device as
persistent memory.
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
In the later case, we are interested in the locality of the
device to an established numa node. In the above example, if the last row represents a
persistent memory device/resource, NUMA node number 40 will be used when using the device
as regular memory and NUMA node number 0 will be the device numa node when using it as
a persistent memory device.
I don't really get what you mean by "locality of the device to an
established numa node". Or at least how that's different from
anything else we're handling here.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
quoted
Persistent memory devices
can also be used as regular memory using DAX KMEM driver and primary domainID indicates
the numa node number OS should use when using these devices as regular memory. Secondary
domainID is the numa node number that should be used when using this device as
persistent memory.
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node 1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
quoted
In the later case, we are interested in the locality of the
device to an established numa node. In the above example, if the last row represents a
persistent memory device/resource, NUMA node number 40 will be used when using the device
as regular memory and NUMA node number 0 will be the device numa node when using it as
a persistent memory device.
I don't really get what you mean by "locality of the device to an
established numa node". Or at least how that's different from
anything else we're handling here.
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource
You describe form 2 below as well, which contradicts this.
Fixed as below.
PAPR interface currently supports different ways of communicating these resource
grouping details to the OS. These are referred to as Form 0, Form 1 and Form2
associativity grouping. Form 0 is the older format and is now considered deprecated.
Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
A value of 1 indicates the usage of Form 1 associativity. For Form 2 associativity
bit 2 of byte 5 in the "ibm,architecture-vec-5" property is used.
LGTM.
quoted
quoted
quoted
+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.
+Form 0 is the older format and is now considered deprecated.
+
+Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
+A value of 1 indicates the usage of Form 1 associativity.
+
+Form 0
+-----
+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).
+
+Form 1
+-----
+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity
+device tree properties are used to determine the NUMA distance between resource groups/domains.
+
+The “ibm,associativity” property contains one or more lists of numbers (domainID)
+representing the resource’s platform grouping domains.
+
+The “ibm,associativity-reference-points” property contains one or more list of numbers
+(domain index) that represents the 1 based ordinal in the associativity lists of the most
+significant boundary, with subsequent entries indicating progressively less significant boundaries.
+
+Linux kernel uses the domain id of the most significant boundary (aka primary domain)
I thought we used the *least* significant boundary (the smallest
grouping, not the largest). That is, the last index, not the first.
Actually... come to think of it, I'm not even sure how to interpret
"most significant". Does that mean a change in grouping at that "most
significant" level results in the largest perfomance difference?
PAPR defines "most significant" as below
When the “ibm,architecture-vec-5” property byte 5 bit 0 has the value of one, the “ibm,associativ-
ity-reference-points” property indicates boundaries between associativity domains presented by the
“ibm,associativity” property containing “near” and “far” resources. The
first such boundary in the list represents the 1 based ordinal in the
associativity lists of the most significant boundary, with subsequent
entries indicating progressively less significant boundaries
No... that's not a definition. Like your draft PAPR uses the term
while entirely failing to define it. From what I can tell about how
it is used the "most significant" boundary corresponds to what Linux
simply thinks of as the node id. But intuitively, I'd think of that
as the "least significant" boundary, since that's basically the
smallest granularity at which we care about NUMA distances.
quoted
I would interpret it as the boundary where we start defining NUMA
nodes.
That isn't any clearer to me.
How about calling it least significant boundary then?
The “ibm,associativity-reference-points” property contains one or more list of numbers
(domainID index) that represents the 1 based ordinal in the associativity lists of the
least significant boundary, with subsequent entries indicating progressively higher
significant boundaries.
ex:
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Linux kernel uses the domainID of the least significant boundary (aka primary domain)
as the NUMA node id. Linux kernel computes NUMA distance between two domains by
recursively comparing if they belong to the same higher-level domains. For mismatch
at every higher level of the resource group, the kernel doubles the NUMA distance between
the comparing domains.
quoted
quoted
quoted
+as the NUMA node id. Linux kernel computes NUMA distance between two domains by
+recursively comparing if they belong to the same higher-level domains. For mismatch
+at every higher level of the resource group, the kernel doubles the NUMA distance between
+the comparing domains.
+
+Form 2
+-------
+Form 2 associativity format adds separate device tree properties representing NUMA node distance
+thereby making the node distance computation flexible. Form 2 also allows flexible primary
+domain numbering. With numa distance computation now detached from the index value of
+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the
+same domain index representing resource groups of different
performance/latency characteristics.
The meaning of "domain index" is not clear to me here.
Sorry for the confusion there. domain index is the index where domainID
is appearing. W.r.t "ibm,associativity" we have
Ok, I think I eventually deduced that. We should start out clearly
defining both domainID and index here.
Also.. I think we need to find more distinct terms, because "index" is
being used for both where the ID appears in an associativity array,
and also when an ID appears in the Form2 "lookup-index-table" and the
two usages are totally unconnected.
quoted
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains. If we can look at
an example property.
{ 4, 6, 7, 0, 0}
{ 4, 6, 7, 0, 40}
With Form 1 both NUMA node 0 and 40 will appear at the same distance.
They both are at domain index 4. With Form 2 we can represent them with
different NUMA distance values.
Ok. Note that PAPR was never clear about what space domain IDs need
to be unique within: do they need to be (a) globally unique (not true
in practice), (b) unique at their index level or (c) unique only
within their "parent" node at the previous index level.
We should take the opportunity with Form2 to make that clearer.
My understanding is that with Form2 it should be entirely feasible to
built a dt have associativity arrays that are always of length 1. Is
that correct?
Correct, unless you have persistent memory device attached in which case
you need two entries.
quoted
quoted
quoted
+
+Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the
+"ibm,architecture-vec-5" property.
+
+"ibm,numa-lookup-index-table" property contains one or more list numbers representing
+the domainIDs present in the system. The offset of the domainID in this property is considered
+the domainID index.
