Hi,
This is a repost of RFCv3 [1] rebased using current David's ppc-for-6.2.
The reason of this repost is that the kernel side changes got accepted
in the maintainer's tree [2], meaning that this series is now an implementation
of the future FORM2 kernel support in the pseries kernel.
In a POWER9 host with Annesh's patches from [2] and a QEMU with this
series, and a guest with the following NUMA distances:
(...)
-object memory-backend-ram,id=mem0,size=4G -numa node,memdev=mem0,cpus=0-1,nodeid=0 \
-object memory-backend-ram,id=mem1,size=4G -numa node,memdev=mem1,cpus=2-3,nodeid=1 \
-object memory-backend-ram,id=mem2,size=4G -numa node,memdev=mem2,cpus=4-5,nodeid=2 \
-object memory-backend-ram,id=mem3,size=2G -numa node,memdev=mem3,cpus=6-7,nodeid=3 \
-numa node,nodeid=4 \
-numa dist,src=0,dst=1,val=22 -numa dist,src=0,dst=2,val=22 \
-numa dist,src=0,dst=3,val=22 -numa dist,src=0,dst=4,val=22 \
-numa dist,src=1,dst=0,val=44 -numa dist,src=1,dst=2,val=44 \
-numa dist,src=1,dst=3,val=44 -numa dist,src=1,dst=4,val=44 \
-numa dist,src=2,dst=0,val=66 -numa dist,src=2,dst=1,val=66 \
-numa dist,src=2,dst=3,val=66 -numa dist,src=2,dst=4,val=66 \
-numa dist,src=3,dst=0,val=88 -numa dist,src=3,dst=1,val=88 \
-numa dist,src=3,dst=2,val=88 -numa dist,src=3,dst=4,val=88
This is the output of 'numactl -H' in the guest:
# numactl -H
available: 4 nodes (0-3)
node 0 cpus: 0 1
node 0 size: 3987 MB
node 0 free: 3362 MB
node 1 cpus: 2 3
node 1 size: 4090 MB
node 1 free: 4071 MB
node 2 cpus: 4 5
node 2 size: 4090 MB
node 2 free: 4075 MB
node 3 cpus: 6 7
node 3 size: 2027 MB
node 3 free: 2010 MB
node distances:
node 0 1 2 3
0: 10 22 22 22
1: 44 10 44 44
2: 66 66 10 66
3: 88 88 88 10
We're now able to set asynchronous distances between nodes while also
not being constrained to any fixed values (10, 20, 40, 80 ...) like we
currently are when using FORM1.
[1] https://lists.gnu.org/archive/html/qemu-devel/2021-07/msg03622.html
[2] https://lore.kernel.org/linuxppc-dev/20210812132223.225214-1-aneesh.kumar@linux.ibm.com/
Daniel Henrique Barboza (5):
spapr_numa.c: split FORM1 code into helpers
spapr: move NUMA data init to post-CAS
spapr_numa.c: base FORM2 NUMA affinity support
spapr: simplify spapr_numa_associativity_init params
spapr: move memory/cpu less check to spapr_numa_FORM1_affinity_init()
hw/ppc/spapr.c | 60 ++++------
hw/ppc/spapr_hcall.c | 4 +
hw/ppc/spapr_numa.c | 224 +++++++++++++++++++++++++++++++++---
include/hw/ppc/spapr.h | 1 +
include/hw/ppc/spapr_numa.h | 3 +-
include/hw/ppc/spapr_ovec.h | 1 +
6 files changed, 236 insertions(+), 57 deletions(-)
--
2.31.1