#
…

2021-11-01

00:09marcansven: hey it protects it RAM at least :p
00:09marcan(but not from DMA)
00:12user982492 has joined #asahi-dev
00:23Augur[m] has joined #asahi-dev
00:39goldsoultheory_ has joined #asahi-dev
00:44goldsoultheory has quit [Ping timeout: 480 seconds]
01:37user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
01:44user982492 has joined #asahi-dev
01:56rohin2 has quit [Ping timeout: 480 seconds]
01:59j_eymarcan: still planning the strream to start now?
02:05user982492 has quit [Read error: Connection reset by peer]
02:05user982492_ has joined #asahi-dev
02:18marcanhaving one of those bad technology days, but yes
02:18robinp[m] has joined #asahi-dev
02:18j_ey(yes according to twitter from 30mins ago)
02:19rohin2 has joined #asahi-dev
02:27phiretypical technology, always letting us down
02:39amwmarcan: Take your time, get it right or make it a feature and talk us through the issues :-)
02:44chadmednew stream suggestion: brining up work stream on youtube
02:45marcanreenber that stream chat is in #asahi-stream
02:45marcan*remember
02:45chadmedoops yeah wrong chat :s
03:10l3k has joined #asahi-dev
03:24l3k has quit [Quit: leaving]
03:34robinp[m] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
03:59mika_ has joined #asahi-dev
04:00kov has quit [Quit: Coyote finally caught me]
04:00kov has joined #asahi-dev
04:01phiologe has joined #asahi-dev
04:01mika_ has quit []
04:02mika_ has joined #asahi-dev
04:02robinp[m] has joined #asahi-dev
04:04PhilippvK has quit [Ping timeout: 480 seconds]
04:15user982492_ has quit [Remote host closed the connection]
04:17robinp[m] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
04:19user982492 has joined #asahi-dev
04:36mika_ has quit [Quit: mika_]
04:43robinp[m] has joined #asahi-dev
04:44robinp[m] has quit []
04:47robinp[m] has joined #asahi-dev
05:01robinp[m] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
05:12rohin2 has quit [Ping timeout: 480 seconds]
05:22robinp[m] has joined #asahi-dev
06:48user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
06:55user982492 has joined #asahi-dev
07:33goldsoultheory_ has quit [Quit: Leaving]
08:46FireFox317 has quit [Quit: ZNC 1.7.2+deb3 - https://znc.in]
08:48FireFox317 has joined #asahi-dev
09:18gabuscus has quit [Remote host closed the connection]
09:22gabuscus has joined #asahi-dev
09:25bps2 has joined #asahi-dev
09:36user982492_ has joined #asahi-dev
09:36user982492 has quit [Remote host closed the connection]
09:40user982492_ has quit []
10:43aleasto has joined #asahi-dev
10:51VinDuv_ has joined #asahi-dev
10:52VinDuv has quit [Ping timeout: 480 seconds]
10:52VinDuv_ is now known as VinDuv
11:58marcansven: so yeah AIC changed and I can't get it to work yet :p
11:58marcanneed to poke around more
11:59svenbut i heard you made DART and USB work
12:01marcansven: pushed
12:03svenhrm, so different addr shift in the PTE and mandatory subpage protection
12:04marcanapparently
12:04svenah, and usb just works once you use the new dart
13:05VinDuvHmm, I updated m1n1 on the MBP and can now get the USB-serial working (yay!), but only with a USB-A to USB-C cable. The Thunderbolt cable I use with macvdmtool doesn’t seem to work. Is this expected?
13:09VinDuvoh wait I’m an idiot
13:09VinDuvI was plugging it on the debug port
13:49j_eyanyone know what might be wrong with these few lines https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/pinctrl/pinctrl-apple-gpio.c#n112 I got a comment saying "This is incorrect.", just havent figured out why
14:01gpanders[m] has joined #asahi-dev
14:32svenI guess they changed the PTE shift because they have phys addrs now that are larger than 40 bits now
14:33j_eyim confused by #define DART_PTE_OFFSET_SHIFT 14
14:34j_eyisnt that meant to be 10 for t600?