You haven't really introduced the term "domainID". Is "domainID
index" the same as "domain index" above? It's not clear to me.
The earlier part of the documented said
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains.
I will update domain index to domainID index.
quoted
The distinction between "domain index" and "primary domain id" is also
not clear to me.
primary domain id is the domainID appearing in the primary domainID
index. Linux kenrel also use that as the NUMA node number.
nit s/kenrel/kernel/
quoted
Primary domainID index is defined by ibm,associativity-reference-points
and we consider that as the most significant resource group boundary.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Ok, explicitly stating that in the doc would help a lot.
quoted
quoted
quoted
+prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by
+N domainID encoded as with encode-int
+
+For ex:
+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.
Above you say "Form 2 allows a large number of primary domain ids at
the same domain index", but this encoding doesn't appear to permit
that.
I didn't follow that question.
Ah, that's because I was thinking of index here as the index within
the lookup-index-table, not the index within the associativity
arrays.
quoted
quoted
quoted
+"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA
+distance between resource groups/domains present in the system.
+
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,numa-lookup-index-table = {3, 0, 8, 40}
+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}
+
+ | 0 8 40
+--|------------
+ |
+0 | 10 20 80
+ |
+8 | 20 10 160
+ |
+40| 80 160 10
What's the reason for multiplying the values by 10 in the expanded
table version?
That was me missing a document update. Since we used 8 bits to encode
distance at some point we were looking at a SCALE factor. But later
realized other architectures also restrict distance to 8 bits. I will
update ibm,numa-distance-table in the document.
Ok.
quoted
quoted
quoted
+
+With Form2 "ibm,associativity" for resources is listed as below:
+
+"ibm,associativity" property for resources in node 0, 8 and 40
+{ 4, 6, 7, 0, 0}
+{ 4, 6, 9, 8, 8}
+{ 4, 6, 7, 0, 40}
+
+With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }
+
+With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
What the heck is the secondary domainID
domainID appearing the secondary domainID index.
I understand that from the clarifications you've made about, but
second domainID index wasn't any more defined in the original draft.
quoted
ibm,associativity-reference-points gives an indication of different
hierachy of resource grouping as below.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
quoted
quoted
+the kernel should use when using persistent memory devices. Persistent memory devices
+can also be used as regular memory using DAX KMEM driver and primary domainID indicates
+the numa node number OS should use when using these devices as regular memory. Secondary
+domainID is the numa node number that should be used when using this device as
+persistent memory. In the later case, we are interested in the locality of the
+device to an established numa node. In the above example, if the last row represents a
+persistent memory device/resource, NUMA node number 40 will be used when using the device
+as regular memory and NUMA node number 0 will be the device numa node when using it as
+a persistent memory device.
+
+Each resource (drcIndex) now also supports additional optional device tree properties.
+These properties are marked optional because the platform can choose not to export
+them and provide the system topology details using the earlier defined device tree
+properties alone. The optional device tree properties are used when adding new resources
+(DLPAR) and when the platform didn't provide the topology details of the domain which
+contains the newly added resource during boot.
+
+"ibm,numa-lookup-index" property contains a number representing the domainID index to be used
+when building the NUMA distance of the numa node to which this resource belongs. The domain id
+of the new resource can be obtained from the existing "ibm,associativity" property. This
+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.
+The value is 1 based array index value.
Am I correct in thinking that if we have an entirely form2 world, we'd
only need this and the ibm,associativity properties could be dropped?
Not really. ibm,numa-lookup-index-table was added to have a concise
representation of numa distance via ibm,numa-distance-table.
For ex: With domainID 0, 4, 5 we could do a 5x5 matrix to represent the
numa distance. Instead ibm,numa-lookup-index-table allows us to present
the same in a 3x3 matrix distance[index0][index1] is the distance
between NUMA node 0 and 4 and distance[index0][index2] is the distance
between NUMA node 0 and 5
Right, I get the purpose of it, and I realized I misphrashed my
question. My point is that in a Form2 world, the *only* thing the
associativity array is used for is to deduce its position in
lookup-index-table. Once you have have that for each resource, you
have everything you need, yes?
ibm,associativity is used find the domainID/NUMA node id of the
resource.
ibm,lookup-index-table is used compute the distance information between
NUMA nodes using ibm,numa-distance-table.
quoted
quoted
quoted
+
+prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
+
+"ibm,numa-distance" property contains one or more list of numbers presenting the NUMA distance
+from this resource domain to other resources.
IIUC this is about extending the global distance table with
information for a new node. Is that correct?
correct.
quoted
The global distance table allows for the possibility of asymmetric
distances between nodes, but this does not. Is that intentional?
This also does, For example with 3 nodes currently present and 4 node
getting added ibm,numa-distance have 8 elements enabling us to have
distance[Node0][Node50] being different from distance[Node50][Node0]
as shown below.
Ok that's not clear from the above. Rather than "one or more lists of
numbers" I think you want to explicitly give two options. Either one
list, which gives symmetric distances, or two which gives distances
to, then distance from.
quoted
quoted
quoted
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,associativity = { 4, 5, 6, 7, 50}
+ibm,numa-lookup-index = { 4 }
+ibm,numa-distance = {8, 16, 32, 8, 1, 16, 32, 8, 1}
+
+resulting in a new toplogy as below.
+ | 0 8 40 50
+--|------------------
+ |
+0 | 10 20 80 160
+ |
+8 | 20 10 160 320
+ |
+40| 80 160 10 80
+ |
+50| 160 320 80 10
On Mon, Jun 14, 2021 at 10:09:56PM +0530, Aneesh Kumar K.V wrote:
quoted
No functional change in this patch.
I think this needs a rationale as to why 'primary_domain_index' is a
better name than 'min_common_depth'. The meaning isn't obvious to me
from either name.
