Ensure there is some space on JS stack available for bootstrapping.

Ensure there is some space on JS stack available for bootstrapping.

Committed: https://crrev.com/339ac2700decfeab70af8335aa57fe766278ba8c
Cr-Commit-Position: refs/heads/master@{#29256}

[Igor Sheludko, 2015-06-24 14:07:13]

Patchset #1 (id:1) has been deleted

[Igor Sheludko, 2015-06-24 14:08:17]

ishell@chromium.org changed reviewers:
+ yangguo@chromium.org

[Igor Sheludko, 2015-06-24 14:08:17]

PTAL

[Yang, 2015-06-24 14:11:20]

On 2015/06/24 14:08:17, Igor Sheludko wrote:
> PTAL

I assume that this can only happen if we actually run scripts during
bootstrapping, That may not be the case if we only deserialize from the
snapshot. Can you move the check to CompileNative, where we already have another
check as well?

Also, do we have a test case?

[Igor Sheludko, 2015-06-24 14:18:57]

On 2015/06/24 14:11:20, Yang wrote:
> On 2015/06/24 14:08:17, Igor Sheludko wrote:
> > PTAL
> 
> I assume that this can only happen if we actually run scripts during
> bootstrapping, That may not be the case if we only deserialize from the
> snapshot. Can you move the check to CompileNative, where we already have
another
> check as well?
Done.


> Also, do we have a test case?

Yes, that's the reason why arm64 nosnap currently fails on
test/mjsunit/regress/regress-crbug-501711.js.
I think it makes sense to require some stack space when we create a new realm
even if we have a snapshot.

[Yang, 2015-06-24 14:27:02]

On 2015/06/24 14:18:57, Igor Sheludko wrote:
> On 2015/06/24 14:11:20, Yang wrote:
> > On 2015/06/24 14:08:17, Igor Sheludko wrote:
> > > PTAL
> > 
> > I assume that this can only happen if we actually run scripts during
> > bootstrapping, That may not be the case if we only deserialize from the
> > snapshot. Can you move the check to CompileNative, where we already have
> another
> > check as well?
> Done.
> 
> 
> > Also, do we have a test case?
> 
> Yes, that's the reason why arm64 nosnap currently fails on
> test/mjsunit/regress/regress-crbug-501711.js.
> I think it makes sense to require some stack space when we create a new realm
> even if we have a snapshot.

lgtm. is there a line in mjsunit.status we can remove?

[Igor Sheludko, 2015-06-24 14:29:25]

On 2015/06/24 14:27:02, Yang wrote:
> On 2015/06/24 14:18:57, Igor Sheludko wrote:
> > On 2015/06/24 14:11:20, Yang wrote:
> > > On 2015/06/24 14:08:17, Igor Sheludko wrote:
> > > > PTAL
> > > 
> > > I assume that this can only happen if we actually run scripts during
> > > bootstrapping, That may not be the case if we only deserialize from the
> > > snapshot. Can you move the check to CompileNative, where we already have
> > another
> > > check as well?
> > Done.
> > 
> > 
> > > Also, do we have a test case?
> > 
> > Yes, that's the reason why arm64 nosnap currently fails on
> > test/mjsunit/regress/regress-crbug-501711.js.
> > I think it makes sense to require some stack space when we create a new
realm
> > even if we have a snapshot.
> 
> lgtm. is there a line in mjsunit.status we can remove?

No. That bot test just waits for this fix ))

[Igor Sheludko, 2015-06-24 14:29:33]

The CQ bit was checked by ishell@chromium.org

[commit-bot: I haz the power, 2015-06-24 14:29:53]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1203873005/40001

[commit-bot: I haz the power, 2015-06-24 14:57:44]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-06-24 14:58:03]

Patchset 2 (id:??) landed as
https://crrev.com/339ac2700decfeab70af8335aa57fe766278ba8c
Cr-Commit-Position: refs/heads/master@{#29256}