linux-kernel - Re: [PATCH bpf-next v3 1/3] libbpf: Refactor libbpf_probe_bpf

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <aedece31-5531-4ffe-9967-5906ec971cec@gmail.com>
Date: Sat, 25 Jan 2025 23:05:23 +0800
From: Tao Chen <chen.dylane@...il.com>
To: Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc: ast@...nel.org, daniel@...earbox.net, andrii@...nel.org,
 eddyz87@...il.com, haoluo@...gle.com, jolsa@...nel.org, qmo@...nel.org,
 bpf@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH bpf-next v3 1/3] libbpf: Refactor libbpf_probe_bpf_helper

在 2025/1/25 02:44, Andrii Nakryiko 写道:
> On Fri, Jan 24, 2025 at 6:44 AM Tao Chen <chen.dylane@...il.com> wrote:
>>
>> Extract the common part as probe_func_comm, which will be used in
>> both libbpf_probe_bpf_{helper, kfunc}
>>
>> Signed-off-by: Tao Chen <chen.dylane@...il.com>
>> ---
>>   tools/lib/bpf/libbpf_probes.c | 38 ++++++++++++++++++++++++-----------
>>   1 file changed, 26 insertions(+), 12 deletions(-)
>>
>> diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
>> index 9dfbe7750f56..b73345977b4e 100644
>> --- a/tools/lib/bpf/libbpf_probes.c
>> +++ b/tools/lib/bpf/libbpf_probes.c
>> @@ -413,22 +413,20 @@ int libbpf_probe_bpf_map_type(enum bpf_map_type map_type, const void *opts)
>>          return libbpf_err(ret);
>>   }
>>
>> -int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type, enum bpf_func_id helper_id,
>> -                           const void *opts)
>> +static int probe_func_comm(enum bpf_prog_type prog_type, struct bpf_insn insn,
>> +                          char *accepted_msgs, size_t msgs_size)
>>   {
>>          struct bpf_insn insns[] = {
>> -               BPF_EMIT_CALL((__u32)helper_id),
>> +               BPF_EXIT_INSN(),
>>                  BPF_EXIT_INSN(),
>>          };
>>          const size_t insn_cnt = ARRAY_SIZE(insns);
>> -       char buf[4096];
>> -       int ret;
>> +       int err;
>>
>> -       if (opts)
>> -               return libbpf_err(-EINVAL);
>> +       insns[0] = insn;
>>
>>          /* we can't successfully load all prog types to check for BPF helper
>> -        * support, so bail out with -EOPNOTSUPP error
>> +        * and kfunc support, so bail out with -EOPNOTSUPP error
>>           */
>>          switch (prog_type) {
>>          case BPF_PROG_TYPE_TRACING:
> 
> there isn't much logic that you will extract here besides this check
> whether program type can even be successfully loaded, so I wouldn't
> extract probe_func_comm(), but rather extract just the check:
> 
> static bool can_probe_prog_type(enum bpf_prog_type prog_type)
> {
>          /* we can't successfully load all prog types to check for BPF
> helper/kfunc
>           * support, so check this early and bail
>           */
>          switch (prog_type) {
>              ...: return false
>          default:
>              return true;
> }
> 
> 
> And just check that can_probe_prog_type() inside
> libbpf_probe_bpf_helper and libbpf_probe_bpf_kfunc
> 
> pw-bot: cr
> 

Hi Andrii,
Thank you for your review, jiri also suggested putting the insn part 
back into libbpf_bpf_probe_{helper, kfunc}, so I'll make the 
modifications as you suggested in v4.

>> @@ -440,10 +438,26 @@ int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type, enum bpf_func_id helpe
>>                  break;
>>          }
>>
>> -       buf[0] = '\0';
>> -       ret = probe_prog_load(prog_type, insns, insn_cnt, buf, sizeof(buf));
>> -       if (ret < 0)
>> -               return libbpf_err(ret);
>> +       accepted_msgs[0] = '\0';
>> +       err = probe_prog_load(prog_type, insns, insn_cnt, accepted_msgs, msgs_size);
>> +       if (err < 0)
>> +               return libbpf_err(err);
>> +
>> +       return 0;
>> +}
>> +
>> +int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type, enum bpf_func_id helper_id,
>> +                           const void *opts)
>> +{
>> +       char buf[4096];
>> +       int ret;
>> +
>> +       if (opts)
>> +               return libbpf_err(-EINVAL);
>> +
>> +       ret = probe_func_comm(prog_type, BPF_EMIT_CALL((__u32)helper_id), buf, sizeof(buf));
>> +       if (ret)
>> +               return ret;
>>
>>          /* If BPF verifier doesn't recognize BPF helper ID (enum bpf_func_id)
>>           * at all, it will emit something like "invalid func unknown#181".
>> --
>> 2.43.0
>>


-- 
Best Regards
Dylane Chen