The documentation added in patch 7 explains the name to some extent. Is
that sufficient?
-aneesh
On Mon, Jun 14, 2021 at 10:09:57PM +0530, Aneesh Kumar K.V wrote:
quoted
No functional change in this patch
As with 1/8 an explanation of what this actually means and therefore
why this is a better name would be very helpful.
quoted
how about max_associativity_domain_index? The documentation added in
patch 7 explains the details. Let me know if we need to add more in
this patch.
-aneesh
On Mon, Jun 14, 2021 at 10:09:59PM +0530, Aneesh Kumar K.V wrote:
quoted
The associativity details of the newly added resourced are collected from
the hypervisor via "ibm,configure-connector" rtas call. Update the numa
distance details of the newly added numa node after the above call. In
later patch we will remove updating NUMA distance when we are looking
for node id from associativity array.
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 41 +++++++++++++++++++
arch/powerpc/platforms/pseries/hotplug-cpu.c | 2 +
.../platforms/pseries/hotplug-memory.c | 2 +
arch/powerpc/platforms/pseries/pseries.h | 1 +
4 files changed, 46 insertions(+)
@@ -287,6 +287,47 @@ int of_node_to_nid(struct device_node *device)}EXPORT_SYMBOL(of_node_to_nid);+staticvoid__initialize_form1_numa_distance(const__be32*associativity)+{+inti,nid;++if(of_read_number(associativity,1)>=primary_domain_index){+nid=of_read_number(&associativity[primary_domain_index],1);++for(i=0;i<max_domain_index;i++){+const__be32*entry;++entry=&associativity[be32_to_cpu(distance_ref_points[i])];+distance_lookup_table[nid][i]=of_read_number(entry,1);+}+}+}
This logic is almost identicaly to initialize_distance_lookup_table()
- it would be good if they could be consolidated, so it's clear that
coldplugged and hotplugged nodes are parsing the NUMA information in
the same way.
initialize_distance_lookup_table() gets removed in the next patch.
-aneesh
@@ -0,0 +1,139 @@+============================+NUMA resource associativity+=============================++Associativity represents the groupings of the various platform resources into+domains of substantially similar mean performance relative to resources outside+of that domain. Resources subsets of a given domain that exhibit better+performance relative to each other than relative to other resources subsets+are represented as being members of a sub-grouping domain. This performance+characteristic is presented in terms of NUMA node distance within the Linux kernel.+From the platform view, these groups are also referred to as domains.++PAPR interface currently supports two different ways of communicating these resource
You describe form 2 below as well, which contradicts this.
Fixed as below.
PAPR interface currently supports different ways of communicating these resource
grouping details to the OS. These are referred to as Form 0, Form 1 and Form2
associativity grouping. Form 0 is the older format and is now considered deprecated.
Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
A value of 1 indicates the usage of Form 1 associativity. For Form 2 associativity
bit 2 of byte 5 in the "ibm,architecture-vec-5" property is used.
LGTM.
quoted
quoted
quoted
+grouping details to the OS. These are referred to as Form 0 and Form 1 associativity grouping.
+Form 0 is the older format and is now considered deprecated.
+
+Hypervisor indicates the type/form of associativity used via "ibm,arcitecture-vec-5 property".
+Bit 0 of byte 5 in the "ibm,architecture-vec-5" property indicates usage of Form 0 or Form 1.
+A value of 1 indicates the usage of Form 1 associativity.
+
+Form 0
+-----
+Form 0 associativity supports only two NUMA distance (LOCAL and REMOTE).
+
+Form 1
+-----
+With Form 1 a combination of ibm,associativity-reference-points and ibm,associativity
+device tree properties are used to determine the NUMA distance between resource groups/domains.
+
+The “ibm,associativity” property contains one or more lists of numbers (domainID)
+representing the resource’s platform grouping domains.
+
+The “ibm,associativity-reference-points” property contains one or more list of numbers
+(domain index) that represents the 1 based ordinal in the associativity lists of the most
+significant boundary, with subsequent entries indicating progressively less significant boundaries.
+
+Linux kernel uses the domain id of the most significant boundary (aka primary domain)
I thought we used the *least* significant boundary (the smallest
grouping, not the largest). That is, the last index, not the first.
Actually... come to think of it, I'm not even sure how to interpret
"most significant". Does that mean a change in grouping at that "most
significant" level results in the largest perfomance difference?
PAPR defines "most significant" as below
When the “ibm,architecture-vec-5” property byte 5 bit 0 has the value of one, the “ibm,associativ-
ity-reference-points” property indicates boundaries between associativity domains presented by the
“ibm,associativity” property containing “near” and “far” resources. The
first such boundary in the list represents the 1 based ordinal in the
associativity lists of the most significant boundary, with subsequent
entries indicating progressively less significant boundaries
No... that's not a definition. Like your draft PAPR uses the term
while entirely failing to define it. From what I can tell about how
it is used the "most significant" boundary corresponds to what Linux
simply thinks of as the node id. But intuitively, I'd think of that
as the "least significant" boundary, since that's basically the
smallest granularity at which we care about NUMA distances.
quoted
I would interpret it as the boundary where we start defining NUMA
nodes.
That isn't any clearer to me.
How about calling it least significant boundary then?
Heck, no. My whole point here is that the meaning is unclear: my
first guess at the meaning is different from whoever wrote that text.
We need to come up with a way of describing it that's clearer.
