Oilpan: Enable lazy sweeping on non-oilpan builds

Oilpan: Enable lazy sweeping on non-oilpan builds

Currently lazy sweeping is enabled only on oilpan builds. Since the lazy sweeping has been stable on the oilpan builds for >3 months, I think it's now safe to enable it on non-oilpan builds as well.

BUG=480837
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=194808

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195022

[haraken, 2015-04-24 09:02:38]

haraken@chromium.org changed reviewers:
+ oilpan-reviews@chromium.org

[haraken, 2015-04-24 09:03:05]

Not yet for review.

We need to address https://codereview.chromium.org/1107573003/ first.

[haraken, 2015-04-24 13:52:06]

> We need to address https://codereview.chromium.org/1107573003/ first.

The above issue is gone (or we need to address it in a different way in the test
infrastructure), so this CL is ready for review. PTAL.

[haraken, 2015-05-01 10:48:41]

On 2015/04/24 13:52:06, haraken wrote:
> > We need to address https://codereview.chromium.org/1107573003/ first.
> 
> The above issue is gone (or we need to address it in a different way in the
test
> infrastructure), so this CL is ready for review. PTAL.

I think we should fix the ASan poisoning crashes before landing this CL.

[sof, 2015-05-01 10:52:49]

On 2015/05/01 10:48:41, haraken wrote:
> On 2015/04/24 13:52:06, haraken wrote:
> > > We need to address https://codereview.chromium.org/1107573003/ first.
> > 
> > The above issue is gone (or we need to address it in a different way in the
> test
> > infrastructure), so this CL is ready for review. PTAL.
> 
> I think we should fix the ASan poisoning crashes before landing this CL.

That might take a while -- I'd be positively surprised if the switch away from
one-shot timers to tasks will be a straightforward one.

[haraken, 2015-05-01 10:55:45]

On 2015/05/01 10:52:49, sof wrote:
> On 2015/05/01 10:48:41, haraken wrote:
> > On 2015/04/24 13:52:06, haraken wrote:
> > > > We need to address https://codereview.chromium.org/1107573003/ first.
> > > 
> > > The above issue is gone (or we need to address it in a different way in
the
> > test
> > > infrastructure), so this CL is ready for review. PTAL.
> > 
> > I think we should fix the ASan poisoning crashes before landing this CL.
> 
> That might take a while -- I'd be positively surprised if the switch away from
> one-shot timers to tasks will be a straightforward one.

Yeah, sounds plausible. Maybe it's better to land this one, keep it in trunk and
see how it goes (i.e., observe crash reports).

[haraken, 2015-05-01 10:56:44]

On 2015/05/01 10:55:45, haraken wrote:
> On 2015/05/01 10:52:49, sof wrote:
> > On 2015/05/01 10:48:41, haraken wrote:
> > > On 2015/04/24 13:52:06, haraken wrote:
> > > > > We need to address https://codereview.chromium.org/1107573003/ first.
> > > > 
> > > > The above issue is gone (or we need to address it in a different way in
> the
> > > test
> > > > infrastructure), so this CL is ready for review. PTAL.
> > > 
> > > I think we should fix the ASan poisoning crashes before landing this CL.
> > 
> > That might take a while -- I'd be positively surprised if the switch away
from
> > one-shot timers to tasks will be a straightforward one.
> 
> Yeah, sounds plausible. Maybe it's better to land this one, keep it in trunk
and
> see how it goes (i.e., observe crash reports).

Yes, that seems better to me. PTAL?

[sof, 2015-05-01 11:04:01]

On 2015/05/01 10:56:44, haraken wrote:
> On 2015/05/01 10:55:45, haraken wrote:
> > On 2015/05/01 10:52:49, sof wrote:
> > > On 2015/05/01 10:48:41, haraken wrote:
> > > > On 2015/04/24 13:52:06, haraken wrote:
> > > > > > We need to address https://codereview.chromium.org/1107573003/
first.
> > > > > 
> > > > > The above issue is gone (or we need to address it in a different way
in
> > the
> > > > test
> > > > > infrastructure), so this CL is ready for review. PTAL.
> > > > 
> > > > I think we should fix the ASan poisoning crashes before landing this CL.
> > > 
> > > That might take a while -- I'd be positively surprised if the switch away
> from
> > > one-shot timers to tasks will be a straightforward one.
> > 
> > Yeah, sounds plausible. Maybe it's better to land this one, keep it in trunk
> and
> > see how it goes (i.e., observe crash reports).
> 
> Yes, that seems better to me. PTAL?

Just to verify, you want to do it now -- on a Friday before Golden Week and
before the M44 branch cut ? :-)

I don't mind flushing out issues earlier, but just want to make sure timing has
been considered.

[sof, 2015-05-01 11:11:01]

sigbjornf@opera.com changed reviewers:
+ sigbjornf@opera.com

[sof, 2015-05-01 11:11:02]

lgtm

[haraken, 2015-05-01 12:26:19]

On 2015/05/01 11:04:01, sof wrote:
> On 2015/05/01 10:56:44, haraken wrote:
> > On 2015/05/01 10:55:45, haraken wrote:
> > > On 2015/05/01 10:52:49, sof wrote:
> > > > On 2015/05/01 10:48:41, haraken wrote:
> > > > > On 2015/04/24 13:52:06, haraken wrote:
> > > > > > > We need to address https://codereview.chromium.org/1107573003/
> first.
> > > > > > 
> > > > > > The above issue is gone (or we need to address it in a different way
> in
> > > the
> > > > > test
> > > > > > infrastructure), so this CL is ready for review. PTAL.
> > > > > 
> > > > > I think we should fix the ASan poisoning crashes before landing this
CL.
> > > > 
> > > > That might take a while -- I'd be positively surprised if the switch
away
> > from
> > > > one-shot timers to tasks will be a straightforward one.
> > > 
> > > Yeah, sounds plausible. Maybe it's better to land this one, keep it in
trunk
> > and
> > > see how it goes (i.e., observe crash reports).
> > 
> > Yes, that seems better to me. PTAL?
> 
> Just to verify, you want to do it now -- on a Friday before Golden Week and
> before the M44 branch cut ? :-)
> 
> I don't mind flushing out issues earlier, but just want to make sure timing
has
> been considered.

