Improve console message about deferring long-running timers.

Improve console message about deferring long-running timers.

BUG=574343
Committed: https://crrev.com/097f2cb6d4af73a27d0456a97bf5053f7f7db913
Cr-Commit-Position: refs/heads/master@{#395341}

[Jonathan Garbee, 2016-05-23 11:58:36]

Description was changed from

==========
Improve console message about deferring long-running timers.

BUG=
==========

to

==========
Improve console message about deferring long-running timers.

BUG=574343
==========

[Jonathan Garbee, 2016-05-23 11:58:37]

jonathan.garbee@chromium.org changed reviewers:
+ skyostil@chromium.org

[Jonathan Garbee, 2016-05-23 12:00:38]

Attempting to improve the console warning on timer tasks. It has been causing a
bit of confusion as developers see it.

1. Slightly tweaked wording to make it understandable. Please verify the message
still conveys what is intended.
2. Updated the crbug link to first be an actual link. Then it will also go
directly to comment #40 in the issue which includes the information to help
developers diagnose the issue.

PTAL

[Sami, 2016-05-23 13:16:37]

Thanks for the patch.

https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
File components/scheduler/renderer/renderer_scheduler_impl.cc (right):

https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
long-running timer task(s) to improve scrolling smoothness. "
What the message is trying to communicate is that the *browser* deferred some
tasks in favor of making scrolling smoother. The guidance for the web developer
should basically be: make sure your timer tasks take less than 50ms to run. (The
limit is about 6ms during user gestures, but I think we might want to avoid
printing the warning in that case since it is pretty spammy right now).

[Jonathan Garbee, 2016-05-23 15:05:21]

On 2016/05/23 at 13:16:37, skyostil wrote:
> Thanks for the patch.
> 
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> 
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
long-running timer task(s) to improve scrolling smoothness. "
> What the message is trying to communicate is that the *browser* deferred some
tasks in favor of making scrolling smoother. The guidance for the web developer
should basically be: make sure your timer tasks take less than 50ms to run. (The
limit is about 6ms during user gestures, but I think we might want to avoid
printing the warning in that case since it is pretty spammy right now).

Ooh, that means my wording is way off. However, an improvement over the current
could be...

"Chrome deferred a task in order to make scrolling smoother. Your timer tasks
should take less than 50ms to run to avoid this. Please see
https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
https://crbug.com/574343#c40 for more information."

I hate throwing a full MDN link in though, it is quite large. However, as a
console warning I don't see that as a big issue. The primary focus here is to
let developers understand the intent of the message and to get a good view of
how to fix it. "Timer tasks" is quasi-vague terminology unless people are
*really* into JS, so the MDN link can help offset that by letting them know
exactly what type of functions to look at when they are profiling and analyzing.

Does that sound good?

[Sami, 2016-05-23 15:25:38]

On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> On 2016/05/23 at 13:16:37, skyostil wrote:
> > Thanks for the patch.
> > 
> >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> > 
> >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
> long-running timer task(s) to improve scrolling smoothness. "
> > What the message is trying to communicate is that the *browser* deferred
some
> tasks in favor of making scrolling smoother. The guidance for the web
developer
> should basically be: make sure your timer tasks take less than 50ms to run.
(The
> limit is about 6ms during user gestures, but I think we might want to avoid
> printing the warning in that case since it is pretty spammy right now).
> 
> Ooh, that means my wording is way off. However, an improvement over the
current
> could be...
> 
> "Chrome deferred a task in order to make scrolling smoother. Your timer tasks
> should take less than 50ms to run to avoid this. Please see
> https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> https://crbug.com/574343#c40 for more information."
> 
> I hate throwing a full MDN link in though, it is quite large. However, as a
> console warning I don't see that as a big issue. The primary focus here is to
> let developers understand the intent of the message and to get a good view of
> how to fix it. "Timer tasks" is quasi-vague terminology unless people are
> *really* into JS, so the MDN link can help offset that by letting them know
> exactly what type of functions to look at when they are profiling and
analyzing.
> 
> Does that sound good?

That's definitely an improvement. How about linking here instead of MDN to give
a little more guidance?

https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...

I think I'll put together another patch which disables alerting for shorter
tasks.

[Jonathan Garbee, 2016-05-23 15:40:51]

On 2016/05/23 at 15:25:38, skyostil wrote:
> On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> > On 2016/05/23 at 13:16:37, skyostil wrote:
> > > Thanks for the patch.
> > > 
> > >
> >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> > > 
> > >
> >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
> > long-running timer task(s) to improve scrolling smoothness. "
> > > What the message is trying to communicate is that the *browser* deferred
some
> > tasks in favor of making scrolling smoother. The guidance for the web
developer
> > should basically be: make sure your timer tasks take less than 50ms to run.
(The
> > limit is about 6ms during user gestures, but I think we might want to avoid
> > printing the warning in that case since it is pretty spammy right now).
> > 
> > Ooh, that means my wording is way off. However, an improvement over the
current
> > could be...
> > 
> > "Chrome deferred a task in order to make scrolling smoother. Your timer
tasks
> > should take less than 50ms to run to avoid this. Please see
> > https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> > https://crbug.com/574343#c40 for more information."
> > 
> > I hate throwing a full MDN link in though, it is quite large. However, as a
> > console warning I don't see that as a big issue. The primary focus here is
to
> > let developers understand the intent of the message and to get a good view
of
> > how to fix it. "Timer tasks" is quasi-vague terminology unless people are
> > *really* into JS, so the MDN link can help offset that by letting them know
> > exactly what type of functions to look at when they are profiling and
analyzing.
> > 
> > Does that sound good?
> 
> That's definitely an improvement. How about linking here instead of MDN to
give a little more guidance?
> 
>
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...
> 
> I think I'll put together another patch which disables alerting for shorter
tasks.

Can the shorter task alert simply be throttled? Say to only go off once within a
minute? Probably too much, but worth asking.

Patch is updated as well now. The DGC link makes the line go to 99 columns. I'd
prefer we not split this up simply in case anyone is grepping for it to be able
to find it as a whole easily. Is that fine within the code style or should I
break out the end around 80 columns onto its own line with no space?

[Jonathan Garbee, 2016-05-23 15:45:22]

On 2016/05/23 at 15:40:51, Jonathan Garbee wrote:
> On 2016/05/23 at 15:25:38, skyostil wrote:
> > On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> > > On 2016/05/23 at 13:16:37, skyostil wrote:
> > > > Thanks for the patch.
> > > > 
> > > >
> > >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> > > > 
> > > >
> > >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
> > > long-running timer task(s) to improve scrolling smoothness. "
> > > > What the message is trying to communicate is that the *browser* deferred
some
> > > tasks in favor of making scrolling smoother. The guidance for the web
developer
> > > should basically be: make sure your timer tasks take less than 50ms to
run. (The
> > > limit is about 6ms during user gestures, but I think we might want to
avoid
> > > printing the warning in that case since it is pretty spammy right now).
> > > 
> > > Ooh, that means my wording is way off. However, an improvement over the
current
> > > could be...
> > > 
> > > "Chrome deferred a task in order to make scrolling smoother. Your timer
tasks
> > > should take less than 50ms to run to avoid this. Please see
> > > https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> > > https://crbug.com/574343#c40 for more information."
> > > 
> > > I hate throwing a full MDN link in though, it is quite large. However, as
a
> > > console warning I don't see that as a big issue. The primary focus here is
to
> > > let developers understand the intent of the message and to get a good view
of
> > > how to fix it. "Timer tasks" is quasi-vague terminology unless people are
> > > *really* into JS, so the MDN link can help offset that by letting them
know
> > > exactly what type of functions to look at when they are profiling and
analyzing.
> > > 
> > > Does that sound good?
> > 
> > That's definitely an improvement. How about linking here instead of MDN to
give a little more guidance?
> > 
> >
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...
> > 
> > I think I'll put together another patch which disables alerting for shorter
tasks.
> 
> Can the shorter task alert simply be throttled? Say to only go off once within
a minute? Probably too much, but worth asking.
> 
> Patch is updated as well now. The DGC link makes the line go to 99 columns.
I'd prefer we not split this up simply in case anyone is grepping for it to be
able to find it as a whole easily. Is that fine within the code style or should
I break out the end around 80 columns onto its own line with no space?

