#
…

2023-04-28

00:11brolin has quit [Ping timeout: 480 seconds]
00:39mazI'm running 6.3++ on my mini with mass storage on the USB-A ports, and it works just fine.
00:46brolin has joined #asahi-dev
01:19eiln has joined #asahi-dev
01:48elvishjerricco has joined #asahi-dev
01:54elvishjerricco has quit [Read error: Connection reset by peer]
01:55brolin has quit [Ping timeout: 480 seconds]
01:56brolin has joined #asahi-dev
02:00gabuscus has quit []
02:06benpoulson has joined #asahi-dev
02:08elvishjerricco has joined #asahi-dev
02:12elvishjerricco has quit [Read error: Connection reset by peer]
02:14benpoulson has quit [Ping timeout: 480 seconds]
02:14elvishjerricco has joined #asahi-dev
02:17gabuscus has joined #asahi-dev
02:17elvishjerricco has quit [Read error: Connection reset by peer]
02:17elvishjerricco has joined #asahi-dev
02:23elvishjerricco has quit [Read error: Connection reset by peer]
02:31_rudi1 is now known as _rudi
02:32elvishjerricco has joined #asahi-dev
02:38benpoulson has joined #asahi-dev
02:47benpoulson has quit [Ping timeout: 480 seconds]
02:52brolin has quit [Ping timeout: 480 seconds]
03:00elvishjerricco has quit [Read error: Connection reset by peer]
03:07brolin has joined #asahi-dev
03:09elvishjerricco has joined #asahi-dev
03:12elvishjerricco has quit [Read error: Connection reset by peer]
03:15elvishjerricco has joined #asahi-dev
03:57brolin has quit [Ping timeout: 480 seconds]
04:19marcanif I had to guess this is a bad interaction between a change in 6.3 and that very specific USB device
04:20marcancould even be a firmware bug in the device itself that just coincidentally is getting hit now
04:20marcanone of the things we changed in 6.3 is cpuidle, so it could even be just a random timing difference
04:25marcanthe logs just sound like the device itself is resetting, there's no actual errors
04:28marcanI don't see any suspicious commits in usb core or xhci
04:29marcanif I could repro it locally I could outright use a USB analyzer but...
05:04greguu has joined #asahi-dev
05:45sjg has quit [Read error: Connection reset by peer]
05:45msteffen has quit [Read error: Connection reset by peer]
05:45sjg has joined #asahi-dev
05:45jbowen has quit [Write error: connection closed]
05:45msteffen has joined #asahi-dev
05:45joshtaylor has quit [Read error: Connection reset by peer]
05:45joshtaylor has joined #asahi-dev
05:45jbowen has joined #asahi-dev
05:59cylm_ has joined #asahi-dev
06:48steven has quit [Quit: ZNC 1.8.2 - https://znc.in]
06:49steven has joined #asahi-dev
07:12kettenis has joined #asahi-dev
07:15rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
07:15Guest12357 has quit [Quit: Bridge terminating on SIGTERM]
07:18Jamie has joined #asahi-dev
07:18rhysmdnz has joined #asahi-dev
07:18Jamie is now known as Guest12394
07:24cylm_ has quit [Ping timeout: 480 seconds]
07:38kettenis has quit [Ping timeout: 480 seconds]
08:06drubrkletern has quit [Remote host closed the connection]
08:09bps has joined #asahi-dev
08:37nsklaus has joined #asahi-dev
09:00mattgirv has quit [Ping timeout: 480 seconds]
10:03hightower3 has joined #asahi-dev
10:09kettenis has joined #asahi-dev
10:09hightower2 has quit [Ping timeout: 480 seconds]
10:57maz has quit [Ping timeout: 480 seconds]
10:59linamarcan: Umm, your cpuidle stuff is making lockdep very angry ^^;;
11:07povikseems like PM core makes DART go suspend even with active consumers, if those consumers don't support runtime PM
11:07povikthat goes away when a driver for the consumer runs pm_runtime_set_active explicitly
11:07povikcould this be the same issue as the one being worked around for PCIe?
11:08povikthis happens with SIO (yes, that was the SIO boot breakage...) for which pm_runtime_enabled is false though
11:08povikthat's at probe time, maybe pm_runtime_enabled is true, like for those PCIe devices, when the device link is being set up
11:09povikjannau: ^
11:44jannaupovik: did you already rebase on asahi-wip?
11:45jannauis the problem dart-sio gets probed, goes into runtime suspend, power-domain gets shutdown, sio probes?
