#
…

2023-04-12

00:04deteg1337 has quit [Ping timeout: 480 seconds]
00:12deteg1337 has joined #asahi-dev
00:15sam_- has quit []
00:15ChaosPrincessIdk, maybe. The mipi im talking about is the one used to connect the touchbar screen.
00:17sam_ has joined #asahi-dev
00:20kettenis_ has quit [Ping timeout: 480 seconds]
00:21deteg1337 has quit [Ping timeout: 480 seconds]
00:28deteg1337 has joined #asahi-dev
00:37deteg1337 has quit [Ping timeout: 480 seconds]
00:42Thsf has joined #asahi-dev
00:45deteg1337 has joined #asahi-dev
00:54deteg1337 has quit [Ping timeout: 480 seconds]
00:57A_L_I_C_E has quit [Remote host closed the connection]
01:02deteg1337 has joined #asahi-dev
01:15Swiftloke has quit [Quit: ABANDON SHIP!]
01:15Swiftloke has joined #asahi-dev
01:16Thsf has quit [Ping timeout: 480 seconds]
01:32raveling has quit [Ping timeout: 480 seconds]
01:58marcanif it has a dart yeah, known problem
01:58marcanjust work around it for now
01:58marcanI'll fix this in core code at some point
01:58marcanit's biting us everywhere, moreso since 6.2
01:59marcanChaosPrincess: ^
01:59marcanjannau: yes, DART probing first is what causes it
01:59marcanthis needs to be fixed in core genpd code to refuse to power down any PDs that have fwlinks to devices that haven't probed yet if they were on at boot time
02:00marcanthat's the only sane solution here
02:00marcanChaosPrincess: just mark the relevant PDs as apple,always-on; for now
02:01marcanCoolStar: webcam is behind a giant firmware blob so it doesn't matter much what link it uses behind the scenes
02:01marcanI think the socs support MIPI and DP
02:47darkapex1 has quit []
02:48darkapex has joined #asahi-dev
03:30deteg1337 has quit [Ping timeout: 480 seconds]
03:39deteg1337 has joined #asahi-dev
03:47deteg1337 has quit [Ping timeout: 480 seconds]
03:54deteg1337 has joined #asahi-dev
04:03deteg1337 has quit [Ping timeout: 480 seconds]
04:11CoolStarmarcan: I see; I guess even if it's MIPI, yeah it doesn't matter if the CIO2 and IPU are both on the same controller (unlike Intel where it's separate)
04:12deteg1337 has joined #asahi-dev
04:21deteg1337 has quit [Ping timeout: 480 seconds]
04:38deteg1337 has joined #asahi-dev
04:46deteg1337 has quit [Ping timeout: 480 seconds]
04:51newiz has joined #asahi-dev
04:53deteg1337 has joined #asahi-dev
05:01deteg1337 has quit [Ping timeout: 480 seconds]
05:10deteg1337 has joined #asahi-dev
05:11A_L_I_C_E has joined #asahi-dev
05:14marcanCoolStar: in general you can access the hardware directly if you want, but all of Apple's coprocessors also access the hardware they are responsible for directly and expect to do so
05:15marcanso unless we're bypassing firmware (which I'm very reluctant to do, both because then we don't have a consistent RE story and because firmware usually helps upwards compat), we don't have to care
05:15marcanI think the only case where we've had a plausible "byassing firmware is OK" story so far is ANE, though I still have my doubts about that one
05:18deteg1337 has quit [Ping timeout: 480 seconds]
05:25chadmed_R and Jamie were going down a path of doing so with AV{D,E} too
05:26deteg1337 has joined #asahi-dev
05:35deteg1337 has quit [Ping timeout: 480 seconds]
05:38chadmed_ has quit [Remote host closed the connection]
05:39cylm has quit [Ping timeout: 480 seconds]
05:43deteg1337 has joined #asahi-dev
05:43chadmed_ has joined #asahi-dev
05:48marcanAVD I can see yeah, AVE I'm not so sure
05:52deteg1337 has quit [Ping timeout: 480 seconds]
05:55kedde has quit [Ping timeout: 480 seconds]
06:00deteg1337 has joined #asahi-dev
06:08deteg1337 has quit [Ping timeout: 480 seconds]
06:26kedde has joined #asahi-dev
06:31chadmed_ has quit [Quit: Page closed]
06:57chipxxx has joined #asahi-dev
06:58chipxxx has quit [Remote host closed the connection]
06:58chipxxx has joined #asahi-dev
07:00deteg1337 has joined #asahi-dev
07:02deteg1337 has quit []
07:28svenfwiw, technically pcie should be a clock controller and dart should have that APPCLK as a dependency
07:28svenit doesn’t matter for the pcie darts so far because appclk is already enabled inside m1n1
07:34newiz has quit [Quit: Konversation terminated!]
