[PATCH] D98213: [InlineCost] Enable the cost benefit analysis on FDO

[Kazu Hirata via Phabricator via llvm-commits]

kazu added a comment.

>   // We use 128-bit APInt here to avoid potential overflow.  This variable
>   // should stay well below 10^^24 (or 2^^80) in practice.  This "worst" case
>   // assumes that we can avoid or fold a billion instructions, each with a
>   // profile count of 10^^15 -- roughly the number of cycles for a 24-hour
>   // period on a 4GHz machine.
>
> If you potentially need a 128 bit integer to store your "cycle savings", you should not be using that value to compare against a size cost. A sufficiently good "saving" will absolutely override a huge size cost.

Keep in mind that the cycle savings is divided by `PSI->getOrCompHotCountThreshold()` (although the actual implementation avoids divisions by doing "A*D >= B*C" instead of "A/B >= C/D").  We are not directly comparing the cycle savings to the size cost, which would not make sense as you have pointed out.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D98213/new/

https://reviews.llvm.org/D98213