Skip to content
This repository was archived by the owner on Nov 1, 2026. It is now read-only.
This repository was archived by the owner on Nov 1, 2026. It is now read-only.

Cannot run iOS with Segmentation failure #295

Closed

Description

@SharkZubat

Operating System

Ubuntu 25.10

What is your CPU's model?

Intel(R) Core(TM) i5-2450M (4)

Describe the unexpected behaviour in detail.

I just completed the setup but if I run iOS, it just showed and error (Segmentation fault (core dumped)). I'm doing to correctly at this time. (command args:

./Inferno/build/qemu-system-aarch64 -M t8030,trustcache=./Restore/Firmware/038-44135-124.dmg.trustcache,ticket=root_ticket.der,sep-fw=sep-firmware.n104.RELEASE.new.img4,sep-rom=AppleSEPROM-Cebu-B1,kaslr-off=true \
-kernel ./Restore/kernelcache.research.iphone12b -dtb ./Restore/Firmware/all_flash/DeviceTree.n104ap.im4p \
-append "tlto_us=-1 mtxspin=-1 agm-genuine=1 agm-authentic=1 agm-trusted=1 serial=3 wdt=-1 -vm_compressor_wk_sw" \
-smp 6 -m 4G -serial mon:stdio \
-drive file=sep_nvram,if=pflash,format=raw \
-drive file=sep_ssc,if=pflash,format=raw \
-drive file=root,format=raw,if=none,id=root -device nvme-ns,drive=root,bus=nvme-bus.0,nsid=1,nstype=1,logical_block_size=4096,physical_block_size=4096 \
-drive file=firmware,format=raw,if=none,id=firmware -device nvme-ns,drive=firmware,bus=nvme-bus.0,nsid=2,nstype=2,logical_block_size=4096,physical_block_size=4096 \
-drive file=syscfg,format=raw,if=none,id=syscfg -device nvme-ns,drive=syscfg,bus=nvme-bus.0,nsid=3,nstype=3,logical_block_size=4096,physical_block_size=4096 \
-drive file=ctrl_bits,format=raw,if=none,id=ctrl_bits -device nvme-ns,drive=ctrl_bits,bus=nvme-bus.0,nsid=4,nstype=4,logical_block_size=4096,physical_block_size=4096 \
-drive file=nvram,if=none,format=raw,id=nvram -device apple-nvram,drive=nvram,bus=nvme-bus.0,nsid=5,nstype=5,id=nvram,logical_block_size=4096,physical_block_size=4096 \
-drive file=effaceable,format=raw,if=none,id=effaceable -device nvme-ns,drive=effaceable,bus=nvme-bus.0,nsid=6,nstype=6,logical_block_size=4096,physical_block_size=4096 \
-drive file=panic_log,format=raw,if=none,id=panic_log -device nvme-ns,drive=panic_log,bus=nvme-bus.0,nsid=7,nstype=8,logical_block_size=4096,physical_block_size=4096 \
-initrd ./Restore/038-44135-124.dmg

)

What should've happened instead?

Well, there are the log:

info: iOS 14.0.0
info: Kernel virtual low: 0xfffffff006000000
info: Kernel virtual high: 0xfffffff00a4cd4f0
warning: ck_kp_root_auth_callback: failed to find pacibsp, trying to find old logic
info: `bypass root authentication` patch applied in `__TEXT_EXEC`.
info: `allow mounting root as R/W` patch applied in `__TEXT_EXEC`.
`bypass root hash authentication` patch did not apply in `__TEXT_EXEC`.
info: `disable APFS snapshots` patch applied in `__TEXT`.
info: `allow SHA1 signatures in AMFI` patch applied in `__TEXT_EXEC`.
`all binaries in TrustCache (AMFI)` patch did not apply in `__TEXT_EXEC`.
info: `increase SCOT size to 0x10000 to use it as TRAC` patch applied in `__TEXT_EXEC`.
info: `allow unsigned firmware in img4_firmware_evaluate` patch applied in `__TEXT_EXEC`.
`allow remounting rootfs, union mounts (old)` patch did not apply in `__TEXT_EXEC`.
info: `allow remounting rootfs, union mounts (new)` patch applied in `__TEXT_EXEC`.
info: `force enable kprintf` patch applied in `__TEXT_EXEC`.
`disable AMX (new)` patch did not apply in `__TEXT_EXEC`.
info: `disable AMX` patch applied in `__TEXT_EXEC`.
info: `bypass code signature checks` patch applied in `__TEXT_EXEC`.
info: `enable hactivation` patch applied in `__TEXT`.
info: ck_kp_tc_callback: found AMFI (Routine)
info: `all binaries in trustcache` patch applied in `__PPLTEXT`.
info: `bypass pmap_cs_enforce` patch applied in `__PPLTEXT`.
warning: `reg` len mismatch; wanted: 8, actual: 16
warning: `reg` len mismatch; wanted: 8, actual: 32
warning: `interrupts` len mismatch; wanted: 4, actual: 108
warning: `interrupts` len mismatch; wanted: 4, actual: 104
warning: `interrupts` len mismatch; wanted: 4, actual: 108
info: `memcmp_validstrs30` patch applied in `seprom`.
info: `memcmp_validstrs14` patch applied in `seprom`.
info: `verify_rsa_signature` patch applied in `seprom`.
info: Boot Mode: 0
info: Auto Boot: false
info: Boot Args: [-restore rd=md0 nand-enable-reformat=1 tlto_us=-1 mtxspin=-1 agm-genuine=1 agm-authentic=1 agm-trusted=1 serial=3 wdt=-1 -vm_compressor_wk_sw]
Segmentation fault (core dumped)

Attach any logs related to this issue.

No response

Issue Report Terms

  • You have made sure that there are no other reports about this bug.
  • You do not use any software that might interfere with correct function of this software. For Hackintosh users, that could be OCLP, WEG, etc.
  • You are on the latest commit of the emulator.
  • You acknowledge that violation of these terms may result in your issue being closed or you getting blocked from the organisation.

Activity

  1. added theissue type on Jun 12, 2026
  2. VisualEhrmanntraut commented on Jun 12, 2026

    @VisualEhrmanntraut
    Member

    get the backtrace using gdb or lldb

  3. chris-pcguy commented on Jun 12, 2026

    @chris-pcguy
    Member

    Also, why are you taking away an AP core? With the emulated SEP setting, it's -smp AP_CORES+SEP_CORES.
    Currently, SEP_CORES is fixed to 1, but some things with newer SEP versions make me believe that there might be some SEP's with more than a single core coming (or possibly already there, I don't have the time to look).
    Would probably be good for the performance, even on actual Apple Silicon.

  4. SharkZubat commented on Jun 12, 2026

    @SharkZubat
    Author

    get the backtrace using gdb or lldb

    I ran it through lldb and got a segmentation fault pointing to libnettle. It seems to happen during crypto/image parsing:

    Process 19075 stopped
    * thread #1, name = 'qemu-system-aar', stop reason = signal SIGSEGV: address not mapped to object (fault address=0x20)
        frame #0: 0x00007ffff656988d libnettle.so.9`_nettle_write_be32(length=32, dst="", src=0x0000555558239240) at write-be32.c:54:5
    

    My arguments actually include: sep-fw=sep-firmware.n104.RELEASE.new.img4, sep-rom=AppleSEPROM-Cebu-B1, trustcache=./Restore/Firmware/038-44135-124.dmg.trustcache.
    It looks like an empty destination pointer (dst="") is being passed to nettle during the boot sequence or firmware parsing block.

  5. VisualEhrmanntraut commented on Jun 12, 2026

    @VisualEhrmanntraut
    Member

    No, I said backtrace. write bt.

  6. SharkZubat commented on Jun 12, 2026

    @SharkZubat
    Author

    No, I said backtrace. write bt.

    Oh, okay. Trying out

  7. SharkZubat commented on Jun 12, 2026

    @SharkZubat
    Author

    No, I said backtrace. write bt.

