[turbofan] Check node input/use consistency for changed operators and new nodes.

[turbofan] Check node input/use consistency for changed operators and new nodes.

Verifies consistency of node inputs and uses:
- node inputs should agree with the input count computed from the node's operator.
- effect inputs should have effect outputs (or be a sentinel).
- control inputs should have control outputs (or be a sentinel).
- frame state inputs should be frame states (or be a sentinel).
- if the node has control uses, it should produce control.
- if the node has effect uses, it should produce effect.
- if the node has frame state uses, it must be a frame state.

I also removed some tests, either because they did not seem to be useful (scheduler) or they tested dead functionality (diamond effect phi).

Committed: https://crrev.com/1c2867c0dac589300817f430bf13a84eba4e8c07
Cr-Commit-Position: refs/heads/master@{#30927}

[Jarin, 2015-09-25 07:30:48]

jarin@chromium.org changed reviewers:
+ bmeurer@chromium.org, mstarzinger@chromium.org

[Jarin, 2015-09-25 07:38:57]

Could you take a look, please?

Michi, could you help me with fixing the control equivalence test? It connects
an End node to another End as input (see the ComputeEquivalence method), which
seems wrong. I tried to use the incoming end node as an end node of the graph,
but that did not work.

[Michael Starzinger, 2015-09-25 08:06:49]

https://codereview.chromium.org/1368913002/diff/40001/test/unittests/compiler...
File test/unittests/compiler/control-equivalence-unittest.cc (left):

https://codereview.chromium.org/1368913002/diff/40001/test/unittests/compiler...
test/unittests/compiler/control-equivalence-unittest.cc:110: Node* end =
End(start);
Removing this is fine with me. If you want to preserve it I suggest to introduce
a {Merge1} helper and build a single-linked chain "Start->Merge->End" instead of
the borked "Start->End->End" here.

[Jarin, 2015-09-25 08:13:33]

Resurrected the test, as you suggested.

https://codereview.chromium.org/1368913002/diff/40001/test/unittests/compiler...
File test/unittests/compiler/control-equivalence-unittest.cc (left):

https://codereview.chromium.org/1368913002/diff/40001/test/unittests/compiler...
test/unittests/compiler/control-equivalence-unittest.cc:110: Node* end =
End(start);
On 2015/09/25 08:06:49, Michael Starzinger wrote:
> Removing this is fine with me. If you want to preserve it I suggest to
introduce
> a {Merge1} helper and build a single-linked chain "Start->Merge->End" instead
of
> the borked "Start->End->End" here.

Done.

[Michael Starzinger, 2015-09-25 08:16:45]

LGTM from my end, just a nit.

https://codereview.chromium.org/1368913002/diff/60001/src/compiler/node-prope...
File src/compiler/node-properties.h (right):

https://codereview.chromium.org/1368913002/diff/60001/src/compiler/node-prope...
src/compiler/node-properties.h:110: static void Verify(Node* node);
nit: This is not really a "miscellaneous mutator" I think, can we move it out of
this section into the "miscellaneous utilities" section below?

[Jarin, 2015-09-25 08:20:06]

Good catch, moved now.

https://codereview.chromium.org/1368913002/diff/60001/src/compiler/node-prope...
File src/compiler/node-properties.h (right):

https://codereview.chromium.org/1368913002/diff/60001/src/compiler/node-prope...
src/compiler/node-properties.h:110: static void Verify(Node* node);
On 2015/09/25 08:16:45, Michael Starzinger wrote:
> nit: This is not really a "miscellaneous mutator" I think, can we move it out
of
> this section into the "miscellaneous utilities" section below?

Done.

[Jarin, 2015-09-25 08:25:53]

The CQ bit was checked by jarin@chromium.org

[Jarin, 2015-09-25 08:25:53]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1368913002/#ps80001
(title: "Tweak")

[commit-bot: I haz the power, 2015-09-25 08:26:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1368913002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1368913002/80001

[commit-bot: I haz the power, 2015-09-25 08:42:58]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-09-25 08:43:20]

Patchset 5 (id:??) landed as
https://crrev.com/1c2867c0dac589300817f430bf13a84eba4e8c07
Cr-Commit-Position: refs/heads/master@{#30927}