From: Niklas Cassel <Niklas.Cassel@wdc.com>
To: "bugzilla-daemon@kernel.org" <bugzilla-daemon@kernel.org>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>
Subject: Re: [Bug 217914] scsi_eh_1 process high cpu after upgrading to 6.5
Date: Mon, 13 Nov 2023 11:53:37 +0000 [thread overview]
Message-ID: <ZVIONaawuLOojZin@x1-carbon> (raw)
In-Reply-To: <bug-217914-11613-kPEwzlBcn3@https.bugzilla.kernel.org/>
On Mon, Nov 13, 2023 at 03:30:57AM +0000, bugzilla-daemon@kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=217914
>
> Christian Kujau (kernel@nerdbynature.de) changed:
>
> What |Removed |Added
> ----------------------------------------------------------------------------
> CC| |kernel@nerdbynature.de
>
> --- Comment #7 from Christian Kujau (kernel@nerdbynature.de) ---
> Noticed the same here when upgrading from 6.1.0-13-amd64 to
> 6.5.0-0.deb12.1-amd64 (both Debian kernels) earlier this month:
>
> ============================================================================
> $ sar -f /var/log/sysstat/sa20231108
> CPU %user %nice %system %iowait %steal %idle
> [...]
> 18:30:03 all 1.03 0.00 0.45 0.04 0.00 98.47
> 18:40:01 all 1.07 0.00 0.52 0.05 0.00 98.36
> 18:50:01 all 1.07 0.00 0.53 0.04 0.00 98.37
> 19:00:01 all 1.35 0.00 0.69 0.08 0.00 97.88
> 19:10:04 all 1.09 0.00 0.52 0.07 0.00 98.31
> 19:20:02 all 1.14 0.00 0.51 0.05 0.00 98.30
> 19:30:04 all 1.62 0.00 0.65 0.08 0.00 97.65
> Average: all 1.06 0.00 0.50 0.06 0.00 98.38
>
> 19:32:27 LINUX RESTART (2 CPU)
>
> 19:40:03 CPU %user %nice %system %iowait %steal %idle
> 19:50:00 all 2.27 0.00 3.23 57.40 0.00 37.11
> 20:00:02 all 1.29 0.00 2.70 59.27 0.00 36.75
> 20:10:03 all 1.48 0.00 2.93 58.38 0.00 37.21
> 20:20:03 all 1.40 0.00 2.94 58.93 0.00 36.73
> 20:30:02 all 1.39 0.00 2.87 59.99 0.00 35.74
> 20:40:03 all 1.48 0.00 3.44 59.83 0.00 35.26
> 20:50:00 all 1.29 0.00 2.88 60.84 0.00 34.98
> 21:00:03 all 1.31 0.00 2.63 59.81 0.00 36.25
> 21:10:03 all 1.33 0.00 2.72 59.85 0.00 36.09
> 21:20:01 all 1.31 0.00 2.82 59.28 0.00 36.59
> 21:30:01 all 1.39 0.00 2.92 60.51 0.00 35.18
> 21:40:01 all 1.34 0.00 3.04 60.04 0.00 35.57
> 21:50:03 all 1.29 0.00 2.51 59.79 0.00 36.41
> 22:00:03 all 1.36 0.00 3.23 59.81 0.00 35.59
> 22:10:03 all 1.37 0.00 2.56 59.13 0.00 36.93
> 22:20:03 all 1.36 0.00 2.88 58.46 0.00 37.29
> 22:30:03 all 1.31 0.00 2.65 59.07 0.00 36.97
> 22:40:00 all 1.32 0.00 2.72 59.61 0.00 36.35
> 22:50:01 all 1.32 0.00 2.72 59.35 0.00 36.61
> 23:00:03 all 1.29 0.00 2.68 59.30 0.00 36.72
> 23:10:03 all 1.35 0.00 2.62 60.11 0.00 35.91
> 23:20:02 all 1.29 0.00 2.91 59.55 0.00 36.25
> 23:30:03 all 1.32 0.00 2.72 58.37 0.00 37.59
> 23:40:01 all 1.34 0.00 2.97 57.74 0.00 37.95
> 23:50:00 all 1.33 0.00 2.54 59.90 0.00 36.24
> Average: all 1.38 0.00 2.83 59.37 0.00 36.41
>
> $ last -n 3 reboot
> reboot system boot 6.5.0-0.deb12.1- Wed Nov 8 19:32 still running
> reboot system boot 6.1.0-13-amd64 Mon Oct 9 23:14 - 19:32 (29+21:17)
> reboot system boot 6.1.0-13-amd64 Mon Oct 9 22:39 - 23:14 (00:35)
> ============================================================================
>
> And top only shows a single [scsi_eh_2] thread using ~50% of CPU time,
> sometimes there's an events thread too.
