views: Remove content dependency from views package. Try 3!

views: Remove content dependency from views package. Try 3!

BUG=105557, 141187
TBR=ben@chromium.org

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=157019

[tfarina, 2012-09-09 19:10:09]

Dear chrome_frame guys!

Can you help me diagnose/understand WHY in THE HELL, chrome_frame_net_tests does
not pass with this ONE LINE CHANGE?

E:\b\build\slave\win\build\src\build\Debug\chrome_frame_net_tests.exe
--gtest_print_time
[0909/120534:ERROR:scoped_com_initializer.h(56)] Multiple CoInitialize() called
for thread 55672
[0909/120534:ERROR:proxy_service_factory.cc(88)] Cannot use V8 Proxy resolver in
single process mode.
[0909/120534:ERROR:proxy_service_factory.cc(88)] Cannot use V8 Proxy resolver in
single process mode.
[0909/120534:ERROR:scoped_com_initializer.h(56)] Multiple CoInitialize() called
for thread 55672
[0909/120540:ERROR:scoped_com_initializer.h(56)] Multiple CoInitialize() called
for thread 53192
4 new files were left in c:\users\chrome~2\appdata\local\temp: Fix the tests to
clean up themselves.

This is barring more I have to do!

Thanks!

[tfarina, 2012-09-09 19:10:53]

On 2012/09/09 19:10:09, tfarina wrote:
> This is barring more I have to do!
> 
more WORK, I mean.

[grt (UTC plus 2), 2012-09-09 22:22:45]

On 2012/09/09 19:10:53, tfarina wrote:
> On 2012/09/09 19:10:09, tfarina wrote:
> > This is barring more I have to do!
> > 
> more WORK, I mean.

you may be running into http://crbug.com/144649.  does this pass on the win_rel
bot?

[tfarina, 2012-09-09 22:26:39]

On 2012/09/09 22:22:45, grt wrote:
> you may be running into http://crbug.com/144649.  does this pass on the
win_rel
> bot?
Yep. In patch set 1 it passed.

http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[tommi (sloooow) - chröme, 2012-09-10 08:29:26]

lgtm as long as it builds! :)

[tfarina, 2012-09-10 12:11:45]

On Monday, September 10, 2012, wrote:

> lgtm as long as it builds! :)
>
>
It builds, but chrome_frame_net_tests does not pass in anyway.

That is why i need someone familiar with chrome_frame to discovery what's
going on.

http://codereview.chromium.**org/10915159/<http://codereview.chromium.org/109...
>


-- 
Thiago

[tommi (sloooow) - chröme, 2012-09-10 13:00:50]

I think Greg has replied to the proxy_service_factory problem and the
CoInitialize checks that were added recently seem to be the other problem
(unless I'm missing something?).  Neither of which is related to your
change and should be happening already on trunk.


On Mon, Sep 10, 2012 at 2:11 PM, Thiago Farina <tfarina@chromium.org> wrote:

>
>
> On Monday, September 10, 2012, wrote:
>
>> lgtm as long as it builds! :)
>>
>>
> It builds, but chrome_frame_net_tests does not pass in anyway.
>
> That is why i need someone familiar with chrome_frame to discovery what's
> going on.
>
>
http://codereview.chromium.**org/10915159/<http://codereview.chromium.org/109...
>>
>
>
> --
> Thiago
>

[tommi (sloooow) - chröme, 2012-09-10 13:08:36]

ah, sorry, I see that there's a crash.  I was just looking at the log.
(I'm looking at this now:
http://build.chromium.org/p/tryserver.chromium/builders/win/builds/26927/step...
)

The process does end with exit code -1073741819 or 0xc0000005, which means
'access denied'/'access violation'.  I don't suppose bit9 is running on the
bot!? I kid I kid.  It does sound like you're hitting what Greg is
describing, so that's unrelated to your change.


On Mon, Sep 10, 2012 at 3:00 PM, Tommi <tommi@chromium.org> wrote:

> I think Greg has replied to the proxy_service_factory problem and the
> CoInitialize checks that were added recently seem to be the other problem
> (unless I'm missing something?).  Neither of which is related to your
> change and should be happening already on trunk.
>
>
> On Mon, Sep 10, 2012 at 2:11 PM, Thiago Farina <tfarina@chromium.org>wrote:
>
>>
>>
>> On Monday, September 10, 2012, wrote:
>>
>>> lgtm as long as it builds! :)
>>>
>>>
>> It builds, but chrome_frame_net_tests does not pass in anyway.
>>
>> That is why i need someone familiar with chrome_frame to discovery what's
>> going on.
>>
>>
http://codereview.chromium.**org/10915159/<http://codereview.chromium.org/109...
>>>
>>
>>
>> --
>> Thiago
>>
>
>

[grt (UTC plus 2), 2012-09-10 16:01:06]

I finally got a local build, and can confirm that this test doesn't run in Debug
mode for reasons outside of your particular change.  We'll make fixing this a
priority so that other developers don't lose time on it.

[tommi (sloooow) - chröme, 2012-09-10 17:44:30]

There is a bug in the v8 code (ToAscii iirc) which assumes signed pointer
arithmetic. When it is passed a null ptr and zero length, it subtracts
sizeof (uintptr) from 0 and subsequently accesses a bogus address.  I fixed
this locally and verified that it fixes this problem but had to leave the
office before I could send an email to you and someone on the v8 team.

Sent from my phone.
On Sep 10, 2012 6:01 PM, <grt@chromium.org> wrote:

> I finally got a local build, and can confirm that this test doesn't run in
> Debug
> mode for reasons outside of your particular change.  We'll make fixing
> this a
> priority so that other developers don't lose time on it.
>
>
http://codereview.chromium.**org/10915159/<http://codereview.chromium.org/109...
>

[commit-bot: I haz the power, 2012-09-15 09:51:05]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/tfarina@chromium.org/10915159/2004

[commit-bot: I haz the power, 2012-09-15 10:22:04]

Step "update" is always a major failure.
Look at the try server FAQ for more details.