https://bugs.kde.org/show_bug.cgi?id=399087
--- Comment #5 from Julian Seward <jsew...@acm.org> ---
(In reply to John Reiser from comment #0)
It's clear that the thing has gone off of the rails somehow, but I'm not
sure I agree with your conclusion here:
> EXPECTED RESULT
> "add r15, r6, r10" is recognized as a computed goto. Emulated control
> should remain inside the user program, or else be diagnosed as a "wild jump"
> error.
If you look at this (from your log) we have
(arm) 0x21F20: add r15, r6, r10
------ IMark(0x21F20, 4, 0) ------
t17 = GET:I32(32) // t17 = value of r6, I assume
t19 = GET:I32(48) // t19 = value of r10
t18 = t19 // r18 = value of r10
t20 = Add32(t17,t18) // t20 = r6 + r10
PUT(68) = t20 // VexGuestARMState offset 68 is offset of PC
PUT(68) = GET:I32(68) // pointless
PUT(68) = GET:I32(68); exit-Boring // "continue at address in (68)"
so that looks correct to me.
Did I miss something? I must say this stuff was pretty hairy to implement
in the ARM front end, so I can believe I got something wrong. OTOH the
arm32 front end has been around easily half a decade now without anyone
reporting anything like this before.
--
You are receiving this mail because:
You are watching all bug changes.
