<phire>
weapon_of_nix[m]: the /usr/bin/powermetric tool in macos (you need sudo) gives a lot of insight into what's going on
<phire>
I've been playing around with it for a few days
vx has quit [Remote host closed the connection]
vx has joined #asahi
<gmarx>
yes, is like `top` tool in terminal
<phire>
and if the power usage breakdowns are anywhere near accurate, it seems the most important part of getting low power draw is minimising the amount of time the DRAM is active.
<phire>
On my m1 pro max, a task that uses 7% of an ecore and a few percent of GPU at 60fps was reported as using something like 7mw of cpu power, 30mw of gpu power and almost 500 mw extra of DRAM power
jluthra has quit [Remote host closed the connection]
jluthra has joined #asahi
SuperED has joined #asahi
gabuscus_ has joined #asahi
gabuscus has quit [Ping timeout: 480 seconds]
gmarx has quit [Remote host closed the connection]
<marcan>
jannau: that's because the images aren't updated
<marcan>
all of them don't have it yet
gmarx has joined #asahi
gmarx has quit [Ping timeout: 480 seconds]
julio7359 has quit [Ping timeout: 480 seconds]
julio7359 has joined #asahi
pyropeter2 has joined #asahi
Stroller has quit [Quit: Stroller]
pyropeter1 has quit [Ping timeout: 480 seconds]
gmarx__ has joined #asahi
gmarx__ has quit []
marvin24 has joined #asahi
marvin24_ has quit [Ping timeout: 480 seconds]
gmarx1 has joined #asahi
vx has quit [Quit: G-line: User has been permanently banned from this network.]
gmarx1 is now known as gmarx
c10l has quit [Quit: Bye o/]
c10l has joined #asahi
vx has joined #asahi
vx has quit [Quit: G-line: User has been permanently banned from this network.]
gladiac has joined #asahi
dustnz has joined #asahi
Matt[m]1234 has joined #asahi
dustnz has quit [Quit: leaving]
the_lanetly_052__ has joined #asahi
vx has joined #asahi
the_lanetly_052 has quit [Ping timeout: 480 seconds]
gladiac is now known as Guest956
gladiac has joined #asahi
Guest956 has quit [Ping timeout: 480 seconds]
vx has quit [Quit: G-line: User has been permanently banned from this network.]
gladiac has quit [Quit: k thx bye]
<_jannau_>
understood, the AsahiLinux installs update themself though
<Ry_Darcy_>
@<phire> what tool/utility did you use to get these measurements (DRAM active, GPU, ecore) ?
<phire>
sudo /usr/bin/powermetrics
<phire>
(on macos)
jluthra has quit [Remote host closed the connection]
jluthra_ has joined #asahi
jluthra_ has quit [Ping timeout: 480 seconds]
vx has joined #asahi
<kujeger[m]>
is hibernate/suspend to disk support something that is likely to happen before suspend-to-ram, or is it dependent on the same driver and infrastructure support?
<Glanzmann>
marcan: ^
<marcan>
hibernate/suspend to disk is unlikely to be supported any time soon with real drivers due to Apple's firmware design
<kujeger[m]>
same situation for both modes then, thanks :)
<marcan>
it requires full save/restore of firmware-dependent state which is a *giant* pain for DCP/AGX
<marcan>
no, not at all
<marcan>
suspend-to-ram is *much* easier
<marcan>
the coprocessors support it natively
<kujeger[m]>
oh, that surprised me
<kujeger[m]>
I naively imagined it would be the other way around
<marcan>
suspend-to-ram isn't even a single thing, we already support it partially (wifi/pcie are the only real blocker) for s2idle mode
jluthra_ has joined #asahi
<marcan>
it's not even entirely clear exactly how hibernate is intended to work on these platforms, so there's a whole lot of platform expectation reverse engineering to be done there
<marcan>
but s2idle you almost get for free on modern platforms, and we know it works well enough since the machines already get like 16+ hour runtimes without even deep sleep
<kujeger[m]>
oo, that's nice to hear! When I decided to make asahi/m1 my daily driver I kind of decided to assume everything listed as not supported might be a year or two away
<kujeger[m]>
I personally haven't really used hibernation instead of suspend-to-ram in years, so it was mostly a curiosity thing
<marcan>
I expect that once we introduce CPU deep sleep, and fix the wifi/pcie issue, that will get us 24+ hours of idle time (== s2idle suspend time) and a lot of people ought to be happy with that level. then we can chip away at turning more things off to extend that, and eventually try out the "full" sleep mode.
