The differences of bit width and endianness are already being taken care of by the current code.
The differences in ABI for ILP32 are the only ones missing and so the new version[1] (sorry, not a series since it makes sense as a single change anyway) should be enough to fix any inconsistencies. With this change, all linux-user versions will show the same consistent error when asked to load an incorrect (wrong architecture, endianness or bit width) binary Carlo [1] https://patchwork.ozlabs.org/project/qemu-devel/patch/20200823101703.18451-1-care...@gmail.com/
