Bug#816176: linux-image-4.3.0-0.bpo.1-amd64: xhci_hub_control causes abnormally high CPU usage when no USB devices attached

2021-05-06 Thread Yves-Alexis Perez
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On Thu, 2021-05-06 at 20:11 +0200, Moritz Mühlenhoff wrote:
> Is that still an issue with current kernels?

Hi Moritz, I don't think so. It's been a while since that original bug report,
but I don't think I've recently seem some high CPU usage related to the usb
drive (I can't believe I have that laptop since 6y already :)

Regards,
- -- 
Yves-Alexis
-BEGIN PGP SIGNATURE-

iQEzBAEBCAAdFiEE8vi34Qgfo83x35gF3rYcyPpXRFsFAmCUOEIACgkQ3rYcyPpX
RFtlLwf+N2xMDGSm+11cYH51sXF5jLry9R6/QaIz50WyGAy+JtPyzTFJ/+TLaI9r
n6qCUhFul0WJME8c5IQPEq5PP8r0rQkgmkczUmcF+tIAkUFd2T4g0tWbEqp+MJqc
GeUhwCSgRQ+41frz1BKsmUjTsLfLCgQxVR1KuttFQM8rkVPbYDWnd7WdKtr5Y+DC
QQUA5IIiOuU4OM9sttkiYu/aEE2nL93xXKA8WOjWjCluIIRhe6mNg+bqhiIxmTYm
IBpHkR17Bj+hdypprBR2hOT3+0uQtIc9c03FqF5SqGzhIkg8M5zZRoN81vmP+q1z
rHF1kubxQYEp5ic1wKGYIWEDkcEWvg==
=b2Et
-END PGP SIGNATURE-



Bug#816176: linux-image-4.3.0-0.bpo.1-amd64: xhci_hub_control causes abnormally high CPU usage when no USB devices attached

2021-05-06 Thread Moritz Mühlenhoff
Am Tue, Sep 25, 2018 at 08:19:13PM +0200 schrieb Yves-Alexis Perez:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
> 
> On Sun, 28 Feb 2016 12:19:43 +0100 Anders Nylander 
> wrote:
> 
> > Following the removal of of the USB device, I noticed very high CPU usage
> > being caused by kworker and ksoftirqd.
> > Using perf to log 10s of data and then reading the report, I noticed that
> > xhci_hcd/xhci_hub_control was generating a LOT of work for the kworker
> > (~42%).
> 
> I'm experiencing a very similar behavior on my ThinkPad X250 attached to a
> dock. It doesn't happen every time I resume, but quite often these days (on
> 4.18.8-1). Unplugging an USB3 disk plugged to the dock fixed the problem for
> me, and replugging it worked as well, so maybe there's some kind of loop I
> managed to break with that.
> 
> I also noticed https://bugzilla.redhat.com/show_bug.cgi?id=1325488 on the
> exact same setup, but there's no other info there.
> 
> I'll try to dig upstream, but unfortunately it's not that easily reproducible
> here.

Is that still an issue with current kernels?

Cheers,
Moritz



Bug#816176: linux-image-4.3.0-0.bpo.1-amd64: xhci_hub_control causes abnormally high CPU usage when no USB devices attached

2018-09-25 Thread Yves-Alexis Perez
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On Sun, 28 Feb 2016 12:19:43 +0100 Anders Nylander 
wrote:

> Following the removal of of the USB device, I noticed very high CPU usage
> being caused by kworker and ksoftirqd.
> Using perf to log 10s of data and then reading the report, I noticed that
> xhci_hcd/xhci_hub_control was generating a LOT of work for the kworker
> (~42%).

I'm experiencing a very similar behavior on my ThinkPad X250 attached to a
dock. It doesn't happen every time I resume, but quite often these days (on
4.18.8-1). Unplugging an USB3 disk plugged to the dock fixed the problem for
me, and replugging it worked as well, so maybe there's some kind of loop I
managed to break with that.