The “ibm,associativity-reference-points” property contains one or more list of numbers
(domainID index) that represents the 1 based ordinal in the associativity lists of the
least significant boundary, with subsequent entries indicating progressively higher
significant boundaries.
ex:
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Linux kernel uses the domainID of the least significant boundary (aka primary domain)
as the NUMA node id. Linux kernel computes NUMA distance between two domains by
recursively comparing if they belong to the same higher-level domains. For mismatch
at every higher level of the resource group, the kernel doubles the NUMA distance between
the comparing domains.
quoted
quoted
quoted
quoted
+as the NUMA node id. Linux kernel computes NUMA distance between two domains by
+recursively comparing if they belong to the same higher-level domains. For mismatch
+at every higher level of the resource group, the kernel doubles the NUMA distance between
+the comparing domains.
+
+Form 2
+-------
+Form 2 associativity format adds separate device tree properties representing NUMA node distance
+thereby making the node distance computation flexible. Form 2 also allows flexible primary
+domain numbering. With numa distance computation now detached from the index value of
+"ibm,associativity" property, Form 2 allows a large number of primary domain ids at the
+same domain index representing resource groups of different
performance/latency characteristics.
The meaning of "domain index" is not clear to me here.
Sorry for the confusion there. domain index is the index where domainID
is appearing. W.r.t "ibm,associativity" we have
Ok, I think I eventually deduced that. We should start out clearly
defining both domainID and index here.
Also.. I think we need to find more distinct terms, because "index" is
being used for both where the ID appears in an associativity array,
and also when an ID appears in the Form2 "lookup-index-table" and the
two usages are totally unconnected.
quoted
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains. If we can look at
an example property.
{ 4, 6, 7, 0, 0}
{ 4, 6, 7, 0, 40}
With Form 1 both NUMA node 0 and 40 will appear at the same distance.
They both are at domain index 4. With Form 2 we can represent them with
different NUMA distance values.
Ok. Note that PAPR was never clear about what space domain IDs need
to be unique within: do they need to be (a) globally unique (not true
in practice), (b) unique at their index level or (c) unique only
within their "parent" node at the previous index level.
We should take the opportunity with Form2 to make that clearer.
My understanding is that with Form2 it should be entirely feasible to
built a dt have associativity arrays that are always of length 1. Is
that correct?
Correct, unless you have persistent memory device attached in which case
you need two entries.
quoted
quoted
quoted
quoted
+
+Hypervisor indicates the usage of FORM2 associativity using bit 2 of byte 5 in the
+"ibm,architecture-vec-5" property.
+
+"ibm,numa-lookup-index-table" property contains one or more list numbers representing
+the domainIDs present in the system. The offset of the domainID in this property is considered
+the domainID index.
You haven't really introduced the term "domainID". Is "domainID
index" the same as "domain index" above? It's not clear to me.
The earlier part of the documented said
The “ibm,associativity” property contains one or more lists of numbers (domainID)
representing the resource’s platform grouping domains.
I will update domain index to domainID index.
quoted
The distinction between "domain index" and "primary domain id" is also
not clear to me.
primary domain id is the domainID appearing in the primary domainID
index. Linux kenrel also use that as the NUMA node number.
nit s/kenrel/kernel/
quoted
Primary domainID index is defined by ibm,associativity-reference-points
and we consider that as the most significant resource group boundary.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Ok, explicitly stating that in the doc would help a lot.
quoted
quoted
quoted
+prop-encoded-array: The number N of the domainIDs encoded as with encode-int, followed by
+N domainID encoded as with encode-int
+
+For ex:
+ibm,numa-lookup-index-table = {4, 0, 8, 250, 252}, domainID index for domainID 8 is 1.
Above you say "Form 2 allows a large number of primary domain ids at
the same domain index", but this encoding doesn't appear to permit
that.
I didn't follow that question.
Ah, that's because I was thinking of index here as the index within
the lookup-index-table, not the index within the associativity
arrays.
quoted
quoted
quoted
+"ibm,numa-distance-table" property contains one or more list of numbers representing the NUMA
+distance between resource groups/domains present in the system.
+
+prop-encoded-array: The number N of the distance values encoded as with encode-int, followed by
+N distance values encoded as with encode-bytes. The max distance value we could encode is 255.
+
+For ex:
+ibm,numa-lookup-index-table = {3, 0, 8, 40}
+ibm,numa-distance-table = {9, 1, 2, 8, 2, 1, 16, 8, 16, 1}
+
+ | 0 8 40
+--|------------
+ |
+0 | 10 20 80
+ |
+8 | 20 10 160
+ |
+40| 80 160 10
What's the reason for multiplying the values by 10 in the expanded
table version?
That was me missing a document update. Since we used 8 bits to encode
distance at some point we were looking at a SCALE factor. But later
realized other architectures also restrict distance to 8 bits. I will
update ibm,numa-distance-table in the document.
Ok.
quoted
quoted
quoted
+
+With Form2 "ibm,associativity" for resources is listed as below:
+
+"ibm,associativity" property for resources in node 0, 8 and 40
+{ 4, 6, 7, 0, 0}
+{ 4, 6, 9, 8, 8}
+{ 4, 6, 7, 0, 40}
+
+With "ibm,associativity-reference-points" { 0x4, 0x3, 0x2 }
+
+With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
What the heck is the secondary domainID
domainID appearing the secondary domainID index.
I understand that from the clarifications you've made about, but
second domainID index wasn't any more defined in the original draft.
quoted
ibm,associativity-reference-points gives an indication of different
hierachy of resource grouping as below.
ibm,associativity-reference-points can be looked at as
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
quoted
quoted
+the kernel should use when using persistent memory devices. Persistent memory devices
+can also be used as regular memory using DAX KMEM driver and primary domainID indicates
+the numa node number OS should use when using these devices as regular memory. Secondary
+domainID is the numa node number that should be used when using this device as
+persistent memory. In the later case, we are interested in the locality of the
+device to an established numa node. In the above example, if the last row represents a
+persistent memory device/resource, NUMA node number 40 will be used when using the device
+as regular memory and NUMA node number 0 will be the device numa node when using it as
+a persistent memory device.
