linux-kernel - Re: Peanut gallery 2c

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>] [day] [month] [year] [list]

Message-ID: <cijxbws4ngjmfj7chgz7pyjelc4lpz2uqqo655sc6rlyoai6og@ysy65pdg56fz>
Date: Tue, 12 Aug 2025 12:26:47 -0400
From: Kent Overstreet <kent.overstreet@...ux.dev>
To: Johannes Thumshirn <jth@...nel.org>
Cc: James Lawrence <jalexanderlawrence@...il.com>, tytso@....edu, 
	admin@...inas.su, gbcox@....us, josef@...icpanda.com, 
	linux-bcachefs@...r.kernel.org, linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org, 
	list-bcachefs@...lthompson.net, malte.schroeder@...ip.de, sashal@...nel.org, 
	torvalds@...ux-foundation.org
Subject: Re: Peanut gallery 2c

On Tue, Aug 12, 2025 at 11:24:18AM +0200, Johannes Thumshirn wrote:
> On Mon, Aug 11, 2025 at 12:08:12PM -0400, Kent Overstreet wrote:
> > I think bcachefs has been a bit of a kick in the pants for them, they've
> > taken some stuff directly from bcachefs - e.g. I believe they took the
> > basic design of raid5 v2, the stripes tree, directly from bcachefs. A
> > btrfs engineer asked me and I explained the design at a conference some
> > years back [...]
> 
> Just one single mail and I will not reply to any follow ups on this. But I
> feel the need to clarify things as people have approached me. Yes Kent and I
> have talked about RAID. This has been at LSFMM 2024 in Salt Lake City. 
> 
> I can't remember if we talked about stripe trees, I only remember us talking
> about erasure coding, which Kent has done for bcachefs and I was planning to
> do for btrfs as well.
> 
> But the 1st presentations I did on the idea of the stripe tree have been Lund
> Linux Conference and Plumbers in 2022 [1]. This is easy to look up. In the
> beginning the RAID stripe tree didn't have anything to do with RAID5 and the
> write hole at all (this is just a nice side effect) but with doing RAID on
> zoned block devices and ZONE APPEND writing.
> 
> [1] https://lpc.events/event/16/contributions/1235/attachments/1111/2132/BTRFS%20RAID-DP.pdf

Thanks for the correction :)