08:58 <mwalle> Kubu_work: yes, that's the "workaround". I don't think that's really a workaround but the normal way to do it

08:58 <mwalle> Kubu_work: grep system-power-controller arch/arm64/boot/dts/ti

09:02 <mwalle> Kubu_work: and yeah, you ran into the same problem as we did. I've tried pushing TI to not hardcode the power-off sequence into PM firmware, but so far, no success.

09:23 <Kubu_work> mwalle: I may have another issue on the custom board, with your patch applied I got "ti-sci 44083000.system-controller: Mbox timedout in resp"

09:23 <Kubu_work> Thanks for your reply :)

09:23 <Kubu_work> Anyway I'm still using a kernel 6.1.y, I didn't tried ti-6.12.y yet (nor even latest upstream)

09:29 <mwalle> Kubu_work: maybe another power-off handler is invoked, there is a chain of them

09:35 <Kubu_work> I don't know, Actually I'm not 100% sure that the power-off PSCI method is really disabled

09:36 <mwalle> Kubu_work: No I don't think it is disabled, it's just another power-off handler. I had some dt patches so you can specify a priority, but it worked for my case, so I dropped them

09:49 <Kubu_work> mwalle: Ok, I'll have a look maybe later next week