+
+Each resource (drcIndex) now also supports additional optional device tree properties.
+These properties are marked optional because the platform can choose not to export
+them and provide the system topology details using the earlier defined device tree
+properties alone. The optional device tree properties are used when adding new resources
+(DLPAR) and when the platform didn't provide the topology details of the domain which
+contains the newly added resource during boot.
+
+"ibm,numa-lookup-index" property contains a number representing the domainID index to be used
+when building the NUMA distance of the numa node to which this resource belongs. The domain id
+of the new resource can be obtained from the existing "ibm,associativity" property. This
+can be used to build distance information of a newly onlined NUMA node via DLPAR operation.
+The value is 1 based array index value.
Am I correct in thinking that if we have an entirely form2 world, we'd
only need this and the ibm,associativity properties could be dropped?
Not really. ibm,numa-lookup-index-table was added to have a concise
representation of numa distance via ibm,numa-distance-table.
For ex: With domainID 0, 4, 5 we could do a 5x5 matrix to represent the
numa distance. Instead ibm,numa-lookup-index-table allows us to present
the same in a 3x3 matrix distance[index0][index1] is the distance
between NUMA node 0 and 4 and distance[index0][index2] is the distance
between NUMA node 0 and 5
Right, I get the purpose of it, and I realized I misphrashed my
question. My point is that in a Form2 world, the *only* thing the
associativity array is used for is to deduce its position in
lookup-index-table. Once you have have that for each resource, you
have everything you need, yes?
ibm,associativity is used find the domainID/NUMA node id of the
resource.
ibm,lookup-index-table is used compute the distance information between
NUMA nodes using ibm,numa-distance-table.
I get that you need to use lookup-index-table to work out how to
interpret numa-distance-table. My point is that IIUC once you've done
the lookup in lookup-index-table once for each associativity array
value, the number you get out (which just a compacted version of the
node id) should be all you need ever again.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
From: David Gibson <hidden> Date: 2021-06-17 10:20:26
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
My point is that reusing associativity-reference-points for something
with completely unrelated semantics seems like a very poor choice.
quoted
quoted
Persistent memory devices
can also be used as regular memory using DAX KMEM driver and primary domainID indicates
the numa node number OS should use when using these devices as regular memory. Secondary
domainID is the numa node number that should be used when using this device as
persistent memory.
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node 1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
On Tue, Jun 15, 2021 at 01:10:27PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
....
quoted
quoted
quoted
PAPR defines "most significant" as below
When the “ibm,architecture-vec-5” property byte 5 bit 0 has the value of one, the “ibm,associativ-
ity-reference-points” property indicates boundaries between associativity domains presented by the
“ibm,associativity” property containing “near” and “far” resources. The
first such boundary in the list represents the 1 based ordinal in the
associativity lists of the most significant boundary, with subsequent
entries indicating progressively less significant boundaries
No... that's not a definition. Like your draft PAPR uses the term
while entirely failing to define it. From what I can tell about how
it is used the "most significant" boundary corresponds to what Linux
simply thinks of as the node id. But intuitively, I'd think of that
as the "least significant" boundary, since that's basically the
smallest granularity at which we care about NUMA distances.
quoted
I would interpret it as the boundary where we start defining NUMA
nodes.
That isn't any clearer to me.
How about calling it least significant boundary then?
Heck, no. My whole point here is that the meaning is unclear: my
first guess at the meaning is different from whoever wrote that text.
We need to come up with a way of describing it that's clearer.
quoted
The “ibm,associativity-reference-points” property contains one or more list of numbers
(domainID index) that represents the 1 based ordinal in the associativity lists of the
least significant boundary, with subsequent entries indicating progressively higher
significant boundaries.
ex:
{ primary domainID index, secondary domainID index, tertiary domainID index.. }
Linux kernel uses the domainID of the least significant boundary (aka primary domain)
as the NUMA node id. Linux kernel computes NUMA distance between two domains by
recursively comparing if they belong to the same higher-level domains. For mismatch
at every higher level of the resource group, the kernel doubles the NUMA distance between
the comparing domains.
Any suggestion on how to reword the above section then? We could say
associativity-reference-points is list of domainID index representing
increasing hierarchy of resource group. I am not sure that explains it
any better?
....
quoted
quoted
quoted
For ex: With domainID 0, 4, 5 we could do a 5x5 matrix to represent the
numa distance. Instead ibm,numa-lookup-index-table allows us to present
the same in a 3x3 matrix distance[index0][index1] is the distance
between NUMA node 0 and 4 and distance[index0][index2] is the distance
between NUMA node 0 and 5
Right, I get the purpose of it, and I realized I misphrashed my
question. My point is that in a Form2 world, the *only* thing the
associativity array is used for is to deduce its position in
lookup-index-table. Once you have have that for each resource, you
have everything you need, yes?
ibm,associativity is used find the domainID/NUMA node id of the
resource.
ibm,lookup-index-table is used compute the distance information between
NUMA nodes using ibm,numa-distance-table.
I get that you need to use lookup-index-table to work out how to
interpret numa-distance-table. My point is that IIUC once you've done
the lookup in lookup-index-table once for each associativity array
value, the number you get out (which just a compacted version of the
node id) should be all you need ever again.
That is correct. We will continue to use the index to nodeid map during
DLPAR, if such an operation adds a new numa node. update_numa_distance()
shows the detail.
-aneesh
From: Daniel Henrique Barboza <hidden> Date: 2021-06-17 10:53:39
On 6/17/21 4:46 AM, David Gibson wrote:
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
My point is that reusing associativity-reference-points for something
with completely unrelated semantics seems like a very poor choice.
I agree that this reuse can be confusing. I could argue that there is
precedent for that in PAPR - FORM0 puts a different spin on the same
property as well - but there is no need to keep following existing PAPR
practices in new spec (and some might argue it's best not to).
