While uc_mem_unmap does unmap memory regions from Unicorn, it does not
free the memory. It accumulates over time when reusing a single Unicorn
instance.
When attempting to set all flags except trap flag, the EFLAGS value
should be 0x00244ED7, I've tested this on Windows and here
https://ideone.com/WQAvk1 which is presumably Linux.
Unicorn however has the value 0x00000ED7, bits 11-21 are not set. Bit 21
is the ID bit indicating whether or not CPUID is available.
The opcode 0x00 translates as `add byte ptr ds:[eax],al`, which leads to
a segfault in Unicorn.
Using a debugger, I believe the crash is located in
`qemu/translate-all.c` in function `page_flush_tb_1`, more specifically
this code `pd[i].first_tb = NULL;`
- Allow to register handler separately for invalid memory access
- Add new memory events for hooking:
- UC_MEM_READ_INVALID, UC_MEM_WRITE_INVALID, UC_MEM_FETCH_INVALID
- UC_HOOK_MEM_READ_PROT, UC_HOOK_MEM_WRITE_PROT, UC_HOOK_MEM_FETCH_PROT
- Rename UC_ERR_EXEC_PROT to UC_ERR_FETCH_PROT
- Change API uc_hook_add() so event type @type can be combined from hooking types
