Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:5 as permitted sender) client-ip=2620:137:e000::3:5;
Date:   Fri, 13 Oct 2023 21:01:02 -0300
From:   Paulo Alcantara <pc@manguebit.com>
To:     matoro <matoro_mailinglist_kernel@matoro.tk>,
        "Dr. Bernd Feige" <bernd.feige@uniklinik-freiburg.de>
CC:     tom@talpey.com, smfrench@gmail.com, paul@darkrain42.org,
        linux-cifs@vger.kernel.org, bagasdotme@gmail.com,
        regressions@lists.linux.dev, ronniesahlberg@gmail.com,
        nspmangalore@gmail.com, brian.pardy@gmail.com,
        bharathsm@microsoft.com, linux-kernel@vger.kernel.org
Subject: =?US-ASCII?Q?Re=3A_Possible_bug_report=3A_kern?= =?US-ASCII?Q?el_6=2E5=2E0/6=2E5=2E1_high_load_when?= =?US-ASCII?Q?_CIFS_share_is_mounted_=28cifsd?= =?US-ASCII?Q?-cfid-laundromat_in=22D=22_state=29?=
In-Reply-To: <E1F307C7-9B1E-40F6-860B-6050856E8395@manguebit.com>
References: <CAO+kfxTwOvaxYV0ZRESxZB-4LHsF9b_VBjAKahhwUm5a1_c4ug@mail.gmail.com> <ZPfPfyIoVxw5L6El@debian.me> <CAO+kfxQgXOsx6u+xLKGJe0KDiFsRAGstSpnrwxjQF6udgz5HFQ@mail.gmail.com> <CAO+kfxTvA6N=i+jGf0XbSyqf85i=q+vR6R9d_42OWfM2sWWXaA@mail.gmail.com> <CAH2r5mtUedfLSv81Z-Yb3_=AbD_QpT3tVbU1PRzMTituaw7bgA@mail.gmail.com> <CAH2r5mt6YzapEKDo=hQ64yvBn7=jwMmY1c85NOABKcMPKPp3KA@mail.gmail.com> <CAO+kfxQtOKoKdb+LtMeFxgu8VXa73nbmTPSfscbdwjUXM7ME_A@mail.gmail.com> <CAH2r5msNf9WDHrBZSi5FhHDSewSNxMAuXTetMJDnoNh3CF_oMA@mail.gmail.com> <a895f860-11fa-e6d9-d042-a32bd08f9e9d@talpey.com> <CAH2r5mszCxPtdURenMVgeVDX5zc8knumH=ASXyUufPa7SxbJBw@mail.gmail.com> <ZRN9MtBqYnT6oX60@vaarsuvius> <85d538fec5a086acf62d5a803056586a6c00e4bd.camel@uniklinik-freiburg.de> <83d00d50bc628a85db71adb440d8afb5@matoro.tk> <E1F307C7-9B1E-40F6-860B-6050856E8395@manguebit.com>
Message-ID: <CA6E0F87-65FD-4672-AA0C-A761E5006B7D@manguebit.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

You probably want these two as well

https://git=2Esamba=2Eorg/?p=3Dsfrench/cifs-2=2E6=2Egit;a=3Dcommit;h=3D2da=
338ff752a2789470d733111a5241f30026675

https://git=2Esamba=2Eorg/?p=3Dsfrench/cifs-2=2E6=2Egit;a=3Dcommit;h=3D3b8=
bb3171571f92eda863e5f78b063604c61f72a

as directory leases isn't supported in SMB1, so no waste of system resourc=
es by having those kthreads running=2E

On 13 October 2023 20:52:11 GMT-03:00, Paulo Alcantara <pc@manguebit=2Ecom=
> wrote:
>Could you please try two commits[1][2] from for-next?
>
>[1] https://git=2Esamba=2Eorg/?p=3Dsfrench/cifs-2=2E6=2Egit;a=3Dcommit;h=
=3De95f3f74465072c2545d8e65a3c3a96e37129cf8
>[2] https://git=2Esamba=2Eorg/?p=3Dsfrench/cifs-2=2E6=2Egit;a=3Dcommit;h=
=3D81ba10959970d15c388bf29866b01b62f387e6a3
>
>On 13 October 2023 20:19:37 GMT-03:00, matoro <matoro_mailinglist_kernel@=
matoro=2Etk> wrote:
>>On 2023-10-05 05:55, Dr=2E Bernd Feige wrote:
>>> Am Dienstag, dem 26=2E09=2E2023 um 17:54 -0700 schrieb Paul Aurich:
>>>> Perhaps the laundromat thread should be using msleep_interruptible()?
>>>>=20
>>>> Using an interruptible sleep appears to prevent the thread from
>>>> contributing
>>>> to the load average, and has the happy side-effect of removing the
>>>> up-to-1s delay
>>>> when tearing down the tcon (since a7c01fa93ae, kthread_stop() will
>>>> return
>>>> early triggered by kthread_stop)=2E
>>>=20
>>> Sorry for chiming in so late - I'm also on gentoo (kernel 6=2E5=2E5-
>>> gentoo), but as a client of Windows AD=2E
>>>=20
>>> Just want to emphasize that using uninterruptible sleep has not just
>>> unhappy but devastating side-effects=2E
>>>=20
>>> I have 8 processors and 16 cifsd-cfid-laundromat processes, so
>>> /proc/loadavg reports a load average of 16 on a totally idle system=2E
>>>=20
>>> This means that load-balancing software will never start additional
>>> tasks on this system - "make -l" but also any other load-dependent
>>> system=2E Just reducing the number of cifsd-cfid-laundromat processes
>>> does not fix this - even a single one makes loadavg report a wrong
>>> result for load balancing=2E
>>>=20
>>> So, if cifsd-cfid-laundromat must really be uninterruptible, the only
>>> solution would be to change the way loadavg is computed by the kernel
>>> to exclude uninterruptible but sleeping processes=2E But must it be
>>> uninterruptible?
>>>=20
>>> Thanks and best regards,
>>> Bernd
>>
>>This is a huge problem here as well, as a client to Samba using SMB1 (fo=
r Unix extensions)=2E
>>
>>For others encountering this problem, I was able to work around it with =
the following snippet:
>>
>>diff --git a/fs/smb/client/cached_dir=2Ec b/fs/smb/client/cached_dir=2Ec
>>index 2d5e9a9d5b8b=2E=2Efc2caccb597a 100644
>>--- a/fs/smb/client/cached_dir=2Ec
>>+++ b/fs/smb/client/cached_dir=2Ec
>>@@ -576,7 +576,7 @@ cifs_cfids_laundromat_thread(void *p)
>>        struct list_head entry;
>>
>>        while (!kthread_should_stop()) {
>>-               ssleep(1);
>>+               msleep_interruptible(1000);
>>                INIT_LIST_HEAD(&entry);
>>                if (kthread_should_stop())
>>                        return 0;