netdev - Re: [RFC PATCH net-next 0/6] devlink: Add device metric support

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:   Sun, 23 Aug 2020 10:04:34 +0300
From:   Ido Schimmel <idosch@...sch.org>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     David Ahern <dsahern@...il.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        netdev@...r.kernel.org, davem@...emloft.net, jiri@...dia.com,
        amcohen@...dia.com, danieller@...dia.com, mlxsw@...dia.com,
        roopa@...dia.com, andrew@...n.ch, vivien.didelot@...il.com,
        tariqt@...dia.com, ayal@...dia.com, mkubecek@...e.cz,
        Ido Schimmel <idosch@...dia.com>
Subject: Re: [RFC PATCH net-next 0/6] devlink: Add device metric support

On Sat, Aug 22, 2020 at 09:27:39AM -0700, Jakub Kicinski wrote:
> On Fri, 21 Aug 2020 19:18:37 -0600 David Ahern wrote:
> > On 8/21/20 6:37 PM, Jakub Kicinski wrote:
> > >>> # cat /proc/net/tls_stat     
> > >>
> > >> I do not agree with adding files under /proc/net for this.  
> > > 
> > > Yeah it's not the best, with higher LoC a better solution should be
> > > within reach.  
> > 
> > The duplicity here is mind-boggling. Tls stats from hardware is on par
> > with Ido's *example* of vxlan stats from an ASIC. You agree that
> > /proc/net files are wrong,
> 
> I didn't say /proc/net was wrong, I'm just trying to be agreeable.
> Maybe I need to improve my command of the English language.
> 
> AFAIK /proc/net is where protocol stats are.
> 
> > but you did it anyway and now you want the
> > next person to solve the problem you did not want to tackle but have
> > strong opinions on.
> 
> I have no need and interest in vxlan stats.
> 
> > Ido has a history of thinking through problems and solutions in a proper
> > Linux Way. netlink is the right API, and devlink was created for
> > 'device' stuff versus 'netdev' stuff. Hence, I agree with this
> > *framework* for extracting asic stats.
> 
> You seem to focus on less relevant points. I primarily care about the
> statistics being defined and identified by Linux, not every vendor for
> themselves.

Trying to understand how we can move this forward. The issue is with the
specific VXLAN metrics, but you generally agree with the need for the
framework? See my two other examples: Cache counters and algorithmic
TCAM counters.

> No question about Ido's ability and contributions, but then again 
> (from the cover letter):
> 
> > This a joint work [...] during a two-day company hackathon.

The *implementation* was done during the hackathon. The *design* was
done before. I sent to the team three design proposals (CLI / netlink
API / device drivers API) before we landed on this. Started thinking
about this idea a few months ago and the hackathon was simply a good
opportunity to implement and showcase it.