Commit a4fccfd8 authored by Sanjoy Das's avatar Sanjoy Das Committed by TensorFlower Gardener
Browse files

[XLA:CPU] Add an explicit code path for the entry computation cycle count 

While this does change the profile counter entry for the entry computation
during AOT compiles (earlier it would always be some non-null llvm::Value, but
now it can be null), it does not change any observable behavior since
RecordCompleteComputation is a no-op for an empty hlo_to_profile_idx_ map.

PiperOrigin-RevId: 176022629
parent 2040491d
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment