Reland: [Autofill] Migrate ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs …

Reland: [Autofill] Migrate ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs to mojo.

Original CL was found breaking browser_test
PasswordManagerBrowserTestBase.InternalsPage on Mac bots.
Fix it and reland.

The original CL:
https://crrev.com/38eab5aeedba6745c0d9f7d7c897c16982299d88

Original CL description:
This CL
-Migrates all ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent
 IPC messages into mojo interfaces.
-Revises related unit tests and browser tests.

TBR=sky@chromium.org,tsepez@chromium.org
TEST=components_unittests, browser_tests
BUG=582391
Committed: https://crrev.com/97e8f751bc527212094f6fe65f1d96f709708991
Cr-Commit-Position: refs/heads/master@{#416593}

[leonhsl(Using Gerrit), 2016-08-24 03:04:00]

leon.han@intel.com changed reviewers:
+ vabr@chromium.org

[leonhsl(Using Gerrit), 2016-08-24 03:04:02]

Hi, Vaclav, I've created this reland CL and started Mac trybots to verify the
fix at ps#2.
Although there have some concerns about the message ordering according Ken's
comments in original CL, I'll wait for your feedback to determine whether we
reland this or not.
Before that, Let's wait for the trybots verification firstly. Thanks ;-)

[leonhsl(Using Gerrit), 2016-08-24 04:47:40]

On 2016/08/24 03:04:02, leonhsl wrote:
> Hi, Vaclav, I've created this reland CL and started Mac trybots to verify the
> fix at ps#2.
> Although there have some concerns about the message ordering according Ken's
> comments in original CL, I'll wait for your feedback to determine whether we
> reland this or not.
> Before that, Let's wait for the trybots verification firstly. Thanks ;-)

OK,, verification failed.. Let me investigate more.

[vabr (Chromium), 2016-08-24 08:12:14]

On 2016/08/24 04:47:40, leonhsl wrote:
> On 2016/08/24 03:04:02, leonhsl wrote:
> > Hi, Vaclav, I've created this reland CL and started Mac trybots to verify
the
> > fix at ps#2.
> > Although there have some concerns about the message ordering according Ken's
> > comments in original CL, I'll wait for your feedback to determine whether we
> > reland this or not.
> > Before that, Let's wait for the trybots verification firstly. Thanks ;-)
> 
> OK,, verification failed.. Let me investigate more.

Thanks for all the info so far, Han Leon.

I just started to look at the post-landing comments at the reverted CL and at
this CL. I have a Mac available, so may try to investigate locally if needed. I
will keep you updated and will be looking forward to hear updates if you have
any.

Cheers,
Vaclav

[vabr (Chromium), 2016-08-24 13:12:21]

On 2016/08/24 08:12:14, vabr (Chromium) wrote:
> On 2016/08/24 04:47:40, leonhsl wrote:
> > On 2016/08/24 03:04:02, leonhsl wrote:
> > > Hi, Vaclav, I've created this reland CL and started Mac trybots to verify
> the
> > > fix at ps#2.
> > > Although there have some concerns about the message ordering according
Ken's
> > > comments in original CL, I'll wait for your feedback to determine whether
we
> > > reland this or not.
> > > Before that, Let's wait for the trybots verification firstly. Thanks ;-)
> > 
> > OK,, verification failed.. Let me investigate more.
> 
> Thanks for all the info so far, Han Leon.
> 
> I just started to look at the post-landing comments at the reverted CL and at
> this CL. I have a Mac available, so may try to investigate locally if needed.
I
> will keep you updated and will be looking forward to hear updates if you have
> any.
> 
> Cheers,
> Vaclav

I'm afraid I did not find for this today (we had a slight pre-branch-point
emergency). I can hopefully try to have a look tomorrow on my Mac, if that would
help.

