[libvirt] [PATCH] qemu: Use the CPU index in capabilities to map NUMA node to cpu list.
On some of the NUMA platforms, the CPU index in each NUMA node grows sequentially. While on other platforms, it can be inconsecutive, E.g. % numactl --hardware available: 4 nodes (0-3) node 0 cpus: 0 4 8 12 16 20 24 28 node 0 size: 131058 MB node 0 free: 86531 MB node 1 cpus: 1 5 9 13 17 21 25 29 node 1 size: 131072 MB node 1 free: 127070 MB node 2 cpus: 2 6 10 14 18 22 26 30 node 2 size: 131072 MB node 2 free: 127758 MB node 3 cpus: 3 7 11 15 19 23 27 31 node 3 size: 131072 MB node 3 free: 127226 MB node distances: node 0 1 2 3 0: 10 20 20 20 1: 20 10 20 20 2: 20 20 10 20 3: 20 20 20 10 This patch is to fix the problem by using the CPU index in caps->host.numaCell[i]->cpus[i] to set the bitmask instead of assuming the CPU index of the NUMA nodes are always sequential. --- src/qemu/qemu_process.c | 11 ++--------- 1 files changed, 2 insertions(+), 9 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 19bb22a..58ba5bf 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -1825,20 +1825,13 @@ qemuProcessInitCpuAffinity(struct qemud_driver *driver, if (vm->def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO) { VIR_DEBUG("Set CPU affinity with advisory nodeset from numad"); /* numad returns the NUMA node list, convert it to cpumap */ - int prev_total_ncpus = 0; for (i = 0; i < driver->caps->host.nnumaCell; i++) { int j; int cur_ncpus = driver->caps->host.numaCell[i]->ncpus; if (nodemask[i]) { - for (j = prev_total_ncpus; - j < cur_ncpus + prev_total_ncpus && - j < maxcpu && - j < VIR_DOMAIN_CPUMASK_LEN; - j++) { - VIR_USE_CPU(cpumap, j); - } + for (j = 0; j < cur_ncpus; j++) + VIR_USE_CPU(cpumap, driver->caps->host.numaCell[i]->cpus[j]); } - prev_total_ncpus += cur_ncpus; } } else { VIR_DEBUG("Set CPU affinity with specified cpuset"); -- 1.7.7.3
On 05/12/2012 06:52 AM, Osier Yang wrote:
On some of the NUMA platforms, the CPU index in each NUMA node grows sequentially. While on other platforms, it can be inconsecutive,
s/inconsecutive/non-consecutive/
This patch is to fix the problem by using the CPU index in caps->host.numaCell[i]->cpus[i] to set the bitmask instead of assuming the CPU index of the NUMA nodes are always sequential. --- src/qemu/qemu_process.c | 11 ++--------- 1 files changed, 2 insertions(+), 9 deletions(-)
ACK. -- Eric Blake eblake@redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On 2012年05月15日 01:03, Eric Blake wrote:
On 05/12/2012 06:52 AM, Osier Yang wrote:
On some of the NUMA platforms, the CPU index in each NUMA node grows sequentially. While on other platforms, it can be inconsecutive,
s/inconsecutive/non-consecutive/
This patch is to fix the problem by using the CPU index in caps->host.numaCell[i]->cpus[i] to set the bitmask instead of assuming the CPU index of the NUMA nodes are always sequential. --- src/qemu/qemu_process.c | 11 ++--------- 1 files changed, 2 insertions(+), 9 deletions(-)
ACK.
Thanks, pushed with the the substitution. Regards, Osier
participants (2)
-
Eric Blake -
Osier Yang