[PATCH] D12991: [DivergenceAnalysis] Separate the definition

[Jingyue Wu via llvm-commits]

jingyue requested changes to this revision.
This revision now requires changes to proceed.

================
Comment at: include/llvm/Analysis/DivergenceAnalysis.h:14
@@ +13,3 @@
+//
+// GPU programs typically use the SIMD execution model, where multiple threads
+// in the same execution group have to execute in lock-step. Therefore, if the
----------------
These detailed header comments should remain in cpp. Add a very brief description in the header file. 

Refer to the header comments in `ScalarEvolution.h` and `ScalarEvolution.cpp`. 


http://reviews.llvm.org/D12991