I also noticed https://bugzilla.redhat.com/show_bug.cgi?id=1325488 on the
exact same setup, but there's no other info there.

I'll try to dig upstream, but unfortunately it's not that easily reproducible
here.

Regards,
- -- 
Yves-Alexis
-BEGIN PGP SIGNATURE-

iQEzBAEBCAAdFiEE8vi34Qgfo83x35gF3rYcyPpXRFsFAluqfCEACgkQ3rYcyPpX
RFtQ6ggA36IjHB87mtIr2ciKDn2DY5afR2oWQ1zlmKXBRQBwVTSxxVFOJ/HP8Mhi
flNOTgChEDjs9+cZiWAPxfJkEpbJBkXxVrios0/eRaV5drpCh5oo3h40ln1E4H+M
6ouvDvKtOo8szqJppE2ATYIZ3uXknaN53b2bDAUvjuPHTp688vk47hTCL3rPjVTL
LQifiMHWQcqixodBZTh+uQxdDDjKrQ7bVY0gUjAEEkQq/2wgAm1RFtYtHAMfvfek
6gWmQ58mRsaEVTWPaLqSBhybEfvw8djlNo4sp2F/rT/0Hzp1OCIrzbQDUnGB9ttP
VF//XAvs8/5ChbhMPNXNRCkS6CFxBw==
=EgQP
-END PGP SIGNATURE-



Bug#816176: linux-image-4.3.0-0.bpo.1-amd64: xhci_hub_control causes abnormally high CPU usage when no USB devices attached

2016-02-28 Thread Anders Nylander
Package: src:linux
Version: 4.3.3-7~bpo8+1
Severity: important

Dear Maintainer,

>From a clean install of Debian Jessie 8.3 on a Lenovo Thinkpad E460 (Intel
Skylake gen.) I upgraded the kernel to 4.3 as made available in
jessie-backports to enable WiFi/Bluetooth.
Fast-forward to yesterday when I was preparing a LiveUSB drive, writing
approx. 4GB of data using cfdisk, mkfs, dd and sync.
Following the removal of of the USB device, I noticed very high CPU usage
being caused by kworker and ksoftirqd.
Using perf to log 10s of data and then reading the report, I noticed that
xhci_hcd/xhci_hub_control was generating a LOT of work for the kworker
(~42%).
Guessing that this had something to do with the USB ports, I tried plugging
in a USB device, and found that plugging in *any* kind of USB device caused
the CPU usage to drop to ~0%. Removing the USB device again caused the CPU
usage to go back to 100% (one core). The same result occured with all 3 USB
ports on this system, with ~0% CPU usage with any USB device connected,
100% CPU usage when no USB devices are connected.
The issue persists through reboots of the system.
The issue persists down to atleast init level 3, and is also present when
booting the system in "Recovery Mode".
Booting the system with the older 3.16 kernel causes the issue to
disappear, but then I lose WiFi/Bluetooth so that isn't an option.
I have not found a workaround to this problem besides leaving atleast one
USB device connected whenever I'm using the system.

-- Package-specific info:
** Version:
Linux version 4.3.0-0.bpo.1-amd64 (debian-kernel@lists.debian.org) (gcc
version 4.9.2 (Debian 4.9.2-10) ) #1 SMP Debian 4.3.3-7~bpo8+1 (2016-01-19)

** Command line:
BOOT_IMAGE=/boot/vmlinuz-4.3.0-0.bpo.1-amd64
root=UUID=8e8f14a8-cdb5-4d74-b834-369017c333fb ro initrd=/install/initrd.gz
quiet