<marcan>
yeah, I haven't used hibernate in ages too, other than when macOS automatically goes into that mode when the battery dies in suspend (which some versions of macOS liked to do due to bugs...)
<marcan>
I suspect once we get sleep times up in the 24-48h range very few people are going to care about hibernate
<kujeger[m]>
yeah, 24h+ seems like it would cover 99% of use cases! And even currently the machine is so fast that I shut it down and turn it on every day anyway; takes less than 20sec from opening the lid to being logged in with my stuff started up
<marcan>
and given I've seen macOS do like a week of sleep time (when it's not bugging out), we ought to be able to achieve the same eventually, though honestly if we can hit 24-48h with just deep cpu idle and not "full" sleep mode, I'd probably backburner that and focus on other things first :)
<rmk>
Hi. question about grub on the m1 mini - does it accept input from a USB keyboard? It doesn't appear to work with my Logitech wireless keyboard which is my only USB keyboard.
<marcan>
on the bootloader, or on linux?
<kujeger[m]>
that sounds really promising, thanks for all the work!
<marcan>
the bootloader only supports the Type C ports right now
<marcan>
linux supports everything
<rmk>
that'll be it then, since it's a type A dongle
julio7359 has quit [Ping timeout: 480 seconds]
<marcan>
try an A-to-C dongle adapter :)
julio7359 has joined #asahi
<rmk>
thanks... I'll see whether Oracle will get me one :D
<marcan>
:)
lgarylmtdlt^ has quit [Remote host closed the connection]
<_jannau_>
24h idle time should not be to hard when the battery life of an unused laptop (display off) is already > 19h (M2 macbook pro 13)
<nsklaus>
i thought currently brightness could not be modified ?
<nsklaus>
is it a full on or full off choice at the moment then ?
<j`ey>
yes
<nsklaus>
i see
<mattgirv>
Full on being whatever it was set to in macos I believe?
<nsklaus>
probably not, i'd guess it's just a power thing, with 100% power given when it's on. but maybe other will correct me if my assumption is not right
<mattgirv>
Well it's definitely a fixed value, pretty sure I've rebooted in to asahi with brightness lowered and it maintains that level
<Glanzmann>
mattgirv: Yes, the last macos value is being used.
<nsklaus>
ah really ? if so nice, that ould already be something. even if quite troublesome to reboot to macos and then back in linux just to change brightness level
<nsklaus>
*Would
<nsklaus>
if it is so, then maybe poking a register, from linux , overriding its previous value, would work then ?
<sven>
it's a dcp function call and not a register
<nsklaus>
i mean, poking at a given memory address
gmarx has quit [Ping timeout: 480 seconds]
<nsklaus>
yes poor choice of words from me
<sven>
no, wrong choice of words
<sven>
there's a huge difference between "poke a register" and "send a DCP IPC call"
<sven>
the first is a single line or so the latter requires the entire DCP driver
<nsklaus>
i see
<_jannau_>
and at least on the 14/16" macbook pros a dcp function call which requires setup (probably at least uploading the display backlight calibration data) which the current dcp driver doesn't do yet
<_jannau_>
so it is even more than a single dcp ipc call
<nsklaus>
what are the current expectation about ETA for a basic dcp driver that would just let us change brightness ? (for base m1 macbook 13")
<j`ey>
none
<sven>
+1 month everytime someone asks for an eta
roxfan2 has joined #asahi
<nsklaus>
:)
<j`ey>
sven: even better!
* nsklaus
refrains asking twelve more times ;)
<nsklaus>
sven: otoh, that value for brightness that macos uses, it must be saved somewhere, what would happen if someone were to find it and attempt to override it manualy ?
<sven>
sure, that might work. will still require a reboot for every change then
<sven>
i'd probably first look into nvram
<j`ey>
nsklaus: feel free to take a look!