Cheers,
Vaclav

[leonhsl(Using Gerrit), 2016-08-25 04:16:32]

On 2016/08/24 13:12:21, vabr (Chromium) wrote:
> On 2016/08/24 08:12:14, vabr (Chromium) wrote:
> > On 2016/08/24 04:47:40, leonhsl wrote:
> > > On 2016/08/24 03:04:02, leonhsl wrote:
> > > > Hi, Vaclav, I've created this reland CL and started Mac trybots to
verify
> > the
> > > > fix at ps#2.
> > > > Although there have some concerns about the message ordering according
> Ken's
> > > > comments in original CL, I'll wait for your feedback to determine
whether
> we
> > > > reland this or not.
> > > > Before that, Let's wait for the trybots verification firstly. Thanks ;-)
> > > 
> > > OK,, verification failed.. Let me investigate more.
> > 
> > Thanks for all the info so far, Han Leon.
> > 
> > I just started to look at the post-landing comments at the reverted CL and
at
> > this CL. I have a Mac available, so may try to investigate locally if
needed.
> I
> > will keep you updated and will be looking forward to hear updates if you
have
> > any.
> > 
> > Cheers,
> > Vaclav
> 
> I'm afraid I did not find for this today (we had a slight pre-branch-point
> emergency). I can hopefully try to have a look tomorrow on my Mac, if that
would
> help.
> 
> Cheers,
> Vaclav

I think it would be very helpful ;-) As I have no idea yet until now..

[vabr (Chromium), 2016-08-25 13:44:32]

On 2016/08/25 04:16:32, leonhsl wrote:
> On 2016/08/24 13:12:21, vabr (Chromium) wrote:
> > On 2016/08/24 08:12:14, vabr (Chromium) wrote:
> > > On 2016/08/24 04:47:40, leonhsl wrote:
> > > > On 2016/08/24 03:04:02, leonhsl wrote:
> > > > > Hi, Vaclav, I've created this reland CL and started Mac trybots to
> verify
> > > the
> > > > > fix at ps#2.
> > > > > Although there have some concerns about the message ordering according
> > Ken's
> > > > > comments in original CL, I'll wait for your feedback to determine
> whether
> > we
> > > > > reland this or not.
> > > > > Before that, Let's wait for the trybots verification firstly. Thanks
;-)
> > > > 
> > > > OK,, verification failed.. Let me investigate more.
> > > 
> > > Thanks for all the info so far, Han Leon.
> > > 
> > > I just started to look at the post-landing comments at the reverted CL and
> at
> > > this CL. I have a Mac available, so may try to investigate locally if
> needed.
> > I
> > > will keep you updated and will be looking forward to hear updates if you
> have
> > > any.
> > > 
> > > Cheers,
> > > Vaclav
> > 
> > I'm afraid I did not find for this today (we had a slight pre-branch-point
> > emergency). I can hopefully try to have a look tomorrow on my Mac, if that
> would
> > help.
> > 
> > Cheers,
> > Vaclav
> 
> I think it would be very helpful ;-) As I have no idea yet until now..

Sorry, I'm still very slow today (and out of office tomorrow). So far I only
managed to confirm that PasswordManagerBrowserTestBase.InternalsPage fails
locally on my Mac and runs fine on my Linux. Will report anything more I find,
but sadly that won't be before Monday.

Cheers,
Vaclav

[vabr (Chromium), 2016-08-29 08:55:42]

As http://crbug.com/640737 shows, the failing test is already flaky without this
patch. I guess the switch to Mojo only amplified what was already wrong with the
test. I'll have a look and update both this CL and that bug.

[leonhsl(Using Gerrit), 2016-08-29 09:10:24]

On 2016/08/29 08:55:42, vabr (Chromium) wrote:
> As http://crbug.com/640737 shows, the failing test is already flaky without
this
> patch. I guess the switch to Mojo only amplified what was already wrong with
the
> test. I'll have a look and update both this CL and that bug.

OK, got it and big thanks for the information sharing!

[vabr (Chromium), 2016-08-29 10:08:13]

Hi Han Leon!