Although... Just thought as well on wording. Should we say "Blink" instead of
"Chrome"? Since if the warning goes off in Opera it would be weird seeing
"Chrome deferred..." since you aren't running Chrome itself. Along with any
other browsers built on top of the engine.

[Sami, 2016-05-23 15:48:17]

On 2016/05/23 15:40:51, Jonathan Garbee wrote:
> On 2016/05/23 at 15:25:38, skyostil wrote:
> > On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> > > On 2016/05/23 at 13:16:37, skyostil wrote:
> > > > Thanks for the patch.
> > > > 
> > > >
> > >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> > > > 
> > > >
> > >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
> > > long-running timer task(s) to improve scrolling smoothness. "
> > > > What the message is trying to communicate is that the *browser* deferred
> some
> > > tasks in favor of making scrolling smoother. The guidance for the web
> developer
> > > should basically be: make sure your timer tasks take less than 50ms to
run.
> (The
> > > limit is about 6ms during user gestures, but I think we might want to
avoid
> > > printing the warning in that case since it is pretty spammy right now).
> > > 
> > > Ooh, that means my wording is way off. However, an improvement over the
> current
> > > could be...
> > > 
> > > "Chrome deferred a task in order to make scrolling smoother. Your timer
> tasks
> > > should take less than 50ms to run to avoid this. Please see
> > > https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> > > https://crbug.com/574343#c40 for more information."
> > > 
> > > I hate throwing a full MDN link in though, it is quite large. However, as
a
> > > console warning I don't see that as a big issue. The primary focus here is
> to
> > > let developers understand the intent of the message and to get a good view
> of
> > > how to fix it. "Timer tasks" is quasi-vague terminology unless people are
> > > *really* into JS, so the MDN link can help offset that by letting them
know
> > > exactly what type of functions to look at when they are profiling and
> analyzing.
> > > 
> > > Does that sound good?
> > 
> > That's definitely an improvement. How about linking here instead of MDN to
> give a little more guidance?
> > 
> >
>
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...
> > 
> > I think I'll put together another patch which disables alerting for shorter
> tasks.
> 
> Can the shorter task alert simply be throttled? Say to only go off once within
a
> minute? Probably too much, but worth asking.

Right we only fire one alert per navigation, which should already limit things
quite a bit. The reason why I wanted to remove the shorter alert completely was
that we'd need to make the message dynamic (threshold is generally <10ms
depending on device speed) and whenever we'd use the lower limit, the deferred
task is going to run very soon anyway (e.g., after the user gesture ends). 

> Patch is updated as well now. The DGC link makes the line go to 99 columns.
I'd
> prefer we not split this up simply in case anyone is grepping for it to be
able
> to find it as a whole easily. Is that fine within the code style or should I
> break out the end around 80 columns onto its own line with no space?

I think it's fine to keep the longer line for greppability. lgtm.

