#asahi-dev · 2026-08-09 09:38
nickchandon't worry about pre M-series AGX too much, it would be... some time at least#asahi-dev · 2026-08-06 16:53
nickchanwith proper patch on the list smartctl works reliably on apple a11 now (well "works" to the extent it is on M1)#asahi-dev · 2026-08-06 16:19
nickchanwhen i try to bring up the controller#asahi-dev · 2026-08-06 16:19
nickchannit: I have seen that panic and warning in nvme on a11 many many times#asahi-dev · 2026-08-06 13:43
nickchanunlike m1 the bit that disable validation don't exist so it never worked before this#asahi-dev · 2026-08-06 13:40
nickchansven: apple a11 nvme also needed same quirk to get the temperature in sensors#asahi-dev · 2026-07-28 14:32
nickchanin any case, 14 commits later i finally did all the prerequesites at least I think so time to lock in on that disk driver#asahi-dev · 2026-07-28 14:29
nickchanand the oslog that does reuse the top 8 bits don't follow this convention, APPLE_RTKIT_OSLOG_SIZE is bit 55:36 so it is irrelvant#asahi-dev · 2026-07-28 14:28
nickchanso im fully convinced that really is supposed to be 4 bit wide#asahi-dev · 2026-07-28 14:28
nickchanthe thing that starts at bit 52 is also used as a message type within that endpoint for syslog, crashlog, and ioreport and none of them exceed 0xf#asahi-dev · 2026-07-28 14:25
nickchanit seems the upper 8 bits did get reused in the OSLOG endpoint though, and that starts at 56#asahi-dev · 2026-07-28 14:25
nickchannone of the fields actually starts at 60#asahi-dev · 2026-07-28 14:18
nickchani sure hope these devices get removed next year because somehow the a8 homepod survives into version 27#asahi-dev · 2026-07-28 14:13
nickchansome homepod and apple tv#asahi-dev · 2026-07-28 14:13
nickchanand the reason is the top 8 bits of the mailbox message is used for endpoint on 64 bit mailboxes and there are such device still supported in latest#asahi-dev · 2026-07-28 14:12
nickchanthe list in linux only goes up to 0xb so it would fit within 4 bits#asahi-dev · 2026-07-28 14:12
nickchanmanagement message type#asahi-dev · 2026-07-28 14:11
nickchansven: have you seen RTKit management message > 0xf ? linux says it is bit [59:52] but i believe it is bit [55:52]#asahi-dev · 2026-07-27 18:33
nickchanthe default 600 mhz is probably too slow#asahi-dev · 2026-07-27 18:33
nickchanand if someone are to check the fast uart on m1, they will need to increase cpu speed first#asahi-dev · 2026-07-27 18:18
nickchanit does work on a11 but i do not have the physical uart on m1 to tell if it could work there#asahi-dev · 2026-07-27 18:18
nickchanand a10 don't want to work for whatever reason that doesn't seem to be cpu speed related#asahi-dev · 2026-07-27 18:17
nickchanon some devices i could make the proxy uart go to 3 mbps but not on every device i have some devices are just too slow to handle that#asahi-dev · 2026-07-27 18:17
nickchanhttps://paste.debian.net/hidden/6e952dc9#asahi-dev · 2026-07-27 18:16
nickchanoversample rate = 0x10 - bit [19:16] of UBDRIV#asahi-dev · 2026-07-27 18:16
nickchanbit [19:16] of UBDRIV seems to control the oversampling rate#asahi-dev · 2026-07-27 18:14
nickchanso in my inability to make usb work properly on the pre-M1 devices i did found a way to make uart go faster than 1.5 mbps#asahi-dev · 2026-07-27 18:13
nickchanthose versions are when sptm is added to the device#asahi-dev · 2026-07-27 15:15
nickchanthe only sensible order is here is iPhone > iPad > Macs, and New > Old#asahi-dev · 2026-07-27 15:12
nickchanmy table on things that could have sptm, but do not have them in every version: https://gist.github.com/asdfugil/101ae32734938cf2c186ff330fba1e97#asahi-dev · 2026-07-26 11:33
nickchanokay and a11 works in polled mode so same as m1 as far as the patch is concerned#asahi-dev · 2026-07-26 11:32
nickchanchaos_princess: "fifo interrupt does not work on m1" ... I suppose it's like this -> [ 1.383385] spmi spmi-0: failed to wait for RX FIFO not empty #asahi-dev · 2026-07-26 11:03
nickchanclang version 22.1.8 (Fedora 22.1.8-4.fc44) asahi fedora clang even#asahi-dev · 2026-07-26 10:50
nickchanit doesn't compile with clang :(#asahi-dev · 2026-07-26 10:46
nickchanI only have that upstream so that's fine#asahi-dev · 2026-07-26 10:46
nickchanthe controller#asahi-dev · 2026-07-26 10:46
nickchanum, it's maxItems: 1 in the bindings#asahi-dev · 2026-07-26 10:45
nickchanclaims to be compatible with apple,t8015-spmi which is upstream so I need to take a look#asahi-dev · 2026-07-26 10:44
nickchanchaos_princess: For your SPMI series how to get spmi bus' interrupt from ADT?#asahi-dev · 2026-07-24 13:17
nickchanit's probably safe to set it everywhere#asahi-dev · 2026-07-24 13:17
nickchan1<<10 in UCON is set on s5l8960x#asahi-dev · 2026-07-23 17:44
nickchanit's been running for 2hrs#asahi-dev · 2026-07-23 17:43
nickchanformat check for 4level paging is still running for some reason#asahi-dev · 2026-07-23 16:42
nickchannot from a video#asahi-dev · 2026-07-23 16:06
nickchan that information is from dosdude1#asahi-dev · 2026-07-23 16:03
nickchanYou need at least 96 if the board does not have a battery#asahi-dev · 2026-07-23 16:02
nickchanor 100#asahi-dev · 2026-07-23 16:02
nickchanbarusu: did you use a 96 watt usb c adapter?#asahi-dev · 2026-07-23 12:27
nickchanyuka: can ignore then I must have made an assumption from the comments saying s5l is "close enough to s3c2410 for earlycon"#asahi-dev · 2026-07-23 07:45
nickchani don't think it's ever figured out but clearing that bit could be interesting for another clock