#
…

2023-04-19

00:03darkapex1 has joined #asahi-dev
00:04darkapex has quit [Ping timeout: 480 seconds]
00:38user982492 has joined #asahi-dev
00:43baozich has joined #asahi-dev
01:00chadmed_ has joined #asahi-dev
01:01CoolStar has quit [Quit: Leaving]
01:17CoolStar has joined #asahi-dev
01:29chadmed has quit [Remote host closed the connection]
01:42hightower2 has quit [Read error: Connection reset by peer]
01:42hightower2 has joined #asahi-dev
02:00gabuscus has quit []
02:13pharonix71 has joined #asahi-dev
02:14gabuscus has joined #asahi-dev
02:42opticron has quit [Ping timeout: 480 seconds]
03:04user982492 has quit [Quit: Textual IRC Client: www.textualapp.com]
03:05CoolStar has quit [Remote host closed the connection]
03:05CoolStar has joined #asahi-dev
03:06opticron has joined #asahi-dev
03:07CoolStar has quit [Remote host closed the connection]
03:08CoolStar has joined #asahi-dev
03:10CoolStar has quit [Remote host closed the connection]
03:10CoolStar has joined #asahi-dev
03:12CoolStar has quit [Remote host closed the connection]
03:12CoolStar has joined #asahi-dev
03:16CoolStar has quit [Read error: Connection reset by peer]
03:16pg12_ has joined #asahi-dev
03:16CoolStar has joined #asahi-dev
03:18pg12 has quit [Ping timeout: 480 seconds]
03:45marcansoo the core shutdown stuff *does* work on M1... it's just that on M2 the non-retention sleep bit seems to be gone?
03:45marcanso it *always* returns from wfi instead of jumping to RVBAR
03:45marcanthat's... interesting.
04:14infinity has joined #asahi-dev
04:18abd has joined #asahi-dev
04:21chadmed_ok something funky is definitely going on with the numbers now. before i was getting almost perfectly quadratic power scaling between pstates and now theres a linear zone and it tapers off at the top????
04:21chadmed_issuing WFI suspend the laws of physics on TSMC N5 it seems
05:01marcanchadmed_: the higher pstates do not increase the voltage, so I would expect linear power on those
05:02marcanI suspect the reason they do that is that they can guarantee sufficient vcore for those iff there is low vdroop from the other cores
05:03abd has quit [Ping timeout: 480 seconds]
05:04chadmed_yeah that makes sense, but i still find it odd that the data has changed so much. my methodology is exactly the same
05:04chadmed_ill do some more digging
05:12marcanre deep sleep, I think what's happening is that on M2 full shutdown only happens with proper full cluster shutdown, and that isn't working properly when I do it (but *is* when macOS does it)
05:13marcanbecause when I shut down the cores after a guest sleep event, MMIO for the cores goes away (there's a bit for that even, which we both set)
05:13marcanbut when I shut them down without macOS involved, they don't
05:13marcanso something else macOS does is allowing the clusters to go into deep sleep
05:14marcanI imagine trying to bring them back after *that* would end up in RVBAR
05:18darkapex1 has quit []
05:18darkapex has joined #asahi-dev
05:20salimterryli has quit [Remote host closed the connection]
05:29cylm has quit [Ping timeout: 480 seconds]
05:38jannauthe drm suspend splat is silly. the suspend/resume helpers disable/enable hpd polling unconditionally. our connectors are setup in a way which excludes polling so we do not init it
05:39jannaua quick grep suggest though that it is universally initialized
05:42jannaunot sure if polling would help us with dp-altmode but I don't think we can use it since our connectors are functionally tied to dcp
05:43salimterryli has joined #asahi-dev
05:43jannauwe don't have any information on the connected display unless dcp is connected to it
05:43nicolas17marcan: your fedi post said only macOS 13 has the charge limiting thing, but I'm on macOS 12.6.5 and the battery icon on the status bar says charging is suspended because I rarely use it on battery
05:44nicolas17(been plugged in for days and it's at 92%)
05:44marcanfirmware-based autonomous chage limiting is only on macOS 13
05:44nicolas17ah so this is being done by macOS?
05:44marcanalways was
05:44marcanthey use (one set of) the binary charge controls
05:44marcan(same one the linux kernel-side limiter does)
05:45nicolas17right, so the OS is telling the SMC "stop charging", while in macOS 13 it can say "stop charging when you reach 80%"
05:45marcanI don't actually know if macOS 13 *uses* that feature, all I know is it's in the firmware
05:45marcanI would sort of expect it's used in suspend at least
05:45jannauor all the charge control daemons written for asahi linux. are we at 3? I lost the count
05:46marcanno, those use a different set of charge controls :)
05:46marcan(there are several)
05:46marcanthe one for optimized battery charging has an auto safety trip at 50% where it does autonomously go back to charging, presumably to prevent the battery from outright dying in sleep mode
05:47marcanthe ones exposed to userspace in sysfs have no safeties, you can ask it to discharge and it will happily run down your battery to emergency shutdown even while plugged in
05:47nicolas17if I close the lid, for hours... in what state is macOS? is it in something closer to Linux's s2idle or s2ram?