[Sami, 2016-05-23 15:48:52]

On 2016/05/23 15:45:22, Jonathan Garbee wrote:
> On 2016/05/23 at 15:40:51, Jonathan Garbee wrote:
> > On 2016/05/23 at 15:25:38, skyostil wrote:
> > > On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> > > > On 2016/05/23 at 13:16:37, skyostil wrote:
> > > > > Thanks for the patch.
> > > > > 
> > > > >
> > > >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > > File components/scheduler/renderer/renderer_scheduler_impl.cc (right):
> > > > > 
> > > > >
> > > >
>
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > > components/scheduler/renderer/renderer_scheduler_impl.cc:1318: "Defer
> > > > long-running timer task(s) to improve scrolling smoothness. "
> > > > > What the message is trying to communicate is that the *browser*
deferred
> some
> > > > tasks in favor of making scrolling smoother. The guidance for the web
> developer
> > > > should basically be: make sure your timer tasks take less than 50ms to
> run. (The
> > > > limit is about 6ms during user gestures, but I think we might want to
> avoid
> > > > printing the warning in that case since it is pretty spammy right now).
> > > > 
> > > > Ooh, that means my wording is way off. However, an improvement over the
> current
> > > > could be...
> > > > 
> > > > "Chrome deferred a task in order to make scrolling smoother. Your timer
> tasks
> > > > should take less than 50ms to run to avoid this. Please see
> > > > https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> > > > https://crbug.com/574343#c40 for more information."
> > > > 
> > > > I hate throwing a full MDN link in though, it is quite large. However,
as
> a
> > > > console warning I don't see that as a big issue. The primary focus here
is
> to
> > > > let developers understand the intent of the message and to get a good
view
> of
> > > > how to fix it. "Timer tasks" is quasi-vague terminology unless people
are
> > > > *really* into JS, so the MDN link can help offset that by letting them
> know
> > > > exactly what type of functions to look at when they are profiling and
> analyzing.
> > > > 
> > > > Does that sound good?
> > > 
> > > That's definitely an improvement. How about linking here instead of MDN to
> give a little more guidance?
> > > 
> > >
>
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...
> > > 
> > > I think I'll put together another patch which disables alerting for
shorter
> tasks.
> > 
> > Can the shorter task alert simply be throttled? Say to only go off once
within
> a minute? Probably too much, but worth asking.
> > 
> > Patch is updated as well now. The DGC link makes the line go to 99 columns.
> I'd prefer we not split this up simply in case anyone is grepping for it to be
> able to find it as a whole easily. Is that fine within the code style or
should
> I break out the end around 80 columns onto its own line with no space?
> 
> Although... Just thought as well on wording. Should we say "Blink" instead of
> "Chrome"? Since if the warning goes off in Opera it would be weird seeing
> "Chrome deferred..." since you aren't running Chrome itself. Along with any
> other browsers built on top of the engine.

Good point, Blink is probably more appropriate.

[Jonathan Garbee, 2016-05-23 15:51:22]

