tree 77da9622b8ef0f143f2c42f78ceba3db5d2a4922
parent 3b92462312126d63ae06af4698e6ae9c57d17116
author Chris Craik <ccraik@google.com> 1606777017 -0800
committer Rahul Ravikumar <rahulrav@google.com> 1606856523 -0800

Add separate Compose macrobenchmark target

Bug: 172659078
Test: Tests in androidx.compose.integration.macrobenchmark

Now Compose/View have vaguely comparable benchmarks
(TrivialStartupBenchmark, SmallListStartupBenchmark). Note that R8 is
not yet enabled, and iteration counts are still low.

Functionality in Compose benchmarks doesn't yet match View equivalents
1-1, e.g. using string resources, or in list case, presence of
constraints or complexity of item layout.

Change-Id: I99370aebfba223de5ab342ddf56d082c8c782b2f