11:46jannauthat should be fixed in asahi-wip
11:47poviki don't think this is the same issue you have in mind
11:47povikthe dart gets suspended/resumed while sio is in probe
11:47povikand of course it gets hit by the dart resume reset like the pcie devices get
11:47povikso PM core doesn
11:47povik* so PM core doesn't keep DART up while SIO is up
11:50jannauare dart-sio and dart in different power-domains? I don't think runtime-pm state of dart-sio matters as long as the pd is on
11:51povikit does because of this reset
11:51jannauthat's not the problem for pcie though
11:51povikand the same DART will get resumed because of ADMAC, which shares it
11:56VinDuv has quit [Quit: ZNC 1.8.2+deb2+b1 - https://znc.in]
11:56VinDuv has joined #asahi-dev
12:07jannauwe might want DL_FLAG_RPM_ACTIVE for every device. I suppose that fixes sio
12:07cafebabe has quit [Quit: Connection closed for inactivity]
12:07jannauafter checking Documentation/driver-api/device_link.rst
12:08povikthat would fix it, but my understanding is DL_FLAG_RPM_ACTIVE would keep the DART up even for devices that do support runtime PM and are suspended
12:10poviki am surprised it doesn't work as is, that the device link to a consumer not supporting runtime PM doesn't prevent the DART from suspending implicitly
12:10jannau"DL_FLAG_RPM_ACTIVE can be specified to runtime resume the supplier and prevent it from suspending before the consumer is runtime suspended."
12:10poviki only read the text above device_link_add
12:10povikwhich wasn't as clear
12:13povikthe qualifier before that makes me a little uneasy though: 'Two other flags are specifically targeted at use cases where the device
12:14poviklink is added from the consumer's ``->probe`` callback:'
12:14povikbut maybe it does work as we need
12:15jannauapple_dart_probe_device() is called from the driver core before the consumer's probe callback
12:42roxfan2 has joined #asahi-dev
12:43roxfan has quit [Ping timeout: 480 seconds]
12:43maz has joined #asahi-dev
13:04bps2 has joined #asahi-dev
13:10bps has quit [Ping timeout: 480 seconds]
13:20abd has joined #asahi-dev
14:08gladiac has joined #asahi-dev
15:00bps2 has quit [Ping timeout: 480 seconds]
15:12ChaosPrincessis dcp the only thing rn that is firmware version dependent?
15:13marcanand some stuff in the ADT bindings but only m1n1 cares about that
15:14ChaosPrincessright, and we are about to support 13.3 firmwares soon(tm)?
15:14marcanlina: argh :(
15:14marcanI'll take a look
15:17ChaosPrincesswill we have a path to update 12.3 machines to newer versions?
15:23marcanonce it's useful, yes
15:23marcaninitially only the new machines will get 13.3
15:23ChaosPrincessim sorta maybe looking at another ip block that may end up being firmware dependent
15:24ChaosPrincessidk for sure yet, but seems to be
15:25_jannau_there goes my plan to support dcpext only on 13.3 or later (more out of annoyance than difficulty to support both)
15:25marcanlina: see https://lore.kernel.org/linux-arm-kernel/20220726145430.bfwidmw6xmeppbfb@bogus/ , looks like it's known broken upstream (turn off CONFIG_PROVE_RAW_LOCK_NESTING to shut it up)
15:26marcan_jannau_: dcpext would be a very good reason to call it useful
15:26cylm_ has joined #asahi-dev
15:26marcanwhat I mean is that we'll probably release in O(1-2 weeks) for M2 Pro/Max only
15:26marcanI don't expect dcpext to be done by then :p
15:31_jannau_not sure, it seems to work fine ignoring type-c/atc-phy. I'm currently down to tipd behaves differently on every other connect but at least consistently
15:31kettenisI should have some time to get u-boot in releasable shape next week
15:32marcanstill need to fix oslog but if you're bored: for MTP, we need to allocate oslog out of the end of the sram region in the DT
15:32_jannau_but yes, 1-2 weeks is not enough time to test it on other systems and fix issues
15:32marcaninstead of RAM
15:32marcansince that sticks in linux, and forwarding RAM mappings is a major PITA
15:32balrog has quit [Quit: Bye]
15:32marcanalso the size field extends down 12 more bits in the message, the DVA isn't as large as we thought
15:32marcansize is in bytes
15:33marcan(not 4k pages)
15:33kettenisthat makes more sense
15:33marcanthere's plenty of space at the end of SRAM so it's perfectly fine to steal some space for oslog there IMO
15:34kettenisyou're going to change the linux driver to do that?
