在 2025/2/25 09:15, Andrii Nakryiko 写道:
On Mon, Feb 24, 2025 at 9:02 AM Tao Chen <chen.dylane@xxxxxxxxx> wrote:
Similarly to libbpf_probe_bpf_helper, the libbpf_probe_bpf_kfunc
used to test the availability of the different eBPF kfuncs on the
current system.
Cc: Tao Chen <dylane.chen@xxxxxxxxxxxxxx>
Reviewed-by: Jiri Olsa <jolsa@xxxxxxxxxx>
Reviewed-by: Eduard Zingerman <eddyz87@xxxxxxxxx>
Signed-off-by: Tao Chen <chen.dylane@xxxxxxxxx>
---
tools/lib/bpf/libbpf.h | 19 ++++++++++++-
tools/lib/bpf/libbpf.map | 1 +
tools/lib/bpf/libbpf_probes.c | 51 +++++++++++++++++++++++++++++++++++
3 files changed, 70 insertions(+), 1 deletion(-)
[...]
+ buf[0] = '\0';
+ ret = probe_prog_load(prog_type, insns, insn_cnt, btf_fd >= 0 ? fd_array : NULL,
+ buf, sizeof(buf));
+ if (ret < 0)
+ return libbpf_err(ret);
+
+ if (ret > 0)
+ return 1; /* assume supported */
+
+ /* If BPF verifier recognizes BPF kfunc but it's not supported for
+ * given BPF program type, it will emit "calling kernel function
+ * <name> is not allowed". If the kfunc id is invalid,
+ * it will emit "kernel btf_id <id> is not a function". If BTF fd
+ * invalid in module BTF, it will emit "invalid module BTF fd specified" or
+ * "negative offset disallowed for kernel module function call". If
+ * kfunc prog not dev buound, it will emit "metadata kfuncs require
+ * device-bound program".
+ */
+ if (strstr(buf, "not allowed") || strstr(buf, "not a function") ||
+ strstr(buf, "invalid module BTF fd") ||
why is invalid module BTF FD not an error (negative return)?
+ strstr(buf, "negative offset disallowed") ||
+ strstr(buf, "device-bound program"))
+ return 0;
+
+ return 1;
+}
+
int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type, enum bpf_func_id helper_id,
const void *opts)
{
--
2.43.0
In probe_prog_load, err will be checked and converted into either 0 or 1.
--
Best Regards
Tao Chen
