netdev - Re: [PATCH v3 net-next 3/7] ipv6: Consolidate option cases in ip6_datagram_send

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]

Date:   Tue, 16 Apr 2019 21:00:40 -0700 (PDT)
From:   David Miller <davem@...emloft.net>
To:     tom@...bertland.com
Cc:     netdev@...r.kernel.org, tom@...ntonium.net
Subject: Re: [PATCH v3 net-next 3/7] ipv6: Consolidate option cases in
 ip6_datagram_send_ctl

From: Tom Herbert <tom@...bertland.com>
Date: Mon, 15 Apr 2019 10:52:16 -0700

>  		case IPV6_2292HOPOPTS:
>  		case IPV6_HOPOPTS:
> -			if (opt->hopopt || cmsg->cmsg_len < CMSG_LEN(sizeof(struct ipv6_opt_hdr))) {
> -				err = -EINVAL;
> -				goto exit_f;
> -			}

You're consolidated version of this code checks opt->hopopt after
things like ns_capable() thus changing the priorities of the error
conditions.  You code get -EPERM when previously -EINVAL would have
been seen.

I really don't like changes like this, it's so irritating double checking
all of the possible user visible side effects in all of these details.