Propagate Deopt reason to cpu-profiler

Propagate DeoptInfo to cpu-profiler

1) Deoptimizer::Reason was replaced with Deoptimizer::DeoptInfo
because it also has raw position. Also the old name clashes with DeoptReason enum.

2) c_entry_fp assignment call was added to EntryGenerator::Generate
So we can calculate sp and have a chance to record the stack for the deopting function.
btw it makes the test stable.

3) new kind of CodeEvents was added to cpu-profiler

4) GetDeoptInfo method was extracted from PrintDeoptLocation.
So it could be reused in cpu profiler.

BUG=452067
LOG=n

Committed: https://crrev.com/86cae1633caa31f2225b6932850fb13d3a21cf59
Cr-Commit-Position: refs/heads/master@{#26545}

[loislo, 2015-02-09 16:41:22]

Patchset #2 (id:20001) has been deleted

[loislo, 2015-02-09 16:41:36]

Patchset #1 (id:1) has been deleted

[loislo, 2015-02-09 17:49:57]

loislo@chromium.org changed reviewers:
+ alph@chromium.org, mstarzinger@chromium.org, svenpanne@chromium.org,
yurys@chromium.org

[loislo, 2015-02-09 17:49:58]

PTAL

[loislo, 2015-02-10 06:18:17]

Patchset #1 (id:40001) has been deleted

[loislo, 2015-02-10 06:18:30]

Patchset #2 (id:80001) has been deleted

[loislo, 2015-02-10 06:18:39]

Patchset #1 (id:60001) has been deleted

[loislo, 2015-02-10 06:18:56]

Patchset #1 (id:100001) has been deleted

[yurys, 2015-02-10 08:27:14]

lgtm

https://codereview.chromium.org/910773002/diff/120001/src/arm64/deoptimizer-a...
File src/arm64/deoptimizer-arm64.cc (right):

https://codereview.chromium.org/910773002/diff/120001/src/arm64/deoptimizer-a...
src/arm64/deoptimizer-arm64.cc:135: __ Mov(x3,
Operand(ExternalReference(Isolate::kCEntryFPAddress, isolate())));
May be use UseScratchRegisterScope to allocate scratch register?

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler-inl.h
File src/cpu-profiler-inl.h (right):

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler-inl.h#...
src/cpu-profiler-inl.h:42: entry->set_deopt_reason(deopt_reason);
Do do we only store last deoptimization reason/position? It'd be worth
mentioning if you are going to change this.

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler.h
File src/cpu-profiler.h (right):

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler.h#newc...
src/cpu-profiler.h:248: virtual void CodeDeoptEvent(Code* code, int bailout_id,
Address from,
from -> pc ?

[loislo, 2015-02-10 08:54:33]

https://codereview.chromium.org/910773002/diff/120001/src/arm64/deoptimizer-a...
File src/arm64/deoptimizer-arm64.cc (right):

https://codereview.chromium.org/910773002/diff/120001/src/arm64/deoptimizer-a...
src/arm64/deoptimizer-arm64.cc:135: __ Mov(x3,
Operand(ExternalReference(Isolate::kCEntryFPAddress, isolate())));
On 2015/02/10 08:27:13, yurys wrote:
> May be use UseScratchRegisterScope to allocate scratch register?

I don't think it makes sense here because the rest part of the code operates
with raw registers.

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler-inl.h
File src/cpu-profiler-inl.h (right):

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler-inl.h#...
src/cpu-profiler-inl.h:42: entry->set_deopt_reason(deopt_reason);
On 2015/02/10 08:27:14, yurys wrote:
> Do do we only store last deoptimization reason/position? It'd be worth
> mentioning if you are going to change this.

This part is not ready yet. The plan is to store deopt reasons in a list in
CpuProfileNode.

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler.h
File src/cpu-profiler.h (right):

https://codereview.chromium.org/910773002/diff/120001/src/cpu-profiler.h#newc...
src/cpu-profiler.h:248: virtual void CodeDeoptEvent(Code* code, int bailout_id,
Address from,
On 2015/02/10 08:27:14, yurys wrote:
> from -> pc ?

Done.

[loislo, 2015-02-10 08:54:53]

New patchsets have been uploaded after l-g-t-m from yurys@chromium.org

[loislo, 2015-02-10 10:32:55]

Patchset #3 (id:160001) has been deleted

[Sven Panne, 2015-02-10 13:13:01]

svenpanne@chromium.org changed reviewers:
+ jarin@chromium.org

[Sven Panne, 2015-02-10 13:13:01]

LGTM from my side, too, but jarin@ (a.k.a. "Mr. Deopt" :-) might want to have a
look, too...

[Jarin, 2015-02-10 14:00:27]

One question:

https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc
File src/deoptimizer.cc (right):

https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc#newc...
src/deoptimizer.cc:597: fp_to_sp_delta_));
I believe it is important that this does not trigger GC. Could we move this a
bit further down, into the DisallowHeapAllocation scope?

[loislo, 2015-02-10 14:07:22]

New patchsets have been uploaded after l-g-t-m from svenpanne@chromium.org

[loislo, 2015-02-10 14:07:29]

https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc
File src/deoptimizer.cc (right):

https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc#newc...
src/deoptimizer.cc:597: fp_to_sp_delta_));
On 2015/02/10 14:00:27, jarin wrote:
> I believe it is important that this does not trigger GC. Could we move this a
> bit further down, into the DisallowHeapAllocation scope?

it shouldn't. Done.

[Jarin, 2015-02-10 14:09:18]

On 2015/02/10 14:07:29, loislo wrote:
> https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc
> File src/deoptimizer.cc (right):
> 
>
https://codereview.chromium.org/910773002/diff/180001/src/deoptimizer.cc#newc...
> src/deoptimizer.cc:597: fp_to_sp_delta_));
> On 2015/02/10 14:00:27, jarin wrote:
> > I believe it is important that this does not trigger GC. Could we move this
a
> > bit further down, into the DisallowHeapAllocation scope?
> 
> it shouldn't. Done.

lgtm. Thanks!

[loislo, 2015-02-10 14:10:18]

The CQ bit was checked by loislo@chromium.org

[commit-bot: I haz the power, 2015-02-10 14:10:33]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/910773002/200001

[commit-bot: I haz the power, 2015-02-10 14:32:50]

Committed patchset #4 (id:200001)

[commit-bot: I haz the power, 2015-02-10 14:33:16]

Patchset 4 (id:??) landed as
https://crrev.com/86cae1633caa31f2225b6932850fb13d3a21cf59
Cr-Commit-Position: refs/heads/master@{#26545}