roxfan has quit [Ping timeout: 480 seconds]
<nsklaus>
sven: hmm i see, if it would require a reboot no matter what, then maybe better to wait until dcp driver implements it properly
<nsklaus>
j`ey: my dev skills would be far too unsufficient to attempt such a thing. i stuggle at making simple 2d platformer in sdl .. i succeed at that, barely. making doubtful quality code even in that simple context.. so anything above that is just way above my reach
roxfan2 has quit [Ping timeout: 480 seconds]
gmarx has joined #asahi
leitao has joined #asahi
gmarx has quit [Ping timeout: 480 seconds]
leitao has quit []
leitao has joined #asahi
leitao has quit [Read error: Connection reset by peer]
leitao has joined #asahi
zotanmew has quit [Quit: goodbye: ~zotan via destiny.zotan.network]
zotanmew has joined #asahi
gmarx has joined #asahi
zotanmew has quit [Quit: goodbye: ~zotan via destiny.zotan.network]
gmarx has quit [Ping timeout: 480 seconds]
zotanmew has joined #asahi
leitao has quit [Ping timeout: 480 seconds]
zotanmew has quit []
zotanmew has joined #asahi
zotanmew has quit []
zotanmew has joined #asahi
Race has joined #asahi
gmarx has joined #asahi
Retr0id6 has joined #asahi
Retr0id has quit [Read error: Connection reset by peer]
Retr0id6 is now known as Retr0id
Retr0id9 has joined #asahi
Retr0id has quit [Read error: Connection reset by peer]
Retr0id9 is now known as Retr0id
gmarx has quit [Ping timeout: 480 seconds]
Retr0id0 has joined #asahi
Retr0id has quit [Read error: Connection reset by peer]
Retr0id0 is now known as Retr0id
Retr0id6 has joined #asahi
Retr0id has quit [Remote host closed the connection]
Retr0id6 is now known as Retr0id
gladiac has joined #asahi
Retr0id has quit [Ping timeout: 480 seconds]
Retr0id has joined #asahi
gmarx has joined #asahi
gmarx has quit [Ping timeout: 480 seconds]
skipwich_ has joined #asahi
refi647 has joined #asahi
kefu_ has joined #asahi
bpye7 has joined #asahi
eroc19909 has joined #asahi
nafod4 has joined #asahi
alethkit_ has joined #asahi
Vaughn_ has joined #asahi
mrCyborg_ has joined #asahi
keegans_ has joined #asahi
d4ve_ has joined #asahi
coder_kalyan_ has joined #asahi
WhyNotHugo_ has joined #asahi
akspecs_ has joined #asahi
tsujp_ has joined #asahi
jleightcap_ has joined #asahi
qyousef_ has joined #asahi
toasterking_ has joined #asahi
cds_ has joined #asahi
tdmm_ has joined #asahi
probie_ has joined #asahi
gpanders_ has joined #asahi
NekomimiScience_ has joined #asahi
rann_ has joined #asahi
aeroraptor_ has joined #asahi
ovf_ has joined #asahi
leo60228- has joined #asahi
jbowen_ has joined #asahi
daniels_ has joined #asahi
mmarchini____ has joined #asahi
hellsenberg has joined #asahi
saintdev__ has joined #asahi
cptcobalt_ has joined #asahi
maennich_ has joined #asahi
Lightsword_ has joined #asahi
nepeat_ has joined #asahi
austriancoder_ has joined #asahi
adamjm_ has joined #asahi
Method_ has joined #asahi
HotSwap` has joined #asahi
jabashque_ has joined #asahi
eric_engestrom_ has joined #asahi
Gaelan_ has joined #asahi
WindowPain_ has joined #asahi
steev_ has joined #asahi
ihaveamac_ has joined #asahi
sorear__ has joined #asahi
eroc1990 has quit [synthon.oftc.net larich.oftc.net]
skipwich has quit [synthon.oftc.net larich.oftc.net]
bredge has quit [synthon.oftc.net larich.oftc.net]
artemist has quit [synthon.oftc.net larich.oftc.net]
princesszoey has quit [synthon.oftc.net larich.oftc.net]
kefu has quit [synthon.oftc.net larich.oftc.net]
toasterking has quit [synthon.oftc.net larich.oftc.net]
dimbulb has quit [synthon.oftc.net larich.oftc.net]
aeroraptor has quit [synthon.oftc.net larich.oftc.net]
saintdev_ has quit [synthon.oftc.net larich.oftc.net]
ihaveamac has quit [synthon.oftc.net larich.oftc.net]
tdmm has quit [synthon.oftc.net larich.oftc.net]
adamjm has quit [synthon.oftc.net larich.oftc.net]
psanford has quit [synthon.oftc.net larich.oftc.net]
Hotswap has quit [synthon.oftc.net larich.oftc.net]
Illya has quit [synthon.oftc.net larich.oftc.net]
WindowPain has quit [synthon.oftc.net larich.oftc.net]
gpanders has quit [synthon.oftc.net larich.oftc.net]
cptcobalt has quit [synthon.oftc.net larich.oftc.net]
kallsyms has quit [synthon.oftc.net larich.oftc.net]
bpye has quit [synthon.oftc.net larich.oftc.net]
