#
✕ nick:flatz
#asahi · 2024-10-07 11:13
flatzand "image list" after app starts contains dyld and app binary only
#asahi · 2024-10-07 11:12
flatzdoes anyone know why breakpoints in lldb in guest os acts weird if xnu is running under m1n1? i can only catch breakpoints in the app, but cannot catch in modules from dyld shared cache, for example
#asahi · 2024-09-26 19:55
flatzI guess it cannot be done using hv_handle_msr_unlocked
#asahi · 2024-09-26 19:55
flatzwould be nice to let XNU think it's non-zero, doesn't matter how (just don't want to patch kernel manually)
#asahi · 2024-09-26 19:54
flatzhow can I hook "msr <reg>, pan" using m1n1's hv? it's needed to bypass XNU kernel code that panics if PAN is 0 (machine_switch_context/machine_stack_handoff)? can be triggered from lldb, for example
#asahi · 2023-08-22 20:29
flatzbtw if it does matter, i'm debugging xnu ^
#asahi · 2023-08-22 20:28
flatzany idea why i'm getting into handle_break() twice a row? so, first time i'm breaking due to h/w breakpoint, then i do single-stepping and landing at handle_step() but after continuing i'm getting back to handle_break() at the same address as before (and sometimes having different cpu id)
#asahi · 2023-06-28 19:20
flatzjannau: thank you a lot, it works now
#asahi · 2023-06-28 18:29
flatzthanks, i'll try soon
#asahi · 2023-06-28 18:15
flatzany ideas?
#asahi · 2023-06-28 18:15
flatzand when i try to boot stock/development macOS kernel (13.2.1) via run_guest.py, i'm getting panic: panic(cpu 1 caller 0xfffffe0016625038): "RTBuddy(DCP)::_setManagedStateGated " "No response received in 20s, DCP not started? (4)" "\nRTBuddy(AP): Debug info: _iopStatus = 0x4 _powerStateChangeLocked = true Mailbox Status: IDLE_STATUS: 0x00000008 INBOX0_CTRL: 0x00200201 OUTBOX0_CTRL: 0x00020001
#asahi · 2023-06-28 18:15
flatzhi. i'm experiencing some problem with DCP on M2 Mini, display is external connected via HDMI. that's logs from m1n1 (first is during cold boot, second is after chainloading of m1n1 itself): https://pastebin.com/2cMzdhUL