benchmark: refactor console benchmark

Fix and refactor the console benchmark. It did not test console
so it got renamed and mainly tests the different ways of passing
arguments through.

PR-URL: https://github.com/nodejs/node/pull/17707
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

1 files changed, 3 insertions, 3 deletions

diff --git a/lib/console.js b/lib/console.js
index 654b5b8280..ae268d8f84 100644
--- a/lib/console.js
+++ b/lib/console.js
@@ -132,12 +132,12 @@ function write(ignoreErrors, stream, string, errorhandler, groupIndent) {
 }
 
 
-// As of v8 5.0.71.32, the combination of rest param, template string
-// and .apply(null, args) benchmarks consistently faster than using
-// the spread operator when calling util.format.
 Console.prototype.log = function log(...args) {
   write(this._ignoreErrors,
         this._stdout,
+        // The performance of .apply and the spread operator seems on par in V8
+        // 6.3 but the spread operator, unlike .apply(), pushes the elements
+        // onto the stack. That is, it makes stack overflows more likely.
         util.format.apply(null, args),
         this._stdoutErrorHandler,
         this[kGroupIndent]);