#
…

2021-09-16

00:03skali has quit [Ping timeout: 480 seconds]
00:06grange_c has quit [Quit: Ping timeout (120 seconds)]
00:06grange_c has joined #asahi
00:09ella-0 has quit [Ping timeout: 480 seconds]
00:22milek7does M1 have accelerated opengl on macos?
00:23milek7because I have report of application running so slow that I think it is falls back to software rendering
00:23i509vcb[m]I believe it's OpenGL implemented ontop of Metal
00:23i509vcb[m]Per vendor info in Minecraft
00:25i509vcb[m]Unless that's a symptom of software renderer?
00:44alyssamilek7: Yes.
00:45alyssai509vcb[m]: This is correct, it's GL-on-Metal, not software.
01:00skali has joined #asahi
01:04Tom__ has quit [Read error: Connection reset by peer]
01:08skali has quit [Ping timeout: 480 seconds]
01:36skali has joined #asahi
01:44skali has quit [Ping timeout: 480 seconds]
02:13skali has joined #asahi
02:21skali has quit [Ping timeout: 480 seconds]
02:48malvo has quit [Ping timeout: 480 seconds]
02:48malvo has joined #asahi
02:49skali has joined #asahi
02:52PhilippvK has joined #asahi
02:55phiologe has quit [Ping timeout: 480 seconds]
02:57skali has quit [Ping timeout: 480 seconds]
03:20borrrden has joined #asahi
03:21marvin24 has joined #asahi
03:25marvin24_ has quit [Ping timeout: 480 seconds]
03:25skali has joined #asahi
03:33skali has quit [Ping timeout: 480 seconds]
03:37yuyichao_ has joined #asahi
03:42borrrden has quit [Quit: WeeChat 2.8]
03:53skali has joined #asahi
04:01skali has quit [Ping timeout: 480 seconds]
04:29skali has joined #asahi
04:37skali has quit [Ping timeout: 480 seconds]
04:56skali has joined #asahi
05:04skali has quit [Ping timeout: 480 seconds]
05:33skali has joined #asahi
05:41skali has quit [Ping timeout: 480 seconds]
05:46skali has joined #asahi
06:50skali has quit [Ping timeout: 480 seconds]
06:56skali has joined #asahi
07:12marcanmaz: I know you expressed some interest in the cpufreq/idle stuff, are you looking into that? otherwise I was thinking of starting to poke around that, build some m1n1 tests to measure relative perf and switching latencies
07:17marcanwell, I'll do the tests anyway, but let me know if you want to poke at the kernel side after that :)
07:17marcanI had a look at how some other socs represent the cluster-level perf states and such
07:22marcanalso, if anyone has shell.py handy, I'd like to know if you get any different values for this:
07:22marcan>>> read32(0x200380dbc); read32(0x200380dc4)
07:22marcan0x1800180
07:22marcan0x813057f
07:23marcanthat should be related to memory controller performance; the ADT describes a higher performance state and the driver apparently just uses the current state as a default, but if that is always the same I might want to just have both described
07:31svensame values on my mac mini
07:42marcanthe perf value is hardcoded in the master adt too, not substituted in, so I think we can safely hardcode these
07:48marcanjust pushed proper support for template ADTs in m1n1.adt
07:50marcanthere's actually relatively little that is per-device in here
07:50marcanat least little that is templated like this
07:50marcanother than product info and MACs, there's:
07:50marcan wifi-rfem-info = << syscfg/RFEM >>
07:50marcan wifi-calibration-msf = << syscfg/WCAL >>
07:50marcan wifi-antenna-sku-info = << syscfg/WSKU >>
07:50marcan bluetooth-tx-calibration = << syscfg/BTTx/0x14 >>
07:50marcanbunch more bluetooth calibration stuff
07:51marcan display-backlight-compensation = << syscfg/DBCl >>
07:51marcan display-temp-compensation = << syscfg/DTCl >>
07:51marcan speaker-trim-gains = << syscfg/SpGa >>
07:51marcan speaker-thiele-small = << syscfg/SpTS >>
07:51marcanso those are the only things m1n1 really needs to gain support to forward into the fdt
07:51marcanbasically wireless, display, and speaker calibration
07:52marcanthe MBP also has some touch bar and mic stuff
07:53marcanand camera
07:53marcanrosaline-calibration-current = << syscfg/RxCL >>
07:53marcan... really? rosaline?
07:53marcanisn't that part of Touch ID?
