ld: Undefined symbols

[waterdragon78]

Hello! I am trying to compile this, but I keep running into this error. Any help would be appreciated. Also, as a side note, what's the status on springboard? I've noticed that recent commits have referenced previous issues so I'm wondering if progress has been made on SEP. Thank you!

ld: warning: -undefined error is deprecated
ld: warning: ignoring duplicate libraries: '-lgcrypt', '-lgpg-error'
ld: Undefined symbols:
_asn1_array2tree, referenced from:
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_asn1_create_element, referenced from:
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_asn1_der_decoding, referenced from:
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_asn1_read_value, referenced from:
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
_extract_im4p_payload in hw_arm_apple-silicon_boot.c.o
clang: error: linker command failed with exit code 1 (use -v to see invocation)
ninja: build stopped: subcommand failed.
make: *** [run-ninja] Error 1

[VisualEhrmanntraut]

@waterdragon78 https://github.com/TrungNguyen1909/qemu-t8030/wiki/Bringing-up-the-emulator#getting-dependencies

[waterdragon78]

I did that, and now I'm getting this
ld: warning: -undefined error is deprecated
ld: warning: ignoring duplicate libraries: '-lc++', '-lgcrypt', '-lglib-2.0', '-lgmodule-2.0', '-lgpg-error', '-lintl'
ld: Undefined symbols:
_asn1_array2tree, referenced from:
_extract_im4p_payload in hw_arm_xnu.c.o
_asn1_create_element, referenced from:
_extract_im4p_payload in hw_arm_xnu.c.o
_asn1_der_decoding, referenced from:
_extract_im4p_payload in hw_arm_xnu.c.o
_asn1_read_value, referenced from:
_extract_im4p_payload in hw_arm_xnu.c.o
_extract_im4p_payload in hw_arm_xnu.c.o
_extract_im4p_payload in hw_arm_xnu.c.o
_extract_im4p_payload in hw_arm_xnu.c.o
_extract_im4p_payload in hw_arm_xnu.c.o
clang: error: linker command failed with exit code 1 (use -v to see invocation)
[3081/3116] Compiling C object tests/qtest/boot-order-test.p/boot-order-test.c.o
[3082/3116] Compiling C object tests/qtest/erst-test.p/erst-test.c.o
[3083/3116] Compiling C object tests/qtest/fw_cfg-test.p/fw_cfg-test.c.o
[3084/3116] Compiling C object tests/qtest/device-plug-test.p/device-plug-test.c.o
[3085/3116] Compiling C object tests/qtest/fdc-test.p/fdc-test.c.o
[3086/3116] Compiling C object tests/qtest/i440fx-test.p/i440fx-test.c.o
[3087/3116] Compiling C object tests/qtest/drive_del-test.p/drive_del-test.c.o
[3088/3116] Compiling C object tests/qtest/cpu-plug-test.p/cpu-plug-test.c.o
[3089/3116] Compiling C object tests/qtest/ide-test.p/ide-test.c.o
[3090/3116] Compiling C object tests/qtest/hd-geo-test.p/hd-geo-test.c.o
[3091/3116] Compiling C object tests/qtest/rtc-test.p/rtc-test.c.o
[3092/3116] Compiling C object tests/qtest/q35-test.p/q35-test.c.o
[3093/3116] Compiling C object tests/qtest/tco-test.p/tco-test.c.o
ninja: build stopped: subcommand failed.
make: *** [run-ninja] Error 1

[VisualEhrmanntraut]

@waterdragon78 You did not install asn1

[waterdragon78]

I did, using venv and pip

[VisualEhrmanntraut]

@waterdragon78 How’s it gonna access the venv

[waterdragon78]

It wouldn't let me install it otherwise

[waterdragon78]

I used the flag, but I still have the same error

[VisualEhrmanntraut]

pyasn1 doesn't have anything to do with the asn1 lib here

[waterdragon78]

I compiled it with Ubuntu which worked just fine, but I'm trying to run the command below to start it but it can't find a property.