15:34jbowen has quit []
15:34marcanthe linux driver doesn't have to do anything :)
15:34marcansince it never chainloads anything else and will inherit that from u-boot if chainloaded
15:34jbowen has joined #asahi-dev
15:35marcan(need to fix the size field though, just that)
15:36_jannau_it would make sense allocate from sram on linux as well so that the behavior between us booting with chainload/HV and user install with u-boot is consistent
15:36ChaosPrincesswhat does sram mean in this context? s=static or s=shared?
15:36marcanboth I guess?
15:36ChaosPrincesss=static sounds expensive
15:37marcan_jannau_: doesn't make much of a difference though
15:37marcanChaosPrincess: eh, it's 1MB
15:37marcanI mean I guess it could be eDRAM but I don't know if they do that
15:38balrog has joined #asahi-dev
15:39gordonfreeman has joined #asahi-dev
15:42gordonfreemanhello! I'm trying to move the single step handler of m1n1 from py to c. single stepping is ok, but I couldn't handle the SSTEP_LOWER exception which comes after stepping.
15:43gordonfreemanI'm actually moving the py code to c verbatim. The issue in the python comment in handle_step happens in C, too: "not sure why MDSCR_EL1.SS needs to be disabled here but otherwise if also SPSR.SS=0 no instruction will be executed after eret and instead a debug exception is generated again"
15:45gordonfreemanwhat I currently do in handle_step C version is: u64 mdscr_el1 = BIT(15); // MDE=1 msr(MDSCR_EL1, mdscr_el1); return true;
15:45gordonfreemanthis prevents the debug exception generation but also the kernel panics after the first handling.
15:45gordonfreemanAm I missing something?
16:00povikjannau: reading further in drivers/base/core.c and power/runtime.c that ACTIVE flag indeed looks like what we want
16:10mkurzmarcan: you say "...release in O(1-2 weeks) for M2 Pro/Max only...", are there also plans to release the gpu/mesa fixes for m1 pro/max as well soon? Thanks!
16:15gladiac has quit [Quit: k thx bye]
16:18svenmarcan: well, I technically have about two weeks of overtime that I can take as vacation now :D
16:19svenprobably won’t do that because I’d go crazy fighting against the atcphy and typec mess for two weeks to get dcpext working ;)
16:26jannausven: do I remember correctly that you got at some point TimingElements with empty ColorModes for dcpext? do you remember what caused that? I see that currently on 13.3
16:38svenyeah, that was when the dptx dance was incorrect
16:38svenI didn’t know about the dcp->ap calls then iirc
16:50jannauhmm, those are still coming through and the identifiers should be correct but I guess some message could have changed as well
16:51svenis that on a m1?
16:52jannauand there is still the sub header version increased to 4 and possibly 2 additional bytes of padding in the epic cmd header
16:52jannauyes, m1 mac mini
16:59crabbedhaloablut has quit []
17:01crabbedhaloablut has joined #asahi-dev
17:13povikwrote up what we know in a patch:
17:28pthariensflame has joined #asahi-dev
17:33jannaufixed, got the reply to the max lane count wrong
17:33pthariensflame has quit [Quit: Textual IRC Client: www.textualapp.com]
17:58brolin has joined #asahi-dev
18:26gordonfreeman has quit [Remote host closed the connection]
19:28brolin has quit [Ping timeout: 480 seconds]
19:30brolin has joined #asahi-dev
19:39brolin has quit [Ping timeout: 480 seconds]
19:40brolin has joined #asahi-dev
20:01brolin has quit [Ping timeout: 480 seconds]
20:12roxfan has joined #asahi-dev
20:15roxfan2 has quit [Ping timeout: 480 seconds]
20:29bcrumb has joined #asahi-dev
20:36abd has quit [Ping timeout: 480 seconds]
20:37bcrumb has quit [Quit: WeeChat 3.8]
21:09abd has joined #asahi-dev
21:34brolin has joined #asahi-dev
21:49kettenis has quit [Ping timeout: 480 seconds]
21:55cylm_ has quit [Ping timeout: 480 seconds]
22:12kettenis has joined #asahi-dev
22:17bcrumb has joined #asahi-dev
22:17bcrumb has quit []
22:18eiln has quit [Read error: Connection reset by peer]
23:36kettenis has quit [Ping timeout: 480 seconds]
…