netdev - Re: [External] Re: [PATCH] lib/bpf: Fix btf_load error lead to enable debug log

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] [day] [month] [year] [list]

Message-ID: <f3f725ac-99a1-be99-8940-a13d00eb8a24@bytedance.com>
Date:   Sun, 1 Aug 2021 13:59:14 +0800
From:   zhoufeng <zhoufeng.zf@...edance.com>
To:     Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc:     Networking <netdev@...r.kernel.org>,
        Stephen Hemminger <stephen@...workplumber.org>,
        kuznet@....inr.ac.ru, duanxiongchun@...edance.com,
        Dongdong Wang <wangdongdong.6@...edance.com>,
        chenzhuowen.simon@...edance.com, cong.wang@...edance.com
Subject: Re: [External] Re: [PATCH] lib/bpf: Fix btf_load error lead to enable
 debug log



在 2021/7/31 上午2:40, Andrii Nakryiko 写道:
> On Wed, Jul 28, 2021 at 9:55 PM Feng zhou <zhoufeng.zf@...edance.com> wrote:
>>
>> From: Feng Zhou <zhoufeng.zf@...edance.com>
>>
>> Use tc with no verbose, when bpf_btf_attach fail,
>> the conditions:
>> "if (fd < 0 && (errno == ENOSPC || !ctx->log_size))"
>> will make ctx->log_size != 0. And then, bpf_prog_attach,
>> ctx->log_size != 0. so enable debug log.
>> The verifier log sometimes is so chatty on larger programs.
>> bpf_prog_attach is failed.
>> "Log buffer too small to dump verifier log 16777215 bytes (9 tries)!"
>>
>> BTF load failure does not affect prog load. prog still work.
>> So when BTF/PROG load fail, enlarge log_size and re-fail with
>> having verbose.
>>
>> Signed-off-by: Feng Zhou <zhoufeng.zf@...edance.com>
>> ---
> 
> This seems to be targeted against iproute2? It would be good to
> specify that as [PATCH iproute2] or something.
> 
> 

Yeah, it's a good idea. I will resend it.


>>   lib/bpf_legacy.c | 4 ++--
>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/lib/bpf_legacy.c b/lib/bpf_legacy.c
>> index 8a03b9c2..d824388c 100644
>> --- a/lib/bpf_legacy.c
>> +++ b/lib/bpf_legacy.c
>> @@ -1531,7 +1531,7 @@ retry:
>>                   * into our buffer. Still, try to give a debuggable error
>>                   * log for the user, so enlarge it and re-fail.
>>                   */
>> -               if (fd < 0 && (errno == ENOSPC || !ctx->log_size)) {
>> +               if (fd < 0 && errno == ENOSPC) {
>>                          if (tries++ < 10 && !bpf_log_realloc(ctx))
>>                                  goto retry;
>>
>> @@ -2069,7 +2069,7 @@ retry:
>>          fd = bpf_btf_load(ctx->btf_data->d_buf, ctx->btf_data->d_size,
>>                            ctx->log, ctx->log_size);
>>          if (fd < 0 || ctx->verbose) {
>> -               if (fd < 0 && (errno == ENOSPC || !ctx->log_size)) {
>> +               if (fd < 0 && errno == ENOSPC) {
>>                          if (tries++ < 10 && !bpf_log_realloc(ctx))
>>                                  goto retry;
>>
>> --
>> 2.11.0
>>