14:35svenDon’t think ao
14:35svenfield_prep takes care of that
14:36j_eyoh i was misunderstanding the 14 there
14:36svenfor t8020 it will be >> 14 and then << 14
14:36svenAnd for t6000 >> 14 and then << 10
14:36j_eyits always 14 for the final addr, 16k pages
14:46jbowen has joined #asahi-dev
14:50svendrivers/pinctrl/pinctrl-apple-gpio.c: In function ‘apple_gpio_irq_set_type’:
14:50sven././include/linux/compiler_types.h:335:38: error: call to ‘__compiletime_assert_289’ declared with attribute error: FIELD_PREP: value too large for the field
14:50svenj_ey: ^-- huh
14:50j_eygcc 9 or 8?
14:50j_eypovik hit that too, but I thought it might be weird/old gcc
14:51svenoh.. i think it's the EINVAL
14:51svenlet me fix that
14:52j_eyI dont understand how this can be a compile time thing..
14:52svengcc figures out the range for apple_gpio_irq_set_type
14:52sventhen there's a compile time check in FIELD_PREP that get trigged because -EINVAL > 0b111
14:53j_eyI guess that makes sense..
14:53kettenisannoying when compilers get too smart
14:53svenj_ey: want me to send a patch or do you want to do it?
14:54j_eysven: either is fine, i was looking at sending a patch with andy's changes anyway
14:54gpanders[m]1 has joined #asahi-dev
14:55j_ey(but send the patch/diff here at least, want to see how you solved it)
14:55svensure :D
14:56j_eyim on gcc 10, don't see it. povik had it go away by changing some random bits, so i thought it was a compiler bug
14:57svenheh, looks like i messed up the encoding
14:59j_eylgtm, might as well send it to linux-gpio now, in case some other people hit it sooner than i do the other fixes
14:59svenlet me just fix that dumb encoding error there
15:01j_eygcc-9-‘
15:19kettenisI'm trying to figure out how to detect the various keyboard layouts
15:20kettenismy MBP has this in the ADT:
15:20ketteniskblang-calibration: 33da1001 00004575 00000000 00000000
15:21j_eykblang-calibration = [0x0000B1A1868B0601, 0x0000000000000000]
15:21ketteniswhat keyboard layout is that?
15:23VinDuvFrench: kblang-calibration = [0x0000B6CC42920201, 0x0000000000000000]
15:23marcanja: m1pro_adt.txt: kblang-calibration = [0x0000C1CB72040301, 0x0000000000000000]
15:25jixde mba: kblang-calibration = [0x00002FC513280101, 0x0000000000000000]
15:26kettenisnot seeing a pattern yet...
15:26kettenisso maybe this property isn't related to the keyboard layout
15:26j_eysven: you might have a de mba, whats your kblang-calib?
15:27marcankblang-calibration = << syscfg/ksku >>
15:27sveni have a US layout i think
15:27marcan"keyboard sku" does sound like it's relevant
15:27sveni'd also have to extract its device tree
15:27marcanyou can look it up with ioreg on macos
15:27marcanioreg -p IODeviceTree -l
15:28sven | | "kblang-calibration" = <0104a7e7af5f00000000000000000000>
15:30marcanI bet it's just that second byte
15:30marcanstrings on AppleHIDKeyboard has a list. Unknown, German, French, Japanese keyboard, US International Keyboard, ...
15:31marcanjix had a 1, VinDuv had a 2, I have a 3, sven has a 4
15:31marcan(convenient how we're the first 4 types...)
15:32jixhow did that order happen though? :D
15:32marcangood question!
15:33kettenisand mine would be a 10?
15:33kettenisor rather 16?
