linux-kernel - Re: Sysfs one-value-per-file (was Re: [PATCH] vmstat: don't auto expand the sysfs files)

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: Thu, 28 Dec 2023 15:50:24 -0500
From: Pasha Tatashin <pasha.tatashin@...een.com>
To: David Rientjes <rientjes@...gle.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>, 
	Linus Torvalds <torvalds@...ux-foundation.org>, 
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>, rafael@...nel.org, surenb@...gle.com, 
	linux-kernel@...r.kernel.org, linux-mm@...ck.org, souravpanda@...gle.com
Subject: Re: Sysfs one-value-per-file (was Re: [PATCH] vmstat: don't auto
 expand the sysfs files)

On Thu, Dec 28, 2023 at 3:43 PM David Rientjes <rientjes@...gle.com> wrote:
>
> On Thu, 28 Dec 2023, Pasha Tatashin wrote:
>
> > > > But for existing files and conventions, I think we should settle it as
> > > > "keep doing what you've been doing for 13+ years" and don't force this
> > > > argument every time a kernel developer wants to just add one more stat.
> > >
> > > Absolutely.  Let's do what makes most sense.  For new things, one value
> > > per file.  For stats which logically group with other existing stats,
> > > do whatever the existing other stats are currently doing.
> >
> > The intent of "[PATCH] vmstat: don't auto expand the sysfs files" is
> > to do exactly this: keep the current fields in
> > /sys/devices/system/node/node*/vmstat as-is, but prevent future fields
> > added to node_stat, numa_stat, zone_stat from showing in vmstat.
> >
>
> This is the opposite of what Andrew and I wrote above.  There should be no
> need to start preventing future stats from being grouped together in
> vmstat if it logically makes sense.
>
> NAK.

Hm, I must have misunderstood what Andrew said.

I am ok with this patch being NAK, as long as the IOMMU observability
patch series and Per-Page Metadata patch can be merged upstream, and
extend vmstat without a NAK from GregKH for expanding a sysfs vmstat
file.

Pasha