Yeah, I think it's better to land this one by M44. From the perspective of when
I can take a revert action quickly, today seems like a good timing :)

[haraken, 2015-05-01 12:26:23]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2015-05-01 12:26:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1055193004/1

[commit-bot: I haz the power, 2015-05-01 12:29:13]

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

[commit-bot: I haz the power, 2015-05-01 12:29:14]

Try jobs failed on following builders:
  mac_blink_compile_dbg on tryserver.blink (JOB_FAILED,
http://build.chromium.org/p/tryserver.blink/builders/mac_blink_compile_dbg/bu...)

[haraken, 2015-05-01 12:34:01]

The CQ bit was checked by haraken@chromium.org

[haraken, 2015-05-01 12:34:02]

The patchset sent to the CQ was uploaded after l-g-t-m from sigbjornf@opera.com
Link to the patchset: https://codereview.chromium.org/1055193004/#ps20001
(title: " ")

[commit-bot: I haz the power, 2015-05-01 12:34:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1055193004/20001

[commit-bot: I haz the power, 2015-05-01 14:38:19]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=194808

[jbroman, 2015-05-01 16:59:46]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/1123443002/ by jbroman@chromium.org.

The reason for reverting is: Broke content_browsertests, preventing Blink from
rolling. 3 tests failed:
    DevToolsAgentTest.DevToolsResumeOnClose
    V8SamplingProfilerTest.V8SamplingJitCodeEventsCollected
    V8SamplingProfilerTest.V8SamplingSamplesCollected

See, for example:
   
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_....

[sof, 2015-05-04 09:07:51]

On 2015/05/01 16:59:46, jbroman wrote:
> A revert of this CL (patchset #2 id:20001) has been created in
> https://codereview.chromium.org/1123443002/ by mailto:jbroman@chromium.org.
> 
> The reason for reverting is: Broke content_browsertests, preventing Blink from
> rolling. 3 tests failed:
>     DevToolsAgentTest.DevToolsResumeOnClose
>     V8SamplingProfilerTest.V8SamplingJitCodeEventsCollected
>     V8SamplingProfilerTest.V8SamplingSamplesCollected
> 
> See, for example:
>    
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_....

With https://codereview.chromium.org/1118383002/ in place, this CL can be
relanded.

[haraken, 2015-05-04 14:14:44]

On 2015/05/04 09:07:51, sof wrote:
> On 2015/05/01 16:59:46, jbroman wrote:
> > A revert of this CL (patchset #2 id:20001) has been created in
> > https://codereview.chromium.org/1123443002/ by mailto:jbroman@chromium.org.
> > 
> > The reason for reverting is: Broke content_browsertests, preventing Blink
from
> > rolling. 3 tests failed:
> >     DevToolsAgentTest.DevToolsResumeOnClose
> >     V8SamplingProfilerTest.V8SamplingJitCodeEventsCollected
> >     V8SamplingProfilerTest.V8SamplingSamplesCollected
> > 
> > See, for example:
> >    
> >
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_....
> 
> With https://codereview.chromium.org/1118383002/ in place, this CL can be
> relanded.

Thanks for all the help! Since I'll be OOO until Wed, let me reland this on Thu.

[haraken, 2015-05-06 23:59:24]

On 2015/05/04 14:14:44, haraken wrote:
> On 2015/05/04 09:07:51, sof wrote:
> > On 2015/05/01 16:59:46, jbroman wrote:
> > > A revert of this CL (patchset #2 id:20001) has been created in
> > > https://codereview.chromium.org/1123443002/ by
mailto:jbroman@chromium.org.
> > > 
> > > The reason for reverting is: Broke content_browsertests, preventing Blink
> from
> > > rolling. 3 tests failed:
> > >     DevToolsAgentTest.DevToolsResumeOnClose
> > >     V8SamplingProfilerTest.V8SamplingJitCodeEventsCollected
> > >     V8SamplingProfilerTest.V8SamplingSamplesCollected
> > > 
> > > See, for example:
> > >    
> > >
> >
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_....
> > 
> > With https://codereview.chromium.org/1118383002/ in place, this CL can be
> > relanded.
> 
> Thanks for all the help! Since I'll be OOO until Wed, let me reland this on
Thu.

Let me reland this :)

[haraken, 2015-05-06 23:59:30]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2015-05-06 23:59:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1055193004/20001

[commit-bot: I haz the power, 2015-05-07 04:36:34]

Committed patchset #2 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=195022

[sof, 2015-05-07 18:07:30]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/1133563004/ by sigbjornf@opera.com.

The reason for reverting is: The use of lazy sweeping appears to be behind this
ASan regression,


https://storage.googleapis.com/chromium-layout-test-archives/WebKit_Linux_ASA...

i.e., it risks lifecycle notification touching already finalized & freed
objects..