Subzero: Register allocator performance improvements and simplifications.

Subzero: Register allocator performance improvements and simplifications.

This removes the redundancy between live ranges stored in the Variable and those stored in Liveness, by removing the Liveness copy.  After liveness analysis, live ranges are constructed directly into the Variable.

Also, the LiveRangeWrapper is removed and Variable * is directly used instead.  The original thought behind LiveRangeWrapper was that it could be extended to include live range splitting.  However, when/if live range splitting is implemented, it will probably involve creating a new variable with its own live range, and carrying around some extra bookkeeping until the split is committed, so such a wrapper probably won't be needed.

BUG= none
R=jvoung@chromium.org

Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=5ce0abb87e7134493febe99117c7a2e88d75efc4

[Jim Stichnoth, 2014-10-15 06:22:27]

stichnot@chromium.org changed reviewers:
+ jvoung@chromium.org

[Jim Stichnoth, 2014-10-15 06:24:48]

This gives just a few percent improvement, so it's more of a cleanup.

[jvoung (off chromium), 2014-10-15 16:42:36]

LGTM otherwise

https://codereview.chromium.org/656023002/diff/1/src/IceOperand.h
File src/IceOperand.h (right):

https://codereview.chromium.org/656023002/diff/1/src/IceOperand.h#newcode415
src/IceOperand.h:415: LiveRange &getLiveRange() { return Live; }
There's also the const LiveRange &getLiveRange() const below. Can that be used
instead?

[Jim Stichnoth, 2014-10-15 16:47:41]

https://codereview.chromium.org/656023002/diff/1/src/IceOperand.h
File src/IceOperand.h (right):

https://codereview.chromium.org/656023002/diff/1/src/IceOperand.h#newcode415
src/IceOperand.h:415: LiveRange &getLiveRange() { return Live; }
On 2014/10/15 16:42:36, jvoung wrote:
> There's also the const LiveRange &getLiveRange() const below. Can that be used
> instead?

Unfortunately, at least one caller needs the const version and at least one
needs the non-const version...

[Jim Stichnoth, 2014-10-15 17:17:06]

Committed patchset #1 (id:1) manually as
5ce0abb87e7134493febe99117c7a2e88d75efc4 (presubmit successful).