#
✕ nick:nnym
#asahi-re · 2026-02-25 19:02
nnymMost sections in the kernel are gone in the kext collection. I'm guessing that it combined or moved them based on their names.
#asahi-re · 2026-02-25 17:29
nnymDoes anybody know what "Missing split seg v2" means? Maybe `kmutil` expects the segments to be split into sections?
#asahi-re · 2026-02-25 16:27
nnymAs it turns out, running `kmutil create` with an m1n1 Mach-O as the kernel tells me what is wrong with it.
#asahi-re · 2026-02-25 15:26
nnymConfirmed: neither `kernel.development.t6041` nor any kext wherewith I linked it has `__PRELINK_INFO` and `kmutil create` produced a kext collection with `__PRELINK_INFO`. The difference is that the section is around 480 bytes in my kext collection and 3.7 M in Apple's.
#asahi-re · 2026-02-25 15:15
nnyms/kernel\ze collection/kext
#asahi-re · 2026-02-25 15:10
nnymXNU's source code contains a single linker option that makes a seemingly empty `__PRELINK_INFO,__info` and it does not seem to put anything there unlike other sections. The property list that I found contains a key `KMBuildVersion` which I suspect to have been generated by `kmutil` because it does not appear in the source code and because of its name. Comments seem to indicate that `__PRELINK_INFO` contains information about th
#asahi-re · 2026-02-25 14:23
nnym`__PRELINK_INFO` is an XML property list in the official kernelcache; perhaps iBoot expects it to be at least an empty XML file.
#asahi-re · 2026-02-25 00:32
nnymSo all of this suggests that it can't load.
#asahi-re · 2026-02-25 00:26
nnymI wouldn't expect a Mach-O that has loaded successfully to behave differently from a raw boot object in this way.
#asahi-re · 2026-02-25 00:24
nnymI just checked: when m1n1 reboots 5 consecutive times, the computer boots into recoveryOS and says that macOS needs to be reinstalled. `iboot-failure-reason` (apparently an NVRAM variable that is unset immediately by recoveryOS) does not appear in the log. `recovery-boot-mode` is `panicmedic`; when the Mach-O fails to load, it's `iboot`.
#asahi-re · 2026-02-24 17:20
nnymiBoot seems to have a CLI? There are some strings that look like command help strings.
#asahi-re · 2026-02-24 16:51
nnym"iboot-failure-reason = 100"
#asahi-re · 2026-02-24 16:44
nnym(Actually, it says to configure a different kernel.)
#asahi-re · 2026-02-24 16:42
nnymIs there nothing that can tell me why the boot failed? recoveryOS just tells me to reinstall macOS or something.
#asahi-re · 2026-02-24 16:36
nnymIn that case, I should see the Apple logo disappear at least for a moment; right? But when it disappears and reappears, it's straight to recovery. Also, the logo seems to go through 3 discrete brightness states per boot and that is all that I see before the computer reboots into recoveryOS.
#asahi-re · 2026-02-24 16:29
nnymYeah but with my Mach-Os only a single attempt is made every time. With raw binaries I get 3 attempts or so.
#asahi-re · 2026-02-24 16:22
nnymIf a Mach-O fails to boot once and the next boot goes straight to recovery, what does that mean exactly? Does it mean that there's a segment that's malformed in a way that looks fine to `kmutil` but breaks when iBoot tries to load it? Or is something that should be there missing?
#asahi-re · 2026-02-19 17:28
nnymJust `__TEXT_EXEC` is fine too.
#asahi-re · 2026-02-19 11:18
nnymDoes anybody have a usable m1n1 Mach-O with `__TEXT_BOOT_EXEC`?
#asahi-re · 2026-02-19 10:26
nnymThanks.
#asahi-re · 2026-02-18 18:08
nnymIs WFI mode 2 deep WFI?
#asahi-re · 2026-02-18 17:35
nnymAlthough there's a write that does not have a substitute so maybe I'm onto nothing.
#asahi-re · 2026-02-18 17:27
nnymThere seems to have been a lot of discussion about `CYC_OVRD` on M4 so I got curious. Based on `_arm64_prepare_for_sleep` in the kernel, s3_1_c15_c7_4 seems to be used instead. And according to my `MSR`-scanning script, it does not occur anywhere else.
#asahi-re · 2026-02-17 16:04
nnymm1n1's exception handler does get called when I try to write to `GXF_CONFIG_EL1`.
#asahi-re · 2026-02-17 16:02
nnym"locked" like the GXF registers or differently?
#asahi-re · 2026-02-17 15:00
nnymSome writes from the proxy cause m1n1 to apparently freeze: there is no response, nothing appears in the framebuffer (it does not appear to be an exception), and the proxy client can't reconnect. For example, this occurs when `chainload.py` sets some cores' RVBARs and when some system registers are written. How can I debug such cases?
#asahi-re · 2026-02-16 13:09
nnymManually?
#asahi-re · 2026-02-16 13:03
nnymHow was `apple_regs.json` generated?
#asahi-re · 2026-02-16 13:03
nnymThanks. I've been trying to find out how to enable GXF on M4 and posted some of my findings here to see whether they were useful; but then I remembered that all Apple extensions are disabled so a GXF register wouldn't be it.
#asahi-re · 2026-02-16 12:18
nnym`sptm.t6041.release.im4p` has a function that looks almost identical. Also, both functions execute 0x00201420 after that. There are 3 registers: 2 apparently contain addresses (exception handlers?) and the other (s3_6_c15_c1_2; I hope that I spelled that correctly) stores `1`.
#asahi-re · 2026-02-16 07:00
nnymActually, it's 25C56.
#asahi-re · 2026-02-16 06:57
nnymThis is the 25A353 kernel.
#asahi-re · 2026-02-16 06:55
nnymI don't know what to make of it.
#asahi-re · 2026-02-16 06:55
nnymI found a function `_gxf_enable` in `kernel.development.t6000` that sets some system registers.
#asahi-re · 2026-02-13 13:46
nnymI thought that relying on `ipsw` to extract all of iBoot from an IPSW surely wouldn't go wrong...
#asahi-re · 2026-02-13 08:55
nnymThe macOS 26.2 IPSW seems to contain only iBootStage2. Does this mean that iBoot1 never changes?
#asahi-re · 2026-02-10 14:30
nnymThe entry point with SPTM enabled seems to be at `__TEXT_BOOT_EXEC` in the 26.2 kernels for T6020 and T6041; in the kernel for T6000 (without SPTM) there is no `__TEXT_BOOT_EXEC`. Has anybody looked at this?
#asahi-re · 2026-02-10 13:29
nnymGood point; thanks.
#asahi-re · 2026-02-10 13:18
nnym`entry` (`__start`) looks like the kernel's entry point but the first parameter—which is supposed to be a pointer to the boot arguments—can be 4. What does this mean?
#asahi-re · 2026-02-10 10:05
nnymI found many occurrences of 2 address adding patterns in ApplePMGR that confuse me a little. I can guess that one adds an offset to a base address and produces an intentionally invalid address if the offset is bigger than 4 GiB; the other checks whether the top 10 bits changed in the sum and sets the top bits to `0xc8a2` but I don't know what that's supposed to do. Can somebody explain? I've ignored them until now as just error
#asahi-re · 2026-02-10 07:03
nnymVirtual UART is exactly what I wanted.
#asahi-re · 2026-02-10 06:32
nnymOh.
#asahi-re · 2026-02-10 06:31
nnymThe diff in `lsusb` is "Bus 005 Device 034: ID 1209:316d Generic m1n1 uartproxy v1.5.0-150-g4562f406-dirty".
#asahi-re · 2026-02-07 11:49
nnymWhat's that? A devicetree in a devicetree?
#asahi-re · 2026-02-04 17:28
nnyms/ps-groups`\zs/ seems/
#asahi-re · 2026-02-04 17:27
nnymApplePMGR seems to determine whether to use `ps-regs` or `ps-groups` (and more) based on the presence and value of the property `edt-property-ver` apparently introduced in M4. The property seems to specify the schema of `/arm-io/pmgr` or some part of it but I don't know what the first "e" means. Like `ps-regs`, `ps-groups` to comprise 12-byte groups (only 4 in T6041).
#asahi-re · 2026-01-14 22:33
nnymNo and they haven't changed since 9 minutes before I first installed m1n1 on October 6.
#asahi-re · 2026-01-14 22:19
nnymhttps://paste.sr.ht/~nnym/81c285012af8fe7e76ea06c8fb2ef6bd9ba963c5
#asahi-re · 2026-01-14 21:34
nnymOr do you mean that only 1 of the boot policy files is necessary?
#asahi-re · 2026-01-14 21:32
nnymhttps://paste.sr.ht/~nnym/780cf5abfe7893a5e67b14d2aa7635877d64998a

plus ›