The test is unrelated to your CL. https://codereview.chromium.org/2287143003/ is
disabling it on Mac. Once that lands, I suggest you delete patch set 2 here and
rerun your trybots on the very original patch (the one which was reverted).

As for the discussion about security concerns, the only one I heard so far about
was that in the direction renderer->browser, the navigation messages could be
overtaken by requests for autofill data, which could result in filling in data
under wrong origin. However, ContentPasswordManagerDriver::OnPasswordFormsParsed
has a check which should prevent this, if I understand correctly. I could not
find any other potential issues, so I have no reason to stop you from relanding
this CL. Once the trybots are green, you have my LGTM to reland the original
patch set (the one which was reverted). According to
https://www.chromium.org/developers/owners-files#TOC-When-to-use-To-Be-Review...,
that should be enough for you to reland the CL. Given the recent discussions on
the original CL, I would still suggest waiting until tomorrow with the actual
landing.

Cheers,
Vaclav

[vabr (Chromium), 2016-08-29 13:08:11]

Hmm, so actually, let me hold this off once more. I checked again, with a fixed
version of the failing test (from https://codereview.chromium.org/2288203002),
and the test still fails. The messages with logs from the renderer site are not
getting through for some reason. I need to investigate more, which I may need to
do tomorrow. Until then, not LGTM, until it is clarified whether the test fails
for a good reason or not.

Sorry about that!
Vaclav

[leonhsl(Using Gerrit), 2016-08-30 02:05:59]

On 2016/08/29 13:08:11, vabr (Chromium) wrote:
> Hmm, so actually, let me hold this off once more. I checked again, with a
fixed
> version of the failing test (from https://codereview.chromium.org/2288203002),
> and the test still fails. The messages with logs from the renderer site are
not
> getting through for some reason. I need to investigate more, which I may need
to
> do tomorrow. Until then, not LGTM, until it is clarified whether the test
fails
> for a good reason or not.
> 
> Sorry about that!
> Vaclav

Got it. Thank you very much for kindly help and all the information sharing. I
have no Mac at hand, but I'm trying to figure out the scenario by reading code.

[leonhsl(Using Gerrit), 2016-08-30 02:10:13]

Patchset #2 (id:20001) has been deleted

[leonhsl(Using Gerrit), 2016-08-30 02:10:22]

The CQ bit was checked by leon.han@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-08-30 02:44:59]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[leonhsl(Using Gerrit), 2016-08-30 03:35:18]

On 2016/08/29 10:08:13, vabr (Chromium) wrote:
> Hi Han Leon!
> 
> The test is unrelated to your CL. https://codereview.chromium.org/2287143003/
is
> disabling it on Mac. Once that lands, I suggest you delete patch set 2 here
and
> rerun your trybots on the very original patch (the one which was reverted).
> 
> As for the discussion about security concerns, the only one I heard so far
about
> was that in the direction renderer->browser, the navigation messages could be
> overtaken by requests for autofill data, which could result in filling in data
> under wrong origin. However,
ContentPasswordManagerDriver::OnPasswordFormsParsed
> has a check which should prevent this, if I understand correctly. I could not
> find any other potential issues, so I have no reason to stop you from
relanding
> this CL. Once the trybots are green, you have my LGTM to reland the original
> patch set (the one which was reverted). According to
>
https://www.chromium.org/developers/owners-files#TOC-When-to-use-To-Be-Review...,
> that should be enough for you to reland the CL. Given the recent discussions
on
> the original CL, I would still suggest waiting until tomorrow with the actual
> landing.
> 
> Cheers,
> Vaclav

I was checking around for crbug.com/641156 and I suppose it relates with the
issue about message ordering with navigation messages mentioned here.. The crash
happened in PasswordAutofillAgent::FillPasswordForm() and seems has disappeared
after the original mojofication CL got reverted.. Maybe it also has some
connections with the failing test InternalsPage here? Thanks~

[commit-bot: I haz the power, 2016-08-30 05:01:17]

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

[commit-bot: I haz the power, 2016-08-30 05:01:18]

Dry run: This issue passed the CQ dry run.

