On 6/13/23 1:50 AM, baomingtong001@xxxxxxxxxx wrote:
Fix the following coccicheck warning:
tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c:28:14-17: Unneeded
variable: "ret".
Return "1".
Signed-off-by: Mingtong Bao <baomingtong001@xxxxxxxxxx>
---
tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
index 4a9f63bea66c..7f0146682577 100644
--- a/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
+++ b/tools/testing/selftests/bpf/progs/tailcall_bpf2bpf6.c
@@ -25,10 +25,9 @@ static __noinline
int subprog_tail(struct __sk_buff *skb)
{
/* Don't propagate the constant to the caller */
- volatile int ret = 1;
bpf_tail_call_static(skb, &jmp_table, 0);
- return ret;
+ return 1;
Please pay attention to the comment:
/* Don't propagate the constant to the caller */
which clearly says 'constant' is not preferred.
The patch introduced this change is:
5e0b0a4c52d30 selftests/bpf: Test tail call counting with bpf2bpf
and data on stack
The test intentionally want to:
'Specifically when the size
of data allocated on BPF stack is not a multiple on 8.'
Note that with volatile and without volatile, the generated
code will be different and it will result in different
verification path.
cc Jakub for further clarification.
}
SEC("tc")
