search

Dasharo / dasharo-issues

The Dasharo issue tracker
https://dasharo.com/
25 stars 0 forks source link

Wake by USB keyboard not working #1044

Open filipleple opened 2 months ago

filipleple commented 2 months ago

Component

Dasharo firmware

Device

MSI PRO Z690-A DDR4

Dasharo version

TBD

Dasharo Tools Suite version

No response

Test case ID

No response

Brief summary

Wake by USB keyboard not working

How reproducible

100%

How to reproduce

Suspend the OS, try to wake it using a USB keyboard

Expected behavior

The platform should wake back up

Actual behavior

No reaction to any keyboard button, can only wake with the power button

Screenshots

No response

Additional context

Solutions you've tried

No response

JanPrusinowski commented 1 month ago

Same issue on MSI PRO Z690-A DDR4 (RPL 14th) on Ubuntu. DUT can be resumed by power button but using keyboard has no effect.

SergiiDmytruk commented 1 month ago

@filipleple, there are 4 firmware variants for this board and while Intel TXT shouldn't make a difference, payload can. Was this tested with EDK2 or SeaBIOS payload on OptiPlex?

filipleple commented 1 month ago

@SergiiDmytruk EDK2 with TXT. Good point, I'll make it clear in this and other recent issues today.

So far we've only released and tested the EDK2 with TXT variant, but that's due to change soon.

krystian-hebel commented 4 weeks ago

@filipleple have you tried waking up using physical keyboard or just PiKVM? Possibly duplicate of https://github.com/Dasharo/dasharo-issues/issues/649

filipleple commented 4 weeks ago

I'm preetty sure I tried USB keyboard as well, having guessed PiKVM could be a factor

philipandag commented 1 week ago

Not waking from suspend on v0.1.0 using a physical USB keyboard. Our PiKVM setup fot this platform does not simulate a keyboard nor a mouse and physical devices have to be used.

Pressing the power button (via osfv_cli rte) does not wake up the platform netiher, but at least produces some output to the serial. The platform "wakes up" fine when the screen is off after a couple hours of inactivity, but according to OS settings it is not suspended, only the screen is turned off. optiplex-suspend-pwr-btn.log That's the log produced when suspending and waking up using FWTS, which works fine: optiplex-suspend-fwts.log The log produced by the FWTS also is perfect with no errors nor warnings. fwts.log

lis 12 08:33:39 3mdeb systemd-sleep[2378]: Entering sleep state 'suspend'...
lis 12 08:34:36 3mdeb systemd-sleep[2378]: System returned from sleep state.
philipandag commented 1 week ago

I think that when the platform is suspended the USB devices are not even powered. All the LEDs, like caps-lock/num-lock/mouse sensor diode are turned off.

I have checked it using a multimeter and when the platform is on there is exactly 5V between the VCC and GND on the USB type A connectors but there is no voltage when the platform is suspended.

miczyg1 commented 1 day ago

I think that when the platform is suspended the USB devices are not even powered. All the LEDs, like caps-lock/num-lock/mouse sensor diode are turned off.

That's normal. The LEDs are always turned off during suspend.

I have checked it using a multimeter and when the platform is on there is exactly 5V between the VCC and GND on the USB type A connectors but there is no voltage when the platform is suspended.

Not sure if the voltage is expected on the ports without any devices connected. You would have to measure the voltage on ports with devices, which were connected before suspending.

miczyg1 commented 1 day ago

I have checked the voltage with a power meter connected in the middle of USB keyboard and the OptiPlex. It shows 5V all the time and USB wakes up the platform. However, coreboot is stuck at NMI sources enabled. So the USb wake works, but firmware has a bug and does not finish the resume process. Removing the OptiPlex label and creating a separate issue