This series allows running the tool on some configurations that do not conform to an assumption each node contains nr_cpus/nr_nodes CPUs at most. Instead, the actual node-to- CPU mapping is acquired dynamically.
patch 1 fixes the described issue patches 2,3 are follow-up fixes Changes since v1: - numa01* and numa02* test names left intact; - "2x3-convergence" fix moved out to separate patch Cc: Peter Zijlstra <pet...@infradead.org> Cc: Ingo Molnar <mi...@redhat.com> Cc: Arnaldo Carvalho de Melo <a...@kernel.org> Cc: Mark Rutland <mark.rutl...@arm.com> Cc: Alexander Shishkin <alexander.shish...@linux.intel.com> Cc: Jiri Olsa <jo...@redhat.com> Cc: Namhyung Kim <namhy...@kernel.org> Alexander Gordeev (3): perf bench numa: use numa_node_to_cpus() to bind tasks to nodes perf bench numa: fix number of processes in "2x3-convergence" test perf bench numa: fix benchmark names tools/perf/bench/numa.c | 64 ++++++++++++++++++++++++------------------------- 1 file changed, 31 insertions(+), 33 deletions(-) -- 1.8.3.1