[vabr (Chromium), 2016-08-30 09:46:56]

So far I saw this:

All wiring seems to work, in particular
ContentPasswordManagerDriver::SendLoggingAvailability gets its message across
into PasswordAutofillAgent::SetLoggingState.

The only issue is that in the test, PasswordAutofillAgent::SetLoggingState
receives it too late, after the renderer tries to log anything. That's why the
test fails. It is not a problem in production -- a human using the internals
page will open the page and then switch back to the browsing tab in a slow human
pace, which will allow all Mojo messages to go through. I'll fix the test, and
once I do, unblock this reland.

Thanks for your patience!

Vaclav

[leonhsl(Using Gerrit), 2016-08-30 10:16:49]

On 2016/08/30 09:46:56, vabr (Chromium) wrote:
> So far I saw this:
> 
> All wiring seems to work, in particular
> ContentPasswordManagerDriver::SendLoggingAvailability gets its message across
> into PasswordAutofillAgent::SetLoggingState.
> 
> The only issue is that in the test, PasswordAutofillAgent::SetLoggingState
> receives it too late, after the renderer tries to log anything. That's why the
> test fails. It is not a problem in production -- a human using the internals
> page will open the page and then switch back to the browsing tab in a slow
human
> pace, which will allow all Mojo messages to go through. I'll fix the test, and
> once I do, unblock this reland.
> 
> Thanks for your patience!
> 
> Vaclav

Ha! Got it and big thanks!

[vabr (Chromium), 2016-08-30 15:06:13]

Hello and sorry for still delaying this.

Interestingly the delay only happens in the test. I have a discussion on
https://codereview.chromium.org/2288203002/ with vasilii@, there are still some
open questions. I'd like to solve that before relanding this patch, but if you
need to land this quickly, we can also disable the test and I'll continue
investigating while your patch is in. If you choose the latter, you have my LGTM
again.

Cheers,
Vaclav

[leonhsl(Using Gerrit), 2016-08-31 03:28:08]

On 2016/08/30 15:06:13, vabr (Chromium) wrote:
> Hello and sorry for still delaying this.
> 
> Interestingly the delay only happens in the test. I have a discussion on
> https://codereview.chromium.org/2288203002/ with vasilii@, there are still
some
> open questions. I'd like to solve that before relanding this patch, but if you
> need to land this quickly, we can also disable the test and I'll continue
> investigating while your patch is in. If you choose the latter, you have my
LGTM
> again.
> 
> Cheers,
> Vaclav

No need to say sorry.. You are helping me a lot as I'm still somehow confused
about the message ordering between navigation events and autofill messages. I
agree to land this CL after we get the test failure fixed, in the meantime I
think I could investigate more inside autofill scenario. ;-) Thanks again for
your kindly consideration.

[leonhsl(Using Gerrit), 2016-09-05 03:04:29]

The CQ bit was checked by leon.han@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-09-05 03:04:40]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-05 03:32:27]

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

[commit-bot: I haz the power, 2016-09-05 03:32:28]

Dry run: Exceeded global retry quota

[leonhsl(Using Gerrit), 2016-09-05 05:50:12]

The CQ bit was checked by leon.han@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-09-05 05:50:29]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-05 05:53:28]

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

[commit-bot: I haz the power, 2016-09-05 05:53:29]

Dry run: This issue passed the CQ dry run.

[leonhsl(Using Gerrit), 2016-09-05 06:08:44]

The CQ bit was checked by leon.han@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-09-05 06:08:58]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-05 06:12:12]

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

[commit-bot: I haz the power, 2016-09-05 06:12:13]

Dry run: This issue passed the CQ dry run.

[leonhsl(Using Gerrit), 2016-09-05 08:00:19]

Uploaded ps#2 to do rebase as all gyp files have been removed completely, Thanks
a lot Vaclav!

