https://bugs.kde.org/show_bug.cgi?id=401828
--- Comment #13 from Mark Wielaard <m...@klomp.org> ---
(In reply to Will Schmidt from comment #12)
> Created attachment 119219 [details]
> vex : sanityCheckFail: exiting due to bad IR
>
> I got this error when regression testing this patch on a P8 BE system.
> Probably obvious, but I don't immediately understand what this is telling
> me. (I'll look more tomorrow)
>
> +IN STATEMENT:
> +
> +t13 =
> ReinterpF64asI64(RoundF64toF32(Xor32(t11,And32(Shl32(t11,0x........:I8),0x...
> .....:I32)),DivF64(Xor32(t11,And32(Shl32(t11,0x........:I8),0x........:I32)),
> F64i{0x........},t9)))
> +
> +ERROR = IRStmt.Put.Tmp: tmp and expr do not match
See just before:
+IR SANITY CHECK FAILURE
+
+IRSB {
+ t0:I64 t1:I64 t2:I64 t3:I64 t4:I64 t5:I64 t6:I64 t7:V128
+ t8:V128 t9:F64 t10:F64 t11:I32 t12:I32 t13:I32 t14:I64
t15:I64
+ t16:I64 t17:V128 t18:I64 t19:I64 t20:I64 t21:I32 t22:I32
t23:I32
+ t24:I64 t25:I32 t26:I32 t27:I32
So t13 has type I32, but ReinterpF64asI64 returns an I64.
I think you had wanted to use Iop_ReinterpF32asI32.
--
You are receiving this mail because:
You are watching all bug changes.
