[PATCH] D113291: [AggressiveInstCombine] Lower Table Based CTTZ and enable it for AARCH64 in -O3

[Dávid Bolvanský via Phabricator via llvm-commits]

xbolva00 added a comment.

I am wondering about general direction..

Is it worth it? On way to become compiler just for benchmarks?

Spend compile time just to optimize one very very specific pattern from spec is bad thing imho.

Can you show us some other real world “hits”? I assume any sane project already uses builtin to compute this value efficiently.



================
Comment at: llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp:380
+    if (Element < InputBits &&
+        (((Mul << Element) & Mask.getZExtValue()) >> Shift) == i)
+      Matched++;
----------------
getZExtValue may assert large ints


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113291/new/

https://reviews.llvm.org/D113291