Linux Kernel Module Cheat

3.4.1. GDB step debug kernel module insmodded by init on ARM

TODO on arm 51e31cdc2933a774c2a0dc62664ad8acec1d2dbe it does not always work, and lx-symbols fails with the message:

loading vmlinux
Traceback (most recent call last):
  File "/linux-kernel-module-cheat//out/arm/buildroot/build/linux-custom/scripts/gdb/linux/symbols.py", line 163, in invoke
    self.load_all_symbols()
  File "/linux-kernel-module-cheat//out/arm/buildroot/build/linux-custom/scripts/gdb/linux/symbols.py", line 150, in load_all_symbols
    [self.load_module_symbols(module) for module in module_list]
  File "/linux-kernel-module-cheat//out/arm/buildroot/build/linux-custom/scripts/gdb/linux/symbols.py", line 110, in load_module_symbols
    module_name = module['name'].string()
gdb.MemoryError: Cannot access memory at address 0xbf0000cc
Error occurred in Python command: Cannot access memory at address 0xbf0000cc

Can’t reproduce on x86_64 and aarch64 are fine.

It is kind of random: if you just insmod manually and then immediately ./run-gdb --arch arm, then it usually works.

But this fails most of the time: shell 1:

./run --arch arm --eval-after 'insmod hello.ko'

shell 2:

./run-gdb --arch arm

then hit Ctrl-C on shell 2, and voila.

Then:

cat /proc/modules

says that the load address is:

0xbf000000

so it is close to the failing 0xbf0000cc.

readelf:

./run-toolchain readelf -- -s "$(./getvar kernel_modules_build_subdir)/hello.ko"

does not give any interesting hits at cc, no symbol was placed that far.