On 2016/05/23 at 15:48:52, skyostil wrote:
> On 2016/05/23 15:45:22, Jonathan Garbee wrote:
> > On 2016/05/23 at 15:40:51, Jonathan Garbee wrote:
> > > On 2016/05/23 at 15:25:38, skyostil wrote:
> > > > On 2016/05/23 15:05:21, Jonathan Garbee wrote:
> > > > > On 2016/05/23 at 13:16:37, skyostil wrote:
> > > > > > Thanks for the patch.
> > > > > > 
> > > > > >
> > > > >
> >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > > > File components/scheduler/renderer/renderer_scheduler_impl.cc
(right):
> > > > > > 
> > > > > >
> > > > >
> >
https://codereview.chromium.org/1997413002/diff/20001/components/scheduler/re...
> > > > > > components/scheduler/renderer/renderer_scheduler_impl.cc:1318:
"Defer
> > > > > long-running timer task(s) to improve scrolling smoothness. "
> > > > > > What the message is trying to communicate is that the *browser*
deferred
> > some
> > > > > tasks in favor of making scrolling smoother. The guidance for the web
> > developer
> > > > > should basically be: make sure your timer tasks take less than 50ms to
> > run. (The
> > > > > limit is about 6ms during user gestures, but I think we might want to
> > avoid
> > > > > printing the warning in that case since it is pretty spammy right
now).
> > > > > 
> > > > > Ooh, that means my wording is way off. However, an improvement over
the
> > current
> > > > > could be...
> > > > > 
> > > > > "Chrome deferred a task in order to make scrolling smoother. Your
timer
> > tasks
> > > > > should take less than 50ms to run to avoid this. Please see
> > > > > https://developer.mozilla.org/en-US/Add-ons/Code_snippets/Timers and
> > > > > https://crbug.com/574343#c40 for more information."
> > > > > 
> > > > > I hate throwing a full MDN link in though, it is quite large. However,
as
> > a
> > > > > console warning I don't see that as a big issue. The primary focus
here is
> > to
> > > > > let developers understand the intent of the message and to get a good
view
> > of
> > > > > how to fix it. "Timer tasks" is quasi-vague terminology unless people
are
> > > > > *really* into JS, so the MDN link can help offset that by letting them
> > know
> > > > > exactly what type of functions to look at when they are profiling and
> > analyzing.
> > > > > 
> > > > > Does that sound good?
> > > > 
> > > > That's definitely an improvement. How about linking here instead of MDN
to
> > give a little more guidance?
> > > > 
> > > >
> >
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...
> > > > 
> > > > I think I'll put together another patch which disables alerting for
shorter
> > tasks.
> > > 
> > > Can the shorter task alert simply be throttled? Say to only go off once
within
> > a minute? Probably too much, but worth asking.
> > > 
> > > Patch is updated as well now. The DGC link makes the line go to 99
columns.
> > I'd prefer we not split this up simply in case anyone is grepping for it to
be
> > able to find it as a whole easily. Is that fine within the code style or
should
> > I break out the end around 80 columns onto its own line with no space?
> > 
> > Although... Just thought as well on wording. Should we say "Blink" instead
of
> > "Chrome"? Since if the warning goes off in Opera it would be weird seeing
> > "Chrome deferred..." since you aren't running Chrome itself. Along with any
> > other browsers built on top of the engine.
> 
> Good point, Blink is probably more appropriate.

Updated to Blink and publishing. Let's hope this lands for the M52 branch... :D

Thank you!

[Jonathan Garbee, 2016-05-23 15:51:30]

The CQ bit was checked by jonathan.garbee@chromium.org

[Jonathan Garbee, 2016-05-23 15:51:30]

The patchset sent to the CQ was uploaded after l-g-t-m from
skyostil@chromium.org
Link to the patchset: https://codereview.chromium.org/1997413002/#ps80001
(title: "Chrome > Blink")

[commit-bot: I haz the power, 2016-05-23 15:51:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1997413002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1997413002/80001

[Sami, 2016-05-23 15:54:17]

On 2016/05/23 15:51:22, Jonathan Garbee wrote:
> Updated to Blink and publishing. Let's hope this lands for the M52 branch...
:D

M52 already branched, but we can always backport this.

[Jonathan Garbee, 2016-05-23 15:56:03]

On 2016/05/23 at 15:54:17, skyostil wrote:
> On 2016/05/23 15:51:22, Jonathan Garbee wrote:
> > Updated to Blink and publishing. Let's hope this lands for the M52 branch...
:D
> 
> M52 already branched, but we can always backport this.

Once it lands safely let's look into that (I've never gone through that process
before.) I've had to explain this message a few times in the past week so people
are noticing it now and asking what it is. I'd like us to get the updated
message out ASAP so it is more useful.

[Sami, 2016-05-23 16:00:34]

On 2016/05/23 15:56:03, Jonathan Garbee wrote:
> Once it lands safely let's look into that (I've never gone through that
process
> before.) I've had to explain this message a few times in the past week so
people
> are noticing it now and asking what it is. I'd like us to get the updated
> message out ASAP so it is more useful.

Sure thing. By the way, if you've seen cases where the warning has actually been
useful (e.g., the page broke because a task was deferred), I'd love to hear
about it. So far the warning has mainly been extra noise as far as I can tell.

