[flang][perf] Adding noalias for function arguments affects function specialization

[vzakhari]

CPU2017/548 slowed down after a0d699a, so this change was partially reverted by d16ecad

I could reproduce it on neoverse-v1 with -Ofast -flto -mcpu=neoverse-v1 -fuse-ld=lld -mmlir -force-no-alias=false/true.

It looks like the digits_2 function does not pass the function specialization threshold on the Latency bonus.

Without noalias:

FnSpecialization: Specialization bonus {Latency = 2424 (71%)}

With noalias:

FnSpecialization: Specialization bonus {Latency = 1098 (34%)}
FnSpecialization: No possible specializations found for _QMbrute_forcePdigits_2

The current threshold is 40%.

It looks like the big portion of the latency bonus comes from the multiple mod(row, 3) expressions, which are not CSEd without noalias - so function specialization computes a lot of bonus for them when row is a constant. With noalias, there is a single srem instruction that computes mod(row, 3) at the beginning of the function.

I will investigate more to see if the function specialization misses to account for some bonuses.

[vzakhari]

It looks like the threshold has been lowered once (also due to the improved alias analysis): 2fb51fb#diff-066363256b7b4164e66b28a3028b2cb9e405c9136241baa33db76ebd2edb87cd