I used:

qemu-t8030/build/qemu-system-aarch64 -s -M t8030,trustcache-filename=Firmware/038-44135-124.dmg.trustcache,ticket-filename=root_ticket.der -kernel kernelcache.research.iphone12b -dtb Firmware/all_flash/DeviceTree.n104ap.im4p -append "debug=0x14e kextlog=0xffff serial=3 -v wdt=-1" -initrd 038-44135-124.dmg -cpu max -smp 4 -m 4G -serial mon:stdio -drive file=nvme.1,format=raw,if=none,id=drive.1 -device nvme-ns,drive=drive.1,bus=nvme-bus.0,nsid=1,nstype=1,logical_block_size=4096,physical_block_size=4096 -drive file=nvme.2,format=raw,if=none,id=drive.2 -device nvme-ns,drive=drive.2,bus=nvme-bus.0,nsid=2,nstype=2,logical_block_size=4096,physical_block_size=4096 -drive file=nvme.3,format=raw,if=none,id=drive.3 -device nvme-ns,drive=drive.3,bus=nvme-bus.0,nsid=3,nstype=3,logical_block_size=4096,physical_block_size=4096 -drive file=nvme.4,format=raw,if=none,id=drive.4 -device nvme-ns,drive=drive.4,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=nvme.6,format=raw,if=none,id=drive.6 -device nvme-ns,drive=drive.6,bus=nvme-bus.0,nsid=6,nstype=6,logical_block_size=4096,physical_block_size=4096 -drive file=nvme.7,format=raw,if=none,id=drive.7 -device nvme-ns,drive=drive.7,bus=nvme-bus.0,nsid=7,nstype=8,logical_block_size=4096,physical_block_size=4096 -monitor telnet:127.0.0.1:1235,server,nowait
(I renamed it to qemu-t8030 for ease of access, its the dev-sw-sep branch)

Which gives:

Property 't8030-machine.trustcache-filename' not found

[VisualEhrmanntraut]

@waterdragon78 Why are you using the dev branch

[waterdragon78]

Because it has more commits and therefore progress

[VisualEhrmanntraut]

but it doesn't restore because it's a W.I.P so what's the point

[waterdragon78]

Where does it state that?

[VisualEhrmanntraut]

and where does it state the opposite

[waterdragon78]

That’s a valid point, mb. Do both dev branches not work?

[VisualEhrmanntraut]

my local changes on the SW SEP branch reach 92% on restore but fails on update_sep because of some weird issue :(

entering update_sep
libMobileGestalt MobileGestalt.c:711: RestoreLog: Client Query: Image4CryptoHashMethod
2024-03-01 20:47:14.469213+0000 restored_external[8:433] RestoreLog: Client Query: Image4CryptoHashMethod
libMobileGestalt MobileGestalt.c:743: RestoreLog: Client Response: Image4CryptoHashMethod : sha2-384
2024-03-01 20:47:14.473331+0000 restored_external[8:433] RestoreLog: Client Response: Image4CryptoHashMethod : sha2-384
ramrod_copy_manifest_digest_data_from_img4: failed to parse img4Data as Img4
Fri Mar  1 20:47:14 2024 localhost com.apple.xpc.launchd[1] <Error>: Host-level exception raised: pid = 8, thread = 0xe33, exception type = 0xd, codes = { 8589934603 }, states = { 0 }
[20:47:14.0506-GMT]{1>4} CHECKPOINT NOTICE: Post-engines NVRAM variable: restore-outcome=initial_engine_no_return
[20:47:14.0539-GMT]{1>4} CHECKPOINT PROGRESS: NO_REBOOT_RETRY (initial_monitor_no_return) -> (recovery_mode)
restore-child-failures = {initial:SIGNAL(11)[REBOOT]}
executing /usr/sbin/nvram restore-child-failures={initial:SIGNAL(11)[REBOOT]}
restore-outcome = recovery_mode
executing /usr/sbin/nvram restore-outcome=recovery_mode
restored exited with status 0xb due to signal 11 - rebooting

[waterdragon78]

I’m not an expert but I think it’s failing some sort of hash and then failing to parse img4data as img4? I’m not sure what that is lol but maybe it’s expecting it to be unchanged if it is changed.

[VisualEhrmanntraut]

I think idevicerestore is sending something wrong but idk the code is kind of unreadable lol

[waterdragon78]

Idk assuming I can fix the unknown property issue and get it running I’ll try it on Linux. Maybe that’ll do something assuming yall are on Mac? Idk, maybe something’s different in the Linux idevicerestore. Slim chance but can’t hurt to test it. Just gotta get it running first tho lol.

[waterdragon78]

I still have no idea what's going on, is there any way to fix the unknown property?

[VisualEhrmanntraut]

@waterdragon78 remove -filename from every property

[VisualEhrmanntraut]

You need to find a decrypted T8020 SEPROM and SEPFW. You have to figure out the rest, Google is your friend.

[waterdragon78]

Gotcha, I can do that. Whats the arguement to use for that?

[VisualEhrmanntraut]

... seprom and sepfw

[waterdragon78]

I only did the seprom because that's all I could find but I still got the apple logo and restore logo. Would that still boot?

[VisualEhrmanntraut]

As I said previously, dev will NOT boot; restore will fail prematurely.