07:38kettenis has joined #asahi-dev
07:59bps has joined #asahi-dev
08:01CoolStarmarcan: yeah understandable given there can be performance penalties from bypassing coprocessors (I tried bitbanging I2S on Skylake once -- it didn't look pretty in terms of how inefficient it was)
08:03ChaosPrincessthese machines do not have it, but on x86 dont you also have a problem that smm might come and say hello, and break all your timings.
08:04CoolStarChaosPrincess: yes, but the machines I've dealt with run coreboot so SMM is fairly minimal
08:05i509vcb has quit [synthon.oftc.net weber.oftc.net]
08:05nafod has quit [synthon.oftc.net weber.oftc.net]
08:05Hibyehello_ has quit [synthon.oftc.net weber.oftc.net]
08:05richyliu2 has quit [synthon.oftc.net weber.oftc.net]
08:05skipwich has quit [synthon.oftc.net weber.oftc.net]
08:05nopeslide13 has quit [synthon.oftc.net weber.oftc.net]
08:05compassion has quit [synthon.oftc.net weber.oftc.net]
08:05systwi has quit [synthon.oftc.net weber.oftc.net]
08:05Chainsaw has quit [synthon.oftc.net weber.oftc.net]
08:05snek has quit [synthon.oftc.net weber.oftc.net]
08:05Puto has quit [synthon.oftc.net weber.oftc.net]
08:05elvishjerricco has quit [synthon.oftc.net weber.oftc.net]
08:05amada95 has quit [synthon.oftc.net weber.oftc.net]
08:05cptcobalt has quit [synthon.oftc.net weber.oftc.net]
08:05enron has quit [synthon.oftc.net weber.oftc.net]
08:05rbenua has quit [synthon.oftc.net weber.oftc.net]
08:05azenla has quit [synthon.oftc.net weber.oftc.net]
08:05cds has quit [synthon.oftc.net weber.oftc.net]
08:05supay has quit [synthon.oftc.net weber.oftc.net]
08:05dubiousness has quit [synthon.oftc.net weber.oftc.net]
08:05tom-w has quit [synthon.oftc.net weber.oftc.net]
08:05krbtgt has quit [synthon.oftc.net weber.oftc.net]
08:05coder_kalyan has quit [synthon.oftc.net weber.oftc.net]
08:05jabashque_ has quit [synthon.oftc.net weber.oftc.net]
08:05qyousef has quit [synthon.oftc.net weber.oftc.net]
08:05jesse-s has quit [synthon.oftc.net weber.oftc.net]
08:05djorz has quit [synthon.oftc.net weber.oftc.net]
08:05cafebabe has quit [synthon.oftc.net weber.oftc.net]
08:05opticron has quit [synthon.oftc.net weber.oftc.net]
08:05psanford has quit [synthon.oftc.net weber.oftc.net]
08:05skippypaws has quit [synthon.oftc.net weber.oftc.net]
08:05d4ve has quit [synthon.oftc.net weber.oftc.net]
08:05handlerug has quit [synthon.oftc.net weber.oftc.net]
08:05alethkit has quit [synthon.oftc.net weber.oftc.net]
08:05NekomimiScience has quit [synthon.oftc.net weber.oftc.net]
08:05khilman has quit [synthon.oftc.net weber.oftc.net]
08:05_alice has quit [synthon.oftc.net weber.oftc.net]
08:05mmarchini_____ has quit [synthon.oftc.net weber.oftc.net]
08:05robher has quit [synthon.oftc.net weber.oftc.net]
08:05jonmasters__ has quit [synthon.oftc.net weber.oftc.net]
08:05tardyp has quit [synthon.oftc.net weber.oftc.net]
08:05piroko has quit [synthon.oftc.net weber.oftc.net]
08:05Manouchehri_ has quit [synthon.oftc.net weber.oftc.net]
08:05abby has quit [synthon.oftc.net weber.oftc.net]
08:05dk_ has quit [synthon.oftc.net weber.oftc.net]
