Re: [PATCH bpf-next v4 3/6] selftests/bpf: netns_new() and netns_free() helpers.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On 7/31/24 14:02, Stanislav Fomichev wrote:
On 07/31, Kui-Feng Lee wrote:
netns_new()/netns_free() create/delete network namespaces. They support the
option '-m' of test_progs to start/stop traffic monitor for the network
namespace being created for matched tests.

Signed-off-by: Kui-Feng Lee <thinker.li@xxxxxxxxx>
---
  tools/testing/selftests/bpf/network_helpers.c | 26 ++++++
  tools/testing/selftests/bpf/network_helpers.h |  2 +
  tools/testing/selftests/bpf/test_progs.c      | 80 +++++++++++++++++++
  tools/testing/selftests/bpf/test_progs.h      |  4 +
  4 files changed, 112 insertions(+)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index a3f0a49fb26f..f2cf43382a8e 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -432,6 +432,32 @@ char *ping_command(int family)
  	return "ping";
  }
+int make_netns(const char *name)
+{

[..]

+	char cmd[128];
+	int r;
+
+	snprintf(cmd, sizeof(cmd), "ip netns add %s", name);
+	r = system(cmd);

I doubt that we're gonna see any real problems with that in the tests,
but maybe easier to use apsrint and avoid dealing with fixed 128-byte
string?

asprintf? Sure!


+	if (r > 0)
+		/* exit code */
+		return -r;
+	return r;
+}
+
+int remove_netns(const char *name)
+{
+	char cmd[128];
+	int r;
+
+	snprintf(cmd, sizeof(cmd), "ip netns del %s >/dev/null 2>&1", name);
+	r = system(cmd);
+	if (r > 0)
+		/* exit code */
+		return -r;
+	return r;
+}
+
  struct nstoken {
  	int orig_netns_fd;
  };
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index cce56955371f..f8aa8680a640 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -93,6 +93,8 @@ struct nstoken;
  struct nstoken *open_netns(const char *name);
  void close_netns(struct nstoken *token);
  int send_recv_data(int lfd, int fd, uint32_t total_bytes);
+int make_netns(const char *name);
+int remove_netns(const char *name);
static __u16 csum_fold(__u32 csum)
  {
diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index 95643cd3119a..f86d47efe06e 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -1074,6 +1074,86 @@ int compare_stack_ips(int smap_fd, int amap_fd, int stack_trace_len)
  	return err;
  }
+struct netns_obj {
+	char nsname[128];
+	struct tmonitor_ctx *tmon;
+	struct nstoken *nstoken;
+};
+
+/* Create a new network namespace with the given name.
+ *
+ * Create a new network namespace and set the network namespace of the
+ * current process to the new network namespace if the argument "open" is
+ * true. This function should be paired with netns_free() to release the
+ * resource and delete the network namespace.
+ *
+ * It also implements the functionality of the option "-m" by starting
+ * traffic monitor on the background to capture the packets in this network
+ * namespace if the current test or subtest matching the pattern.
+ *
+ * name: the name of the network namespace to create.
+ * open: open the network namespace if true.
+ *
+ * Return: the network namespace object on success, NULL on failure.
+ */
+struct netns_obj *netns_new(const char *name, bool open)
+{
+	struct netns_obj *netns_obj = malloc(sizeof(*netns_obj));
+	int r;
+
+	if (!netns_obj)
+		return NULL;
+	memset(netns_obj, 0, sizeof(*netns_obj));
+
+	strncpy(netns_obj->nsname, name, sizeof(netns_obj->nsname));
+	netns_obj->nsname[sizeof(netns_obj->nsname) - 1] = '\0';

Same here. Seems easier to have "char *nsname" and do
netns_obj->nsname = strdup(name) here. Trimming the name, in theory,
is problematic because do do remove_netns(netns_obj->nsname) later
on (with potentially trimmed name).

You are right!


But, again, probably not a huge deal in the selftests. So up to you on
whether you want to address it or not.




[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux