Re-reland: Remove register index/code indirection

Re-reland: Remove register index/code indirection

Previous to this patch, both the lithium and TurboFan register
allocators tracked allocated registers by "indices", rather than
the register codes used elsewhere in the runtime. This patch
ensures that codes are used everywhere, and in the process cleans
up a bunch of redundant code and adds more structure to how the
set of allocatable registers is defined.

Some highlights of changes:

* TurboFan's RegisterConfiguration class moved to V8's top level
  so that it can be shared with Crankshaft.
* Various "ToAllocationIndex" and related methods removed.
* Code that can be easily shared between Register classes on
  different platforms is now shared.
* The list of allocatable registers on each platform is declared
  as a list rather than implicitly via the register index <->
  code mapping.

Committed: https://crrev.com/80bc6f6e11f79524e3f1ad05579583adfd5f18b2
Cr-Commit-Position: refs/heads/master@{#30913}

Committed: https://crrev.com/7b7a8205d9a00c678fb7a6e032a55fecbc1509cf
Cr-Commit-Position: refs/heads/master@{#31075}

Committed: https://crrev.com/5cf1c0bcf6f10ff61a95a00fa302eeee7c390885
Cr-Commit-Position: refs/heads/master@{#31087}

[danno, 2015-08-21 10:29:15]

danno@chromium.org changed reviewers:
+ bmeurer@chromium.org, jarin@chromium.org, titzer@chromium.org

[danno, 2015-08-21 10:29:16]

This is cleanup/refactoring that makes it possible to use register codes in the
register allocator rather than allocation indices.

Although I think the cleanup is useful, it has turned into a bit of a monster
and would like to get your opinion before going further, e.g. getting the MIPS
folks to port.

[Benedikt Meurer, 2015-08-24 04:37:56]

Looking good. I think it's a useful refactoring.

[akos.palfi.imgtec, 2015-09-14 13:25:09]

akos.palfi@imgtec.com changed reviewers:
+ akos.palfi@imgtec.com

[akos.palfi.imgtec, 2015-09-14 13:25:10]

Hi Danno - We've created the MIPS ports
(https://codereview.chromium.org/1343533002), but there are a few cctest failure
the we need to investigate. We're working on the fix.

[paul.l..., 2015-09-15 02:09:45]

paul.lind@imgtec.com changed reviewers:
+ paul.lind@imgtec.com

[paul.l..., 2015-09-15 02:09:46]

Hey Danno - Akos mentioned the MIPS port CL. We still have bugs in
test-run-native-calls. If you want to land this now, you can add that to the
skip list for mips & mips64:
  'test-run-native-calls/*': [SKIP],

Everything else builds/tests fine. Hopefully we'll get this one fixed tomorrow.

https://codereview.chromium.org/1287383003/diff/240001/src/arm/deoptimizer-ar...
File src/arm/deoptimizer-arm.cc (right):

https://codereview.chromium.org/1287383003/diff/240001/src/arm/deoptimizer-ar...
src/arm/deoptimizer-arm.cc:1: Fdeop  // Copyright 2012 the V8 project authors.
All rights reserved.
extra chars, won't build.

[danno, 2015-09-15 15:43:45]

On 2015/09/15 at 02:09:46, paul.lind wrote:
> Hey Danno - Akos mentioned the MIPS port CL. We still have bugs in
test-run-native-calls. If you want to land this now, you can add that to the
skip list for mips & mips64:
>   'test-run-native-calls/*': [SKIP],
> 
> Everything else builds/tests fine. Hopefully we'll get this one fixed
tomorrow.
> 
>
https://codereview.chromium.org/1287383003/diff/240001/src/arm/deoptimizer-ar...
> File src/arm/deoptimizer-arm.cc (right):
> 
>
https://codereview.chromium.org/1287383003/diff/240001/src/arm/deoptimizer-ar...
> src/arm/deoptimizer-arm.cc:1: Fdeop  // Copyright 2012 the V8 project authors.
All rights reserved.
> extra chars, won't build.

Thanks for working on the port. There is no huge rush, let me know when you've
gotten all the tests to pass.

[akos.palfi.imgtec, 2015-09-15 22:19:32]

Hi Danno,

We've fixed the remaining bugs, the latest patchset shows no test failures. I
also had to modify the test-run-native-calls.cc to use the allocatable reg codes
everywhere instead of the raw indices (on MIPS the 0 and 1 indices are zero_reg
and at). PTAL at those changes.

https://codereview.chromium.org/1343533002

[danno, 2015-09-17 11:03:32]

On 2015/09/15 at 22:19:32, akos.palfi wrote:
> Hi Danno,
> 
> We've fixed the remaining bugs, the latest patchset shows no test failures. I
also had to modify the test-run-native-calls.cc to use the allocatable reg codes
everywhere instead of the raw indices (on MIPS the 0 and 1 indices are zero_reg
and at). PTAL at those changes.
> 
> https://codereview.chromium.org/1343533002

Hi Akos,

Thanks a lot for the patch! I'll integrate it into this one.

Danno

[danno, 2015-09-17 11:54:58]

With the MIPS port complete, I think this is ready for a first review, PTAL

[Benedikt Meurer, 2015-09-22 17:41:16]

bmeurer@chromium.org changed required reviewers:
+ jarin@chromium.org, titzer@chromium.org

[Benedikt Meurer, 2015-09-22 17:41:17]

As discussed offline earlier, I'm fine with the patch overall; it'll simplify
things long-term and I think it's important that we tackle such issues. I'm not
sure about some of the details, but I feel like I have to see this "in the wild"
to come up with useful suggestions (read: I might propose some small
refactorings later).

LGTM from me.

[danno, 2015-09-24 07:05:57]

On 2015/09/22 at 17:41:17, bmeurer wrote:
> As discussed offline earlier, I'm fine with the patch overall; it'll simplify
things long-term and I think it's important that we tackle such issues. I'm not
sure about some of the details, but I feel like I have to see this "in the wild"
to come up with useful suggestions (read: I might propose some small
refactorings later).
> 
> LGTM from me.

OK, thanks. Unless Jaro or Ben have any objections, I'll land this later today.

[danno, 2015-09-24 08:39:59]

The CQ bit was checked by danno@chromium.org

[danno, 2015-09-24 08:39:59]

The patchset sent to the CQ was uploaded after l-g-t-m from bmeurer@chromium.org
Link to the patchset: https://codereview.chromium.org/1287383003/#ps300001
(title: "Merge with ToT")

[commit-bot: I haz the power, 2015-09-24 08:40:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1287383003/300001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1287383003/300001

[commit-bot: I haz the power, 2015-09-24 08:40:26]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-24 08:40:27]

All required reviewers (with asterisk prefixes) have not yet approved this CL.

No L-G-T-M from a valid reviewer yet. Only full committers are accepted.
Even if an L-G-T-M may have been provided, it was from a non-committer,
_not_ a full super star committer.
See http://www.chromium.org/getting-involved/become-a-committer
Note that this has nothing to do with OWNERS files.

[titzer, 2015-09-24 09:54:45]

https://codereview.chromium.org/1287383003/diff/300001/test/mjsunit/regress/r...
File test/mjsunit/regress/regress-crbug-513507.js (right):

https://codereview.chromium.org/1287383003/diff/300001/test/mjsunit/regress/r...
test/mjsunit/regress/regress-crbug-513507.js:22: makeFun()(7);  // Warm up.
What's up here?

[titzer, 2015-09-24 09:56:37]

LGTM other than the previous comment.

[danno, 2015-09-24 10:43:26]

Stray test case change fixed, Jaro, any feedback?

https://codereview.chromium.org/1287383003/diff/300001/test/mjsunit/regress/r...
File test/mjsunit/regress/regress-crbug-513507.js (right):

https://codereview.chromium.org/1287383003/diff/300001/test/mjsunit/regress/r...
test/mjsunit/regress/regress-crbug-513507.js:22: makeFun()(7);  // Warm up.
On 2015/09/24 at 09:54:45, titzer wrote:
> What's up here?

The test change was probably a strange merge hiccup. Fixed.

[Jarin, 2015-09-24 12:29:39]

lgtm

[danno, 2015-09-24 12:30:34]

The CQ bit was checked by danno@chromium.org

[danno, 2015-09-24 12:30:34]

The patchset sent to the CQ was uploaded after l-g-t-m from
bmeurer@chromium.org, titzer@chromium.org
Link to the patchset: https://codereview.chromium.org/1287383003/#ps320001
(title: "Revert stray change")

[commit-bot: I haz the power, 2015-09-24 12:31:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1287383003/320001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1287383003/320001

[commit-bot: I haz the power, 2015-09-24 12:53:03]

Committed patchset #17 (id:320001)

[commit-bot: I haz the power, 2015-09-24 12:53:33]

Patchset 17 (id:??) landed as
https://crrev.com/80bc6f6e11f79524e3f1ad05579583adfd5f18b2
Cr-Commit-Position: refs/heads/master@{#30913}

[danno, 2015-09-24 13:38:00]

A revert of this CL (patchset #17 id:320001) has been created in
https://codereview.chromium.org/1365073002/ by danno@chromium.org.

The reason for reverting is: Failures on greedy RegAlloc, Fuzzer.

[danno, 2015-10-02 13:55:47]

The CQ bit was checked by danno@chromium.org

[danno, 2015-10-02 13:55:47]

The patchset sent to the CQ was uploaded after l-g-t-m from
bmeurer@chromium.org, titzer@chromium.org, jarin@chromium.org
Link to the patchset: https://codereview.chromium.org/1287383003/#ps380001
(title: "Fix MIPS")

[commit-bot: I haz the power, 2015-10-02 13:56:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1287383003/380001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1287383003/380001

[commit-bot: I haz the power, 2015-10-02 13:59:01]

Committed patchset #20 (id:380001)

[commit-bot: I haz the power, 2015-10-02 13:59:14]

Patchset 20 (id:??) landed as
https://crrev.com/7b7a8205d9a00c678fb7a6e032a55fecbc1509cf
Cr-Commit-Position: refs/heads/master@{#31075}

[danno, 2015-10-02 15:35:47]

A revert of this CL (patchset #20 id:380001) has been created in
https://codereview.chromium.org/1380863004/ by danno@chromium.org.

The reason for reverting is: Failures on MIPS.

[danno, 2015-10-02 15:58:10]

The CQ bit was checked by danno@chromium.org

[danno, 2015-10-02 15:58:11]

The patchset sent to the CQ was uploaded after l-g-t-m from
bmeurer@chromium.org, titzer@chromium.org, jarin@chromium.org
Link to the patchset: https://codereview.chromium.org/1287383003/#ps400001
(title: "Fix MIPS tests again")

[commit-bot: I haz the power, 2015-10-02 15:58:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1287383003/400001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1287383003/400001

[paul.l..., 2015-10-02 16:11:30]

Thanks for the fixes Danno! We were just starting to look after the revert, but
you beat us to it.

[commit-bot: I haz the power, 2015-10-02 16:55:19]

Committed patchset #21 (id:400001)

[commit-bot: I haz the power, 2015-10-02 16:55:38]

Patchset 21 (id:??) landed as
https://crrev.com/5cf1c0bcf6f10ff61a95a00fa302eeee7c390885
Cr-Commit-Position: refs/heads/master@{#31087}