>
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 336 root 20 0 0.0m 0.0m 0.0m D 50.0 0.0 53,37
> [scsi_eh_2]
> 3794126 root 20 0 0.0m 0.0m 0.0m I 25.7 0.0 0:06.27
> [kworker/0:0-events]
>
> This is a Debian/amd64 VM running on a VMware ESX host. There's a virtual CDROM
> drive, but nothing is attached here and I'm not using it, all:
>
>
> ============================================================================
> $ lsblk -d --scsi
> NAME HCTL TYPE VENDOR MODEL REV SERIAL
> TRAN
> sda 0:0:0:0 disk VMware Virtual disk 2.0
> sr0 2:0:0:0 rom NECVMWar VMware Virtual IDE CDROM Drive 1.00
> 10000000000000000001 ata
>
>
> $ dmesg -t | grep -i scsi
> Block layer SCSI generic (bsg) driver version 0.4 loaded (major 247)
> SCSI subsystem initialized
> VMware PVSCSI driver - version 1.0.7.0-k
> vmw_pvscsi: using 64bit dma
> vmw_pvscsi: max_id: 65
> vmw_pvscsi: setting ring_pages to 32
> vmw_pvscsi: enabling reqCallThreshold
> vmw_pvscsi: driver-based request coalescing enabled
> vmw_pvscsi: using MSI-X
> scsi host0: VMware PVSCSI storage adapter rev 2, req/cmp/msg rings: 32/32/1
> pages, cmd_per_lun=254
> vmw_pvscsi 0000:03:00.0: VMware PVSCSI rev 2 host #0
> scsi 0:0:0:0: Direct-Access VMware Virtual disk 2.0 PQ: 0 ANSI: 6
> sd 0:0:0:0: [sda] Attached SCSI disk
> sd 0:0:0:0: Attached scsi generic sg0 type 0
> scsi host1: ata_piix
> scsi host2: ata_piix
> scsi 2:0:0:0: CD-ROM NECVMWar VMware IDE CDR10 1.00 PQ: 0 ANSI: 5
> scsi 2:0:0:0: Attached scsi generic sg1 type 5
> sr 2:0:0:0: [sr0] scsi3-mmc drive: 1x/1x writer dvd-ram cd/rw xa/form2 cdda
> tray
> sr 2:0:0:0: Attached scsi CD-ROM sr0
> ============================================================================
>
> Will that "scsi: Do no try to probe for CDL on old drives" patch land in
> mainline or is this still under discussion?
It has been in mainline for a while:
2132df16f53b ("scsi: core: ata: Do no try to probe for CDL on old drives")
$ git tag --contains 2132df16f53b
v6.6
v6.6-rc4
v6.6-rc5
v6.6-rc6
v6.6-rc7
v6.6.1
v6.7-rc1
Kind regards,
Niklas
next prev parent reply other threads:[~2023-11-13 11:53 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-15 19:33 [Bug 217914] New: scsi_eh_1 process high cpu after upgrading to 6.5 bugzilla-daemon
2023-09-15 20:42 ` Bart Van Assche
2023-09-15 20:55 ` Laurence Oberman
2023-09-15 22:01 ` Niklas Cassel
2023-09-15 22:10 ` Damien Le Moal
2023-09-15 22:46 ` Damien Le Moal
2023-09-15 20:42 ` [Bug 217914] " bugzilla-daemon
2023-09-15 20:55 ` bugzilla-daemon
2023-09-15 21:06 ` bugzilla-daemon
2023-09-15 22:01 ` bugzilla-daemon
2023-09-15 22:10 ` bugzilla-daemon
2023-09-15 22:46 ` bugzilla-daemon
2023-11-13 3:30 ` bugzilla-daemon
2023-11-13 11:53 ` Niklas Cassel [this message]
2023-11-13 11:53 ` bugzilla-daemon
2023-11-13 12:02 ` bugzilla-daemon
2023-11-27 9:50 ` bugzilla-daemon
2024-11-21 9:32 ` bugzilla-daemon
2024-11-21 19:30 ` bugzilla-daemon
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZVIONaawuLOojZin@x1-carbon \
--to=niklas.cassel@wdc.com \
--cc=bugzilla-daemon@kernel.org \
--cc=linux-scsi@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.