nepeat has quit [synthon.oftc.net larich.oftc.net]
emptynine has quit [synthon.oftc.net larich.oftc.net]
Poplar has quit [synthon.oftc.net larich.oftc.net]
rbenua has quit [synthon.oftc.net larich.oftc.net]
Erus_Iluvatar has quit [synthon.oftc.net larich.oftc.net]
jleightcap has quit [synthon.oftc.net larich.oftc.net]
keegans has quit [synthon.oftc.net larich.oftc.net]
rossy has quit [synthon.oftc.net larich.oftc.net]
coder_kalyan has quit [synthon.oftc.net larich.oftc.net]
leo60228 has quit [synthon.oftc.net larich.oftc.net]
jabashque has quit [synthon.oftc.net larich.oftc.net]
balrog has quit [synthon.oftc.net larich.oftc.net]
cds has quit [synthon.oftc.net larich.oftc.net]
hell__ has quit [synthon.oftc.net larich.oftc.net]
qyousef has quit [synthon.oftc.net larich.oftc.net]
probie has quit [synthon.oftc.net larich.oftc.net]
Method has quit [synthon.oftc.net larich.oftc.net]
Gaelan has quit [synthon.oftc.net larich.oftc.net]
Lightsword has quit [synthon.oftc.net larich.oftc.net]
Tramtrist has quit [synthon.oftc.net larich.oftc.net]
d4ve has quit [synthon.oftc.net larich.oftc.net]
alethkit has quit [synthon.oftc.net larich.oftc.net]
NekomimiScience has quit [synthon.oftc.net larich.oftc.net]
rann has quit [synthon.oftc.net larich.oftc.net]
nafod has quit [synthon.oftc.net larich.oftc.net]
zv has quit [synthon.oftc.net larich.oftc.net]
Vaughn has quit [synthon.oftc.net larich.oftc.net]
tsujp has quit [synthon.oftc.net larich.oftc.net]
mrCyborg has quit [synthon.oftc.net larich.oftc.net]
refi64 has quit [synthon.oftc.net larich.oftc.net]
sorear_ has quit [synthon.oftc.net larich.oftc.net]
akspecs has quit [synthon.oftc.net larich.oftc.net]
tbodt has quit [synthon.oftc.net larich.oftc.net]
maennich has quit [synthon.oftc.net larich.oftc.net]
ovf has quit [synthon.oftc.net larich.oftc.net]
whynothugo has quit [synthon.oftc.net larich.oftc.net]
jbowen has quit [synthon.oftc.net larich.oftc.net]
austriancoder has quit [synthon.oftc.net larich.oftc.net]
eric_engestrom has quit [synthon.oftc.net larich.oftc.net]
daniels has quit [synthon.oftc.net larich.oftc.net]
steev has quit [synthon.oftc.net larich.oftc.net]
sjs has quit [synthon.oftc.net larich.oftc.net]
rcombs has quit [synthon.oftc.net larich.oftc.net]
JTL has quit [synthon.oftc.net larich.oftc.net]
bpye7 is now known as bpye
adamjm_ is now known as adamjm
aeroraptor_ is now known as aeroraptor
refi647 is now known as refi64
nafod4 is now known as nafod
cptcobalt_ is now known as cptcobalt
saintdev__ is now known as saintdev_
mmarchini___ has quit [synthon.oftc.net larich.oftc.net]
qyousef_ is now known as qyousef
Vaughn_ is now known as Vaughn
NekomimiScience_ is now known as NekomimiScience
steev_ is now known as steev
bredge has joined #asahi
rbenua has joined #asahi
Tramtrist has joined #asahi
princesszoey has joined #asahi
dimbulb has joined #asahi
Erus_Iluvatar has joined #asahi
tbodt has joined #asahi
emptynine has joined #asahi
zv has joined #asahi
Illya has joined #asahi
Poplar has joined #asahi
JTL has joined #asahi
rcombs has joined #asahi
artemist has joined #asahi
coder_kalyan_ is now known as coder_kalyan
mrCyborg_ is now known as mrCyborg
d4ve_ is now known as d4ve
keegans_ is now known as keegans
sjs has joined #asahi
alethkit_ is now known as alethkit
jleightcap_ is now known as jleightcap
toasterking_ is now known as toasterking
gpanders_ is now known as gpanders
probie_ is now known as probie
cds_ is now known as cds
tdmm_ is now known as tdmm
balrog has joined #asahi
austriancoder_ has quit []
austriancoder has joined #asahi
kallsyms has joined #asahi
psanford has joined #asahi
rossy has joined #asahi
gmarx has joined #asahi
gmarx has quit [Ping timeout: 480 seconds]
gmarx has joined #asahi
c10l has quit [Quit: Bye o/]
gmarx has quit [Ping timeout: 480 seconds]
c10l has joined #asahi
gmarx has joined #asahi
<kit_ty_kate1>
hi, I haven't seen anything on the bugtracker so I'm asking here just in case it is a known issue but i just booted my asahilinux partition and past Grub, both the mouse and keyboard were unresponsive and i had to force shutdown and reboot to make it work (pressing the power button did bring up the GDM shutdown prompt though). Is that something anyone else has seen?