05:47marcanthat's... a good question
05:48marcanwell, it depends on config
05:48marcanat least on battery with default config, it's s2ram, but I suspenct macOS still periodically wakes up to do stuff and that's how it implements charge control
05:49marcan(among other things because there have been multiple macOS issues with suspend battery drain, probably caused by such wakeups)
05:49nicolas17it's always responsive to network, but I *know* Apple has separate tricks for that (even on Intel), so it's hard to tell when it's really asleep
05:50marcanthe wifi implements ARP/ping/mDNS directly on the wifi card, so yeah, they're cheating
05:50nicolas17yeah and there's also the "sleep proxy"
05:53nicolas17"I suspenct macOS still periodically wakes up to do stuff" I think this is even done on Intel, called "powernap" in marketing and "dark wake" internally
05:54marcanand the whole s2ram thing would've been way more critical on intel since s2idle on intel is... not great
05:54marcanwhich is probably why it's designed this way
05:54marcanI think if macOS had been designed around AS from the get go they probably wouldn't have bothered with s2ram, and I still think there's a good chance we can get s2idle to be just about as good
05:55marcans2idle isn't just "be idle", you still put all devices into low power mode, we save about 1W over plain system idle right now and I'm sure there's more we can do
05:57nicolas17iPhones have "deep sleep", not sure how that compares to laptop s2ram but afaik they fully power off the main CPUs
05:57marcanultimately the only difference with s2ram is that in s2ram the CPUs go into effectively shutdown and the RAM goes into self-refresh... but there's no reason that wouldn't be achievable with the CPUs in deep sleep if you manage to quiesce all RAM traffic
05:57marcannicolas17: the ACCs have "deep sleep" but it's not yet clear how much power it saves over CPU deep idle (which also powers down the individual cores)
05:58marcanthat's what I enabled in the cpuidle stuff the other day
05:58marcanyou can tell this has all been improving with newer hardware from the xnu dumps
05:58marcanback in the day you'd have needed s2ram for good battery
05:58marcanthese days the CPUs power down any time they're idle
06:03nicolas17hmm do M* SoCs even have the recfg engine?
06:19infinity has quit [Quit: Connection closed for inactivity]
06:29Dementor has quit [Read error: Connection reset by peer]
06:29Dementor has joined #asahi-dev
06:33chadmed_ has quit [Quit: Page closed]
06:36nicolas17 has quit [Ping timeout: 480 seconds]
07:01Mary has quit [Quit: The Lounge - https://thelounge.chat]
07:01Mary has joined #asahi-dev
07:06ChaosPrincessWould appreciate someone reviewing this
07:20mpsChaosPrincess: is the newer m1n1 needed to test it
07:20mpsand CONFIG_DRM_ADP=y is enough for enabling it?
07:32ChaosPrincessconfig drm adp is enough, newer m1n1 and firmware are needed for the digitizer part which is in a separate pr, this is only screen
07:33ChaosPrincessactually, you'd probably have to rebuild m1n1 to get new device trees
07:36mpsChaosPrincess: it is detected and I see another card in /dev/dri
07:37ChaosPrincesstry booting into wayland kde and going to display configuration
07:37mpsbut I have no idea what I can do with it
07:38mpssorry I don't have kde and I don't have any experience with KDE, last time I tried it was before 1998 year
07:43mpsfine, DSI display is there
07:50roxfan has quit [Ping timeout: 480 seconds]
08:17nsklaus has joined #asahi-dev
08:36aleasto- has quit []
08:43aleasto has joined #asahi-dev
08:45bps has joined #asahi-dev
08:55nyilas has joined #asahi-dev
08:58zalyx09 has joined #asahi-dev
08:59abd has joined #asahi-dev
08:59zalyx0 has quit [Read error: Connection reset by peer]
08:59zalyx09 is now known as zalyx0
09:00Cyrinux9 has quit [Remote host closed the connection]
09:21hightower2 has quit [Ping timeout: 480 seconds]
09:32Cyrinux9 has joined #asahi-dev
09:43roxfan has joined #asahi-dev
10:25chadmed has joined #asahi-dev
10:43nela has quit [Quit: bye!]