07:54marcanor maybe that was the ALS
07:55marcanyeah, rosaline is the flood illuminator AIUI... what's this doing on the MBP, that makes no sense
07:56marcanthis is under isp {}, I think it's just junk left over from iPhones
08:15bgb has joined #asahi
08:21Ariadnethis is really impressive to see how far asahi has come in such a short time
08:22Ariadnei think the mac mini, with asahi's work, is going to be the preferred ARM workstation solution
08:23Ariadnethe "competition" is proposing a design with a price tag around $2000
08:24ar🤷 for me the current m1 lineup is undesirable because of its 16G ram limit
08:24areven the lx2k board i have has 2 sodimm slots with 32GB of ram
08:26sveni'm hoping for a m1x/m2/whatever with 32gb ram
08:26marcanpretty sure that's coming
08:26sveni'd be surprised if it isn't :)
08:27marcanin fact the only reason I din't bui 3-4 minis to replace my home server with an actual redundant build/storage cluster is because I'm waiting for the M1X ;)
08:27Emantor[m]I doubt that the layerscape 2160 on lx2k can match the performance of the M1. Lots of ethernet throughput though, which is nice :-)
08:28bgb has quit [Ping timeout: 480 seconds]
08:30arEmantor[m]: single thread, absolutely not
08:31kettenismarcan: not sure what the wifi calibration stuff is for
08:31_jannau_16 core lx2k was iirc factor 4 slower than 8 core m1 (running linux in a VM) NEON heavy workload (AV1 decoding)
08:32kettenisdon't think the current brcmfmac driver has a way to use that stuff
08:44marcanI also see iBoot making quite a few changes outside of those templates
08:44marcanadding tunables etc
08:45marcansven: what version are you on, and would you mind sending me your DT? I wonder how much actually changes between units
08:45marcan(warning: likely has your wifi password in it)
08:45svenno idea what version i'm on
08:45svennot sure i even setup wifi, but let me check
08:46svenhuh... it has my wifi stuff in there
08:46marcanyou can look at system-firmware-version and firmware-version
08:46svenoh... icloud sync probably
08:47svenfirmware-version: iBoot-6723.81.1
08:47svensystem-firmware-version: iBoot-6723.81.1
08:50marcanah, older
08:50marcanmight still be good enough for diffing tunables though
08:50Emantorsystem-firmware-version = iBoot-6723.120.36, firmware-version = b'iBoot-6723.120.36 for my mini.
08:50svenyeah, let me just strip the wifi stuff
08:58marcangoing to stream for a bit, looking at the cpufreq stuff
09:03EmantorLooks like one of the macOS updates pulled in a newer version and i was looking at an older dump, system-firmware-version = iBoot-6723.140.2 & firmware-version = iBoot-6723.120.36 is whats currently on my m1n1.
09:08skali has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
09:18skali has joined #asahi
11:22skali is now known as skali_
11:22skali_ is now known as skali
11:33skali has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
11:44X-Scale` has joined #asahi
11:50X-Scale has quit [Ping timeout: 480 seconds]
12:00X-Scale has joined #asahi
12:05X-Scale` has quit [Ping timeout: 480 seconds]
12:23al3xtjames8 has joined #asahi
12:26al3xtjames has quit [Ping timeout: 480 seconds]
12:26al3xtjames8 is now known as al3xtjames
12:41chadmed has quit [Remote host closed the connection]
12:41chadmed has joined #asahi
13:06aleasto has joined #asahi
13:35nskl has quit [Ping timeout: 480 seconds]
14:17yuyichao_ has quit [Ping timeout: 480 seconds]
14:35psykose has quit [Ping timeout: 480 seconds]
14:36psykose has joined #asahi
14:47nskl has joined #asahi
14:49psykose has quit [Ping timeout: 480 seconds]
14:50marcanI know why I couldn't boost to 3.2
14:50marcanmacos has the same issue
14:50marcanI just tested it on the MBA: the cores claim to be at 3.2, but if all are active, the cluster gets capped at 3.0
14:50marcanso you need to actually put some cores to sleep to reach the higher pstates
14:51marcanwell that was a waste of time on the stream :p
14:52j_ey*actual* sleep?
14:52j_eylike wfi or whatever, or just lower states
14:52marcanthe state is per-cluster
14:52marcancores can only be in sleep states
14:52marcanwhich is two kinds of wfi, the kind that loses state and the kind that does not
14:52j_eyoh right, state is per cluster
14:53marcanwell, either that or it's a thermal issue
14:53marcanthe MBA extremely quickly throttles down from max state to the prior one
14:54marcanbut I think it's not that
15:06kettenisthe mini should be fine; it has a fan
15:09rowang077[m]Do you develop this on an air? I thought you had a mac mini.
15:09yuyichao_ has joined #asahi
15:10j_eyrowang077[m]: marcan has all of them!
15:10psykose has joined #asahi
15:14marcan== Final CPU frequencies ==
15:14marcanCPU 0: 599.99 MHz
15:14marcanCPU 4: 3203.84 MHz
15:15j_eymarcan: please feed yourself now!
15:15marcanswitched the SMP code to use wfi and IPIs to wake up secondaries, and made a deep sleep function
15:19marcanalso only a single register write is needed
15:19marcaneverything I did on the stream was a waste of time :)
15:19j_eylol ;_;
15:19marcan(well, I did want to bring up PMP anyway for power measurements, eventually)
15:30zigmars has joined #asahi
15:34zigmars has quit [Remote host closed the connection]
15:35zigmars has joined #asahi
15:44marcan(pushed)
15:55zigmars has quit [Read error: Connection timed out]
16:00zigmars has joined #asahi
16:09zigmars has quit [Remote host closed the connection]
16:09zigmars has joined #asahi
16:22Andalu30 has joined #asahi
16:33Andalu30 has quit [Ping timeout: 480 seconds]
16:34Andalu30 has joined #asahi
16:36jvoisin has quit [Ping timeout: 480 seconds]
16:37zigmars has quit []
16:45jvoisin has joined #asahi
17:38yuyichao_ has quit [Quit: Konversation terminated!]
