[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in original order.

[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in original order.

This is a complement to
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same
latency, which caused bench_copy.js 4% regression and
bench_skinning.js 5% regression on Atom when enabling
'--turbo-instruction-scheduling' flag according to our
observation.

We submit this patch to sort the nodes with same latency
in original order. It aligns with instruction scheduling
logic before the patch above and fixes these regression.

BUG=

Committed: https://crrev.com/3b083983d066034f4ead229cefcd27b622f7622f
Cr-Commit-Position: refs/heads/master@{#39228}

[shiyu.zhang, 2016-08-30 01:03:37]

Description was changed from

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency in the
list sorted in original order.

This is a small complement to https://codereview.chromium.org/2281523002.
It sorts the nodes with same latency in original order, which aligns with
instruction scheduling logic before the patch above.

It makes sense to obay the original instruction order if the candidate nodes
have same latency.

BUG=
==========

to

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency in the list
sorted in original order.

This is a small complement to https://codereview.chromium.org/2281523002.
It sorts the nodes with same latency in original order, which aligns with
instruction scheduling logic before the patch above.

It makes sense to obay the original instruction order if the candidate nodes
have same latency.

BUG=
==========

[shiyu.zhang, 2016-09-02 09:51:48]

Description was changed from

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency in the list
sorted in original order.

This is a small complement to https://codereview.chromium.org/2281523002.
It sorts the nodes with same latency in original order, which aligns with
instruction scheduling logic before the patch above.

It makes sense to obay the original instruction order if the candidate nodes
have same latency.

BUG=
==========

to

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in
original order.

This is a complement to 
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same 
latency, which caused bench_copy.js 4% regression and 
bench_skinning.js 5% regression on Atom when enabling 
'--turbo-instruction-scheduling' flag according to our 
observation.

We submit this patch to sort the nodes with same latency 
in original order. It aligns with instruction scheduling 
logic before the patch above and fixes these regression.

BUG=
==========

[shiyu.zhang, 2016-09-02 10:07:15]

shiyu.zhang@intel.com changed reviewers:
+ bmeurer@chromium.org, jarin@chromium.org

[shiyu.zhang, 2016-09-06 03:33:06]

PTAL

[Benedikt Meurer, 2016-09-06 03:34:10]

Sorry for the delay, I missed this one.

LGTM.

[shiyu.zhang, 2016-09-07 01:33:25]

The CQ bit was checked by shiyu.zhang@intel.com

[commit-bot: I haz the power, 2016-09-07 01:33:30]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-07 02:11:55]

Description was changed from

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in
original order.

This is a complement to 
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same 
latency, which caused bench_copy.js 4% regression and 
bench_skinning.js 5% regression on Atom when enabling 
'--turbo-instruction-scheduling' flag according to our 
observation.

We submit this patch to sort the nodes with same latency 
in original order. It aligns with instruction scheduling 
logic before the patch above and fixes these regression.

BUG=
==========

to

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in
original order.

This is a complement to 
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same 
latency, which caused bench_copy.js 4% regression and 
bench_skinning.js 5% regression on Atom when enabling 
'--turbo-instruction-scheduling' flag according to our 
observation.

We submit this patch to sort the nodes with same latency 
in original order. It aligns with instruction scheduling 
logic before the patch above and fixes these regression.

BUG=
==========

[commit-bot: I haz the power, 2016-09-07 02:11:56]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-09-07 02:12:33]

Description was changed from

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in
original order.

This is a complement to 
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same 
latency, which caused bench_copy.js 4% regression and 
bench_skinning.js 5% regression on Atom when enabling 
'--turbo-instruction-scheduling' flag according to our 
observation.

We submit this patch to sort the nodes with same latency 
in original order. It aligns with instruction scheduling 
logic before the patch above and fixes these regression.

BUG=
==========

to

==========
[turbofan] Instruction scheduler: keep ready nodes with same latency sorted in
original order.

This is a complement to
https://codereview.chromium.org/2281523002.

The patch above reversed the order of nodes with same
latency, which caused bench_copy.js 4% regression and
bench_skinning.js 5% regression on Atom when enabling
'--turbo-instruction-scheduling' flag according to our
observation.

We submit this patch to sort the nodes with same latency
in original order. It aligns with instruction scheduling
logic before the patch above and fixes these regression.

BUG=

Committed: https://crrev.com/3b083983d066034f4ead229cefcd27b622f7622f
Cr-Commit-Position: refs/heads/master@{#39228}
==========

[commit-bot: I haz the power, 2016-09-07 02:12:34]

Patchset 1 (id:??) landed as
https://crrev.com/3b083983d066034f4ead229cefcd27b622f7622f
Cr-Commit-Position: refs/heads/master@{#39228}