10:54nela has joined #asahi-dev
11:02kenzie795 has quit []
11:18nsklaus has quit [Quit: ZZZzzz…]
11:19cylm has joined #asahi-dev
11:34i509vcb has quit [Quit: Connection closed for inactivity]
11:50bluetail42 has quit [Remote host closed the connection]
11:50thelounge60 has joined #asahi-dev
11:50bluetail42 has joined #asahi-dev
11:52baozich has quit [Remote host closed the connection]
11:56Guest11450 has quit [Ping timeout: 480 seconds]
12:01bluetail42 has quit [Remote host closed the connection]
12:02bluetail42 has joined #asahi-dev
12:02bluetail42 has quit [Remote host closed the connection]
12:06bluetail42 has joined #asahi-dev
12:24stickytoffee has quit [Ping timeout: 480 seconds]
12:24chadmed has quit [Read error: Connection reset by peer]
12:36chadmed has joined #asahi-dev
12:54chadmed_ has joined #asahi-dev
12:54chadmed has quit [Remote host closed the connection]
12:54gabuscus_ has joined #asahi-dev
12:55abd has quit [Ping timeout: 480 seconds]
13:01gabuscus has quit [Ping timeout: 480 seconds]
13:09c10l has joined #asahi-dev
13:12hightower2 has joined #asahi-dev
13:17sosys has joined #asahi-dev
13:17c10l3 has joined #asahi-dev
13:19c10l has quit [Ping timeout: 480 seconds]
13:22c10l3 has quit []
13:22c10l has joined #asahi-dev
13:38hightower2 has quit [Read error: Connection reset by peer]
13:42cylm_ has joined #asahi-dev
13:45cylm__ has joined #asahi-dev
13:48sosys has quit [Remote host closed the connection]
13:49cylm has quit [Ping timeout: 480 seconds]
13:51cylm_ has quit [Ping timeout: 480 seconds]
13:56krbtgt has quit [Ping timeout: 480 seconds]
14:28abd has joined #asahi-dev
14:32cylm has joined #asahi-dev
14:34cylm__ has quit [Ping timeout: 480 seconds]
15:05abd has quit [Ping timeout: 480 seconds]
15:26nsklaus has joined #asahi-dev
15:35bps has quit [Ping timeout: 480 seconds]
15:45m42uko has joined #asahi-dev
15:48m42uko_ has quit [Ping timeout: 480 seconds]
17:17nela has quit [Ping timeout: 480 seconds]
17:19lawrence has quit [Quit: The Lounge - https://thelounge.chat]
17:21lawrence has joined #asahi-dev
17:25nela has joined #asahi-dev
17:29cylm has quit [Ping timeout: 480 seconds]
18:30___nick___ has joined #asahi-dev
18:38nsklaus has quit [Quit: disappeared in a cloud of smoke]
18:39nsklaus has joined #asahi-dev
18:41kaazoo has joined #asahi-dev
19:00___nick___ has quit []
19:01ocrete has joined #asahi-dev
19:02ocreteHeya! Has anyone looked at the video codec ? I'm curious to know what is done in software vs firm/hardware ?
19:02___nick___ has joined #asahi-dev
19:02___nick___ has quit []
19:05___nick___ has joined #asahi-dev
19:05jannauocrete: people started looking into it but there wasn't any progress in the last few months that I'm aware of
19:05ocreteare there notes anywhere ?
19:06jannauthere are one or more video decode unit (h264, h265, vp9), encode (h264, h265, ...?) and prores decoder/encoder
19:08jannaueach block has custom co-processsor which apple's driver talk to over mailbox + shared memory
19:08ChaosPrincessdecode uses some cursed firmware, while encode does rtkit
19:09ChaosPrincessalso there is a m1n1 experiment for prores
19:09jannauthe coprocessor then interacts with the real codec hw
19:10zkrx has quit []
19:10jannauthe decode firmware was so horrible that the persons looked into it decided it would be easier to interact with the hw directly
19:13zkrx has joined #asahi-dev
19:14ChaosPrincesswas it R who looked into avd/ave? or someone else?
19:14jannauyes, R and Jamie
19:20jannaufrom my understanding v4l2's stateless m2m codecs would be a good fit api-wise
19:21ChaosPrincesssomeone also needs to write a vaapi driver that just wraps v4l decode
19:21ChaosPrincesscause both ff and chrome use vaapi to decode, and its a problem on things like raspis that only expose v4l
19:54i509vcb has joined #asahi-dev
20:08___nick___ has quit [Ping timeout: 480 seconds]
20:10WindowPain has quit [Remote host closed the connection]
20:12WindowPain has joined #asahi-dev
20:48chadmed_ is now known as chadmed
21:31abd has joined #asahi-dev
21:48zzywysm has joined #asahi-dev
21:57stipa is now known as Guest11598
21:57stipa has joined #asahi-dev
22:04Guest11598 has quit [Ping timeout: 480 seconds]
22:40pthariensflame has joined #asahi-dev
22:43pthariensflame has quit []
22:52nyilas has quit [Remote host closed the connection]
23:00urothis has quit [Read error: Connection reset by peer]
23:17abd has quit [Remote host closed the connection]
23:22abd has joined #asahi-dev
23:23cylm has joined #asahi-dev
23:37nicolas17 has joined #asahi-dev
23:40kaazoo has left #asahi-dev [#asahi-dev]
23:47nopeslide13 has quit []
…