#asahi-dev · 2023-10-07 17:12
kettenisI think that might be the problem#asahi-dev · 2023-10-07 17:12
kettenisreserved.cnt = 0x8 / max = 0x8#asahi-dev · 2023-10-07 16:40
ketteniscalling lmb_dump_all_force(&lmb) in arch/arm/mach-apple/board.c:board_late_init() might give us some clues about what's going wrong#asahi-dev · 2023-10-07 16:33
kettenisI don't immediately see how a non-page-aligned region would break the lmb code#asahi-dev · 2023-10-07 16:22
kettenisso it seems something was added that knocks out most of the memory#asahi-dev · 2023-10-07 16:21
kettenisthat initializes an lmb taking into account reserved memory regions#asahi-dev · 2023-10-07 16:21
kettenisso these are clearly the blocks we allocate in board_late_init() for loading the kernel, initrc , etc.#asahi-dev · 2023-10-07 16:13
kettenisI vaguely remember hitting something like that before#asahi-dev · 2023-10-07 16:13
kettenisbut yes, I think the assumption is that memory regions in the device tree are page aligned#asahi-dev · 2023-10-07 16:12
kettenisnever seen that#asahi-dev · 2023-10-07 15:59
kettenisthanks, the openbsd driver makes it fatail, but I hacked around that#asahi-dev · 2023-10-07 15:37
kettenisdo you see the same on Linux?#asahi-dev · 2023-10-07 15:36
kettenismarcan: it seems the join_pref command fails on bcm4388#asahi-dev · 2023-10-06 20:33
kettenisit took me far too much time I didn't get any kernel output because I left the usb keyboard plugged in...#asahi-dev · 2023-10-06 20:31
kettenisthanks, I'll give that a go later this weekend#asahi-dev · 2023-10-06 20:27
kettenisI think it doesn't support the standard VESA mode#asahi-dev · 2023-10-06 20:27
kettenisnew m1n1 mostly fails to detect my monitor properly; but I've had issues with it before#asahi-dev · 2023-10-06 20:26
kettenisyay, managed to get my m2 pro mini running again with 13.5 OS firmware#asahi-dev · 2023-10-06 17:45
kettenisso the problem is either in Linux-specific code or in some system library#asahi-dev · 2023-10-06 17:44
kettenisnote that chromium works fine on OpenBSD with BTI#asahi-dev · 2023-10-05 20:27
kettenismarcan: was installer_data.json deliberately not updated to include 13.5 as a supported firmware?#asahi-dev · 2023-10-05 17:52
kettenisah, but I need newer OS firmware for this to actually work on my M2 Pro mini#asahi-dev · 2023-10-05 15:57
kettenisnot powering off any unknown power domains yet on OpenBSD, so that should be fine#asahi-dev · 2023-10-05 15:54
kettenisthat's in m1n1 though I suppose?#asahi-dev · 2023-10-05 15:48
kettenisoh, cool, 4388 support; let's see if I can port that to our bwfm(4) driver...#asahi-dev · 2023-10-02 09:27
kettenisso far I've made only minimal modifications to the code#asahi-dev · 2023-10-02 09:26
kettenisjanneg: I've got my port of the dcp driver exchanging messages with the DCP in OpenBSD#asahi-dev · 2023-09-18 10:21
kettenism2 pro mini has the Tp0* keys#asahi-dev · 2023-09-18 10:09
kettenishmm, my m1 mini doesn't enumerate any of the Tp0* keys#asahi-dev · 2023-09-18 10:03
kettenisguess for the core ones the thing to do is to pin a cpu-intensive process to a particular core and see if the corresponding temp sensor changes?#asahi-dev · 2023-09-18 09:48
kettenisadding a key to my openbsd driver is really easy#asahi-dev · 2023-09-18 09:48
kettenischadmed: I can help with figuring some of this out on machines you don't have#asahi-dev · 2023-09-05 11:32
kettenisj'ey: is that what ahdeb[m] was trying to point out? I guess it is in marcan's usb patches then#asahi-dev · 2023-09-05 09:25
kettenisI think there is still some time to sneak this into 2023.10 and even if it misses that it isn't a big issue#asahi-dev · 2023-09-05 09:24
kettenisno worries#asahi-dev · 2023-09-05 09:19
kettenisbut it needs to happen very early on and there are some restrictions there#asahi-dev · 2023-09-05 09:18
kettenismarcan: yeah, it should be possible to build the map from the device tree#asahi-dev · 2023-09-05 09:18
ketteniswould be great if you could send that to the u-boot mailing list#asahi-dev · 2023-09-05 09:17
kettenisjannau: looks good to me#asahi-dev · 2023-09-05 09:15
kettenisjust noticed that u-boot doesn't have m2 ultra support yet; guess marcan booted kernels directly from m1n1#asahi-dev · 2023-09-04 09:44
ketteniswe've been maintaining local patches to implement this in OpenBSD for ~5 years now#asahi-dev · 2023-09-04 09:44
kettenis__jannau__: really unfortunate that this didn't make any progress#asahi-dev · 2023-08-30 09:45
kettenisso gdb might indeed be lying to you#asahi-dev · 2023-08-30 09:44
kettenisbut having a PLT entry start with NOPs is weird#asahi-dev · 2023-08-30 09:43
kettenis(we have BTI enabled for almost everything on OpenBSD)#asahi-dev · 2023-08-30 09:41
kettenis__jannau__: that suggests it jumped into the middle of the PLT stub?#asahi-dev · 2023-08-30 09:39
kettenisnot sure on what timescale this will be all sorted#asahi-dev · 2023-08-30 09:38
kettenisthe bootstd stuff is still somewhat in flux and buggy#asahi-dev · 2023-08-30 09:33
kettenisjannau: that is puzzling indeed; is it at the first instruction of __libc_start_main@plt?#asahi-dev · 2023-08-04 14:34
ketteniswith "lowest pstate" you mean the lowest state advertised in the app tables in the DT isn't it?