bpf: Unify data extension operation of jited_ksyms and jited_linfo

We found that 32-bit environment can not print BPF line info due to a data
inconsistency between jited_ksyms[0] and jited_linfo[0].

For example:

  jited_kyms[0] = 0xb800067c, jited_linfo[0] = 0xffffffffb800067c

We know that both of them store BPF func address, but due to the different
data extension operations when extended to u64, they may not be the same.
We need to unify the data extension operations of them.

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/CAEf4BzZ-eDcdJZgJ+Np7Y=V-TVjDDvOMqPwzKjyWrh=i5juv4w@mail.gmail.com
Link: https://lore.kernel.org/bpf/20220530092815.1112406-2-pulehui@huawei.com

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 2b69306..aeb3113 100644 (file)
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -4090,14 +4090,15 @@ static int bpf_prog_get_info_by_fd(struct file *file,
                info.nr_jited_line_info = 0;
        if (info.nr_jited_line_info && ulen) {
                if (bpf_dump_raw_ok(file->f_cred)) {
+                       unsigned long line_addr;
                        __u64 __user *user_linfo;
                        u32 i;
 
                        user_linfo = u64_to_user_ptr(info.jited_line_info);
                        ulen = min_t(u32, info.nr_jited_line_info, ulen);
                        for (i = 0; i < ulen; i++) {
-                               if (put_user((__u64)(long)prog->aux->jited_linfo[i],
-                                            &user_linfo[i]))
+                               line_addr = (unsigned long)prog->aux->jited_linfo[i];
+                               if (put_user((__u64)line_addr, &user_linfo[i]))
                                        return -EFAULT;
                        }
                } else {