[TLP 1.7 Beta] Call for testers

[linrunner]

Hi @ALL,

as with every upcoming release, I'm excited to get your help with test coverage!

Test Objectives

1. General regression testing (check if everything still works for you)

   • When changing the power source AC <-> BAT, do the tunables change accordingly (compare output of tlp-stat)?
   • Can the charge thresholds be changed ?
   • Does tlp recalibrate work? - ThinkPads only

2. Battery Care

   • Check if newly supported hardware is able to set start and stop charge threshold:
     • Apple Silicon Macbooks (M*) with MacOS 13.0 (or later) firmware and Asahi Linux kernel 6.6 (or later) - important: the firmware restricts the possible values to 75/80 and 95/100.
     • MSI laptops with kernel 6.3 or later - important: the kernel driver msi_ec only accepts individual models and BIOS versions, for the unsupported ones tlp-stat -b displays Plugin: generic / Supported features: none available
   • ThinkPads with coreboot: does tlp-stat -b show complete battery readings? See https://linrunner.de/tlp/faq/battery.html#are-thinkpads-with-coreboot-supported

3. Adaptive Backlight Modulation on AMD Vega or newer GPUs with kernel 6.9 or later

   • Configure AMDGPU_ABM_LEVEL_ON_BAT=3 and check with tlp-stat -g if the setting applies on battery = /sys/class/drm/card1/card1-eDP-1/amdgpu/panel_power_savings = 3 - on my AMD ThinkPad, it saves 0.5W at medium display brightness with a slight (subjective) loss of brightness, it has no effect at full or minimum display brightness.

4. General features

   • Are error, warning, notice and success messages shown in color?
   • Check the new Fish shell command completion

5. Any other of the new features that you like, check out the changelog

6. Bugfixes

   • xhci_hcd was added to RUNTIME_PM_DRIVER_DENYLIST defaults - do USB devices work properly on battery?

Beta Packages - Arch, Debian/Ubuntu and Fedora
See the download page.

What do I need from the testers?
After installing please start the beta with

sudo tlp start

I'm expecting your unabbreviated outputs of

sudo tlp-stat

on battery and AC (via Gist please).

Bug Reports
If something does not work as expected → please open an issue.

I look forward to your feedback on the beta :-)

References

• https://github.com/linrunner/TLP/blob/main/changelog
• https://github.com/linrunner/TLP/releases/tag/1.7.0-beta.1

[Mylinde]

Let's test.

• BAT https://gist.github.com/Mylinde/61d92a3a5b16c3833c27db51a0818c7a
• AC https://gist.github.com/Mylinde/5dd937813cc1deb747560081806dd079

Battery Care: START_CHARGE_THRESH_BAT0 not working, is set to 75 and has worked in latest release.

https://gist.github.com/Mylinde/1bcf27389a455df92989d05d35417c08

Adaptive Backlight Modulation working:

• BAT https://gist.github.com/Mylinde/5ebc4d6c058f76f8549f5a16071b07f8
• AC https://gist.github.com/Mylinde/3bb2a9e09a7afc7cbec97cb92132526f

[linrunner]

@Mylinde thanks for testing and your outputs. Looks perfect to me.

Battery Care: START_CHARGE_THRESH_BAT0 not working, is set to 75 and has worked in latest release.

Negative. Your memory must be betraying you. The kernel driver asus_wmi has always supported a stop threshold only.

[Mylinde]

... yes will be, 89-asus (also asus-wmi.c) tells you are right. Thanks so far, everythink works as aspected.

[dkwo]

All looks good on Apple silicon https://paste.c-net.org/BrieflyCrushed

[BarbUk]

I have some error during sudo tlp-stat:

❯ sudo tlp-stat  > /dev/null
Use of uninitialized value $bus in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 111.
Use of uninitialized value $dev in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 111.
Use of uninitialized value $desc in substitution (s///) at /usr/share/tlp/tlp-usblist line 112.
Use of uninitialized value $bus in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 115.
Use of uninitialized value $dev in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 115.
Use of uninitialized value $usbid in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 115.
Use of uninitialized value in concatenation (.) or string at /usr/share/tlp/tlp-usblist line 115.
Use of uninitialized value $_[0] in string at /usr/share/tlp/tlp-usblist line 61.

Output of tlp-stat:

• AC
• BAT

EDIT:
battery threshold are working

[linrunner]

@BarbUk please show

lsusb

[BarbUk]

lsusb

My keyboard has an EOL that cause the issue ?
Without my keyboard connected, tlp-stat show no error.

[linrunner]

@BarbUk unexpected linebreak in lsusb's output

Bus 001 Device 006: ID 04d9:a293 Holtek Semiconductor, Inc. OBINS
AnnePro2

Probably needs another plausibility check. Sigh. Stay tuned.

[linrunner]

@BarbUk fix commited.

Please replace /usr/share/tlp/tlp-usblist with https://raw.githubusercontent.com/linrunner/TLP/main/tlp-usblist
and re-test - with keyboard connected of course.

