Skip to contentSkip to conversation

vrf: print BPF log buffer if bpf_program_load fails

PATCH In review Luca Boccassi 2 participants 4 messages Jan 17, 2021 – Jan 18, 2021

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 vrf: print BPF log buffer if bpf_program_load fails

Necessary to understand what is going on when bpf_program_load fails

---
 ip/ipvrf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
ip/ipvrf.c +2 -2
index 42779e5c..91578031 100644
@@ -278,8 +278,8 @@ static int vrf_configure_cgroup(const char *path, int ifindex)
 	 */
 	prog_fd = prog_load(ifindex);
 	if (prog_fd < 0) {
-		fprintf(stderr, "Failed to load BPF prog: '%s'\n",
-			strerror(errno));
+		fprintf(stderr, "Failed to load BPF prog: '%s'\n%s",
+			strerror(errno), bpf_log_buf);
 
 		if (errno != EPERM) {
 			fprintf(stderr,
Signature · 1 line
2.29.2
#1 lore raw

PATCH 2/2 vrf: fix ip vrf exec with libbpf

The size of bpf_insn is passed to bpf_load_program instead of the number of elements as it expects, so ip vrf exec fails with:

$ sudo ip link add vrf-blue type vrf table 10
$ sudo ip link set dev vrf-blue up
$ sudo ip/ip vrf exec vrf-blue ls
Failed to load BPF prog: 'Invalid argument'
last insn is not an exit or jmp
processed 0 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
Kernel compiled with CGROUP_BPF enabled?

https://bugs.debian.org/980046

---
 lib/bpf_glue.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
lib/bpf_glue.c +2 -1
index fa609bfe..d00a0dc1 100644
@@ -14,7 +14,8 @@ int bpf_program_load(enum bpf_prog_type type, const struct bpf_insn *insns,
 		     size_t size_log)
 {
 #ifdef HAVE_LIBBPF
-	return bpf_load_program(type, insns, size_insns, license, 0, log, size_log);
+	return bpf_load_program(type, insns, size_insns / sizeof(struct bpf_insn),
+				license, 0, log, size_log);
 #else
 	return bpf_prog_load_dev(type, insns, size_insns, license, 0, log, size_log);
 #endif
Signature · 1 line
2.29.2
Reviewed-by on v1 2/2
#2 lore raw

Re: [PATCH iproute2 2/2] vrf: fix ip vrf exec with libbpf

Quoted 11 lines · Luca Boccassi

The size of bpf_insn is passed to bpf_load_program instead of the number of elements as it expects, so ip vrf exec fails with:

$ sudo ip link add vrf-blue type vrf table 10
$ sudo ip link set dev vrf-blue up
$ sudo ip/ip vrf exec vrf-blue ls
Failed to load BPF prog: 'Invalid argument'
last insn is not an exit or jmp
processed 0 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
Kernel compiled with CGROUP_BPF enabled?
Quoted 9 lines

https://bugs.debian.org/980046

---
 lib/bpf_glue.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)