netdev - Re: [PATCH iproute2-next] ipaddress: fix label matching

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: <m3y3eh4iya.fsf@luffy.cx>
Date:   Wed, 11 Jul 2018 22:26:53 +0200
From:   Vincent Bernat <vincent@...nat.im>
To:     Stephen Hemminger <stephen@...workplumber.org>
Cc:     netdev@...r.kernel.org, serhe.popovych@...il.com
Subject: Re: [PATCH iproute2-next] ipaddress: fix label matching

 ❦ 11 juillet 2018 13:03 -0700, Stephen Hemminger <stephen@...workplumber.org> :

>> Since 9516823051ce, "ip addr show label lo:1" doesn't work
>> anymore (doesn't show any address, despite a matching label).
>> Reverting to return 0 instead of -1 fix the issue.
>> 
>> However, the condition says: "if we filter by label [...] and the
>> label does NOT match the interface name". This makes little sense to
>> compare the label with the interface name. There is also a logic
>> around filter family being provided or not. The match against the
>> label is done by ifa_label_match_rta() in print_addrinfo() and
>> ipaddr_filter().
>> 
>> Just removing the condition makes "ip addr show" works as expected
>> with or without specifying a label, both when the label is matching
>> and not matching. It also works if we specify a label and the label is
>> the interface name. The flush operation also works as expected.
>> 
>> Fixes: 9516823051ce ("ipaddress: Improve print_linkinfo()")
>> Signed-off-by: Vincent Bernat <vincent@...nat.im>
>> ---
>>  ip/ipaddress.c | 5 -----
>>  1 file changed, 5 deletions(-)
>> 
>> diff --git a/ip/ipaddress.c b/ip/ipaddress.c
>> index 5009bfe6d2e3..20ef6724944e 100644
>> --- a/ip/ipaddress.c
>> +++ b/ip/ipaddress.c
>> @@ -837,11 +837,6 @@ int print_linkinfo(const struct sockaddr_nl *who,
>>  	if (!name)
>>  		return -1;
>>  
>> -	if (filter.label &&
>> -	    (!filter.family || filter.family == AF_PACKET) &&
>> -	    fnmatch(filter.label, name, 0))
>> -		return -1;
>> -
>>  	if (tb[IFLA_GROUP]) {
>>  		int group = rta_getattr_u32(tb[IFLA_GROUP]);
>> 
>
> If this is a regression, it should go to iproute2 not iproute2-next.
>
> Surprised by the solution since it is removing code that was there
> before the commit you referenced in Fixes.

Yes, but as I explain in the commit message, the condition does not make
sense for me: why would we match the label against the interface name?
This code exists since a long time.
-- 
The lunatic, the lover, and the poet,
Are of imagination all compact...
		-- Wm. Shakespeare, "A Midsummer Night's Dream"