Improvements in the Inliner.

- Change from a depth limit to a total number of HInstructions
  inlined limit. Remove the dex2oat depth limit argument.
- Add more stats to diagnose reasons for not inlining.
- Clean up logging to easily parse output.

Individual Ritz benchmarks improve from 3 to 10%.

No change in other heuristics. There was already an instruction budget.
Note that the instruction budget is rarely hit in the "apps" I've tried
with.

Compile-times improve from 5 to 15%.
Code size go from 4% increase (Gms) to 1% decrease (Docs).

bug:35724239
test: test-art-host test-art-target

Change-Id: I5a35c4bd826cf21fead77859709553c5b57608d6

1 files changed, 0 insertions, 5 deletions

diff --git a/compiler/jit/jit_compiler.cc b/compiler/jit/jit_compiler.cc
index 3ae7974038..ad951bcc3f 100644
--- a/compiler/jit/jit_compiler.cc
+++ b/compiler/jit/jit_compiler.cc
@@ -97,7 +97,6 @@ JitCompiler::JitCompiler() {
       CompilerOptions::kDefaultSmallMethodThreshold,
       CompilerOptions::kDefaultTinyMethodThreshold,
       CompilerOptions::kDefaultNumDexMethodsThreshold,
-      CompilerOptions::kDefaultInlineDepthLimit,
       CompilerOptions::kDefaultInlineMaxCodeUnits,
       /* no_inline_from */ nullptr,
       CompilerOptions::kDefaultTopKProfileThreshold,
@@ -177,10 +176,6 @@ JitCompiler::JitCompiler() {
     jit_logger_.reset(new JitLogger());
     jit_logger_->OpenLog();
   }
-
-  size_t inline_depth_limit = compiler_driver_->GetCompilerOptions().GetInlineDepthLimit();
-  DCHECK_LT(thread_count * inline_depth_limit, std::numeric_limits<uint16_t>::max())
-      << "ProfilingInfo's inline counter can potentially overflow";
 }
 
 JitCompiler::~JitCompiler() {