#asahi-dev · 2023-04-18 04:07
marcanso at least *some* device PM is helping significantly already :)#asahi-dev · 2023-04-18 04:07
marcan1.9W in s2idle#asahi-dev · 2023-04-18 04:07
marcanyeah, 3.2W here#asahi-dev · 2023-04-18 04:06
marcanthat sounds more like it#asahi-dev · 2023-04-18 03:53
marcanthe MCC stuff should be at the *lower* level right now, if anything what macOS does is boost it when some cores are fast, but I never saw a difference in benchmarks so I didn't bother#asahi-dev · 2023-04-18 03:52
marcan(s2idle)#asahi-dev · 2023-04-18 03:51
marcan35 hours in sleep mode#asahi-dev · 2023-04-18 03:51
marcanchadmed: now that it's fully charged, I'm getting 22 hours screen off idle (with no peripherals connected) on t6000, so something is wrong if you only get 11h#asahi-dev · 2023-04-17 20:37
marcanso I guess the next step is to work out the DCS counters#asahi-dev · 2023-04-17 20:37
marcanit's possible this is just papering over the problem of some device hogging SoC fabric/DRAM bandwidth by forcing the entire thing to slow down#asahi-dev · 2023-04-17 20:36
marcanalso I'm not sure if this is the right approach, since I don't think macOS does any runtime-PM with this stuff and I think it still gets better power usage. so I won't upstream it for now#asahi-dev · 2023-04-17 20:08
marcan(if you have logging enabled)#asahi-dev · 2023-04-17 20:08
marcanI also changed macsmc-power to log power state on button press, so at least you can see what the latest data was after a suspend cycle (SMC updates once per second so usually you'll catch the previous state)#asahi-dev · 2023-04-17 20:07
marcanchadmed, jannau: enjoy :)#asahi-dev · 2023-04-17 20:07
marcanbut I really should get some sleep#asahi-dev · 2023-04-17 20:06
marcanI think I can start merging all that mess back into the bits branches :p#asahi-dev · 2023-04-17 20:06
marcanpushed to t602x/bringup#asahi-dev · 2023-04-17 20:03
marcan(ignore Tempty, I didn't let it sit there long enough and it's at 62% anyway)#asahi-dev · 2023-04-17 20:03
marcannot sure how much runtime that would be yet, but it's 66% of t602x so we're doing good!#asahi-dev · 2023-04-17 20:02
marcanah yes, it was just the charging :)#asahi-dev · 2023-04-17 20:02
marcan[ 272.730779] macsmc-power macsmc-power: In 0.000W Sys 2.206W 3V8 1.619W MPMU 0.537W SPMU 0.779W CLVR 0.364W CPU 1.475W Batt -2.217W 62% Tempty 467m#asahi-dev · 2023-04-17 20:00
marcancould just be an artifact of the battery charging though#asahi-dev · 2023-04-17 20:00
marcansince it's not the system PMUs or CPUs#asahi-dev · 2023-04-17 20:00
marcanthere's something dumb there using power#asahi-dev · 2023-04-17 20:00
marcan(this one's not charged, but system power usage is ~double)#asahi-dev · 2023-04-17 20:00
marcan[ 77.369426] macsmc-power macsmc-power: In 74.121W Sys 6.605W 3V8 1.722W MPMU 0.581W SPMU 0.799W CLVR 0.416W CPU 1.612W Batt 67.520W 56% Tfull 63m#asahi-dev · 2023-04-17 19:59
marcant6000 not doing as well, but at least I got CLVR pretty low#asahi-dev · 2023-04-17 19:17
marcanhey, it's getting better :p#asahi-dev · 2023-04-17 19:17
marcan34 hours sleep time on t6021#asahi-dev · 2023-04-17 19:17
marcan[ 78.961958] macsmc-power macsmc-power: In 0.203W Sys 3.364W 3V8 122.997W MPMU 0.839W SPMU 1.114W CLVR 0.663W CPU 2.221W Batt -3.148W 100% Tempty 2045m#asahi-dev · 2023-04-17 17:43
marcanaaaand I just found a really dumb bug in m1n1 that may or may not be breaking some things...#asahi-dev · 2023-04-17 17:30
marcannow how do we deal with this...#asahi-dev · 2023-04-17 17:30
marcanfabric and stuff#asahi-dev · 2023-04-17 17:29
marcanI just dropped power consumption by another watt by doing that.#asahi-dev · 2023-04-17 17:29
marcanargh. it's non-CPU power states.#asahi-dev · 2023-04-17 16:40
marcan*xx#asahi-dev · 2023-04-17 16:39
marcanit's a t600x problem with the P core regulators not shutting down#asahi-dev · 2023-04-17 16:24
marcanI think we must be missing something to enable actually powering down idle clusters#asahi-dev · 2023-04-17 16:23
marcanthis doesn't make any sense... going into deep sleep isn't helping at all with power consumption, but also the power usage is already high *before* even starting the secondaries.#asahi-dev · 2023-04-17 16:18
marcanhm, nah, but it's not using deep mode which doesn't seem to be saving any power#asahi-dev · 2023-04-17 15:35
marcanbut right now for me on 13.3 t416 it's instead trying to shut down a CPU, and when I made the HV not die on that, it's failing instead#asahi-dev · 2023-04-17 15:35
marcanbut it only panics if you have too much tracing and it exceeds max power limits and such#asahi-dev · 2023-04-17 15:35
marcanCLPC was always causing issues#asahi-dev · 2023-04-17 15:21
marcanI think we never noticed because CLPC makes that decision and we've always been disabling that in the hypervisor lately.#asahi-dev · 2023-04-17 15:21
marcanokaaay, I think it's actually shutting down CPUs. this may have been changed recently.#asahi-dev · 2023-04-17 14:17
marcanthanks :)#asahi-dev · 2023-04-17 14:10
marcanI guess that's a standard prop for ethernet controllers but not wifi#asahi-dev · 2023-04-17 14:09
marcanhave we always had that one? I swear I saw mention of that before...#asahi-dev · 2023-04-17 14:09
marcan/home/marcan/asahi/git/linux/arch/arm64/boot/dts/apple/t8103-j293.dtb: network@0,0: 'local-mac-address' does not match any of the regexes: 'pinctrl-[0-9]+'#asahi-dev · 2023-04-17 14:07
marcanOK, then I think we're good, I'll try to remember to poke things when I do dtschema updates