17:57 <calebccff> konradybcio: how would one check that cpuidle is working properly? would there be errors during boot?

17:58 <calebccff> i've been running the rb3gen2 with TF-A for a while now without the pm-domain changes that Chrome platforms have and I haven't seen any probe errors at least

18:25 <konradybcio> calebccff: 7280 was updated to get OSI mode support

18:25 <konradybcio> calebccff: as for testing, in this case debugfs of power domain summary and /sys/bus/cpu/devices/cpuN/cpuidle/ should be telling, and of course dmesg errors if any

18:26 <calebccff> konradybcio: ah so it doesnt have this issue

19:34 <bamse> konradybcio: iiuc the power-domain link might give improved latency when it comes to cpuidle... but the big difference would be power-collapse numbers; and to restore the system

19:34 <bamse> althoguh the latency would depend on the psci arguments

19:37 <konradybcio> bamse the link is not an option, the psci genpd driver marks the domain always on with PSCI in PC mode

20:46 <bamse> konradybcio: what i meant was to further define what we're looking for in the "i tested and it was fine" statement

20:46 <bamse> konradybcio: we have a bunch of OSI-mode devices runnning with that power-domain and they work "just fine" as well

22:04 <alexeymin_> hi, do you know if these faults can be fixed? https://paste.sr.ht/~minlexx/50290bd6003aa839e2b2a5d9c0a8740c8b3f92f4

22:06 <alexeymin_> iova=0x9d400000 hints at the framebuffer@9d400000