[PATCH] D38501: [ValueTracking] Fix a misuse of APInt in GetPointerBaseWithConstantOffset

[Michael Ferguson via Phabricator via llvm-commits]

mppf added inline comments.


================
Comment at: lib/Analysis/ValueTracking.cpp:3081
   }
   Offset = ByteOffset.getSExtValue();
   return Ptr;
----------------
hfinkel wrote:
> Do we need to add a check here that the offset can actually be represented in 64 bits? I don't think that a truncated return value here actually makes sense.
> 
getSExtValue does make sure it can be represented in an int64_t. If you were talking about using sextOrTrunc above, I think we need that truncation so that we can handle overflow in GEP operations so that we can handle negative GEP offsets correctly. (see my recent comment in D38499).


https://reviews.llvm.org/D38501