[Jonathan Garbee, 2016-05-23 16:31:28]

On 2016/05/23 at 16:00:34, skyostil wrote:
> On 2016/05/23 15:56:03, Jonathan Garbee wrote:
> > Once it lands safely let's look into that (I've never gone through that
process
> > before.) I've had to explain this message a few times in the past week so
people
> > are noticing it now and asking what it is. I'd like us to get the updated
> > message out ASAP so it is more useful.
> 
> Sure thing. By the way, if you've seen cases where the warning has actually
been useful (e.g., the page broke because a task was deferred), I'd love to hear
about it. So far the warning has mainly been extra noise as far as I can tell.

I'll let you know if I hear anything. Typically I help people figure out how to
use the DevTools to find their problem then I don't hear back. :/ I just hope I
helped them enough to get by. So far no "breakage" either, just developers
confused to what the warning is and why it happens.

IMO this "noise" is acceptable. While it would be better to only do it when
running a Timeline record; the time at which the record is occurring may not
concur with this problem. Therefore leaving developers unaware that it occurs. I
am going to try and discuss with DevRel guys and DevTools a better way to handle
these cases. Right now a console warning is fairly limited and crbug links alone
can be arcane. I'd like to have some simple place to input problem, debug steps,
and the expected solution that is easily linkable to a given topic.

Probably should fit under DGC/web somewhere ideally as you pointed out with the
RAIL intro recommendation for the link on this one. We are looking into more
instructional guides for debugging, so maybe doing a guide on this and linking
to that specifically would be a benefit.

[commit-bot: I haz the power, 2016-05-23 16:44:03]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-05-23 16:46:01]

Description was changed from

==========
Improve console message about deferring long-running timers.

BUG=574343
==========

to

==========
Improve console message about deferring long-running timers.

BUG=574343

Committed: https://crrev.com/097f2cb6d4af73a27d0456a97bf5053f7f7db913
Cr-Commit-Position: refs/heads/master@{#395341}
==========

[commit-bot: I haz the power, 2016-05-23 16:46:02]

Patchset 5 (id:??) landed as
https://crrev.com/097f2cb6d4af73a27d0456a97bf5053f7f7db913
Cr-Commit-Position: refs/heads/master@{#395341}

[abiramithanupillai_gmail.com, 2016-12-16 08:37:45]

Hi all,

I have created SAP UI5 application with swipe feature implemented.. *but 
when we i am debugging android devices i am getting the following warning.. 
after that swipe feature stopped working.. What should i do to make that 
feature working*

*Blink deferred a task in order to make scrolling smoother. Your timer and 
network tasks should take less than 50ms to run to avoid this. Please see 
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...

and https://crbug.com/574343#c40 for more information.*



On Monday, May 23, 2016 at 10:16:02 PM UTC+5:30, commit-bot@chromium.org 
via codereview.chromium.org wrote:
>
> Patchset 5 (id:??) landed as
> https://crrev.com/097f2cb6d4af73a27d0456a97bf5053f7f7db913
> Cr-Commit-Position: refs/heads/master@{#395341}
>
> https://codereview.chromium.org/1997413002/
>

[abiramithanupillai_gmail.com, 2016-12-16 08:37:46]

Hi all,

I have created SAP UI5 application with swipe feature implemented.. *but 
when we i am debugging android devices i am getting the following warning.. 
after that swipe feature stopped working.. What should i do to make that 
feature working*

*Blink deferred a task in order to make scrolling smoother. Your timer and 
network tasks should take less than 50ms to run to avoid this. Please see 
https://developers.google.com/web/tools/chrome-devtools/profile/evaluate-perf...

and https://crbug.com/574343#c40 for more information.*



On Monday, May 23, 2016 at 10:16:02 PM UTC+5:30, commit-bot@chromium.org 
via codereview.chromium.org wrote:
>
> Patchset 5 (id:??) landed as
> https://crrev.com/097f2cb6d4af73a27d0456a97bf5053f7f7db913
> Cr-Commit-Position: refs/heads/master@{#395341}
>
> https://codereview.chromium.org/1997413002/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.