netdev - Re: [PATCH bpf-next 11/11] selftests/bpf: add checks on extack messages for eBPF hw offload tests

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: <45c685c2-cd64-6f88-439c-63d62aa618a2@gmail.com>
Date:   Mon, 15 Jan 2018 18:02:17 -0700
From:   David Ahern <dsahern@...il.com>
To:     Jakub Kicinski <jakub.kicinski@...ronome.com>
Cc:     davem@...emloft.net, daniel@...earbox.net,
        alexei.starovoitov@...il.com, netdev@...r.kernel.org,
        oss-drivers@...ronome.com, jiri@...nulli.us,
        john.fastabend@...il.com, jhs@...atatu.com, gerlitz.or@...il.com,
        aring@...atatu.com, xiyou.wangcong@...il.com,
        Quentin Monnet <quentin.monnet@...ronome.com>
Subject: Re: [PATCH bpf-next 11/11] selftests/bpf: add checks on extack
 messages for eBPF hw offload tests

On 1/15/18 5:55 PM, Jakub Kicinski wrote:
> On Mon, 15 Jan 2018 17:49:07 -0700, David Ahern wrote:
>> On 1/15/18 5:30 PM, Jakub Kicinski wrote:
>>> A new flag ("--skip-extack") is added to the Python script so as to
>>> allow to skip these checks. This is because extack messages cannot be
>>> displayed by tc and ip if tools from iproute2 package were compiled
>>> without libmnl, but we do not want this to prevent users to run the
>>> other checks.  
>>
>> That is unfortunate. Did you consider auto-detecting support? e.g., run
>> a command that is known to fail and return a message. For example,
>>
>> $ ip ro add 1.1.1.1/32 dev eth0 onlink
>> Error: Invalid flags for nexthop - PERVASIVE and ONLINK can not be set.
> 
> That's a good idea.  And then if it fails would you suggest to skip the
> test entirely or just skip extack checks?  With the --skip-extack flag
> the hope was that people would be inclined to install libmnl to save
> themselves the hassle of setting the flag each time.  If it's detected
> automatically there is no hassle/nudge to install libmnl..
> 

My inclination is to avoid unnecessary options and adapt to the
environment. ie., just skip the checks. You can still nudge users with a
warning that it appears ip or tc does not have extack support.