[H-Sachse]

Here's my log files for the Lenovo Edge 11 with AMD Athlon(tm) II Neo K325 Dual-Core Processor
Debian Bookworm with kernel 6.1.106-3

A/C connected:
tlp-stat-AC.log
A/C disconnected:
tlp-stat-BAT.log
calibration in discharge phase:
tlp-stat-cal-dischg.log
calibration, discharge done:
tlp-stat-cal-chg.log
manually fixed charge thresholds for full charge:
tlp-stat-cal-chg2.log

Here's my issues:

• Charge thresholds cannot be set individually both values are the same (e.g. with tlp setcharge 80 85, both values will be 80, except if I use setcharge 96 100 it goes to 256/100 see tlp-stat-cal-chg2.log)
• charge thresholds are not set correctly by "recalibrate" for a full charge (see tlp-stat-cal-chg.log)
• battery was stuck in "idle" after discharge from calibration, had to re-plug the A/C to resolve, but could not reproduce.
• battery state toggles between idle and charging when almost full, but this might have to do with the battery itself and/or the driver

[linrunner]

@H-Sachse thanks for testing and reporting. Your data looks good.

Regarding you issues:

• Charge thresholds cannot be set individually both values are the same (e.g. with tlp setcharge 80 85, both values will be 80, except if I use setcharge 96 100 it goes to 256/100 see tlp-stat-cal-chg2.log)
• charge thresholds are not set correctly by "recalibrate" for a full charge (see tlp-stat-cal-chg.log)

False threshold readings are symptomatic for the Edge series. Nonetheless the configured values are properly written and work afaik -> FAQ.

Lenovo's firmware for the Edge and E series behaves differently to that of all other series. There is no solution other than ignoring the false readings. Just be happy that this beautiful old treasure is alive and kicking.

• battery was stuck in "idle" after discharge from calibration, had to re-plug the A/C to resolve, but could not reproduce.
• battery state toggles between idle and charging when almost full, but this might have to do with the battery itself and/or the driver

Possibly EC firmware issue too.

[tisyang]

Test looks fine.

log ac&bat : https://gist.github.com/tisyang/477bd6e053f9a6f19a9870bc09dc1a77

[BarbUk]

@BarbUk fix commited.

Please replace /usr/share/tlp/tlp-usblist with https://raw.githubusercontent.com/linrunner/TLP/main/tlp-usblist
and re-test - with keyboard connected of course.

Thanks for the fix.

No error with the new commit:

Bus 001 Device 001 ID 1d6b:0002 control = on,   autosuspend_delay_ms =    0 -- Linux Foundation 2.0 root hub (hub)
Bus 001 Device 002 ID 8087:0024 control = auto, autosuspend_delay_ms =    0 -- Intel Corp. Integrated Rate Matching Hub (hub)
Bus 001 Device 003 ID 0cf3:e300 control = auto, autosuspend_delay_ms = 2000 -- Qualcomm Atheros Communications QCA61x4 Bluetooth 4.0 (btusb)
Bus 001 Device 005 ID 04d9:a293 control = on,   autosuspend_delay_ms = 2000 -- Holtek Semiconductor, Inc. OBINS (usbhid)
Bus 002 Device 001 ID 1d6b:0002 control = on,   autosuspend_delay_ms =    0 -- Linux Foundation 2.0 root hub (hub)
Bus 002 Device 002 ID 8087:0024 control = auto, autosuspend_delay_ms =    0 -- Intel Corp. Integrated Rate Matching Hub (hub)
Bus 002 Device 003 ID 04f2:b217 control = auto, autosuspend_delay_ms = 2000 -- Chicony Electronics Co., Ltd Lenovo Integrated Camera (0.3MP) (uvcvideo)
Bus 002 Device 004 ID 046d:c092 control = on,   autosuspend_delay_ms = 2000 -- Logitech, Inc. G102/G203 LIGHTSYNC Gaming Mouse (usbhid)
Bus 003 Device 001 ID 1d6b:0002 control = on,   autosuspend_delay_ms =    0 -- Linux Foundation 2.0 root hub (hub)
Bus 004 Device 001 ID 1d6b:0003 control = on,   autosuspend_delay_ms =    0 -- Linux Foundation 3.0 root hub (hub)

[linrunner]

@BarbUk great. Thanks for the confirmation.

[linrunner]

Beta packages for Fedora 40 are now available. See the download page.

[linrunner]

ThinkPad T14 Gen 4 AMD

• AC
• BAT

[linrunner]

I have just released Beta 2:

• https://github.com/linrunner/TLP/blob/main/changelog
• https://github.com/linrunner/TLP/releases/tag/1.7.0-beta.2
• https://download.linrunner.de/packages/

I would be so grateful if you could retest and post the corresponding outputs. Thank you so much! I really appreciate your help with this.

The packages for the Fedora Beta Repo will take some time.

[BarbUk]

Thanks for the work and the new release 🚀

• AC
• BAT

[linrunner]

The beta test is complete, and TLP 1.7.0 is now available. Have fun!

A huge thank you to all of our wonderful contributors, testers, and bug reporters! 👍