As far as QEMU goes, renaming this property to "numa-associativity-mode"
(just an example) is a quick change to do since we separated FORM1 and FORM2
code over there.
Doing such a rename can also help with the issue of having to describe new
FORM2 semantics using "least significant boundary" or "primary domain" or
any FORM0|FORM1 related terminology.
Thanks,
Daniel
quoted
quoted
quoted
Persistent memory devices
can also be used as regular memory using DAX KMEM driver and primary domainID indicates
the numa node number OS should use when using these devices as regular memory. Secondary
domainID is the numa node number that should be used when using this device as
persistent memory.
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node 1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
...
quoted
quoted
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node 1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
In the above example in order allow use of PMEMB and PMEMD as memory
only NUMA nodes
we need platform to represent them in its own domainID. Let's assume that
platform assigned id 40 and 41 and hence both PMEMB and PMEMD will have
associativity array like below
{ 4, 6, 0} -> PROCA/MEMA
{ 4, 6, 40} -> PMEMB
{ 4, 6, 41} -> PMEMD
{ 4, 6, 1} -> PROCB/MEMB
When we want to use this device PMEMB and PMEMD as fsdax/devdax devices,
we essentially look for the first nearest online node. Which means both
PMEMB and PMEMD will appear as devices attached to node0. That is not
ideal for for many applications.
using secondary domainID index as explained here helps to associate
each PMEM device to the right group. On a non virtualized config or hard
partitioned config such a device tree representation can be looked at as
a hint to identify which socket the actual device is connected to.
-aneesh
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
My point is that reusing associativity-reference-points for something
with completely unrelated semantics seems like a very poor choice.
I agree that this reuse can be confusing. I could argue that there is
precedent for that in PAPR - FORM0 puts a different spin on the same
property as well - but there is no need to keep following existing PAPR
practices in new spec (and some might argue it's best not to).
As far as QEMU goes, renaming this property to "numa-associativity-mode"
(just an example) is a quick change to do since we separated FORM1 and FORM2
code over there.
Doing such a rename can also help with the issue of having to describe new
FORM2 semantics using "least significant boundary" or "primary domain" or
any FORM0|FORM1 related terminology.
It is not just changing the name, we will then have to explain the
meaning of ibm,associativity-reference-points with FORM2 right?
With FORM2 we want to represent the topology better
--------------------------------------------------------------------------------
| domainID 20 |
| --------------------------------------- |
| | NUMA node1 | |
| | | -------------------- |
| | ProcB -------> MEMC | | NUMA node40 | |
| | | | | | |
| | ---------------------------------- |--------> | PMEMD | |
| | | -------------------- |
| | | |
| --------------------------------------- |
--------------------------------------------------------------------------------
ibm,associativity:
{ 20, 1, 40} -> PMEMD
{ 20, 1, 1} -> PROCB/MEMC
is the suggested FORM2 representation.
-aneesh
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
My point is that reusing associativity-reference-points for something
with completely unrelated semantics seems like a very poor choice.
I agree that this reuse can be confusing. I could argue that there is
precedent for that in PAPR - FORM0 puts a different spin on the same
property as well - but there is no need to keep following existing PAPR
practices in new spec (and some might argue it's best not to).
As far as QEMU goes, renaming this property to "numa-associativity-mode"
(just an example) is a quick change to do since we separated FORM1 and FORM2
code over there.
Doing such a rename can also help with the issue of having to describe new
FORM2 semantics using "least significant boundary" or "primary domain" or
any FORM0|FORM1 related terminology.
It is not just changing the name, we will then have to explain the
meaning of ibm,associativity-reference-points with FORM2 right?
With FORM2 we want to represent the topology better
--------------------------------------------------------------------------------
| domainID 20 |
| --------------------------------------- |
| | NUMA node1 | |
| | | -------------------- |
| | ProcB -------> MEMC | | NUMA node40 | |
| | | | | | |
| | ---------------------------------- |--------> | PMEMD | |
| | | -------------------- |
| | | |
| --------------------------------------- |
--------------------------------------------------------------------------------
ibm,associativity:
{ 20, 1, 40} -> PMEMD
{ 20, 1, 1} -> PROCB/MEMC
is the suggested FORM2 representation.
We can simplify this as below too
ibm,associativity:
{ 20, 1, 40} -> PMEMD
{ 20, 1 } -> PROCB/MEMC
-aneesh
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
.....
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
How about the below update?
With Form2 "ibm,associativity" for resources is listed as below:
"ibm,associativity" property for resources in node 0, 8 and 40
{ 3, 6, 7, 0 }
{ 3, 6, 9, 8 }
{ 4, 6, 7, 0, 40}
With "ibm,associativity-reference-points" { 0x3, 0x2 }
Form2 adds additional property which can be used with devices like persistence
memory devices which would also like to be presented as memory-only NUMA nodes.
"ibm,associativity-memory-node-reference-point" property contains a number
representing the domainID index to be used to find the domainID that should be used
when using the resource as memory only NUMA node. The NUMA distance information
w.r.t this domainID will take into consideration the latency of the media. A
high latency memory device will have a large NUMA distance value assigned w.r.t
the domainID found at at "ibm,associativity-memory-node-reference-point" domainID index.
prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
In the above example:
"ibm,associativity-memory-node-reference-point" { 0x4 }
ex:
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
--------------------------------------------------------------------------------
| domainID 20 |
| --------------------------------------- |
| | NUMA node0 | |
| | | -------------------- |
| | ProcA -------> MEMA | | NUMA node40 | |
| | | | | | |
| | ---------------------------------- |--------> | PMEMB | |
| | | -------------------- |
| | | |
| --------------------------------------- |
| |
| --------------------------------------- |
| | NUMA node1 | |
| | | |
| | ProcB -------> MEMC | ------------------- |
| | | | | NUMA node41 | |
| | --------------------------------------------> | PMEMD | |
| | | ------------------- |
| | | |
| --------------------------------------- |
| |
--------------------------------------------------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
"ibm,associativity":
PROCA/MEMA -> { 2, 20, 0 }
PROCB/MEMC -> { 2, 20, 1 }
PMEMB -> { 3, 20, 0, 40}
PMEMB -> { 3, 20, 1, 41}
"ibm,associativity-reference-points" -> { 2, 1 }
"ibm,associativity-memory-node-reference-points" -> { 3 }
-aneesh
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
.....
quoted
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
How about the below update?
