Separate the array/string .length load from the bounds check.

Separate the array/string .length load from the bounds check.

This is done in preparation to inline more array and array view
operations.

Since the length is immutable for fixed length arrays, it allows
hoisting the load out of loops.
Committed: https://code.google.com/p/dart/source/detail?r=17984

[Vyacheslav Egorov (Google), 2013-02-01 14:44:30]

lgtm

[srdjan, 2013-02-01 19:01:28]

DBC

https://codereview.chromium.org/12088108/diff/1/runtime/vm/flow_graph_optimiz...
File runtime/vm/flow_graph_optimizer.cc (right):

https://codereview.chromium.org/12088108/diff/1/runtime/vm/flow_graph_optimiz...
runtime/vm/flow_graph_optimizer.cc:555: const bool is_immutable = (class_id !=
kGrowableObjectArrayCid);
It may be better to compute is_immutable by checking what class_id IS and not
what it IS NOT, since that could cause problems when adding new types in
TryReplaceWithStoreIndexed.

https://codereview.chromium.org/12088108/diff/1/runtime/vm/intermediate_langu...
File runtime/vm/intermediate_language.cc (right):

https://codereview.chromium.org/12088108/diff/1/runtime/vm/intermediate_langu...
runtime/vm/intermediate_language.cc:1672: case
MethodRecognizer::kStringBaseLength:
Why not add the byte arrays as well?

[Florian Schneider, 2013-02-06 13:08:19]

https://codereview.chromium.org/12088108/diff/1/runtime/vm/flow_graph_optimiz...
File runtime/vm/flow_graph_optimizer.cc (right):

https://codereview.chromium.org/12088108/diff/1/runtime/vm/flow_graph_optimiz...
runtime/vm/flow_graph_optimizer.cc:555: const bool is_immutable = (class_id !=
kGrowableObjectArrayCid);
On 2013/02/01 19:01:28, srdjan wrote:
> It may be better to compute is_immutable by checking what class_id IS and not
> what it IS NOT, since that could cause problems when adding new types in
> TryReplaceWithStoreIndexed.

Agree. I'll change it.

https://codereview.chromium.org/12088108/diff/1/runtime/vm/intermediate_langu...
File runtime/vm/intermediate_language.cc (right):

https://codereview.chromium.org/12088108/diff/1/runtime/vm/intermediate_langu...
runtime/vm/intermediate_language.cc:1672: case
MethodRecognizer::kStringBaseLength:
On 2013/02/01 19:01:28, srdjan wrote:
> Why not add the byte arrays as well?

Byte arrays are handled by kByteArrayBaseLength above.