[vabr (Chromium), 2016-09-05 14:44:52]

Thanks for the rebase, Han Leon!

https://codereview.chromium.org/2288203002 just landed, which should mean that
the InternalsPage test should not fail when you reland this (in the unlikely
event that it will, http://crbug.com/640737#c8 has a plan B).

Given what
https://www.chromium.org/developers/owners-files#TOC-When-to-use-To-Be-Review...
says about relanding, you should be able to TBR owners for the changes here and
land right away, because you have another committers approval (mine: LGTM) and
the only changes compared to the original CL are rebasing. Technically,
components/autofill/content/public/interfaces/*.mojom also have whitespace
changes coming from trimming the blank lines to 0 characters, but that's minor.
If you want to be super careful, you can ask the security owners to review the
diff (patch set 1 vs. 2), but that seems optional.

Thanks!
Vaclav

[leonhsl(Using Gerrit), 2016-09-06 02:59:49]

On 2016/09/05 14:44:52, vabr (Chromium) wrote:
> Thanks for the rebase, Han Leon!
> 
> https://codereview.chromium.org/2288203002 just landed, which should mean that
> the InternalsPage test should not fail when you reland this (in the unlikely
> event that it will, http://crbug.com/640737#c8 has a plan B).
> 
> Given what
>
https://www.chromium.org/developers/owners-files#TOC-When-to-use-To-Be-Review...
> says about relanding, you should be able to TBR owners for the changes here
and
> land right away, because you have another committers approval (mine: LGTM) and
> the only changes compared to the original CL are rebasing. Technically,
> components/autofill/content/public/interfaces/*.mojom also have whitespace
> changes coming from trimming the blank lines to 0 characters, but that's
minor.
> If you want to be super careful, you can ask the security owners to review the
> diff (patch set 1 vs. 2), but that seems optional.
> 
> Thanks!
> Vaclav

Got it and thanks for your great support on this! Landing this CL now with TBR
other reviewers. Will fight for the final step to remove autofill_messages.h
completely~

[leonhsl(Using Gerrit), 2016-09-06 03:11:32]

Description was changed from

==========
Reland: [Autofill] Migrate
ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs to mojo.

Original CL was found breaking browser_test
PasswordManagerBrowserTestBase.InternalsPage on Mac bots.
Fix it and reland.

The original CL:
https://crrev.com/38eab5aeedba6745c0d9f7d7c897c16982299d88

Original CL description:
This CL
-Migrates all ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent
 IPC messages into mojo interfaces.
-Revises related unit tests and browser tests.

TEST=components_unittests, browser_tests
BUG=582391
==========

to

==========
Reland: [Autofill] Migrate
ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs to mojo.

Original CL was found breaking browser_test
PasswordManagerBrowserTestBase.InternalsPage on Mac bots.
Fix it and reland.

The original CL:
https://crrev.com/38eab5aeedba6745c0d9f7d7c897c16982299d88

Original CL description:
This CL
-Migrates all ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent
 IPC messages into mojo interfaces.
-Revises related unit tests and browser tests.

TBR=sky@chromium.org,tsepez@chromium.org
TEST=components_unittests, browser_tests
BUG=582391
==========

[leonhsl(Using Gerrit), 2016-09-06 03:12:18]

The CQ bit was checked by leon.han@intel.com

[commit-bot: I haz the power, 2016-09-06 03:12:31]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-06 04:34:17]

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

[commit-bot: I haz the power, 2016-09-06 04:34:18]

Try jobs failed on following builders:
  linux_android_rel_ng on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

[leonhsl(Using Gerrit), 2016-09-06 05:37:56]

The CQ bit was checked by leon.han@intel.com

[commit-bot: I haz the power, 2016-09-06 05:38:18]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-06 06:13:39]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2016-09-06 06:15:03]

Description was changed from

==========
Reland: [Autofill] Migrate
ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs to mojo.

Original CL was found breaking browser_test
PasswordManagerBrowserTestBase.InternalsPage on Mac bots.
Fix it and reland.

