_mjit_compile_send.erb: inline recursive call

mjit_compile.c: propagate funcname to compile_insn

test_jit.rb: add test covering this behavior

* Benchmark

```
require 'benchmark_driver'

Benchmark.driver(runner: :time, repeat_count: 4) do |x|
  x.prelude %{
    def fib(x)
      return x if x == 0 || x == 1
      fib(x-1) + fib(x-2)
    end
  }
  x.report 'fib(40)'
  x.loop_count 1

  x.rbenv(
    'before,--jit',
    'before',
    'after,--jit',
    'after',
  )
  x.verbose
end
```

```
before,--jit: ruby 2.6.0dev (2018-05-08 trunk 63349) +JIT [x86_64-linux]
before: ruby 2.6.0dev (2018-05-08 trunk 63349) [x86_64-linux]
after,--jit: ruby 2.6.0dev (2018-05-08 trunk 63349) +JIT [x86_64-linux]
last_commit=_mjit_compile_send.erb: inline recursive call
after: ruby 2.6.0dev (2018-05-08 trunk 63349) [x86_64-linux]
last_commit=_mjit_compile_send.erb: inline recursive call
Calculating -------------------------------------
                     before,--jit      before  after,--jit       after
             fib(40)        2.886       8.685        2.562       8.800 s -       1.000 times

Comparison:
                          fib(40)
         after,--jit:         2.6 s
        before,--jit:         2.9 s - 1.13x  slower
              before:         8.7 s - 3.39x  slower
               after:         8.8 s - 3.44x  slower

```

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63350 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

3 files changed, 15 insertions, 0 deletions

diff --git a/mjit_compile.c b/mjit_compile.c
index 4f87550e78..e01f6b3cda 100644
--- a/mjit_compile.c
+++ b/mjit_compile.c
@@ -27,6 +27,7 @@ struct compile_status {
     /* If TRUE, JIT-ed code will use local variables to store pushed values instead of
        using VM's stack and moving stack pointer. */
     int local_stack_p;
+    const char *funcname; /* the method name which is being compiled */
 };
 
 /* Storage to keep data which is consistent in each conditional branch.
@@ -193,6 +194,7 @@ mjit_compile(FILE *f, const struct rb_iseq_constant_body *body, const char *func
     status.success = TRUE;
     status.local_stack_p = !body->catch_except_p;
     status.stack_size_for_pos = ALLOC_N(int, body->iseq_size);
+    status.funcname = funcname;
     memset(status.stack_size_for_pos, NOT_COMPILED_STACK_SIZE, sizeof(int) * body->iseq_size);
 
     /* For performance, we verify stack size only on compilation time (mjit_compile.inc.erb) without --jit-debug */
diff --git a/test/ruby/test_jit.rb b/test/ruby/test_jit.rb
index 891b7f496f..844f36a98f 100644
--- a/test/ruby/test_jit.rb
+++ b/test/ruby/test_jit.rb
@@ -322,6 +322,16 @@ class TestJIT < Test::Unit::TestCase
 
   def test_compile_insn_opt_send_without_block
     assert_compile_once('print', result_inspect: 'nil', insns: %i[opt_send_without_block])
+
+    # recursive inline
+    assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '120', success_count: 1, min_calls: 2)
+    begin;
+      def fact(i)
+        return i if i <= 1
+        i * fact(i - 1)
+      end
+      print fact(5)
+    end;
   end
 
   def test_compile_insn_invokesuper
diff --git a/tool/ruby_vm/views/_mjit_compile_send.erb b/tool/ruby_vm/views/_mjit_compile_send.erb
index cb7c1f40f8..b1dc0e72c8 100644
--- a/tool/ruby_vm/views/_mjit_compile_send.erb
+++ b/tool/ruby_vm/views/_mjit_compile_send.erb
@@ -58,6 +58,9 @@
                 fprintf(f, "            VM_ENV_FLAGS_SET(ec->cfp->ep, VM_FRAME_FLAG_FINISH);\n");
                 fprintf(f, "            v = vm_exec(ec, TRUE);\n");
             }
+            else if (body == iseq->body) { /* inline recursive call */
+                fprintf(f, "            v = %s(ec, ec->cfp);\n", status->funcname);
+            }
             else {
                 fprintf(f, "            if ((v = mjit_exec(ec)) == Qundef) {\n");
                 fprintf(f, "                VM_ENV_FLAGS_SET(ec->cfp->ep, VM_FRAME_FLAG_FINISH);\n"); /* This is vm_call0_body's code after vm_call_iseq_setup */