[PATCH] D55529: [InstCombine] try to convert x86 movmsk intrinsic to generic IR (PR39927)

[Craig Topper via Phabricator via llvm-commits]

craig.topper added inline comments.


================
Comment at: lib/Transforms/InstCombine/InstCombineCalls.cpp:774
 
-    if ((CInt && CInt->isNegative()) || (CFp && CFp->isNegative()))
-      Result.setBit(I);
+  // If the argument is bitcast, look through that, but make sure the source of
+  // that bitcast is still a vector with the same number of elements.
----------------
Should we have a comment here that describes what we're trying to do before we start talking about peeking through a bitcast. At first glance this looks like the comment for the whole block, but its really not.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D55529/new/

https://reviews.llvm.org/D55529