docs/specs/ppc-spapr-numa.rst

12 --------------------------------------------
19 bit 0 of byte 5 of the ibm,architecture-vec-5 property. The format with
23 with the highest level grouping up to the smallest. Considering the
28     Mem M1 ---- Proc P1    |
29     -----------------      | Socket S1  ---|
32     Mem M2 ---- Proc P2    |               |
33     -----------------      | Socket S2  ---|
46 Relative Performance Distance and ibm,associativity-reference-points
47 --------------------------------------------------------------------
49 The ibm,associativity-reference-points property is an array that is used
54 of the ibm,architecture-vec-5 property. The format with bit 0 equal to zero
56 ibm,associativity-reference-points represents an 1 based ordinal index (i.e.
67 three NUMA levels:
69 * ibm,associativity-reference-points = {0x3, 0x2, 0x1}
71 The first NUMA level (0x3) is interpreted as the third element of each
72 ibm,associativity array, the second level is the second element and
73 the third level is the first element. Let's also consider that elements
74 belonging to the first NUMA level have distance equal to 10 from each
75 other, and each NUMA level doubles the distance from the previous. This
76 means that the second would be 20 and the third level 40. For the P1 and
81   * ibm,associativity-reference-points = {0x3, 0x2, 0x1}
85   First NUMA level (0x3) => associativity[2] = C1
86   Second NUMA level (0x2) => associativity[1] = S1
87   Third NUMA level (0x1) => associativity[0] = MOD1
91   First NUMA level (0x3) => associativity[2] = C2
92   Second NUMA level (0x2) => associativity[1] = S2
93   Third NUMA level (0x1) => associativity[0] = MOD1
95   P1 and P2 have the same third NUMA level, MOD1: Distance between them = 40
97 Changing the ibm,associativity-reference-points array changes the performance
103   * ibm,associativity-reference-points = {0x2}
107   First NUMA level (0x2) => associativity[1] = S1
111   First NUMA level (0x2) => associativity[1] = S2
113   P1 and P2 does not have a common performance boundary. Since this is a one level
115   level, 20.
123   * ibm,associativity-reference-points = {0x1}
127   First NUMA level (0x1) => associativity[0] = MOD0
131   First NUMA level (0x1) => associativity[0] = MOD0
144 provides a common-performance hierarchy,  and the ibm,associativity-reference-points
145 array that tells which level of associativity is considered to be relevant
149 as it sees fit. For the pseries Linux guest, each level of NUMA duplicates
150 the distance of the previous level, and the maximum amount of levels is
154 * both resources in the first NUMA level: 10
155 * resources one NUMA level apart: 20
157 * resources three NUMA levels apart: 80
167 * ibm,associativity-reference-points is set to {0x4, 0x3, 0x2, 0x1}, allowing
170 * ibm,max-associativity-domains supports multiple associativity domains in all
175 These changes are only effective for pseries-5.2 and newer machines that are
189 --------------------------------------------
202 distance (10), keeping it exclusive to the 4th NUMA level (which is still
241 --------------------------------------
274 NUMA level matches (considering the reference points {0x4, 0x3, 0x2, 0x1}) for
284 the NUMA level matches. Between 0 and 1, a match in 0x2, we'll also copy
302 the NUMA level match:
329 Note that this is not what the user wanted - the desired distance between
340 ---------------------------------
349   any A-B pair where the distance back and forth is asymmetric. For example, the
360 * 'non-transitive' topologies will be poorly translated to the guest. This is the
392 the default ibm,associativity-reference-points being used in the pseries
395 ibm,associativity-reference-points = {0x4, 0x4, 0x2}
397 The first and second level are equal, 0x4, and a third one was added in
400 the device tree, the pseries Linux guest will only recognize three scenarios
403 * if the resources belongs to the same first NUMA level = 10
404 * second level is skipped since it's equal to the first
406   to the same third NUMA level, having distance = 40