Remove BootUpMemoryUse test.

Remove BootUpMemoryUse test.

BUG=
R=bmeurer@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=21073

[Benedikt Meurer, 2014-04-30 08:18:54]

Yay! LGTM

[Hannes Payer (out of office), 2014-04-30 08:19:44]

Committed patchset #1 manually as r21073 (presubmit successful).

[Jakob Kummerow, 2014-05-05 08:28:41]

DBC: I'm not convinced that removing this test is a good idea.
Yes, bumping the limit every so often is annoying and kind of defeats the
purpose of the test. On the other hand, we put in a lot of work to get boot-up
memory use low (IIRC the effort at the time brought it from ~20 to ~2 MB), and
as we know, every metric we don't track is bound to regress over time. So I
think in spite of its drawbacks, this test did serve a valid purpose. I vote for
putting it back in.

[Dmitry Lomov (no reviews), 2014-05-05 08:52:24]

On 2014/05/05 08:28:41, Jakob wrote:
> DBC: I'm not convinced that removing this test is a good idea.
> Yes, bumping the limit every so often is annoying and kind of defeats the
> purpose of the test. On the other hand, we put in a lot of work to get boot-up
> memory use low (IIRC the effort at the time brought it from ~20 to ~2 MB), and
> as we know, every metric we don't track is bound to regress over time. So I
> think in spite of its drawbacks, this test did serve a valid purpose. I vote
for
> putting it back in.

+1 to keeping us honest about boot-up memory

[Hannes Payer (out of office), 2014-05-05 17:10:30]

On 2014/05/05 08:52:24, Dmitry Lomov (chromium) wrote:
> On 2014/05/05 08:28:41, Jakob wrote:
> > DBC: I'm not convinced that removing this test is a good idea.
> > Yes, bumping the limit every so often is annoying and kind of defeats the
> > purpose of the test. On the other hand, we put in a lot of work to get
boot-up
> > memory use low (IIRC the effort at the time brought it from ~20 to ~2 MB),
and
> > as we know, every metric we don't track is bound to regress over time. So I
> > think in spite of its drawbacks, this test did serve a valid purpose. I vote
> for
> > putting it back in.
> 
> +1 to keeping us honest about boot-up memory

Feel free to revert this change and bump up the limit to something acceptable...