    Oh, okay. Trying out

    there it is:

    (lldb) bt
    * thread #1, name = 'qemu-system-aar', stop reason = signal SIGSEGV: address not mapped to object (fault address=0x20)
      * frame #0: 0x00007ffff656988d libnettle.so.9`_nettle_write_be32(length=32, dst="", src=0x00005555580dee90) at write-be32.c:54:5
        frame #1: 0x00007ffff655c516 libnettle.so.9`sha256_write_digest(ctx=<unavailable>, length=<unavailable>, digest=<unavailable>) at sha256.c:148:3 [artificial]
        frame #2: 0x00007ffff655c78a libnettle.so.9`nettle_sha256_digest(ctx=0x00005555580dee90, digest=<unavailable>) at sha256.c:155:3
        frame #3: 0x0000555556086fcd qemu-system-aarch64`qcrypto_nettle_hash_finalize + 77
        frame #4: 0x000055555608134d qemu-system-aarch64`qcrypto_hash_bytesv + 93
        frame #5: 0x00005555560813be qemu-system-aarch64`qcrypto_hash_bytes + 46
        frame #6: 0x0000555555d9c689 qemu-system-aarch64`apple_boot_populate_dt + 521
        frame #7: 0x0000555555db43e1 qemu-system-aarch64`t8030_memory_setup + 1601
        frame #8: 0x0000555555db4bc4 qemu-system-aarch64`t8030_init_done + 20
        frame #9: 0x00005555561d4ccf qemu-system-aarch64`notifier_list_notify + 47
        frame #10: 0x00005555559c6c88 qemu-system-aarch64`qdev_machine_creation_done + 120
        frame #11: 0x0000555555bfaa8e qemu-system-aarch64`qmp_x_exit_preconfig.part.0 + 350
        frame #12: 0x0000555555bfe420 qemu-system-aarch64`qemu_init + 12816
        frame #13: 0x000055555590ffdd qemu-system-aarch64`main + 29
        frame #14: 0x00007ffff602a575 libc.so.6`__libc_start_call_main(main=(qemu-system-aarch64`main), argc=49, argv=0x00007fffffffd4e8) at libc_start_call_main.h:58:16
        frame #15: 0x00007ffff602a628 libc.so.6`__libc_start_main_impl(main=(qemu-system-aarch64`main), argc=49, argv=0x00007fffffffd4e8, init=<unavailable>, fini=<unavailable>, rtld_fini=<unavailable>, stack_end=0x00007fffffffd4d8) at libc-start.c:360:3
        frame #16: 0x00005555559110c5 qemu-system-aarch64`_start + 37
    
  8. VisualEhrmanntraut commented on Jun 12, 2026

    @VisualEhrmanntraut
    Member

    Uhm, but this bug was fixed in ec89fc7

  9. SharkZubat commented on Jun 12, 2026

    @SharkZubat
    Author

    Uhm, but this bug was fixed in ec89fc7

    Yeah, I cloned the repo with bugfix already. So it doesn't work or maybe I'll check it if I fail or not

  10. chris-pcguy commented on Jun 12, 2026

    @chris-pcguy
    Member

    His Ubuntu version is still using nettle3, and even the not yet released version is still using nettle3. So it can't be that.
    Also, it has been tested with nettle3 and nettle4, so it's somewhat likely to be a user error.

  11. VisualEhrmanntraut commented on Jun 12, 2026

    @VisualEhrmanntraut
    Member

    it says "libnettle.so.9" which is nettle 4 so he likely replaced it by the steps in the aside.

  12. chris-pcguy commented on Jun 12, 2026

    @chris-pcguy
    Member

    Damn, I didn't notice that. If he really wants nettle4 and even use it globally without breaking half of his system or recompiling everything, he should use Arch (but please stay away from AUR for the time being), since it already includes it, Gentoo not yet.

  13. SharkZubat commented on Jun 13, 2026

    @SharkZubat
    Author

    Damn, I didn't notice that. If he really wants nettle4 and even use it globally without breaking half of his system or recompiling everything, he should use Arch (but please stay away from AUR for the time being), since it already includes it, Gentoo not yet.

    Yeah I will try learning to install inferno for the second time when I use arch

  14. locked and limited conversation to collaborators on Jun 13, 2026
  15. converted this issue into a discussion #296 on Jun 13, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    Fields

    Priority

    None yet

    Effort

    None yet

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions