Implement %TypedArray%.prototype.sort

Implement %TypedArray%.prototype.sort

The sort method of TypedArrays sorts in numerical order by default.
This patch implements sorting based on Arrays and adds a test.
The length of %TypedArray%.prototype.sort, like Array.prototype.sort,
seems to be unspecified in ES6, so this patch lets it have the value
1, to match our interpretation for Array.prototype.sort (though 0
would also be a sensible length).

R=arv@chromium.org
BUG=v8:3578
LOG=Y

Committed: https://chromium.googlesource.com/v8/v8/+/540bb0c80beb66578c42e71378de81b40455a83a

[arv (Not doing code reviews), 2015-05-19 15:40:11]

The length is specified to 1.

https://people.mozilla.org/~jorendorff/es6-draft.html#sec-ecmascript-standard...

[arv (Not doing code reviews), 2015-05-19 15:49:31]

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js
File src/harmony-typedarray.js (right):

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:94: function TypedArrayComparefn(x, y) {
This is OK for now but I think we might want to inline this.

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:97: } else if ($isNaN(x)) {
no else after return

The order can be changed to reduce the number of tests:

if ($isNaN(x)) {
  return $isNaN(y) ? 0 : 1;
}
if ($isNaN(y)) {
  return -1;
}

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:102: return -1;
Do we care about -1? Generally this just has to be a negative number which
allows us to do:

return x - y;

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:108: } else if (!%_IsMinusZero(x) && %_IsMinusZero(y))
{
reorg to reduce test?

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
File test/mjsunit/harmony/typedarray-sort.js (right):

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
test/mjsunit/harmony/typedarray-sort.js:26: assertTrue(Object.is(expected[i],
value[i]), "expected " + expected[i] + " found " + value[i]);
assertSameValue should also work.

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
test/mjsunit/harmony/typedarray-sort.js:38: if (constructor == Float32Array ||
constructor == Float64Array) {
===

[dehrenberg, 2015-05-19 16:39:09]

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js
File src/harmony-typedarray.js (right):

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:94: function TypedArrayComparefn(x, y) {
On 2015/05/19 15:49:31, arv wrote:
> This is OK for now but I think we might want to inline this.

Acknowledged.

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:97: } else if ($isNaN(x)) {
On 2015/05/19 15:49:31, arv wrote:
> no else after return
> 
> The order can be changed to reduce the number of tests:
> 
> if ($isNaN(x)) {
>   return $isNaN(y) ? 0 : 1;
> }
> if ($isNaN(y)) {
>   return -1;
> }

Done.

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:102: return -1;
On 2015/05/19 15:49:31, arv wrote:
> Do we care about -1? Generally this just has to be a negative number which
> allows us to do:
> 
> return x - y;

Done.

https://codereview.chromium.org/1148513002/diff/1/src/harmony-typedarray.js#n...
src/harmony-typedarray.js:108: } else if (!%_IsMinusZero(x) && %_IsMinusZero(y))
{
On 2015/05/19 15:49:31, arv wrote:
> reorg to reduce test?

Done.

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
File test/mjsunit/harmony/typedarray-sort.js (right):

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
test/mjsunit/harmony/typedarray-sort.js:26: assertTrue(Object.is(expected[i],
value[i]), "expected " + expected[i] + " found " + value[i]);
On 2015/05/19 15:49:31, arv wrote:
> assertSameValue should also work.

Done.

https://codereview.chromium.org/1148513002/diff/1/test/mjsunit/harmony/typeda...
test/mjsunit/harmony/typedarray-sort.js:38: if (constructor == Float32Array ||
constructor == Float64Array) {
On 2015/05/19 15:49:31, arv wrote:
> ===

Done.

[arv (Not doing code reviews), 2015-05-19 16:41:26]

LGTM

[dehrenberg, 2015-05-19 16:52:03]

The CQ bit was checked by dehrenberg@chromium.org

[commit-bot: I haz the power, 2015-05-19 16:52:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1148513002/20001

[commit-bot: I haz the power, 2015-05-19 16:53:52]

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

[commit-bot: I haz the power, 2015-05-19 16:53:54]

Try jobs failed on following builders:
  v8_linux64_avx2_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel/builds/508)

[dehrenberg, 2015-05-20 00:10:21]

The CQ bit was checked by dehrenberg@chromium.org

[dehrenberg, 2015-05-20 00:10:21]

The patchset sent to the CQ was uploaded after l-g-t-m from arv@chromium.org
Link to the patchset: https://codereview.chromium.org/1148513002/#ps40001
(title: "rebase")

[commit-bot: I haz the power, 2015-05-20 00:20:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1148513002/40001

[commit-bot: I haz the power, 2015-05-20 00:23:03]

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

[commit-bot: I haz the power, 2015-05-20 00:23:04]

Try jobs failed on following builders:
  v8_linux_arm64_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel/builds/5834)
  v8_mac_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel/builds/5930)

[commit-bot: I haz the power, 2015-05-20 00:56:43]

Patchset 4 (id:??) landed as
https://crrev.com/540bb0c80beb66578c42e71378de81b40455a83a
Cr-Commit-Position: refs/heads/master@{#28494}

[adamk, 2015-05-20 00:56:44]

Committed patchset #4 (id:60001) manually as
540bb0c80beb66578c42e71378de81b40455a83a (presubmit successful).