The original CL:
https://crrev.com/38eab5aeedba6745c0d9f7d7c897c16982299d88

Original CL description:
This CL
-Migrates all ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent
 IPC messages into mojo interfaces.
-Revises related unit tests and browser tests.

TBR=sky@chromium.org,tsepez@chromium.org
TEST=components_unittests, browser_tests
BUG=582391
==========

to

==========
Reland: [Autofill] Migrate
ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent IPCs to mojo.

Original CL was found breaking browser_test
PasswordManagerBrowserTestBase.InternalsPage on Mac bots.
Fix it and reland.

The original CL:
https://crrev.com/38eab5aeedba6745c0d9f7d7c897c16982299d88

Original CL description:
This CL
-Migrates all ContentPasswordManagerDriver<-->Password{Autofill,Generation}Agent
 IPC messages into mojo interfaces.
-Revises related unit tests and browser tests.

TBR=sky@chromium.org,tsepez@chromium.org
TEST=components_unittests, browser_tests
BUG=582391

Committed: https://crrev.com/97e8f751bc527212094f6fe65f1d96f709708991
Cr-Commit-Position: refs/heads/master@{#416593}
==========

[commit-bot: I haz the power, 2016-09-06 06:15:05]

Patchset 2 (id:??) landed as
https://crrev.com/97e8f751bc527212094f6fe65f1d96f709708991
Cr-Commit-Position: refs/heads/master@{#416593}

[jam, 2016-09-08 06:11:14]

there are missing dependencies and it's causing build failures, i.e.
https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64...

are you fixing this?

FAILED: obj/chrome/browser/test_support_ui/password_manager_test_base.obj 
ninja -t msvc -e environment.x64 -- E:\b\c\cipd\goma/gomacc.exe
"E:\b\depot_tools\win_toolchain\vs_files\d5dc33b15d1b2c086f2f6632e2fd15882f80dbd3\VC\bin\amd64/cl.exe"
/nologo /showIncludes /FC
@obj/chrome/browser/test_support_ui/password_manager_test_base.obj.rsp /c
../../chrome/browser/password_manager/password_manager_test_base.cc
/Foobj/chrome/browser/test_support_ui/password_manager_test_base.obj
/Fd"obj/chrome/browser/test_support_ui_cc.pdb"
e:\b\c\b\win\src\components\password_manager\content\browser\content_password_manager_driver_factory.h(14):
fatal error C1083: Cannot open include file:
'components/autofill/content/public/interfaces/autofill_driver.mojom.h': No such
file or directory

[leonhsl(Using Gerrit), 2016-09-08 06:23:29]

On 2016/09/08 06:11:14, jam wrote:
> there are missing dependencies and it's causing build failures, i.e.
>
https://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64...
> 
> are you fixing this?
> 
> FAILED: obj/chrome/browser/test_support_ui/password_manager_test_base.obj 
> ninja -t msvc -e environment.x64 -- E:\b\c\cipd\goma/gomacc.exe
>
"E:\b\depot_tools\win_toolchain\vs_files\d5dc33b15d1b2c086f2f6632e2fd15882f80dbd3\VC\bin\amd64/cl.exe"
> /nologo /showIncludes /FC
> @obj/chrome/browser/test_support_ui/password_manager_test_base.obj.rsp /c
> ../../chrome/browser/password_manager/password_manager_test_base.cc
> /Foobj/chrome/browser/test_support_ui/password_manager_test_base.obj
> /Fd"obj/chrome/browser/test_support_ui_cc.pdb"
>
e:\b\c\b\win\src\components\password_manager\content\browser\content_password_manager_driver_factory.h(14):
> fatal error C1083: Cannot open include file:
> 'components/autofill/content/public/interfaces/autofill_driver.mojom.h': No
such
> file or directory

Sorry,, yeah https://codereview.chromium.org/2319763003/ is trying to solve this
problem. I'm confirming it now.