08:05steev has quit [synthon.oftc.net weber.oftc.net]
08:05linxz has quit [synthon.oftc.net weber.oftc.net]
08:05arnd has quit [synthon.oftc.net weber.oftc.net]
08:05isinyaaa has quit [synthon.oftc.net weber.oftc.net]
08:05eric_engestrom has quit [synthon.oftc.net weber.oftc.net]
08:05esden has quit [synthon.oftc.net weber.oftc.net]
08:05austriancoder has quit [synthon.oftc.net weber.oftc.net]
08:05daniels has quit [synthon.oftc.net weber.oftc.net]
08:05Lightsword has quit [synthon.oftc.net weber.oftc.net]
08:05steven has quit [synthon.oftc.net weber.oftc.net]
08:05pzc has quit [synthon.oftc.net weber.oftc.net]
08:05lucic71 has quit [synthon.oftc.net weber.oftc.net]
08:05hays has quit [synthon.oftc.net weber.oftc.net]
08:06i509vcb has joined #asahi-dev
08:06nafod has joined #asahi-dev
08:06Hibyehello_ has joined #asahi-dev
08:06richyliu2 has joined #asahi-dev
08:06skipwich has joined #asahi-dev
08:06compassion has joined #asahi-dev
08:06nopeslide13 has joined #asahi-dev
08:06systwi has joined #asahi-dev
08:06elvishjerricco has joined #asahi-dev
08:06dubiousness has joined #asahi-dev
08:06steven has joined #asahi-dev
08:06lucic71 has joined #asahi-dev
08:06abby has joined #asahi-dev
08:06alethkit has joined #asahi-dev
08:06cds has joined #asahi-dev
08:06amada95 has joined #asahi-dev
08:06handlerug has joined #asahi-dev
08:06dk_ has joined #asahi-dev
08:06coder_kalyan has joined #asahi-dev
08:06d4ve has joined #asahi-dev
08:06skippypaws has joined #asahi-dev
08:06krbtgt has joined #asahi-dev
08:06enron has joined #asahi-dev
08:06psanford has joined #asahi-dev
08:06hays has joined #asahi-dev
08:06pzc has joined #asahi-dev
08:06Manouchehri_ has joined #asahi-dev
08:06djorz has joined #asahi-dev
08:06opticron has joined #asahi-dev
08:06Lightsword has joined #asahi-dev
08:06piroko has joined #asahi-dev
08:06arnd has joined #asahi-dev
08:06supay has joined #asahi-dev
08:06Puto has joined #asahi-dev
08:06tom-w has joined #asahi-dev
08:06cafebabe has joined #asahi-dev
08:06qyousef has joined #asahi-dev
08:06jonmasters__ has joined #asahi-dev
08:06khilman has joined #asahi-dev
08:06daniels has joined #asahi-dev
08:06mmarchini_____ has joined #asahi-dev
08:06austriancoder has joined #asahi-dev
08:06_alice has joined #asahi-dev
08:06robher has joined #asahi-dev
08:06esden has joined #asahi-dev
08:06cptcobalt has joined #asahi-dev
08:06tardyp has joined #asahi-dev
08:06snek has joined #asahi-dev
08:06eric_engestrom has joined #asahi-dev
08:06jesse-s has joined #asahi-dev
08:06azenla has joined #asahi-dev
08:06NekomimiScience has joined #asahi-dev
08:06linxz has joined #asahi-dev
08:06steev has joined #asahi-dev
08:06Chainsaw has joined #asahi-dev
08:06isinyaaa has joined #asahi-dev
08:06rbenua has joined #asahi-dev
08:06jabashque_ has joined #asahi-dev
08:27c10l has quit [Quit: Bye o/]
08:28c10l has joined #asahi-dev
08:28chadmed has quit [Ping timeout: 480 seconds]
08:57chadmed has joined #asahi-dev
09:09hightower2 has quit [Ping timeout: 480 seconds]
09:12nyilas has joined #asahi-dev
09:33nsklaus has joined #asahi-dev
