Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 22 Feb 2021 13:31:18 -0800 (PST)
From:   Hugh Dickins <hughd@google.com>
To:     Jens Axboe <axboe@kernel.dk>
cc:     Andrew Morton <akpm@linux-foundation.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] percpu_counter: increase batch count
In-Reply-To: <919b3aa5-7332-0817-3b70-4f20ae6410a8@kernel.dk>
Message-ID: <alpine.LSU.2.11.2102221309570.4728@eggly.anvils>
References: <0bf90e07-8758-b238-b3f3-a330725a1134@kernel.dk> <20210218151644.df430e4f77f763b7db2a004f@linux-foundation.org> <919b3aa5-7332-0817-3b70-4f20ae6410a8@kernel.dk>
User-Agent: Alpine 2.11 (LSU 23 2013-08-11)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk

On Thu, 18 Feb 2021, Jens Axboe wrote:
> On 2/18/21 4:16 PM, Andrew Morton wrote:
> > On Thu, 18 Feb 2021 14:36:31 -0700 Jens Axboe <axboe@kernel.dk> wrote:
> > 
> >> Currently we cap the batch count at max(32, 2*nr_online_cpus), which these
> >> days is kind of silly as systems have gotten much bigger than in 2009 when
> >> this heuristic was introduced.
> >>
> >> Bump it to capping it at 256 instead. This has a noticeable improvement
> >> for certain io_uring workloads, as io_uring tracks per-task inflight count
> >> using percpu counters.

I want to quibble with the word "capping" here, it's misleading -
but I'm sorry I cannot think of the right word.

The macro is max() not min(): you're making an improvement for
certain io_uring workloads on machines with 1 to 15 cpus, right?
Does "bigger than in 2009" apply to those?

Though, io_uring could as well use percpu_counter_add_batch() instead?

(Yeah, this has nothing to do with me really, but I was looking at
percpu_counter_compare() just now, for tmpfs reasons, so took more
interest.  Not objecting to a change, but the wording leaves me
wondering if the patch does what you think - or, not for the
first time, I'm confused.)

Hugh

> >>
> > 
> > It will also make percpu_counter_read() and
> > percpu_counter_read_positive() more inaccurate than at present.  Any
> > effects from this will take a while to discover.
> 
> It will, but the value of 32 is very low, especially when you are potentially
> doing millions of these per second. So I do think it should track the times
> a bit.
> 
> > But yes, worth trying - I'll add it to the post-rc1 pile.
> 
> Thanks!
> 
> -- 
> Jens Axboe