[PATCH] D109883: [Analysis] Add support for vscale in computeKnownBitsFromOperator

[Dave Green via Phabricator via llvm-commits]

dmgreen added a comment.

Sounds good. I assume that unsetting low bits from the min value does not help much?



================
Comment at: llvm/lib/Analysis/ValueTracking.cpp:1705
+        unsigned FirstZeroBit = 64 - countLeadingZeros(MaxVScale);
+        unsigned VScaleTypeSize = Q.DL.getTypeSizeInBits(II->getType()->getScalarType());
+        if (FirstZeroBit < VScaleTypeSize)
----------------
Will VScaleTypeSize be BitWidth?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D109883/new/

https://reviews.llvm.org/D109883