Strange counter-performance in an alternative `decimalLength9` function

[kinke]

On Friday, 28 February 2020 at 10:11:23 UTC, Bruce Carneal wrote:
> On Friday, 28 February 2020 at 06:50:55 UTC, 9il wrote:
>> bsr can be done in one/two CPU operation, quite quick. But 
>> core.bitop.bsr wouldn't be inlined. Instead, mir-core 
>> (mir.bitop: ctlz) or LDC intrinsics llvm_ctlz can be used for 
>> to get code with inlining.
>
> That's surprising.  I just got ldc to inline core.bitop.bsr on 
> run.dlang.io using ldc -O3 -mcpu=native.

These tiny core.bitop wrappers around LLVM intrinsics are always 
inlined (`pragma(inline, true)`), i.e., you don't even need -O.