ROG Ally does not sleep successfully with 20250904.1000 main update

[eskimododo]

Your system information

• Steam client version:
• SteamOS version: 3.8
• Opted into Steam client beta?: [Yes]
• Opted into SteamOS beta?: [Yes]
• Have you checked for updates in Settings > System?: [Yes]

Please describe your issue in as much detail as possible:

on the latest main updates 20250904.1000, when ROG Ally is put to sleep, the screen turns off and activity indicator flashes as usual, but after a minute the unit wakes itself up without intervention. ROG ally is unable to fully enter sleep.

Additional info: I have simpledeckyTDP installed with Rog Ally Extreme Power Save (aka MCU Powersave) option enabled and up to date firmware installed.

Steps for reproducing this issue:

1. Set unit to sleep or press power button for sleep
2. Wait
3. ROG Ally will wake up on its own.

[matte-schwartz]

sigh... the fun never stops with this device. can you confirm whether or not it happens if you leave the unit unplugged?

[eskimododo]

sigh... the fun never stops with this device. can you confirm whether or not it happens if you leave the unit unplugged?

I just tried, sleep works fine and as intended on battery.. but as soon as I plugged in, the same symptoms happen..

[matte-schwartz]

Okay, thanks for confirming. I can reproduce the same issue then and have a temproary workaround that you can try. it will require you to run a couple of commands in a terminal:

1. If you haven't already set a sudo password, do so with passwd
2. run this command: echo disabled | sudo tee /sys/bus/usb/devices/1-3/power/wakeup

Try suspending the device while on a charger after doing this. You will have to repeat the command if you reboot the device but for now, confirmation about whether or not it helps would be appreciated and then we can go from there.

[eskimododo]

Okay, thanks for confirming. I can reproduce the same issue then and have a temproary workaround that you can try. it will require you to run a couple of commands in a terminal:

1. If you haven't already set a sudo password, do so with passwd
2. run this command: echo disabled | sudo tee /sys/bus/usb/devices/1-3/power/wakeup

Try suspending the device while on a charger after doing this. You will have to repeat the command if you reboot the device but for now, confirmation about whether or not it helps would be appreciated and then we can go from there.

Thank you for the quick response and workaround.. I switched to desktop mode, ran the commands in terminal and switched back to game mode to sleep the device.. I can confirm that the workaround works while on charger.. have been waiting for 10min and the device does not wake up with the charger still plugged in..

[matte-schwartz]

Okay cool, glad that's working for you. It seems to be passing my own sleep tests as well, so I will move forward with sending a patch for this to Valve.

[matte-schwartz]

pm-resume-debug-oldpatchset-good.txt pm-resume-debug-currentpatchset-bad.txt

just going to leave this here for future reference if a kernel developer investigates the root cause of the issue, pm_debug messages from the old "good" patchset (that did not have this wakeup issue) and the current "bad" patchset (without any patches/workarounds.)

[ionis789]

I’m experiencing the same issue. The terminal command resolved it, thank you. However, a hotfix is definitely needed, as I can’t run this command every time I reboot my ROG Ally.

[matte-schwartz]

sorry, i never heard back so not sure what the status is on this :/

for now you can use a udev rule which should make that command persistent:

/etc/udev/rules.d/99-ally-wakeup.rules and then in that file put:

# Disable USB device wakeup for device 1-3
SUBSYSTEM=="usb", KERNEL=="1-3", ATTR{power/wakeup}="disabled"

you can whitelist it in /etc/atomic-update.conf.d/

[Ashnobe]

Thank you very much for solution, Matte!:) just as a side note, even after the latest update (October 3rd 25), the issue still persists.

[matte-schwartz]

I'm on vacation atm but when I have some time I'll pull request the patch I mentioned above.

[matte-schwartz]

fix is queued for the next kernel release

[firrae]

I am running a ROG Ally X with the latest Main release and am getting the same issue. I have upgraded the SSD to be a Samsung 990 Pro 4TB and have an SD Card in my device but it's not formatted for Steam to use. I submitted a system report, my Steam username is also Firrae.

