* magic data is architecture independent
* fcnsign data is architecture independent
* opcode data is architecture independent
* syscall data is architecture independent
* hud data is architecture independent
without these guards the linker complains:
p/asm_z80.o:(.data.rel+0x0): multiple definition of `radare_plugin'
p/asm_m68k_cs.o:(.data.rel+0x0): first defined here
collect2: error: ld returned 1 exit status
* Test for valid numbers in arm getnum function
Using strtod it's possible to test if the string passed in is a valid
number
Also handles hexvalues
* KISS the fix
$ rasm2 -a x86 -b 64 'mov byte [rbp - 0x100], 2'
c68500ffffff02
Oddly this also compiles correctly before this patch with keystone
installed. Not sure as to why yet.
* added sorted parameter in r_list
* use r_list_sort in r_range_sort
* some clean up
* added is_data into RBinSection
* use tinyrange by default to improve speed
* Move z80 non-commercial to r2e
Rename z80-cr to z80 as a consequence as it will be the default z80 disassembler
* Add GPL z80 assembler back in
* Fix Clang build
* Fix dd IX instructions
* Fix segfault for dd and de ops
Although only seemed to happen on OSX, not on linux.
* Fix ed op
* Remove invalid instruction "in f, [c]" which is not referenced in
http://clrhome.org/table/
* fix bad offsets to ops following removed one
* Fix fd ops with bad type flags
* Fix ed and fdcb ops
Remove debug prints
* Add support for seg registers
Support added for mov and pop instructions
[fs] is treated as [fs:0] which is not equivalent
Accepts sreg:[x] and [sreg:x] syntax
* Error if mov op reg sizes differ
also add support for 16bit register mov ops
* Improve segment register handling for mov op
* Fix opcode for 64bit instr with 32bits regs
* Refactor parsing of segment reg offset
