On 4/7/2014 10:00 AM, Peter Maydell wrote: > It seems a touch dubious to assume the area of memory > being dissassembled is necessarily the same endianness > the CPU happens to be currently...
I don't disagree but it is less dubious than relying on TARGET_WORDS_BIGENDIAN, which is what the code currently does. I think the only other alternative is to put it back on the user via options on the command, e.g. xp/8il xp/8ib But, I would contend that the using the MSR[LE] bit would still be a better default than TARGET_WORDS_BIGENDIAN if an option is not specified. Thanks for the other comments ... I will incorporate.