With Form2 "ibm,associativity" for resources is listed as below:
"ibm,associativity" property for resources in node 0, 8 and 40
{ 3, 6, 7, 0 }
{ 3, 6, 9, 8 }
{ 4, 6, 7, 0, 40}
With "ibm,associativity-reference-points" { 0x3, 0x2 }
Form2 adds additional property which can be used with devices like persistence
memory devices which would also like to be presented as memory-only NUMA nodes.
"ibm,associativity-memory-node-reference-point" property contains a number
representing the domainID index to be used to find the domainID that should be used
when using the resource as memory only NUMA node. The NUMA distance information
w.r.t this domainID will take into consideration the latency of the media. A
high latency memory device will have a large NUMA distance value assigned w.r.t
the domainID found at at "ibm,associativity-memory-node-reference-point" domainID index.
prop-encoded-array: An integer encoded as with encode-int specifying the domainID index
In the above example:
"ibm,associativity-memory-node-reference-point" { 0x4 }
ex:
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
--------------------------------------------------------------------------------
| domainID 20 |
| --------------------------------------- |
| | NUMA node0 | |
| | | -------------------- |
| | ProcA -------> MEMA | | NUMA node40 | |
| | | | | | |
| | ---------------------------------- |--------> | PMEMB | |
| | | -------------------- |
| | | |
| --------------------------------------- |
| |
| --------------------------------------- |
| | NUMA node1 | |
| | | |
| | ProcB -------> MEMC | ------------------- |
| | | | | NUMA node41 | |
| | --------------------------------------------> | PMEMD | |
| | | ------------------- |
| | | |
| --------------------------------------- |
| |
--------------------------------------------------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
"ibm,associativity":
PROCA/MEMA -> { 2, 20, 0 }
PROCB/MEMC -> { 2, 20, 1 }
PMEMB -> { 3, 20, 0, 40}
PMEMB -> { 3, 20, 1, 41}
"ibm,associativity-reference-points" -> { 2, 1 }
"ibm,associativity-memory-node-reference-points" -> { 3 }
Another option is to make sure that numa-distance-value is populated
such that PMEMB distance indicates it is closer to node0 when compared
to node1. ie, node_distance[40][0] < node_distance[40][1]. One could
possibly infer the grouping based on the distance value and not deepend
on ibm,associativity for that purpose.
-aneesh
From: Daniel Henrique Barboza <hidden> Date: 2021-06-17 20:00:41
On 6/17/21 8:11 AM, Aneesh Kumar K.V wrote:
Daniel Henrique Barboza [off-list ref] writes:
quoted
On 6/17/21 4:46 AM, David Gibson wrote:
quoted
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Mon, Jun 14, 2021 at 10:10:03PM +0530, Aneesh Kumar K.V wrote:
quoted
FORM2 introduce a concept of secondary domain which is identical to the
conceept of FORM1 primary domain. Use secondary domain as the numa node
when using persistent memory device. For DAX kmem use the logical domain
id introduced in FORM2. This new numa node
Signed-off-by: Aneesh Kumar K.V <redacted>
---
arch/powerpc/mm/numa.c | 28 +++++++++++++++++++++++
arch/powerpc/platforms/pseries/papr_scm.c | 26 +++++++++++++--------
arch/powerpc/platforms/pseries/pseries.h | 1 +
3 files changed, 45 insertions(+), 10 deletions(-)
@@ -265,6 +265,34 @@ static int associativity_to_nid(const __be32 *associativity)returnnid;}+intget_primary_and_secondary_domain(structdevice_node*node,int*primary,int*secondary)+{+intsecondary_index;+const__be32*associativity;++if(!numa_enabled){+*primary=NUMA_NO_NODE;+*secondary=NUMA_NO_NODE;+return0;+}++associativity=of_get_associativity(node);+if(!associativity)+return-ENODEV;++if(of_read_number(associativity,1)>=primary_domain_index){+*primary=of_read_number(&associativity[primary_domain_index],1);+secondary_index=of_read_number(&distance_ref_points[1],1);
Secondary ID is always the second reference point, but primary depends
on the length of resources? That seems very weird.
primary_domain_index is distance_ref_point[0]. With Form2 we would find
both primary and secondary domain ID same for all resources other than
persistent memory device. The usage w.r.t. persistent memory is
explained in patch 7.
Right, I misunderstood
quoted
With Form2 the primary domainID and secondary domainID are used to identify the NUMA nodes
the kernel should use when using persistent memory devices.
This seems kind of bogus. With Form1, the primary/secondary ID are a
sort of heirarchy of distance (things with same primary ID are very
close, things with same secondary are kinda-close, etc.). With Form2,
it's referring to their effective node for different purposes.
Using the same terms for different meanings seems unnecessarily
confusing.
They are essentially domainIDs. The interpretation of them are different
between Form1 and Form2. Hence I kept referring to them as primary and
secondary domainID. Any suggestion on what to name them with Form2?
My point is that reusing associativity-reference-points for something
with completely unrelated semantics seems like a very poor choice.
I agree that this reuse can be confusing. I could argue that there is
precedent for that in PAPR - FORM0 puts a different spin on the same
property as well - but there is no need to keep following existing PAPR
practices in new spec (and some might argue it's best not to).