<kit_ty_kate1>
(it worked fine after a reboot somehow)
<kit_ty_kate1>
i haven't seen that issue in the few weeks i've been using it daily and i installed the latest kernel/uboot upgrades last night before powering off so maybe it's related?
gmarx has quit [Ping timeout: 480 seconds]
gmarx has joined #asahi
gmarx has quit [Quit: WeeChat 3.6]
<tpw_rules>
that happened to me last night on my m1 max mbp and nixos
<tpw_rules>
but i was futzing around with the boot selector so i wonder if it didn't shut down the keyboard properly or osmething
<kit_ty_kate1>
tpw_rules: in grub you mean? if so i remember pressing Enter a bit later during the countdown so maybe there is a race condition somewhere there
<rmk>
I got some weirdness after the first boot - I logged in fine, was able to open a terminal. I sudo su'd, and got sshd running. Logged in over the network, everything seemed fine. A while later (so the screen blanker had kicked in, I woke it up, it kicked in again a while later), I tried to open firefox by clicking on the icon. Nothing happened. I tried opening a new konsole. Nothing happened. I closed the
<rmk>
existing one and tried to log out... nope, couldn't. So I tried opening a konsole from the ssh connection using "DISPLAY=:0 konsole" and that told me it couldn't authorise with Xorg. Eventually the screen blanker tried to kick in and displayed a message saying it couldn't.
<rmk>
I checked the sddm authority file, and that matched ~/.Xauthority
<rmk>
(so Xorg's -auth file matched)
<rmk>
eventually, I had to use loginctl to terminate the session, and haven't yet had a repeat of the problem
<rmk>
I have, however, noticed some rendering issues with the background
<rmk>
(like it doesn't always re-paint it when changing some settings)
<rmk>
so you end up with a black background but with a rectangle of the correct background in the top left
<rmk>
so, I'm wondering if there's some data corruption going on
<marcan>
I think I've seen the repainting issues once or twice, I suspect it's just some bug in simpledrmfb or modesetting when used in this way
<marcan>
I've never seen things break the way they did for you though
<marcan>
it sounds like specifically something went very wrong with the xauth
<marcan>
is it possible you did something with SSH with x11 forwarding enabled that somehow messed around with the x11 auth cookie?
<rmk>
no - X forwarding by ssh is disabled by default (I run debian...), the ssh connection never had a DISPLAY env set on the M1 end, and the timestamp of ~/.Xauthority matched the Xorg's -auth file - as did its contents.
<rmk>
that was also my first thought too... which is why I investigated that specifically while the problem existed.
the_lanetly_052__ has quit [Ping timeout: 480 seconds]
roxfan has joined #asahi
<aechaare[m]>
<opticron> "Go ahead and describe the..." <- SOLVED- package corruption. Libinput fixed it
nicolas17 has joined #asahi
eroc1990 has joined #asahi
eroc19909 has quit [Read error: Connection reset by peer]
eroc1990 is now known as eroc19909
Gaspare has quit [Quit: Gaspare]
Gaspare has joined #asahi
buh0 has joined #asahi
riker77 has quit [Quit: Quitting IRC - gone for good...]
riker77 has joined #asahi
buh0 has quit [Quit: Bye!]
caef^ has joined #asahi
Stroller has quit [Ping timeout: 480 seconds]
rayanamukami has quit [Ping timeout: 480 seconds]
rayanamukami has joined #asahi
caef^ has quit [Remote host closed the connection]
markwang0 has joined #asahi
<markwang0>
hi all, is there any battery charge limiting software that works on Asahi? I'm looking for something similar to AlDente https://github.com/davidwernhart/AlDente. I'd like to force the battery to stop charging and use A/C power at 80%
<milek7>
echo inhibit-charge | sudo tee /sys/class/power_supply/macsmc-battery/charge_behaviour