Dear Greg,
Thank you for your quick response.
Am 22.08.24 um 01:31 schrieb Greg KH:
On Wed, Aug 21, 2024 at 11:32:04PM +0200, Paul Menzel wrote:
On the Intel Kaby Lake laptop Dell XPS 13 9360 with Debian sid/unstable and
*powertop* 2.15-3, connecting a USB-C adapter like Dell DA300 or LMP USB-C
mini Dock (P/N 15954) [1] and connecting only an Ethernet cable (module
r8152 is used), the adapter gets very hot, and according to PowerTOP it uses
over 5 Watts – almost more as the laptop idling.
$ lsusb # Dell DA300
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 002: ID 0cf3:e300 Qualcomm Atheros Communications QCA61x4 Bluetooth 4.0
Bus 001 Device 003: ID 04f3:2234 Elan Microelectronics Corp. Touchscreen
Bus 001 Device 004: ID 0c45:670c Microdia Integrated Webcam HD
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 002: ID 2109:2820 VIA Labs, Inc. VL820 Hub
Bus 003 Device 003: ID 06c4:c412 Bizlink International Corp. DELL DA300
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 004 Device 002: ID 2109:0820 VIA Labs, Inc. VL820 Hub
Bus 004 Device 003: ID 0bda:8153 Realtek Semiconductor Corp. RTL8153 Gigabit Ethernet Adapter
With `LANG= sudo powertop --auto-tune` it stays high.
PowerTOP:
```
The battery reports a discharge rate of 6.01 W
The energy consumed was 146 J
The estimated remaining time is 3 hours, 51 minutes
Summary: 384.6 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and
8.5% CPU use
Power est. Usage Events/s Category Description
5.94 W 0.0% Device Display backlight
5.23 W 100.0% Device USB device: USB Optical Mouse (Logitech)
4.62 W 66.1% Device USB device: USB 10/100/1000 LAN (Realtek)
205 mW 100.0% Device USB device: Fujitsu Keyboard (Fujitsu)
14.1 mW 13.5 ms/s 0.9 kWork intel_atomic_commit_work
```
At another time:
```
The battery reports a discharge rate of 10.5 W
The energy consumed was 235 J
The estimated remaining time is 2 hours, 20 minutes
Summary: 395.8 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and 23.8% CPU use
Power est. Usage Events/s Category Description
7.13 W 100.0% Device USB device: USB 10/100/1000 LAN (Realtek)
3.92 W 15.8% Device Display backlight
320 mW 0.0 us/s 0.00 Process [PID 1349] /usr/bin/pipewire
63.6 mW 65.4 ms/s 0.5 Process [PID 4982] /usr/lib/thunderbird/thunderbird
24.9 mW 25.6 ms/s 6.7 Process [PID 37753] /usr/lib/firefox-nightly/firefox-bin -contentproc -isForBrowser -prefsLen 36793 -prefMapSize 265654 -jsInitLe
14.7 mW 15.1 ms/s 0.5 kWork intel_atomic_commit_work
```
The heat of the USB-C adapter might suggest, that it draws that much power.
What is your experience? Can you suggest something?
Buy a different adapter? That seems like something is really wrong with
it. Does other devices also suck that much power from that port on the
laptop?
It happens with two Dell DA300 adapters and two LMP USB-C mini Dock (P/N
15954, 12-22 Rev. 3):
$ lsusb # LMP USB-C
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 002: ID 0cf3:e300 Qualcomm Atheros Communications
QCA61x4 Bluetooth 4.0
Bus 001 Device 003: ID 04f3:2234 Elan Microelectronics Corp.
Touchscreen
Bus 001 Device 004: ID 0c45:670c Microdia Integrated Webcam HD
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 002: ID 2109:2817 VIA Labs, Inc. USB2.0 Hub
Bus 003 Device 003: ID 2109:2817 VIA Labs, Inc. USB2.0 Hub
Bus 003 Device 005: ID 2109:8817 VIA Labs, Inc. USB Billboard Device
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 004 Device 002: ID 2109:0817 VIA Labs, Inc. USB3.0 Hub
Bus 004 Device 003: ID 2109:0817 VIA Labs, Inc. USB3.0 Hub
Bus 004 Device 004: ID 058f:8468 Alcor Micro Corp. Mass Storage Device
Bus 004 Device 005: ID 0bda:8153 Realtek Semiconductor Corp.
RTL8153 Gigabit Ethernet Adapter
Both use a Realtek RTL8153 Ethernet adapter.
### LMP device
With *no* auto-tuning:
```
>> Bad VM writeback timeout
Bad NMI watchdog should be turned off
Bad Autosuspend for USB device USB Billboard Device
[VIA Labs, Inc. ]
Bad Autosuspend for USB device Mass Storage Device [Generic]
Bad Autosuspend for USB device Touchscreen [ELAN]
Bad Autosuspend for USB device USB 10/100/1000 LAN [Realtek]
Bad Runtime PM for PCI Device SK hynix PC300 NVMe Solid
State Drive 512GB
Bad Runtime PM for disk sda
Bad Runtime PM for disk sdb
Bad Runtime PM for PCI Device Intel Corporation Sunrise
Point-LP PCI Express Root Port #1
Bad Runtime PM for PCI Device Qualcomm Atheros QCA6174
802.11ac Wireless Network Adapter
Bad Runtime PM for PCI Device Intel Corporation DSL6340
Thunderbolt 3 Bridge [Alpine Ridge 2C 2015]
Bad Runtime PM for PCI Device Intel Corporation Sunrise
Point-LP LPC Controller
Bad Runtime PM for PCI Device Intel Corporation DSL6340
Thunderbolt 3 Bridge [Alpine Ridge 2C 2015]
```
```
The battery reports a discharge rate of 8.89 W
The energy consumed was 243 J
The estimated remaining time is 0 hours, 42 minutes
Summary: 572.3 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and
54.8% CPU use
Power est. Usage Events/s Category Description
6.00 W 5.9% Device Display backlight
2.33 W 100.0% Device USB device: USB
Billboard Device (VIA Labs, Inc. )
```
After `powertop --auto-tune`:
```
The battery reports a discharge rate of 8.58 W
The energy consumed was 213 J
The estimated remaining time is 0 hours, 39 minutes
Summary: 509.3 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and
34.3% CPU use
Power est. Usage Events/s Category Description
8.23 W 5.9% Device Display backlight
6.21 W 7938 pkts/s Device Network
interface: enx00e04ceabb21 (r8152)
```
But it also shows:
```
The battery reports a discharge rate of 9.54 W
The energy consumed was 189 J
The estimated remaining time is 0 hours, 33 minutes
Summary: 509.0 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and
44.0% CPU use
Power est. Usage Events/s Category Description
7.39 W 5.9% Device Display backlight
776 mW 12391 pkts/s Device Network
interface: enx00e04ceabb21 (r8152)
210 mW 357.9 ms/s 0.4 kWork
intel_atomic_commit_work
```
So measuring energy consumption, and attributing it to devices, also
seems to be hard and sometimes unreliable.
Therefore, I’d be interested, it what numbers to expect, and also, if
the developers have other methods and tools for measuring this.
Kind regards,
Paul
