code review 61600043: bench: add "garbage" benchmark

bench: add "garbage" benchmark
It's based on the famous test/bench/garbage/parser benchmark.
But parses only net/http for better stability.

[dvyukov, 2014-02-10 15:51:03]

Hello adg@golang.org (cc: golang-codereviews@googlegroups.com),

I'd like you to review this change to
https://code.google.com/p/go.benchmarks

[dvyukov, 2014-02-10 16:07:33]

On 2014/02/10 15:51:03, dvyukov wrote:
> Hello mailto:adg@golang.org (cc: mailto:golang-codereviews@googlegroups.com),
> 
> I'd like you to review this change to
> https://code.google.com/p/go.benchmarks

It's still not as stable as I would like to, but I don't know how to improve
it...

[adg, 2014-02-11 05:28:40]

old chunk mismatch
please re-upload

[dvyukov, 2014-02-11 09:09:49]

On 2014/02/11 05:28:40, adg wrote:
> old chunk mismatch
> please re-upload

Done

[adg, 2014-02-25 05:43:13]

LGTM

https://codereview.appspot.com/61600043/diff/80001/garbage/garbage.go
File garbage/garbage.go (right):

https://codereview.appspot.com/61600043/diff/80001/garbage/garbage.go#newcode90
garbage/garbage.go:90: runtime.GC()
why twice?

[dvyukov, 2014-02-25 12:26:13]

https://codereview.appspot.com/61600043/diff/80001/garbage/garbage.go
File garbage/garbage.go (right):

https://codereview.appspot.com/61600043/diff/80001/garbage/garbage.go#newcode90
garbage/garbage.go:90: runtime.GC()
On 2014/02/25 05:43:13, adg wrote:
> why twice?

added comment:
	// One GC does not give precise results,
	// because concurrent sweep may be still in progress.

[dvyukov, 2014-02-25 12:27:01]

*** Submitted as
https://code.google.com/p/go/source/detail?r=b9f69a8adb51&repo=benchmarks ***

bench: add "garbage" benchmark
It's based on the famous test/bench/garbage/parser benchmark.
But parses only net/http for better stability.

LGTM=adg
R=adg
CC=golang-codereviews
https://codereview.appspot.com/61600043