As far as QEMU goes, renaming this property to "numa-associativity-mode"
(just an example) is a quick change to do since we separated FORM1 and FORM2
code over there.
Doing such a rename can also help with the issue of having to describe new
FORM2 semantics using "least significant boundary" or "primary domain" or
any FORM0|FORM1 related terminology.
It is not just changing the name, we will then have to explain the
meaning of ibm,associativity-reference-points with FORM2 right?
Hmmmm why? My idea over there was to add a new property that indicates that
resource might have a different NUMA affinity based on the mode of operation
(like PMEM), and get rid of ibm,associativity-reference-points altogether.
The NUMA distances already express the topology. Closer distances indicates
closer proximity, larger distances indicates otherwise. Having
"associativity-reference-points" to reflect a associativity domain
relationship, when you already have all the distances from each node, is
somewhat redundant.
The concept of 'associativity domain' was necessary in FORM1 because we had no
other way of telling distance between NUMA nodes. We needed to rely on these
overly complex and convoluted subdomain abstractions to say that "nodeA belongs
to the same third-level domain as node B, and in the second-level domain with
node C". The kernel would read that and calculate that each level is doubling
the distance from the level before and local_distance is 10, so:
distAA = 10 distAB= 20 distAC = 40
With FORM2, if this information is already explicit in ibm,numa-distance-table,
why bother calculating associativity domains? If you want to know whether
PROCA is closer to PROCB or PROCX, just look at the NUMA distance table and
see which one is closer.
The way I see it, the '20' over there is not needed at all. What utility it
brings? And why create an associativity domain '1' in the MEMC associativity
at 0x3?
What the current QEMU FORM2 implementation is doing would be this:
{ 0, 0, 1, 40} -> PMEMD
{ 0, 0, 0, 1} -> PROCB/MEMC
PMEMD has a pointer to the NUMA node in which it would run as persistent
memory, node 1. All the memory/cpu nodes of node1 would be oblivious
to what PMEMD is doing.
I don't see the need of creating an associativity domain between node1
and node40 in 0x3. Besides, if a device_add operation of a PMEM that wants
to use nodeN as the node for persistent memory would trigger a massive
ibm,associativity update, on all LMBs that belongs to nodeN, because then
everyone needs to have the same third level associativity domain as the
hotplugged PMEM. To avoid that, if the idea is to 'just duplicate the
logical_domain_id in 0x3 for all non-PMEM devices' then what's the
difference of looking into the logical_numa_id at 0x4 in the first
place?
In fact, the more I speak about this PMEM scenario the more I wonder:
why doesn't the PMEM driver, when switching from persistent to regular
memory and vice-versa, take care of all the necessary updates in the
numa-distance-table and kernel internals to reflect the current distances
of its current mode? Is this a technical limitation?
Thanks
Daniel
On 6/18/21 1:30 AM, Daniel Henrique Barboza wrote:
On 6/17/21 8:11 AM, Aneesh Kumar K.V wrote:
quoted
Daniel Henrique Barboza [off-list ref] writes:
quoted
On 6/17/21 4:46 AM, David Gibson wrote:
quoted
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
In fact, the more I speak about this PMEM scenario the more I wonder:
why doesn't the PMEM driver, when switching from persistent to regular
memory and vice-versa, take care of all the necessary updates in the
numa-distance-table and kernel internals to reflect the current distances
of its current mode? Is this a technical limitation?
I sent v4 doing something similar to this .
-aneesh
From: David Gibson <hidden> Date: 2021-06-24 03:19:26
On Thu, Jun 17, 2021 at 04:29:01PM +0530, Aneesh Kumar K.V wrote:
On 6/17/21 1:16 PM, David Gibson wrote:
quoted
On Tue, Jun 15, 2021 at 12:35:17PM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
quoted
On Tue, Jun 15, 2021 at 11:27:50AM +0530, Aneesh Kumar K.V wrote:
quoted
David Gibson [off-list ref] writes:
...
quoted
quoted
quoted
It's weird to me that you'd want to consider them in different nodes
for those different purposes.
--------------------------------------
| NUMA node0 |
| ProcA -----> MEMA |
| | |
| | |
| -------------------> PMEMB |
| |
---------------------------------------
---------------------------------------
| NUMA node1 |
| |
| ProcB -------> MEMC |
| | |
| -------------------> PMEMD |
| |
| |
---------------------------------------
For a topology like the above application running of ProcA wants to find out
persistent memory mount local to its NUMA node. Hence when using it as
pmem fsdax mount or devdax device we want PMEMB to have associativity
of NUMA node0 and PMEMD to have associativity of NUMA node 1. But when
we want to use it as memory using dax kmem driver, we want both PMEMB
and PMEMD to appear as memory only NUMA node at a distance that is
derived based on the latency of the media.
I'm still not understanding why the latency we care about is different
in the two cases. Can you give an example of when this would result
in different actual node assignments for the two different cases?
In the above example in order allow use of PMEMB and PMEMD as memory only
NUMA nodes
we need platform to represent them in its own domainID. Let's assume that
platform assigned id 40 and 41 and hence both PMEMB and PMEMD will have
associativity array like below
{ 4, 6, 0} -> PROCA/MEMA
{ 4, 6, 40} -> PMEMB
{ 4, 6, 41} -> PMEMD
{ 4, 6, 1} -> PROCB/MEMB
When we want to use this device PMEMB and PMEMD as fsdax/devdax devices, we
essentially look for the first nearest online node. Which means both PMEMB
and PMEMD will appear as devices attached to node0. That is not ideal for
for many applications.
Not if you actually look at the distance table which tells you that
PMEMB is closer to node0 and PMEMD is closer to node1. That's exactly
what the distance table is for - making this information explicit,
rather than intuited from a confusing set of nested domains.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson