[PATCH] D24818: Using branch probability to guide critical edge splitting.

[David Li via llvm-commits]

davidxl added inline comments.


> davidxl wrote in code-placement.ll:35
> The layout change in function 't2' is weird -- there are unnecessary branch back and forth introduced. Can you check why (not directly related to this patch -- but triggers some bugs).

I looked at this a little more.  The layout change is caused by loop rotation (in MBP) exposed. The loop rotation here is beneficial because the Freq of the outer loop's back edge is larger than the sum of the exit edge of the inner loop and the incoming edge of the outer loop.

https://reviews.llvm.org/D24818