[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250915195633.96236cecebd8777243a770bc@linux-foundation.org>
Date: Mon, 15 Sep 2025 19:56:33 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Suren Baghdasaryan <surenb@...gle.com>
Cc: kent.overstreet@...ux.dev, vbabka@...e.cz, hannes@...xchg.org,
usamaarif642@...il.com, rientjes@...gle.com, roman.gushchin@...ux.dev,
harry.yoo@...cle.com, shakeel.butt@...ux.dev, 00107082@....com,
pyyjason@...il.com, pasha.tatashin@...een.com, souravpanda@...gle.com,
linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/1] alloc_tag: mark inaccurate allocation counters
in /proc/allocinfo output
On Mon, 15 Sep 2025 19:48:14 -0700 Suren Baghdasaryan <surenb@...gle.com> wrote:
>
> > Perhaps we can tell people what accurate:no actually means. It is a
> > rather disturbing thing to see! How worried should our users be about
> > it?
>
> Right. How about adding a section like this:
>
> Supported markers in v2:
> accurate:no
> Absolute values of the counters in this line are not
> accurate because of the failure to allocate storage required
> to track some of the allocations made at this location.
> Deltas in these counters are accurate, therefore counters
> can be used to track allocation size and count changes.
>
>
> If this looks good,
looks awesome ;)
> could you fold it into the existing patch or
> should I respin?
A little fixlet would be preferred (by me, at least).
Powered by blists - more mailing lists