** Tainted: WO (4608)
 * Taint on warning.
 * Out-of-tree module has been loaded.

** Kernel log:
[   10.267305]  [] ? __pm_runtime_resume+0x48/0x70
[   10.267309]  [] ? azx_pcm_open+0x1d4/0x350
[snd_hda_codec]
[   10.267312]  [] ? snd_pcm_open_substream+0x7f/0x140
[snd_pcm]
[   10.267314]  [] ? snd_pcm_open+0xd1/0x260 [snd_pcm]
[   10.267316]  [] ? wake_up_q+0x60/0x60
[   10.267318]  [] ? snd_pcm_playback_open+0x40/0x70
[snd_pcm]
[   10.267320]  [] ? chrdev_open+0xbc/0x190
[   10.267322]  [] ? cdev_put+0x20/0x20
[   10.267323]  [] ? do_dentry_open+0x1be/0x2d0
[   10.267325]  [] ? path_openat+0x20a/0x1560
[   10.267326]  [] ? __wake_up+0x34/0x50
[   10.267328]  [] ? do_filp_open+0x91/0x100
[   10.267330]  [] ? snd_card_file_remove+0x7d/0xe0 [snd]
[   10.267332]  [] ? do_sys_open+0x13a/0x230
[   10.267333]  [] ? system_call_fast_compare_end+0xc/0x6b
[   10.267334] ---[ end trace c26fb6e00392cc00 ]---
[   10.267335] [ cut here ]
[   10.267343] WARNING: CPU: 2 PID: 1555 at
/build/linux-kTc2b3/linux-4.3.3/drivers/gpu/drm/i915/intel_csr.c:462
skl_set_power_well+0x773/0x9d0 [i915]()
[   10.267343] CSR HTP Not fine
[   10.267344] Modules linked in: ctr ccm binfmt_misc bnep nfsd auth_rpcgss
oid_registry nfs_acl nfs lockd grace fscache sunrpc joydev uvcvideo
videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common btusb btrtl
videodev btbcm btintel media bluetooth snd_hda_codec_hdmi msr intel_rapl
iosf_mbi acpi_call(O) x86_pkg_temp_thermal intel_powerclamp coretemp
kvm_intel kvm crct10dif_pclmul crc32_pclmul jitterentropy_rng arc4
sha256_ssse3 sha256_generic hmac iTCO_wdt iTCO_vendor_support drbg
snd_hda_codec_conexant snd_hda_codec_generic ansi_cprng aesni_intel
aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd iwlmvm psmouse
snd_hda_intel mac80211 pcspkr serio_raw thinkpad_acpi snd_hda_codec iwlwifi
snd_hda_core battery ac nvram snd_hwdep mei_me mei snd_pcm cfg80211
snd_timer tpm_crb snd rfkill tpm soundcore
[   10.267429]  i2c_i801 shpchp evdev processor fuse parport_pc ppdev lp
parport autofs4 ext4 crc16 mbcache jbd2 sg sd_mod uas usb_storage
crc32c_intel e1000e ptp pps_core amdkfd ahci libahci amdgpu i915 xhci_pci
ttm i2c_algo_bit libata xhci_hcd drm_kms_helper usbcore scsi_mod drm
usb_common thermal wmi video button
[   10.267451] CPU: 2 PID: 1555 Comm: pulseaudio Tainted: GW  O
4.3.0-0.bpo.1-amd64 #1 Debian 4.3.3-7~bpo8+1
[   10.267452] Hardware name: LENOVO 20ETCTO1WW/20ETCTO1WW, BIOS R00ET37W
(1.12 ) 01/17/2016
[   10.267452]   ccc62a2a 812e1889
880237e87978
[   10.267454]  81074451 88023807 880237e879d0
300f
[   10.267455]  0001 8800360e7800 810744dc
a0343032
[   10.267457] Call Trace:
[   10.267458]  [] ? dump_stack+0x40/0x57
[   10.267462]  [] ? warn_slowpath_common+0x81/0xb0
[   10.267463]  [] ? warn_slowpath_fmt+0x5c/0x80
[   10.267471]  [] ? skl_set_power_well+0x773/0x9d0 [i915]
[   10.267478]  [] ? intel_display_power_get+0x9c/0xe0
[i915]
[   10.267481]  [] ? snd_hdac_display_power+0x74/0x100
[snd_hda_core]
[   10.267484]  [] ? hda_codec_runtime_resume+0x13/0x