[PATCH] D57335: [IR] Don't assume all functions are 4 byte aligned

[Eli Friedman via Phabricator via llvm-commits]

efriedma added reviewers: efriedma, chandlerc.
efriedma added a comment.

Missing update to PPC backend.

I'm not convinced that we should be modifying the datalayout here, as opposed to adding some sort of target hook for the optimizer.

It's a little unfortunate that there isn't any way to represent that the alignment of a function pointer is the same as the alignment of the function, but has some known minimum value.



================
Comment at: llvm/docs/LangRef.rst:2007
+``F<abi>``
+    This specifies the alignment for function pointers.
 ``m:<mangling>``
----------------
This needs more description.


================
Comment at: llvm/lib/IR/ConstantFold.cpp:1083
+                  ? GV->getPointerAlignment(GV->getParent()->getDataLayout())
+                  : 0U;
+
----------------
If you fail to find a datalayout, you have to assume the worst case (that the pointer isn't aligned at all).


================
Comment at: llvm/lib/IR/ConstantFold.cpp:1087
+          if (GVAlign == 0U && isa<Function>(GV))
+            GVAlign = 4U;
 
----------------
Using "4" as a default is dangerous; on x86, the alignment of the pointer corresponds to the alignment of the function, but might be less than 4 if it isn't explicitly specified.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57335/new/

https://reviews.llvm.org/D57335