Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
cmd/compile: byte type conversion optimizations #18575
What did you do?
I noticed that rewriting a busy-loop with cast-then-compare (ie BenchmarkU64) instead of compare-then-cast (ie BenchmarkByte) I was getting a noticeable throughput difference.
Here is a benchmark that shows the difference:
What did you expect to see?
Same benchmark performance.
What did you see instead?
BenchmarkU64 is >10% faster than BenchmarkByte.
Does this issue reproduce with the latest release (go1.7.4)?
Reproduced with 1.8beta2