Skip to contentSkip to conversation

selftests: net: reuseport_bpf_numa: consider cpuless numa node

PATCH Superseded Feng Tang 3 participants 4 messages Aug 3 – Aug 6

Subsystems (4)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Feng Tang

author
#0 lore raw

PATCH v2 1/1 selftests: net: reuseport_bpf_numa: consider cpuless numa node

reuseport_bpf_numa case failed when testing on a platform with CXL memory:

#./reuseport_bpf_numa
---- IPv4 UDP ----
send node 0, receive socket 0
./reuseport_bpf_numa: failed to pin to node: Invalid argument

The root cause is that the platform has 2 numa nodes: node 0 has both cpu and memory, while node 1 is a CXL node which only has memory, and caused numa_run_on_node() to fail.

Add sanity check to skip cpuless numa node for the numa binding test.

---
Changlog:

since v1:
* address comments from Sashiko bot
https://lore.kernel.org/all/[email protected]/
* lift the cpuless node out of send_from_node()

 .../selftests/net/reuseport_bpf_numa.c        | 24 +++++++++++++++++++
 1 file changed, 24 insertions(+)
tools/testing/selftests/net/reuseport_bpf_numa.c +24 -0
index 8ec52fc5ef41..b5fb576fa61a 100644
@@ -104,6 +104,26 @@ static void attach_bpf(int fd)
 	close(bpf_fd);
 }
 
+/*
+ * Return true if it is a cpuless node. Return false if it isn't or any
+ * error (very unlikeyly) happens during the libnuma calls.
+ */
+static inline bool is_cpuless_node(int node_id)
+{
+	struct bitmask *cpumask;
+	bool ret = false;
+
+	cpumask = numa_allocate_cpumask();
+	if (!cpumask)
+		return ret;
+
+	if (!numa_node_to_cpus(node_id, cpumask) && !numa_bitmask_weight(cpumask))
+		ret = true;
+
+	numa_bitmask_free(cpumask);
+	return ret;
+}
+
 static void send_from_node(int node_id, int family, int proto)
 {
 	struct sockaddr_storage saddr, daddr;
@@ -213,6 +233,8 @@ static void test(int *rcv_fd, int len, int family, int proto)
 	for (node = 0; node < len; ++node) {
 		if (!numa_bitmask_isbitset(numa_nodes_ptr, node))
 			continue;
+		if (is_cpuless_node(node))
+			continue;
 		send_from_node(node, family, proto);
 		receive_on_node(rcv_fd, len, epfd, node, proto);
 	}
@@ -221,6 +243,8 @@ static void test(int *rcv_fd, int len, int family, int proto)
 	for (node = len - 1; node >= 0; --node) {
 		if (!numa_bitmask_isbitset(numa_nodes_ptr, node))
 			continue;
+		if (is_cpuless_node(node))
+			continue;
 		send_from_node(node, family, proto);
 		receive_on_node(rcv_fd, len, epfd, node, proto);
 	}
Signature · 1 line
2.39.5 (Apple Git-154)
#1 lore raw
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Low] Spelling mistake in newly added comment: 'unlikeyly' instead of 'unlikely'. --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Spelling mistake in newly added comment: 'unlikeyly' instead of 'unlikely'. --

--- Patch [1]: [PATCH net v2] selftests: net: reuseport_bpf_numa: consider cpuless numa node --- commit 4e53579a82f841c838d35fa42c245feb267f6bcb
Author: Feng Tang <[email protected]>

selftests: net: reuseport_bpf_numa: consider cpuless numa node

This commit fixes a test failure on platforms with CXL memory where a NUMA node might only have memory and no CPUs. It adds a sanity check to skip cpuless NUMA nodes during the NUMA binding test to prevent numa_run_on_node() from failing.

Quoted 4 lines
tools/testing/selftests/net/reuseport_bpf_numa.c +0 -0
index 8ec52fc5ef41b..b5fb576fa61af 100644

[ ... ]

Quoted 3 lines

+/*
+ * Return true if it is a cpuless node. Return false if it isn't or any
+ * error (very unlikeyly) happens during the libnuma calls.

[Severity: Low]
This isn't a bug, but there is a typo here: "unlikeyly" instead of "unlikely".

Quoted 5 lines

+ */

tools/testing/selftests/net/reuseport_bpf_numa.c +4 -0
+static inline bool is_cpuless_node(int node_id)
+{
+	struct bitmask *cpumask;
+	bool ret = false;

[ ... ]

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
maintainer
#2 lore raw
In reply to Feng Tang
Quoted 1 line · On Mon, 3 Aug 2026 09:42:14 +0800 Feng Tang wrote:

+ * error (very unlikeyly) happens during the libnuma calls.

spelling - "unlikeyly" instead of "unlikely"

when you repost please tag as net-next, rather than net

Signature · 1 line
pw-bot: cr
maintainer
#3 lore raw
In reply to Feng Tang
Quoted 1 line · On Mon, 3 Aug 2026 09:42:14 +0800 Feng Tang wrote:

+static inline bool is_cpuless_node(int node_id)

Also - why is this "inline" ? It's a test.