[asm] Fix lots of invalid asm.js tests

[asm] Fix lots of invalid asm.js tests

I identified lots of asm.js tests that are actually not valid according
to the spec, hence they execute in default-javascript-mode.
This CL fixes most of them by adding additional type annotations.

The atomic tests are totally non-spec-compliant by expecting a fourth
argument, and infinite-loops-taken expects a function-type parameter,
so I did not fix those.
I also did not fix the regression tests.

R=titzer@chromium.org, bradnelson@chromium.org
BUG=v8:4203
Review-Url: https://codereview.chromium.org/2663243002
Cr-Commit-Position: refs/heads/master@{#43179}
Committed: https://chromium.googlesource.com/v8/v8/+/be1a5f7551268a0f5cb9119e04af9834125132d0

[Clemens Hammacher, 2017-01-31 16:49:35]

The CQ bit was checked by clemensh@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-31 16:49:40]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Clemens Hammacher, 2017-01-31 16:50:56]

This is blocked on http://crbug.com/v8/5912 (do-while-false.js does not
terminate).

[Clemens Hammacher, 2017-01-31 16:57:54]

The CQ bit was checked by clemensh@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-31 16:58:00]

Dry run: 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, 2017-01-31 17:11:47]

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

[commit-bot: I haz the power, 2017-01-31 17:11:49]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)
  v8_linux64_gyp_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng_trigg...)
  v8_win_nosnap_shared_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_nosnap_shared_rel_ng...)
  v8_win_nosnap_shared_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_nosnap_shared_rel_ng...)

[bradnelson, 2017-01-31 17:16:08]

lgtm

These tests predate the validator. I'd avoided changing these to valid asm way
back because v1 of the typer was more permissive (due to using types.h), and
because I didn't want to reduce coverage on the existing pipeline.
But now seems a good time to update them.

While you're at it, maybe we should also assert that they validate?
assertTrue(%IsWasmCode(Module));

https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
File test/mjsunit/asm/float32array-negative-offset.js (right):

https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
test/mjsunit/asm/float32array-negative-offset.js:15: return +MEM32[(i | 0) >>
2];
You shouldn't need the i | 0 (the >> can take an intish)

[Clemens Hammacher, 2017-01-31 18:26:23]

On 2017/01/31 at 17:16:08, bradnelson wrote:
> While you're at it, maybe we should also assert that they validate?
> assertTrue(%IsWasmCode(Module));

That would be nice, but the tests current execute both with --validate-asm and
without it. We would need another runtime function to do this check
conditionally.

https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
File test/mjsunit/asm/float32array-negative-offset.js (right):

https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
test/mjsunit/asm/float32array-negative-offset.js:15: return +MEM32[(i | 0) >>
2];
On 2017/01/31 at 17:16:08, bradnelson wrote:
> You shouldn't need the i | 0 (the >> can take an intish)

Done.

[titzer, 2017-01-31 20:34:30]

On 2017/01/31 18:26:23, Clemens Hammacher wrote:
> On 2017/01/31 at 17:16:08, bradnelson wrote:
> > While you're at it, maybe we should also assert that they validate?
> > assertTrue(%IsWasmCode(Module));
> 
> That would be nice, but the tests current execute both with --validate-asm and
> without it. We would need another runtime function to do this check
> conditionally.

You could, e.g., do something nasty like take the source of the function,
programmatically remove the "use asm" annotation, and then eval() it, thereby
executing in both modes.

> 
>
https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
> File test/mjsunit/asm/float32array-negative-offset.js (right):
> 
>
https://codereview.chromium.org/2663243002/diff/20001/test/mjsunit/asm/float3...
> test/mjsunit/asm/float32array-negative-offset.js:15: return +MEM32[(i | 0) >>
> 2];
> On 2017/01/31 at 17:16:08, bradnelson wrote:
> > You shouldn't need the i | 0 (the >> can take an intish)
> 
> Done.

[Clemens Hammacher, 2017-02-01 17:23:42]

On 2017/01/31 at 20:34:30, titzer wrote:
> On 2017/01/31 18:26:23, Clemens Hammacher wrote:
> > On 2017/01/31 at 17:16:08, bradnelson wrote:
> > > While you're at it, maybe we should also assert that they validate?
> > > assertTrue(%IsWasmCode(Module));
> > 
> > That would be nice, but the tests current execute both with --validate-asm
and
> > without it. We would need another runtime function to do this check
> > conditionally.
> 
> You could, e.g., do something nasty like take the source of the function,
programmatically remove the "use asm" annotation, and then eval() it, thereby
executing in both modes.

The point is that if we execute in --no-validate-asm mode, then
%IsWasmCode(Module) will be false. And AFAIC there is no way to check the
validate-asm flag from JS. We would have to introduce a IsValidAsmWasm runtime
function which checks IsWasmCode if validate-asm is on and returns true
otherwise.

[bradn, 2017-02-01 17:49:57]

bradnelson@google.com changed reviewers:
+ bradnelson@google.com

[bradn, 2017-02-01 17:49:58]

I think adding a runtime-test method to know would be useful.
But no need to do that in this change.

[Clemens Hammacher, 2017-02-13 18:24:07]

The CQ bit was checked by clemensh@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-13 18:24:20]

Dry run: 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, 2017-02-13 18:50:50]

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

[commit-bot: I haz the power, 2017-02-13 18:50:51]

Dry run: This issue passed the CQ dry run.

[Clemens Hammacher, 2017-02-14 09:25:00]

The CQ bit was checked by clemensh@chromium.org

[Clemens Hammacher, 2017-02-14 09:25:01]

The patchset sent to the CQ was uploaded after l-g-t-m from
bradnelson@chromium.org
Link to the patchset: https://codereview.chromium.org/2663243002/#ps40001
(title: "Rebase")

[commit-bot: I haz the power, 2017-02-14 09:25:09]

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, 2017-02-14 09:26:52]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1487064300525450,
"parent_rev": "0b287bd4ea5b799a6686c456ba4e877e65c94c4a", "commit_rev":
"be1a5f7551268a0f5cb9119e04af9834125132d0"}

[commit-bot: I haz the power, 2017-02-14 09:26:57]

Description was changed from

==========
[asm] Fix lots of invalid asm.js tests

I identified lots of asm.js tests that are actually not valid according
to the spec, hence they execute in default-javascript-mode.
This CL fixes most of them by adding additional type annotations.

The atomic tests are totally non-spec-compliant by expecting a fourth
argument, and infinite-loops-taken expects a function-type parameter,
so I did not fix those.
I also did not fix the regression tests.

R=titzer@chromium.org, bradnelson@chromium.org
BUG=v8:4203
==========

to

==========
[asm] Fix lots of invalid asm.js tests

I identified lots of asm.js tests that are actually not valid according
to the spec, hence they execute in default-javascript-mode.
This CL fixes most of them by adding additional type annotations.

The atomic tests are totally non-spec-compliant by expecting a fourth
argument, and infinite-loops-taken expects a function-type parameter,
so I did not fix those.
I also did not fix the regression tests.

R=titzer@chromium.org, bradnelson@chromium.org
BUG=v8:4203

Review-Url: https://codereview.chromium.org/2663243002
Cr-Commit-Position: refs/heads/master@{#43179}
Committed:
https://chromium.googlesource.com/v8/v8/+/be1a5f7551268a0f5cb9119e04af9834125...
==========

[commit-bot: I haz the power, 2017-02-14 09:26:58]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/v8/v8/+/be1a5f7551268a0f5cb9119e04af9834125...