[PATCH bpf-next] libbpf: Fix unintentional success return code in bpf_object__load

[PATCH bpf-next] libbpf: Fix unintentional success return code in bpf_object__load

[Alex Gartrell]

There are code paths where EINVAL is returned directly without setting
errno. In that case, errno could be 0, which would mask the
failure. For example, if a careless programmer set log_level to 10000
out of laziness, they would have to spend a long time trying to figure
out why.

Fixes: 4f33ddb4e3e2 ("libbpf: Propagate EPERM to caller on program load")
Signed-off-by: Alex Gartrell <alexgartrell@gmail.com>
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 2e2523d8bb6d..8f9e7d281225 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -6067,7 +6067,7 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 		free(log_buf);
 		goto retry_load;
 	}
-	ret = -errno;
+	ret = errno ? -errno : -LIBBPF_ERRNO__LOAD;
 	cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
 	pr_warn("load bpf program failed: %s\n", cp);
 	pr_perm_msg(ret);
-- 
2.26.0

Re: [PATCH bpf-next] libbpf: Fix unintentional success return code in bpf_object__load

[Alexei Starovoitov]

On Wed, Aug 26, 2020 at 12:56 AM Alex Gartrell <alexgartrell@gmail.com> wrote:
>
> There are code paths where EINVAL is returned directly without setting
> errno. In that case, errno could be 0, which would mask the
> failure. For example, if a careless programmer set log_level to 10000
> out of laziness, they would have to spend a long time trying to figure
> out why.
>
> Fixes: 4f33ddb4e3e2 ("libbpf: Propagate EPERM to caller on program load")
> Signed-off-by: Alex Gartrell <alexgartrell@gmail.com>

Applied. Thanks