09:35sventhere might be power consumption penalties if we bypassed co-processors but I doubt we'd see a measurable performance penalty
09:37sventhe biggest issues are that reverse engineering the firmware is usually much more challenging and time consuming than just reverse engineering its interface and that the firmware interface probably helps us when supporting new chip revisions
09:55nyilas has quit [Remote host closed the connection]
09:56nyilas has joined #asahi-dev
10:08cylm has joined #asahi-dev
10:27kettenisspeaking about firmware: I probably should work on getting the asmedia firmware loaded in u-boot
10:27kettenisthat'll save me having to implement it in OpenBSD
10:39Thsf has joined #asahi-dev
10:47Thsf has quit [Ping timeout: 480 seconds]
12:07nsklaus has quit [Ping timeout: 480 seconds]
12:08kettenishmm, the m2 pro mini should not inherit the wifi bits from t600x-j375.dtsi
12:25nsklaus has joined #asahi-dev
12:31raveling has joined #asahi-dev
12:47DarkShadow4444 has quit [Quit: ZNC - https://znc.in]
12:47DarkShadow44 has joined #asahi-dev
13:06hightower2 has joined #asahi-dev
13:25ez has joined #asahi-dev
13:26ez has quit [Remote host closed the connection]
13:36psykose has quit [Remote host closed the connection]
13:37psykose has joined #asahi-dev
14:19raveling has quit [Ping timeout: 480 seconds]
14:27hightower2 has quit [Ping timeout: 480 seconds]
14:28jannaukettenis: it should overwrite the relevant bits but I forgot to look for the board type
14:34A_L_I_C_E has quit [Ping timeout: 480 seconds]
14:35A_L_I_C_E has joined #asahi-dev
14:45nsklaus has quit [Quit: disappears in a cloud of smoke]
15:00kedde has quit [Ping timeout: 480 seconds]
15:10kettenisjannau: the compatibles are also wrong
15:11stipa is now known as Guest10748
15:11stipa has joined #asahi-dev
15:13Guest10748 has quit [Read error: Connection reset by peer]
15:14nsklaus has joined #asahi-dev
15:15gladiac has joined #asahi-dev
15:24raveling has joined #asahi-dev
15:30stipa is now known as Guest10750
15:30stipa has joined #asahi-dev
15:32Guest10750 has quit [Ping timeout: 480 seconds]
16:06bps has quit [Ping timeout: 480 seconds]
16:54cylm has quit [Ping timeout: 480 seconds]
17:00eric_engestromdoes `perf` work for you guys? no matter what I do `perf record` gives me empty records
17:00eric_engestrom> The perf.data data has no samples!
17:06jannaukettenis: t602x/bringup wifi/bt fixes are in https://github.com/AsahiLinux/linux/pull/134
17:06A_L_I_C_E has quit [Ping timeout: 480 seconds]
17:06jannaueric_engestrom: we have seen that before
17:07A_L_I_C_E has joined #asahi-dev
17:07eric_engestromjannau: glad to know I'm not alone :]
17:07eric_engestromwas there a fix or workaround?
17:14jannauif I remembered. with hw counters you need to taskset either efficiency or performance cores
17:14jannauthat said, a plain `perf record sleep 1` produces samples for me on m1 max
17:16linuxgemini13 has quit []
17:16linuxgemini13 has joined #asahi-dev
17:16linuxgemini13 is now known as linuxgemini
17:16jannaudoes perf record -e cpu-clock ... work?
