On 5/15/24 6:13 PM, Geliang Tang wrote:
From: Geliang Tang <tanggeliang@xxxxxxxxxx>
Many BPF selftests create new test network namespaces by using CLONE_NEWNET
flag to unshare the network namespace, so that the calling process is moved
into a new network namespace which is not shared with any previously
existing process. So this patch adds a new helper in network_helpers.c
named unshare_netns(), which create a new network namespace and set the
net device lo up.
Signed-off-by: Geliang Tang <tanggeliang@xxxxxxxxxx>
---
tools/testing/selftests/bpf/network_helpers.c | 19 +++++++++++++++++++
tools/testing/selftests/bpf/network_helpers.h | 1 +
2 files changed, 20 insertions(+)
diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 35250e6cde7f..919bb2a0c6a6 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -670,3 +670,22 @@ int send_recv_data(int lfd, int fd, uint32_t total_bytes)
return err;
}
+
+int unshare_netns(void)
+{
+ int err;
+
+ err = unshare(CLONE_NEWNET);
+ if (err) {
+ log_err("unshare netns failed");
+ return err;
+ }
+
+ err = SYS_NOFAIL("ip link set dev lo up");
Losing useful error output is not good. Please find a way to keep them available
such that it is easier to debug. The same goes for similar cases in this patch
set. I think the only exception could be the "ip netns del" during cleanup.
Also, bpf CI reports a crash (should be the test #18 bpf_ip_check_defrag). Not
sure how this set indirectly triggered it more easily:
https://github.com/kernel-patches/bpf/actions/runs/9290828674/job/25569002645
pw-bot: cr
+ if (err) {
+ log_err("set dev lo up failed");
+ return err;
+ }
+
+ return 0;
+}
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 883c7ea9d8d5..b087330f2861 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -82,6 +82,7 @@ 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 unshare_netns(void);
static __u16 csum_fold(__u32 csum)
{