15:33VinDuvI bet it’s a code dating back from early Mac OS
15:34marcankettenis: 16: Dutch
15:35rohin2 has joined #asahi-dev
15:35j_eyeither he's right, or he's making it up, but it fits
15:35j_eyI have to find out what "Zhuyin Bopomofo
15:36marcanused in taiwan, in particular
15:36kettenisyes, this one is indeed dutch
15:36j_eybopo mofo :D
15:36kettenis(had to check since I didn't order the machine myself)
15:37j_eypovik: hey, whats yours kblang-calib! probably XXXX16XX
15:54daniels has joined #asahi-dev
15:54jixhmm I wonder for autodetection would you want to chose the german pc or the german apple layout, they match for the labels on the keys (at least for apple not labeling the 3rd layer) but even ascii chars like []{}\ and I think some more are in completely different places
15:55bps2 has quit [Ping timeout: 480 seconds]
15:59svenmarcan: https://github.com/AsahiLinux/linux/commits/dart/t6000 completely untested t6000 dart support
15:59marcanwill test once I fix AIC
16:02j_eysven: hm, you had t8xx dart as 36 bits before, but not 40?
16:02j_ey*but now
16:02svenyeah, i think that was wrong
16:02svenit should've been 40 all along
16:03j_eyworth calling out in the commit message
16:03j_eyi know completely untested etc
16:04sventhe commit messages don't even exist yet
16:04rohin2 has quit [Ping timeout: 480 seconds]
16:04j_eysorry, too eager *_*
16:06sveni'm not even sure i like the check to depend on oas. should work though i hope :D
16:09rohin2 has joined #asahi-dev
16:30bps2 has joined #asahi-dev
16:35povikj_ey: if i understand what i should be looking for, you guessed right
16:35povik | | "kblang-calibration" = <0116ef9616ac00000000000000000000>
16:36j_eynice, too many correct to be a coincidence now :P
16:36povikshame i didn't order US layout like i considered, that would have thrown your guess off :-p
16:40kettenisi keep forgetting that Apple's "Dutch" keyboard layout is just "International English" with an additional Euro sign printed in the 2/@ key
16:41kettenis(there is an official ducth keyboard layout that nobody uses)
16:43jbowen has quit [Ping timeout: 480 seconds]
16:44rohin2 has quit [Ping timeout: 480 seconds]
16:45Bey0ndB1nary has joined #asahi-dev
16:52Bey0ndB1nary has quit []
17:00pipcet[m] has left #asahi-dev [#asahi-dev]
17:02jbowen has joined #asahi-dev
17:24bps2 has quit [Ping timeout: 480 seconds]
17:28marcanpushed basic HV support for t6000 (had to add a page table level, but only on this one, t8103 keeps the old setup since it can't support it)
17:29marcanit mostly gets through m1n1 init; needs a bunch more things around MMIO made dynamic, but at least vUART and RAM work
17:49jbowen has quit [Ping timeout: 480 seconds]
18:00jbowen has joined #asahi-dev
18:17___nick___ has joined #asahi-dev
18:18jmr2 has joined #asahi-dev
18:18___nick___ has quit []
18:20___nick___ has joined #asahi-dev
18:20jmr2 has quit []
18:29user982492 has joined #asahi-dev
18:52quentin[m] has joined #asahi-dev
19:38mazmarcan: what's the PA range on this new toy?
19:39j_ey44 bits
19:40mazthat's a lot more than I anticipated. probably means that they expect to have systems with a lot more than the current 64GB.
19:40j_eymemory also starts a lot higher
19:41mazI expect they moved away from the strict 50% split between ram and IO.
19:51mika_ has joined #asahi-dev
19:51kettenisRAM starts at an address that starts with 0x1, so yes
19:51mika_ has quit [Remote host closed the connection]
20:03jacoxon has joined #asahi-dev
20:26aleasto has quit [Remote host closed the connection]
20:29jbowen has quit [Ping timeout: 480 seconds]
21:08___nick___ has quit [Ping timeout: 480 seconds]
21:44user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
21:57jacoxon has quit []
22:00bps2 has joined #asahi-dev
22:09bps2 has quit [Ping timeout: 480 seconds]
22:12bps2 has joined #asahi-dev
22:23bps2 has quit [Ping timeout: 480 seconds]
22:30bps2 has joined #asahi-dev
23:02user982492 has joined #asahi-dev
23:53X-Scale has joined #asahi-dev
23:59X-Scale` has quit [Ping timeout: 480 seconds]
…