<div data-ntes="ntes_mail_body_root" style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial"><div id="spnEditorContent"><p style="margin: 0in; font-family: Calibri; font-size: 11pt;">Thank you for your reply.</p><p style="margin: 0in; font-family: Calibri; font-size: 11pt;">The same problem mentioned by Renmingshuai in this
email refers to the fact that in the DHCPv6 packet processing flow, the
function dhcp6_no_relay adds the stack variable state->tag to
daemon->dhcp_comf->netid->list.</p><p style="margin:0in;font-family:"Microsoft YaHei";font-size:9.0pt"><a href="https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017676.html">https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017676.html</a></p><p style="margin: 0in; font-family: Calibri; font-size: 11pt;">This is the same as the behavior of the function
dhcp_reply in the DHCPv4 packet processing flow, which
adds stack variable netid to
daemon->dhcp_comf->netid->list. When dnsmasq receives a SIGHUP signal
to reload dhcp hostsfile, it will release daemon ->dhcp_comf ->netid in
the function clear_dynamic_conf. In this case, daemon ->dhcp_comf
->netid in stack space is freed as a pointer. Therefore, both DHCPv6
and DHCPv4 have bad-free issues.</p><p style="margin: 0in; font-family: Calibri; font-size: 11pt;">This is also the question I asked in this email,
question one:</p><p style="margin: 0;">
</p><p style="margin:0in;font-family:Calibri;font-size:11.0pt"><a href="https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017730.html">https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017730.html</a></p></div><pre><br>At 2024-09-21 20:57:03, "Geert Stappers" <stappers@stappers.nl> wrote:
>On Sat, Sep 21, 2024 at 07:08:45PM +0800, ºúÒåÕé wrote:
>> I'm renmingshuai's successor,
>
>Pleased to meet you!
>
>
>> a beginner with dnsmasq, and this e-mail is follow-up to
>>
>> https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017664.html
>>
>
>And that email thread is currently stalled
>at https://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/2024q3/017676.html
>
>I have bounced that message to you, to enable you to do an in thread
>follow-up. (mailserver logentry: 2024-09-21T12:41:31.187603+00:00
>postfix/smtp[2522404]: D0C182000E: to=<huyizhen2024@163.com>,
>relay=163mx01.mxmail.netease.com[103.129.252.43]:25, delay=9.3,
>delays=0.08/0.02/4.9/4.3, dsn=2.0.0, status=sent (250 Mail OK queued as
>gzga-mx-mtada-g2-6,_____wDnV2r0vu5mj3s4AA--.27372S3 1726922491) )
>
>
>> Question 1: Why does the dhcp_reply function add the stack variable
>> netid to daemon->dhcp_conf->netid->list, which is a global variable?
>> ...
>> Question 2: Why does the dhcp_reply function use two netid linked lists:
>> netid and tagif_netid? Can't we just use one?
>> ...
>> Question 3: Are there any common test cases for dnsmasq?
>> ...
>
>
>One question got a response with matching Subject line. My estimation
>is that the other two questions wouldn't get a response. My advice is
>to retry in separate emails, each with matching subject line.
>
>
>
>Looking forward to further co-operation, regards
>Geert Stappers
>--
>Silence is hard to parse
</pre></div>