netdev - Re: More painful merges...

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]

Date:   Mon, 30 Oct 2017 09:40:51 -0700
From:   Cong Wang <xiyou.wangcong@...il.com>
To:     David Miller <davem@...emloft.net>
Cc:     Linux Kernel Network Developers <netdev@...r.kernel.org>,
        John Fastabend <john.fastabend@...il.com>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Jiri Pirko <jiri@...lanox.com>
Subject: Re: More painful merges...

On Mon, Oct 30, 2017 at 8:58 AM, Cong Wang <xiyou.wangcong@...il.com> wrote:
> On Mon, Oct 30, 2017 at 5:15 AM, David Miller <davem@...emloft.net> wrote:
>>
>> BPF and tc action bug fixes in 'net' caused some extremely
>> painful merges with 'net-next'.
>>
>> Could the interested parties please take a look at 'net-next'
>> and send me any necessary fixes, as I did the best that I
>> could.
>>
>
> The tcf_block_put_ext() part looks good to me. I will run
> some tests to make sure.
>

Hmm, there is some bug in tcf_unbind_filter() after this
merge, but that is not your fault. In net-next, we use
tp->chain after it is freed:

static inline void
 tcf_unbind_filter(struct tcf_proto *tp, struct tcf_result *r)
 {
+       struct Qdisc *q = tp->chain->block->q;
        unsigned long cl;

+       if (!q)
+               return;
        if ((cl = __cls_set_class(&r->class, 0)) != 0)
-               tp->q->ops->cl_ops->unbind_tcf(tp->q, cl);
+               q->ops->cl_ops->unbind_tcf(q, cl);
 }

I will provide a fix.

Just FYI.