17:42yuyichao has joined #asahi
18:07kettenisstarted upstreaming my u-boot code
18:07j_eykettenis: great!
18:08alyssahas anyone else tried kettenis's u-boot with linux?
18:08alyssaI was having SMP and framebuffer problems
18:09kettenisI may have fixed a framebuffer problem today (if you booted using EFI)
18:12kettenisthat fix would probably only fix color issues though
18:12svennice :)
18:14nskl has quit [Ping timeout: 480 seconds]
18:15kettenissven: The cover letter of the v2 mailbox series said you added some more examples, but I didn't spot those...
18:15sveni meant that i wrote into the description that e.g. m3 one is used for thunderbolt
18:16sveni can add more examples if you think that's helpful though
18:16kettenisso i just misunderstood what you meant
18:16kettenisdon't really think that would be particularly helpful
18:17svenhrm, now that i re-read that latter again it's slightly confusing
18:17sven*letter
18:17alyssaletter latter letter latterletter latterletter latter
18:18svenlooks like macos does a full xhci reset whenever a new device is plugged in fwiw
18:18alyssapsykose: <-- kettle
18:18svenand it looks like that may be required :(
18:18psykosealyssa: :p
18:18svenjust resetting the port in the PHY doesn't always work
18:19alyssapsykose: you are in the irc channel of a linux port. just saying ;-p
18:20psykosea spicy port at that
18:37nskl has joined #asahi
18:42torstenvl has joined #asahi
18:44torstenvl_ has joined #asahi
18:44torstenvl_ has quit []
18:46torstenvl has quit []
18:48torstenvl has joined #asahi
18:51Andalu30 has quit [Quit: Konversation terminated!]
18:55skali has joined #asahi
18:55skali is now known as skali_
18:55skali_ is now known as skali
18:59skali has quit []
19:04malvo has quit [Ping timeout: 480 seconds]
19:04aleasto has quit [Remote host closed the connection]
19:04torstenvl has quit [Remote host closed the connection]
19:06malvo has joined #asahi
19:31torstenvl has joined #asahi
19:57zigmars has joined #asahi
20:37zigmars_ has joined #asahi
20:47zigmarshello, any pointers to what's the simplest (beginner) way to run original OS X kernel under hv after having successfully run m1n1 with debian initrd as per https://github.com/AsahiLinux/docs/wiki/Developer-Quickstart ?
20:48zigmarsI understand that marcan (as seen in worklogs) does a `run_guest.py` with extracted .macho file, so alternatively how & where from would I extract that?
20:52zigmars_ has quit []
20:52j_eyzigmars: if you boot into macOS, you can find a file called `kernelcache` under /System/Volumes/Preboot/
20:55j_eyzigmars: then you need to get https://github.com/tihmstar/img4tool
20:56torstenvl has quit [Remote host closed the connection]
20:57j_eythen you run: img4tool -e -p out.im4p kernelcache ; img4tool -e -o kernel.macho out.im4p
21:01zigmars has quit [Quit: Leaving]
21:01torstenvl has joined #asahi
21:01zigmars has joined #asahi
21:18torstenvl has quit [Remote host closed the connection]
21:28zigmars has quit [Remote host closed the connection]
21:33yuyichao has quit [Quit: Konversation terminated!]
21:36zigmars has joined #asahi
21:40alyssaspeaking of
21:40alyssasomeone who has successfully setup m1n1 and booted linux with usb support and has a flash drive to spare, want to test something for me? for freee!!! 😋
21:41alyssa(I mean. I can provide the kernel image I guess.)
21:44zigmars has quit [Ping timeout: 480 seconds]
21:47zigmars has joined #asahi
22:03zigmars has quit [Remote host closed the connection]
22:04zigmars has joined #asahi
22:05torstenvl has joined #asahi
22:08aleasto has joined #asahi
22:10zigmars has quit []
22:13torstenvl has quit [Ping timeout: 480 seconds]
22:15aleasto has quit [Quit: Konversation terminated!]
22:29artemist has quit [Remote host closed the connection]
22:32artemist has joined #asahi
22:34zigmars has joined #asahi
23:18alyssaI have a few minutes and very little energy. That sounds like a perfect time to rebase my kernel.
23:21psykoserebasing kernels is a health hazard
23:27zigmars has quit [Quit: Leaving]
23:53alyssamaz: just built your branch (+ reboot + DCP patches), everything works as before
23:53alyssain particular hello from m1 ethernet and usb keyboard and mouse :]
23:53alyssaso the pcie series is Tested-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
…