netdev - Re: [PATCH net-next 2/5] eth: fbnic: add initial PHC 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] [day] [month] [year] [list]

Message-ID: <20240911152539.4030764b@kernel.org>
Date: Wed, 11 Sep 2024 15:25:39 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Andrew Lunn <andrew@...n.ch>
Cc: Vadim Fedorenko <vadim.fedorenko@...ux.dev>, David Ahern
 <dsahern@...nel.org>, Paolo Abeni <pabeni@...hat.com>, "David S. Miller"
 <davem@...emloft.net>, Alexander Duyck <alexanderduyck@...com>,
 netdev@...r.kernel.org
Subject: Re: [PATCH net-next 2/5] eth: fbnic: add initial PHC support

On Wed, 11 Sep 2024 22:45:11 +0200 Andrew Lunn wrote:
> > > That's good question. Do we need another set of helpers just because of 
> > > names? Obviously, the internals will be the same sequence magic.  
> > 
> > Good question. To be clear we want a seq lock that goes away on 64b
> > since what it protects is accessed on the fast path (potentially per
> > packet). We could s/u64_stats/u64_seq/ the existing helpers. But that
> > sounds like a lot for a single user. Dunno..  
> 
> It does sound like a lot of a single user.
> 
> And what is the likelihood of this device ever being used on a 32 bit
> system? It is a server class NIC. Are there still 32 bit servers in
> use?
> 
> Maybe "depends on 64BIT" with a good commit message why?

Will this not apply all new MMIO devices? They will either be high
enough class to be only used on 64b systems, or built into a 64b SoC.

No strong feelings either way, but I think we'd need a better defined
guidance on when "depends on 64BIT" is acceptable and when developers
have to go thru the pain of using u64_stats_*.