Typo in ARM opnd_compute_address_priv caught by GCC 7.3.0

[egrimley]

With the GCC in Debian unstable I got this warning:

/.../dynamorio/core/arch/opnd_shared.c:1920:72: error: ?: using integer constants in boolean context [-Werror=int-in-bool-context]
                 (TEST(EFLAGS_C, mc->cpsr) ? (1 << (sizeof(reg_t)*8-1)) : 0);
                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~

GCC deserves a pat on the head for this, because it seems to have caught a real bug: the || at the end of line 1919 should be a |, I think. What the code seems to be doing is computing the effective address when RRX is used in an address calculation: a very weird thing to do, so it's not surprising if it was never tested. (Is there a place where a test for this could easily be added? Not worth adding a new top-level test just for this.)

[derekbruening]

It would be nice to have a test for this RRX code: we could re-open.