[ARM64] Optimize load followed by shift.

[ARM64] Optimize load followed by shift.

    Instead of loading 64 bits and shifting:

      ldr x0, [x1, #offset]
      asr x0, x0, #32

    directly load the interesting 32 bits and sign-extend:

      ldrsw x0, [x1, #offset+4]

BUG=

Committed: https://crrev.com/dc6b5109d7ee98965812c5a30bac413e3c29acfa
Cr-Commit-Position: refs/heads/master@{#38622}

[georgia.kouveli, 2016-08-12 12:41:20]

georgia.kouveli@arm.com changed reviewers:
+ bmeurer@chromium.org, epertoso@chromium.org

[georgia.kouveli, 2016-08-12 12:41:20]

Resurrecting patch set 1 of
https://codereview.chromium.org/2183923003.

I've removed the second pattern from the old patchset as it wasn't getting hit
anymore.

[epertoso, 2016-08-12 13:32:16]

lgtm

Thanks!

[georgia.kouveli, 2016-08-12 13:33:04]

The CQ bit was checked by georgia.kouveli@arm.com

[commit-bot: I haz the power, 2016-08-12 13:33:16]

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, 2016-08-12 13:55:36]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-08-12 13:55:50]

Description was changed from

==========
[ARM64] Optimize load followed by shift.

    Instead of loading 64 bits and shifting:

      ldr x0, [x1, #offset]
      asr x0, x0, #32

    directly load the interesting 32 bits and sign-extend:

      ldrsw x0, [x1, #offset+4]

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

to

==========
[ARM64] Optimize load followed by shift.

    Instead of loading 64 bits and shifting:

      ldr x0, [x1, #offset]
      asr x0, x0, #32

    directly load the interesting 32 bits and sign-extend:

      ldrsw x0, [x1, #offset+4]

BUG=

Committed: https://crrev.com/dc6b5109d7ee98965812c5a30bac413e3c29acfa
Cr-Commit-Position: refs/heads/master@{#38622}
==========

[commit-bot: I haz the power, 2016-08-12 13:55:51]

Patchset 1 (id:??) landed as
https://crrev.com/dc6b5109d7ee98965812c5a30bac413e3c29acfa
Cr-Commit-Position: refs/heads/master@{#38622}