[interpreter] Fix nosnap build for interpreter table generation.

[interpreter] Fix nosnap build for interpreter table generation.

Moves the creation of the interpreter table early on during initialization
to ensure that even on nosnap builds it still gets allocated in the
first page.

BUG=v8:4280
LOG=N

Committed: https://crrev.com/cc74437ba70633fcee817a45aca74b41fec37d38
Cr-Commit-Position: refs/heads/master@{#30096}

[rmcilroy, 2015-08-10 14:13:37]

rmcilroy@chromium.org changed reviewers:
+ hpayer@chromium.org

[rmcilroy, 2015-08-10 14:13:37]

rmcilroy@chromium.org changed reviewers:
+ hpayer@chromium.org

[rmcilroy, 2015-08-10 14:13:37]

Hannes, could you have a quick look please (this is needed to green the tree).
Thanks!

[rmcilroy, 2015-08-10 14:13:37]

Hannes, could you have a quick look please (this is needed to green the tree).
Thanks!

[rmcilroy, 2015-08-10 14:14:01]

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

[commit-bot: I haz the power, 2015-08-10 14:14:06]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1278413002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1278413002/20001

[commit-bot: I haz the power, 2015-08-10 14:39:58]

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

[commit-bot: I haz the power, 2015-08-10 14:39:59]

Dry run: This issue passed the CQ dry run.

[Hannes Payer (out of office), 2015-08-10 15:26:17]

LGTM

[Hannes Payer (out of office), 2015-08-10 16:28:05]

https://codereview.chromium.org/1278413002/diff/40001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/40001/src/interpreter/interpr...
src/interpreter/interpreter.cc:43: if (initialized_) return;
What is the purpose of initialized_?

[rmcilroy, 2015-08-10 16:37:03]

https://codereview.chromium.org/1278413002/diff/40001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/40001/src/interpreter/interpr...
src/interpreter/interpreter.cc:43: if (initialized_) return;
On 2015/08/10 16:28:05, Hannes Payer wrote:
> What is the purpose of initialized_?

It was for the code in test-interpreter to check whether it was initialized
before calling initialize again, but agree, this is no longer useful - removed.

[Hannes Payer (out of office), 2015-08-10 16:50:00]

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
src/interpreter/interpreter.cc:41: void Interpreter::Initialize(bool
create_heap_objects) {
Why would you call this function with create_heap_objects=false?

[rmcilroy, 2015-08-10 16:50:18]

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

[commit-bot: I haz the power, 2015-08-10 16:50:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1278413002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1278413002/60001

[rmcilroy, 2015-08-10 16:54:36]

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
src/interpreter/interpreter.cc:41: void Interpreter::Initialize(bool
create_heap_objects) {
On 2015/08/10 16:50:00, Hannes Payer wrote:
> Why would you call this function with create_heap_objects=false?

It is called with create_heap_objects=false in Isolate::Init() if the
deserializer is non-null. We could possibly drop this for now (since it still
needs to check with IsInterpreterTableInitialized anyway) but I would like to
keep it for the time when the flag is enabled by default and we can remove
IsInterpreterTableInitialized (and only rely on create_heap_objects to decide
whether we need to generate the interpreter table).

[Hannes Payer (out of office), 2015-08-10 16:59:26]

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
src/interpreter/interpreter.cc:41: void Interpreter::Initialize(bool
create_heap_objects) {
On 2015/08/10 16:54:36, rmcilroy wrote:
> On 2015/08/10 16:50:00, Hannes Payer wrote:
> > Why would you call this function with create_heap_objects=false?
> 
> It is called with create_heap_objects=false in Isolate::Init() if the
> deserializer is non-null. We could possibly drop this for now (since it still
> needs to check with IsInterpreterTableInitialized anyway) but I would like to
> keep it for the time when the flag is enabled by default and we can remove
> IsInterpreterTableInitialized (and only rely on create_heap_objects to decide
> whether we need to generate the interpreter table).

Sure, but the semantics right now is strange. It is more like
force_create_heap_objects since it is happily ignored by the || condition. I
would remove it for now. You can always bring it back.

[rmcilroy, 2015-08-10 17:08:43]

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/1278413002/diff/60001/src/interpreter/interpr...
src/interpreter/interpreter.cc:41: void Interpreter::Initialize(bool
create_heap_objects) {
On 2015/08/10 16:59:25, Hannes Payer wrote:
> On 2015/08/10 16:54:36, rmcilroy wrote:
> > On 2015/08/10 16:50:00, Hannes Payer wrote:
> > > Why would you call this function with create_heap_objects=false?
> > 
> > It is called with create_heap_objects=false in Isolate::Init() if the
> > deserializer is non-null. We could possibly drop this for now (since it
still
> > needs to check with IsInterpreterTableInitialized anyway) but I would like
to
> > keep it for the time when the flag is enabled by default and we can remove
> > IsInterpreterTableInitialized (and only rely on create_heap_objects to
decide
> > whether we need to generate the interpreter table).
> 
> Sure, but the semantics right now is strange. It is more like
> force_create_heap_objects since it is happily ignored by the || condition. I
> would remove it for now. You can always bring it back.

Ok, done.

[rmcilroy, 2015-08-10 17:09:01]

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

[commit-bot: I haz the power, 2015-08-10 17:09:08]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1278413002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1278413002/80001

[commit-bot: I haz the power, 2015-08-10 17:45:53]

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

[commit-bot: I haz the power, 2015-08-10 17:45:54]

Dry run: This issue passed the CQ dry run.

[rmcilroy, 2015-08-10 18:20:23]

The CQ bit was checked by rmcilroy@chromium.org

[rmcilroy, 2015-08-10 18:20:23]

The patchset sent to the CQ was uploaded after l-g-t-m from hpayer@chromium.org
Link to the patchset: https://codereview.chromium.org/1278413002/#ps80001
(title: "Remove create_heap_objects")

[commit-bot: I haz the power, 2015-08-10 18:20:31]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1278413002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1278413002/80001

[commit-bot: I haz the power, 2015-08-10 18:22:12]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-08-10 18:22:28]

Patchset 5 (id:??) landed as
https://crrev.com/cc74437ba70633fcee817a45aca74b41fec37d38
Cr-Commit-Position: refs/heads/master@{#30096}