Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3731.600.4\))
Subject: Re: [PATCH v7 00/13] fold per-CPU vmstats remotely
From:   Andrew Theurer <atheurer@redhat.com>
In-Reply-To: <ZD/NbAU8gHILW/6w@tpad>
Date:   Wed, 19 Apr 2023 08:44:23 -0500
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Christoph Lameter <cl@linux.com>,
        Aaron Tomlin <atomlin@atomlin.com>,
        Frederic Weisbecker <frederic@kernel.org>,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        Russell King <linux@armlinux.org.uk>,
        Huacai Chen <chenhuacai@kernel.org>,
        Heiko Carstens <hca@linux.ibm.com>, x86@kernel.org,
        Vlastimil Babka <vbabka@suse.cz>,
        Michal Hocko <mhocko@suse.com>
Content-Transfer-Encoding: quoted-printable
Message-Id: <98E08059-6607-43FF-84E3-BFF3FF193D09@redhat.com>
References: <20230320180332.102837832@redhat.com>
 <20230418150200.027528c155853fea8e4f58b2@linux-foundation.org>
 <ZD/NAaa5TVcL7Mxm@tpad> <ZD/NbAU8gHILW/6w@tpad>
To:     Marcelo Tosatti <mtosatti@redhat.com>
Precedence: bulk


> On Apr 19, 2023, at 6:15 AM, Marcelo Tosatti <mtosatti@redhat.com> =
wrote:
>=20
> On Wed, Apr 19, 2023 at 08:14:09AM -0300, Marcelo Tosatti wrote:
>> On Tue, Apr 18, 2023 at 03:02:00PM -0700, Andrew Morton wrote:
>>> On Mon, 20 Mar 2023 15:03:32 -0300 Marcelo Tosatti =
<mtosatti@redhat.com> wrote:
>>>=20
>>>> This patch series addresses the following two problems:
>>>>=20
>>>> 1. A customer provided evidence indicating that a process
>>>>   was stalled in direct reclaim:
>>>>=20
>>>> ...
>>>>=20
>>>> 2. With a task that busy loops on a given CPU,
>>>>    the kworker interruption to execute vmstat_update
>>>>    is undesired and may exceed latency thresholds
>>>>    for certain applications.
>>>>=20
>>>=20
>>> I don't think I'll be sending this upstream in the next merge =
window.=20
>>> Because it isn't clear that the added complexity in vmstat handling =
is
>>> justified.
>>=20
>> =46rom my POV this is an incorrect statement (that the complexity in
>> vmstat handling is not justified).
>>=20
>> Andrew, this is the 3rd attempt to fix this problem:
>>=20
>> First try:  =
https://lore.kernel.org/lkml/20220127173037.318440631@fedora.localdomain/
>>=20
>> Second try: =
https://patchew.org/linux/20230105125218.031928326@redhat.com/
>>=20
>> Third try: syncing vmstats remotely from vmstat_shepherd (this
>> patchset).
>>=20
>> And also, can you please explain: what is so complicated about the
>> vmstat handling? cmpxchg has been around and is used all over the
>> kernel, and nobody considers "excessively complicated".
>>=20
>>> - Michal's request for more clarity on the end-user requirements
>>>  seems reasonable.
>>=20
>> And i explained to Michal in great detail where the end-user=20
>> requirements come from. For virtualized workloads, there are two
>> types of use-cases:
>>=20
>> 1) For example, for the MAC scheduler processing must occur every =
1ms,
>> and a certain amount of computation takes place (and must finish =
before
>> the next 1ms timeframe). A > 50us latency spike as observed by =
cyclictest
>> is considered a "failure".
>>=20
>> I showed him a 7us trace caused by, and explained that will extend to =
>
>> 50us in the case of virtualized vCPU.
>>=20
>> 2) PLCs. These workloads will also suffer > 50us latency spikes
>> which is undesirable.
>>=20
>> Can you please explain what additional clarity is required?
>>=20
>> RH's performance team, for example, has been performing packet
>> latency tests and waiting for this issue to be fixed for about 2
>> years now.
>>=20
>> Andrew Theurer, can you please explain what problem is the =
vmstat_work
>> interruption causing in your testing?
>=20
> +CC Andrew.

Nearly every telco we work with for 5G RAN is demanding <20 usec CPU =
latency as measured by cyclictest & oslat.  We cannot achieve under 20 =
usec with the vmstats interruption.

-Andrew=