Wenyong Huang

@bxkai seems that the win32 system is little endian while the aot file generated is big endian, no idea about that, could you try modifying aot_loader.c to ignore the check...

@bxkai could you please test whether #855 fixes the issue?

Seems that endian/bit-width check in win32 is invalid, could you please whether patch #886 works?

@bxkai I uploaded a [patch set](https://github.com/bytecodealliance/wasm-micro-runtime/pull/886/commits/842694aaa5ffd31d01a80de61e66277dda3ab477) to fix it, could you please try again?

Hi, do you run iwasm aot on win32 x86_32 target? The relocation operations for Windows x86_32 haven't been implemented yet. Could you please refer to the relocation handlings for Windows...

Hi, the file CMakeLists.txt under product-mini/platforms/{x} is to build the files for that platform only. Could you please try using the file CMakeLists.txt under ${WAMR_ROOT}?

> @lum1n0us In current implementation, I think `NULL` means there is no stack boundary at all. Some platforms doesn't provide the related API to get native stack boundary, so we...

> in that case, such a comparison looks a little wired. It will be passed but ... > > ```c > // if (&module_inst < NULL) > if ((uint8 *)&module_inst...

@amboowang it is not easy to implement the nested attr_container. The attr_container itself can be serialized into a byte array, can you create another attr_container for the nested object, serialize...

Close as issue was fixed.