Dorian Stoll

Suspended again for 10 hours, without S0ix, and it lost 20% of battery life in that time. So 2%/h without S0ix. The CPU still entered PC10 though. Given this result...

@woberts quite the opposite actually! `low_power_idle_system_residency_us` and `slp_s0_residency_usec` not being 0 means that the system reached S0ix successfully. And since both values are approximatey the same as `low_power_idle_cpu_residency_us`, it means...

Could you upload an acpidump (`sudo acpidump > acpidump.out`)? That might help later. For example, the acpidump could show us if IPTS uses a new firmware or one of the...

I can confirm that with your patch applied, the WiFi seems to reliably survive suspending the surface, even without the systemd-sleep script. However, it seems to break S0ix for me...

> @StollD Can you test with the parameter added to see if it will break S0ix states or not? Added the parameter to the kernel commandline and same result: If...

Same result as before: If I disable the sleep script, the wifi still survives the suspend, but the surface doesn't enter S0ix anymore. I can only get S0ix with the...

Yeah, sorry, your patches were never the reason for S0ix failing. Should have clarified that before. I get the same behaviour without these patches and the disabled sleep script.

> @StollD It would be interesting to know which module is preventing S0ix, could you please test if S0ix can be achieved by removing only the mwifiex_pcie module ? No,...

@sebanc That seems to work fine! Wifi doesn't crash after suspend, and the surface reaches S0ix, without having to unload the modules. I will do a longer test and report...

@sebanc Used it for a full day, with multiple 30 min suspends, one 2h suspend and one suspend overnight, same result: Stable Wifi + S0ix