01:17 <digetx> m4t: you need to use the latest grate-kernel to get proper interrupt trigger

01:17 <m4t> hmm

01:17 <digetx> please apply https://dpaste.com/8G4CHPVBT

01:18 <m4t> apply that to the OLD kernel (2021/05/20) ?

01:18 <m4t> i cannot use usb audio on the latest so i just rebuilt the old one

01:19 <digetx> you can use old, but you'll get interrupt storm on trigger

01:20 <digetx> it shouldn't trigger by default

01:20 <m4t> is that sufficient to test, then?

01:22 <digetx> at first we need to fix interrupt, then we can check it

01:22 <digetx> you can use old kernel for that too

01:22 <m4t> i'll just rebuild the latest and unplug usb audio for a minute. do i still need both patches with the latest git?

01:23 <digetx> yes, both needed

01:25 <m4t> ok

01:26 <digetx> it actually could be easier to test using older kernel since it has warning messages that are turned into debug messages now, but you can enable them manually if needed

01:27 <digetx> https://github.com/grate-driver/linux/commit/22f1a97558fabf1a47b6b5bd450b2f6767f9bcf4

01:30 <m4t> ok

01:32 <digetx> if interrupt will work properly at boot, then use this to trigger it:

01:32 <digetx> echo 50000 > /sys/devices/soc0/7000d000.i2c/i2c-0/0-004c/hwmon/hwmon4/temp2_min

01:32 <m4t> i think it idles above 50c

01:33 <m4t> nope

01:33 <m4t> https://imgur.com/a/KAWOUca will just 'openssl speed' or something

01:34 <m4t> what should i expect if it triggers the interrupt?

01:34 <m4t> something in dmesg?

01:35 <digetx> /proc/interrupts will bump and you will get message in dmesg if you'll enable those debug messages

01:35 <m4t> gotcha

01:36 <m4t> yeah i enabled them

01:36 <m4t> will let you know in a minute. rebooting

01:37 <m4t> not correct sysfs path. /sys/devices/soc0/7000d000.i2c/i2c-1/1-004c/hwmon/hwmon2/temp2_min

01:38 <digetx> find /sys | grep hwmon

01:39 <m4t> seems to be working: https://paste.ee/p/2R9hK#qlLFW2rpZcyLUstNInA3HFMdqfTzgTzn

01:40 <digetx> nice

01:40 <m4t> yep

01:40 <digetx> I'll add yours t-b, thank you

01:40 <m4t> cool

01:40 <m4t> going to downgrade so i can keep listening to music heh

01:43 <m4t> oh hrm, temp2_min not temp2_max :P

01:43 <m4t> so it should've triggered < 50c

01:43 <digetx> it did

01:43 <m4t> nct1008_2: +47.2°C (low = +42.0°C, high = +60.0°C)

01:44 <digetx> nct1008_2: +45.8°C (low = +50.0°C, high = +60.0°C) ALARM (LOW)

01:44 <m4t> yes

01:45 <m4t> just the whole openssl speed thing was unnecessary i think

01:45 <m4t> :D

01:46 <digetx> :)

01:50 <digetx> won't hurt to double check the whole pinmux, pgwipeout was assuring that it is correct, but seems not entirely

01:51 <m4t> lots of pins to check

01:52 <m4t> i did run into the mmc thing on the initial kernel inclusion

01:52 <m4t> in hindsight i wish i would've tested that beforehand instead of waiting for it to be merged

01:55 <digetx> it also could be difficult to notice incorrect pinmux, it may result in odd occasional failures