00:59 <robclark> maybe linux needs some battery related throttling.. but AFAIK that doesn't exist

01:00 <robclark> I've not noticed problems but maybe even a full kernel build is fast enough to not run into problems (or at least I tend to reboot btwn kernel builds to test whatever change prompted the recompile so maybe that gives battery enough time to recover)

01:02 <robclark> the whole gig for the oryon core seems to be ability to scale from low on the power/perf curve to high (ie. whey they don't think they need in-order small cores to hit the lower points on the curve)

01:02 <robclark> so maybe this is more of a linux problem than an x1e problem ;-)

01:05 <robclark> x2e maybe gets more efficient from process shrink/etc, but that probably isn't addressing the root issue that linux cpufreq needs to understand more than just thermal limits

06:25 <DeepGaurav[m]> @_oftc_icecream95:matrix.org: Hey thanks a ton for the HID-BPF program for the Vivobook, it's working great.... (full message at <https://matrix.org/oftc/media/v1/media/download/AazA0hq8LbamsNtjUaUwuVfDveSqVFFUhiAfeW21Qyov5F-_GuoWLTcc15_FIwFF7wi8h8KuwPzO7k4FJDTbfORCeWlNJGygAG1hdHJpeC5vcmcvR3dtS3piaVl4WE9HbEdOd2ZLS1h6SkpW>)

09:07 <deathmist> robclark: maybe since X2E is apparently going to support dGPU we'll have unholy laptops with NVIDIA GPUs and large power bricks if nothing else.. just need the 100W+ PSU (with passthrough charging please, but unfortunately we're at the mercy of OEMs in the end) and it'd be able to fully unleash the CPU performance too

09:41 <icecream95> DeepGaurav[m]: I know, it is annoying. Especially since the fans run during suspend and suck up battery...

09:45 <icecream95> I did spend some time trying to work out what was happening in Windows, but never managed to get anywhere. The fans still work in safe mode and WinPE, so it's something relatively low-level, and kernel-debugging is annoying compared to userspace (and the EC resets the system if it you pause the kernel for too long)

09:47 <icecream95> The keyboard (which is still handled by the EC...) was easier to work out, since HID is a more standardised protocol.

09:48 <travmurav[m]> icecream95: the ec is 0x5b right?

09:57 <icecream95> travmurav[m]: Both 0x5b and 0x76 seem to respond to commands... oh hey that's funny the keyboard backlight is on again, did it reboot (or just reset the keyboard)?

09:58 <travmurav[m]> icecream95: I'm wondering if this is 0x76

09:59 <alexVinarskis[m]> On Zenbook i suspect EC is on 0x5b. When attempting to dump it, keyboard backlight went on (instead of flashing) and keyboard stop working. So seems to be the one.

09:59 <travmurav[m]> datasheet for that adc says the address is in the 0x70 to 0x7e range depending on the strapping resistor

09:59 <travmurav[m]> which allows it to be 0x76

09:59 <travmurav[m]> there is a chip id register 0xb2 -- 0x1a

10:00 <icecream95> Oh yup, I think it was poking 0x5b that made the EC reset

10:00 <travmurav[m]> reading 0xb2 should give 0x1a that is

10:00 <travmurav[m]> I guess i could just try measuring the mainboard...

10:05 <icecream95> `sudo i2cget -y 2 0x5b 23` causes the EC reset for me

10:07 <travmurav[m]> sad can't seem to measure it in-board but it's hard to probe tiny chip/passives

10:08 <alexVinarskis[m]> reading 0x5b 23 on zenbook also freaks it out.

10:13 <travmurav[m]> meh no I think I was wrong

10:13 <travmurav[m]> the ads on my board is supposed to be 0x7e so I guess it's not connected to the SoC but to something else (i.e. to the ec?)

10:17 <DeepGaurav[m]> icecream95: I was randomly sending commands to 0x5a, like in the hid-bpf program trying to see if I hit something.... (full message at <https://matrix.org/oftc/media/v1/media/download/AVA2s5o5eoHdrl2y3otcmIw5opDTh8Si1QVF7bzw8RYVBjupJvfvh8-A2ZMmhNelPdVoCM_xXuPKXF2-eQ-HoklCeWlaagmQAG1hdHJpeC5vcmcvRGh3VkRiSkFLdUFJV1BFd1FnZG5MWHF5>)

10:27 <alexVinarskis[m]> hmm on Zenbook that doesn't work. could you share exact command you used? potential use case is the Fn keys, on Zenbook F11 disables/enabled touchpad (for whatever reason)

10:53 <DeepGaurav[m]> alexVinarskis: sure, you can run this in chrome devtools on windows., can do similar thing on Linux with hid-bpf plugin that icecream95 wrote.

10:53 <DeepGaurav[m]> https://pastebin.com/JfUN4p4u

11:06 <alexVinarskis[m]> ahhh got it, thanks! i was trying to write to EC on 0x5B :D

11:06 <alexVinarskis[m]> will try adding it into icecream95's solution, already adapting for zenbook.

13:03 <alexVinarskis[m]> @_oftc_icecream95: tried your plugin, with minor changes [1] got backlight control working on Asus Zenbook A14! Fn lock, brightness does not work, will need to catch magic numbers under windows like you suggested. Deep Gauravunforunately seems touchpad enable/disable also does not work..

13:03 <alexVinarskis[m]> [1] https://github.com/alexVinarskis/udev-hid-bpf/tree/zenbook