2019-08-01 16:31:38

by Jiri Olsa

Subject: [PATCH] perf bench numa: Fix cpu0 binding

Michael reported an issue with perf bench numa failing with
binding to cpu0 with '-0' option.

# perf bench numa mem -p 3 -t 1 -P 512 -s 100 -zZcm0 --thp 1 -M 1 -ddd
# Running 'numa/mem' benchmark:

# Running main, "perf bench numa numa-mem -p 3 -t 1 -P 512 -s 100 -zZcm0 --thp 1 -M 1 -ddd"
binding to node 0, mask: 0000000000000001 => -1
perf: bench/numa.c:356: bind_to_memnode: Assertion `!(ret)' failed.
Aborted (core dumped)

This happens when the cpu0 is not part of node0,
which is the benchmark assumption and we can see
that's not the case for some powerpc servers.

Using correct node for cpu0 binding.

Cc: Satheesh Rajendran <[email protected]>
Reported-by: Michael Petlan <[email protected]>
Link: http://lkml.kernel.org/n/[email protected]
Signed-off-by: Jiri Olsa <[email protected]>
---
tools/perf/bench/numa.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
index a640ca7aaada..513cb2f2fa32 100644
--- a/tools/perf/bench/numa.c
+++ b/tools/perf/bench/numa.c
@@ -379,8 +379,10 @@ static u8 *alloc_data(ssize_t bytes0, int map_flags,

/* Allocate and initialize all memory on CPU#0: */
if (init_cpu0) {
- orig_mask = bind_to_node(0);
- bind_to_memnode(0);
+ int node = numa_node_of_cpu(0);
+
+ orig_mask = bind_to_node(node);
+ bind_to_memnode(node);
}

bytes = bytes0 + HPSIZE;
--
2.21.0

2019-08-08 20:17:33

by tip-bot for Vasyl Gomonovych

Subject: [tip:perf/urgent] perf bench numa: Fix cpu0 binding

Commit-ID: 6bbfe4e602691b90ac866712bd4c43c51e546a60
Gitweb: https://git.kernel.org/tip/6bbfe4e602691b90ac866712bd4c43c51e546a60
Author: Jiri Olsa <[email protected]>
AuthorDate: Thu, 1 Aug 2019 16:26:42 +0200
Committer: Arnaldo Carvalho de Melo <[email protected]>
CommitDate: Thu, 1 Aug 2019 11:34:13 -0300

perf bench numa: Fix cpu0 binding

Michael reported an issue with perf bench numa failing with binding to
cpu0 with '-0' option.

# perf bench numa mem -p 3 -t 1 -P 512 -s 100 -zZcm0 --thp 1 -M 1 -ddd
# Running 'numa/mem' benchmark:

# Running main, "perf bench numa numa-mem -p 3 -t 1 -P 512 -s 100 -zZcm0 --thp 1 -M 1 -ddd"
binding to node 0, mask: 0000000000000001 => -1
perf: bench/numa.c:356: bind_to_memnode: Assertion `!(ret)' failed.
Aborted (core dumped)

This happens when the cpu0 is not part of node0, which is the benchmark
assumption and we can see that's not the case for some powerpc servers.

Using correct node for cpu0 binding.

Reported-by: Michael Petlan <[email protected]>
Signed-off-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Satheesh Rajendran <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
---
tools/perf/bench/numa.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
index a640ca7aaada..513cb2f2fa32 100644
--- a/tools/perf/bench/numa.c
+++ b/tools/perf/bench/numa.c
@@ -379,8 +379,10 @@ static u8 *alloc_data(ssize_t bytes0, int map_flags,

/* Allocate and initialize all memory on CPU#0: */
if (init_cpu0) {
- orig_mask = bind_to_node(0);
- bind_to_memnode(0);
+ int node = numa_node_of_cpu(0);
+
+ orig_mask = bind_to_node(node);
+ bind_to_memnode(node);
}

bytes = bytes0 + HPSIZE;