17:17eric_engestromjannau: nope :/
17:22jannau/proc/sys/kernel/perf_event_paranoid is set to allow user event sampling?
17:27jannausomehow the blizzard pmu went missing on m2 pro
17:31A_L_I_C_E has quit [Ping timeout: 480 seconds]
17:31jannauah, the compatible is missing a 'z'
17:36Stary has quit [Ping timeout: 480 seconds]
17:36jannaueric_engestrom: `perf stat sleep 1` is generally a nice way to test if perf works as expected. can you try that on both the performance and efficiency cores
17:38jannauit should work on one or the other. I'm not sure how perf/ the kernel selects which counters are cycles/instructions but it mappes only to either one
17:39jannausee perf list about which cycle and instruction counters you have
17:39jannauon m2* it's apple_avalanche_pmu/cycles/ and apple_blizzard_pmu/cycles/
17:41Stary has joined #asahi-dev
17:43newiz has joined #asahi-dev
17:47A_L_I_C_E has joined #asahi-dev
17:48kit_ty_kate3 has quit [Quit: WeeChat 3.6]
17:48kit_ty_kate has joined #asahi-dev
17:54eric_engestromjannau: /proc/sys/kernel/perf_event_paranoid was 2, but setting it to 0 makes no difference
17:54eric_engestromthanks for `perf stat`, thats' much simpler than `perf record ... && perf report ...` :)
17:54eric_engestromI'm on an m2 air
17:54jannaueric_engestrom: try -1
17:55eric_engestromnot sure why I typed 0, -1 is what I tried :]
17:55A_L_I_C_E has quit [Ping timeout: 480 seconds]
17:56eric_engestrom"can you try that on both the performance and efficiency cores" how do I do that?
17:56jannau`taskset -c $CORE perf stat sleep 1` with 0 and 4 for $CORE
17:56A_L_I_C_E has joined #asahi-dev
17:57jannauworks for me on m2 mbp 13"
17:58kedde has joined #asahi-dev
17:58eric_engestromcore 4 gives me the same result as nothing, but core 0 allows me to have cycles & instructions counts!
17:58eric_engestromI have to go now, but I'll check tomorrow if that's enough for what I need
17:59eric_engestromthanks jannau :)
18:13A_L_I_C_E has quit [Read error: Connection reset by peer]
18:14newiz has quit [Ping timeout: 480 seconds]
18:18SalimTer- has joined #asahi-dev
18:19salimterryli has quit [Ping timeout: 480 seconds]
18:37gladiac has quit [Quit: k thx bye]
18:42kedde has quit [Ping timeout: 480 seconds]
19:00kedde has joined #asahi-dev
19:07hightower2 has joined #asahi-dev
20:14salimterryli has joined #asahi-dev
20:15raveling has quit [Ping timeout: 480 seconds]
20:19SalimTer- has quit [Ping timeout: 480 seconds]
20:29MajorBiscuit has joined #asahi-dev
20:37raveling has joined #asahi-dev
20:49raveling has quit [Ping timeout: 480 seconds]
20:58raveling has joined #asahi-dev
21:09raveling has quit [Ping timeout: 480 seconds]
21:16nyilas has quit [Remote host closed the connection]
21:18raveling has joined #asahi-dev
21:26raveling has quit [Ping timeout: 480 seconds]
21:28A_L_I_C_E has joined #asahi-dev
21:37raveling has joined #asahi-dev
21:41MajorBiscuit has quit [Quit: WeeChat 3.6]
21:45raveling has quit [Ping timeout: 480 seconds]
21:53raveling has joined #asahi-dev
22:17bps has joined #asahi-dev
22:23nsklaus has quit [Quit: ZZZzzz…]
22:25bps has quit [Ping timeout: 480 seconds]
23:04chadmed_ has joined #asahi-dev
23:47maria285474 has joined #asahi-dev
23:53maria28547 has quit [Ping timeout: 480 seconds]
…