On Fri, Jul 20, 2018 at 11:46 AM, Sebastian Huber <sebastian.hu...@embedded-brains.de> wrote: > ----- Am 20. Jul 2018 um 16:28 schrieb Gedare Bloom ged...@rtems.org: > >> I want to reiterate my comment that some ISAs have valid instructions >> with an encoding of all zeroes. Off the top of my head, MIPS32 of all >> 0s is the encoding for sll $0, $0, 0, which is a nop. > > Yes, I added some special instructions on some architectures, but I don't > know all of them. I will try to use the "illegal" instruction on m68k. > > Do you now an opcode for MIPS? >
I think all 1s is an undefined/illegal opcode for mips, so .word -1 might work >> >> I am not too sure about the purpose of the test anymore, as if an >> illegal instruction exception occurs, the alignment/data access >> exception does not get executed. > > Yes, I will remove this stuff in v2. _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