I have tried the following:

• Plugged in with ethernet and power, press power button, it goes to sleep and then almost immediately wakes back up
• Plugged in with ethernet and power, power menu sleep option, it goes to sleep and then almost immediately wakes back up
• Power plugged in, press power button, it goes to sleep and then almost immediately wakes back up
• Power plugged in, power menu sleep option, it goes to sleep and then almost immediately wakes back up
• On battery, press power button, it goes to sleep and then a few seconds later wakes back up
• On battery, power menu sleep option, it goes to sleep and then few seconds later wakes back up

I also tried the above situations with and without the SD card to the same outcomes. If you want me to test specific cases and submit reports please just let me know.

[matte-schwartz]

@firrae can you share the output of fwupdmgr get-devices | grep -EA2 "Microcontroller"

[matte-schwartz]

Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: pci_pm_suspend(): hcd_pci_suspend returns -16
Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: dpm_run_callback(): pci_pm_suspend returns -16
Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: failed to suspend async: error -16
Oct 10 01:10:52 steamdeck kernel: PM: Some devices failed to suspend, or early wake event detected

seems like the likely suspect in your system report, just not sure why I'm not able to reproduce that yet while on battery. do you have any other external USB devices plugged in while on battery power?

[firrae]

@firrae can you share the output of fwupdmgr get-devices | grep -EA2 "Microcontroller"

Sorry for the delay in response and it being an image, currently travelling.

Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: pci_pm_suspend(): hcd_pci_suspend returns -16
Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: dpm_run_callback(): pci_pm_suspend returns -16
Oct 10 01:10:52 steamdeck kernel: xhci_hcd 0000:64:00.3: PM: failed to suspend async: error -16
Oct 10 01:10:52 steamdeck kernel: PM: Some devices failed to suspend, or early wake event detected

seems like the likely suspect in your system report, just not sure why I'm not able to reproduce that yet while on battery. do you have any other external USB devices plugged in while on battery power?

I have nothing extra plugged in, just the SD card and even removing that still ends up with the issue where it wakes back up almost immediately. The only other mod to my ROG Ally X that I've made is I put in Hall Effect joy sticks from HandheldDIY, but not sure that would cause it to wake from sleep. The sleep issue is also on the desktop, though hibernate seems to work but I can't always reliably wake it back up with just the handheld.

[firrae]

I just tried to submit another system report but it failed @matte-schwartz, while trying a few times it did randomly go to sleep properly but I cannot reliably recreate it. I have saved it to the desktop if there's an alternative way to submit it and you think it's worth looking at.

[matte-schwartz]

any difference if you set LED Mode to Disabled in HueSync? the other thing I see a lot of in your report is [asus.py:67:_set_hardware_color] INFO: set_asus_color: device not ready, I'm wondering if this could be related since I am still not seeing this issue on the same MCU firmware version and BIOS version.

[firrae]

@matte-schwartz that initially seemed to help a lot, I disabled it and turned off "Enabled LED Control" and it seemed much more reliable getting to sleep. Though after a restart it seems sadly back to waking back up almost immediately every time. It still randomly will properly go into sleep but it is rare and I cannot find a way to replicate it.

I will be interested to test the next kernel release to see if it helps the issue.

[firrae]

Sadly even with the most recent OS release (main) this is still an issue for me. I uploaded another system report if it helps.

[minipol89-create]

i think u dont need to search anymore. I was on Main branch bios bios 312 mcu 315 and sleep mode worked perfectly. I just upgraded the oem sticks to the handhelddiy hall effekt/tmr sticks and got the issue too instantly after upgrading the sticks... so i think we found the Problem, the failure came from the sticks with the instant wake after sleep.... i dont know why the sticks can do this but i think thats the problem..

[john270]

Replying to https://github.com/ValveSoftware/SteamOS/issues/2119#issuecomment-3315131551

Thanks. This seems to have fixed the issue for me (Asus Rog Ally z1 extreme, beta steamOS)

[firrae]

i think u dont need to search anymore. I was on Main branch bios bios 312 mcu 315 and sleep mode worked perfectly. I just upgraded the oem sticks to the handhelddiy hall effekt/tmr sticks and got the issue too instantly after upgrading the sticks... so i think we found the Problem, the failure came from the sticks with the instant wake after sleep.... i dont know why the sticks can do this but i think thats the problem..

Mine are the same sticks, so that's certainly interesting

[matehuos]

Just to weight in. The problem is present in both CachyOS handheld edition and on the SteamOS at least on the Z1E. The fix did work though.

[mario-64]

/etc/atomic-update.conf.d/

FWIW, this does not fix the issue on my ROG Ally running the latest stable release.