Adds eventlog provider dll to describe the message types of SYSLOG.

Adds eventlog provider dll to describe the message types of SYSLOG.

Creates a dll file that can be registered with the system event-logging
mechanisms on Windows to make the events generated when using the SYSLOG
macro nicer to read in the Event Viewer snapin.

BUG=642115
TEST=manual

Committed: https://crrev.com/cee58011177e2cf4bb8cfd08e64b0cd48629b0ff
Cr-Commit-Position: refs/heads/master@{#431872}

[pastarmovj, 2016-11-08 13:30:52]

pastarmovj@chromium.org changed reviewers:
+ brettw@chromium.org

[pastarmovj, 2016-11-08 13:30:52]

Hi Brett,
can you please review this CL which adds the provider dll needed for the Event
Log viewer to properly visualize log entries coming from Chrome. We will decide
how will this dll be distributed to the users (as part of the installation or
separately).
Thanks Julian

[pastarmovj, 2016-11-10 17:16:18]

On 2016/11/08 13:30:52, pastarmovj wrote:
> Hi Brett,
> can you please review this CL which adds the provider dll needed for the Event
> Log viewer to properly visualize log entries coming from Chrome. We will
decide
> how will this dll be distributed to the users (as part of the installation or
> separately).
> Thanks Julian

ping :)

[brettw, 2016-11-10 23:07:01]

You mention this makes it nicer to read. In practice, what does the even log
look like with and without this change?

[pastarmovj, 2016-11-10 23:34:10]

On 2016/11/10 23:07:01, brettw (ping on IM after 24h) wrote:
> You mention this makes it nicer to read. In practice, what does the even log
> look like with and without this change?

I sent you a mail with the screenshots showing the before and after as well as
some metrics on the proposed change.

[brettw, 2016-11-11 17:48:15]

LGTM. For shipping this, can you go through the "Simple Launch" process in
go/newChromeFeature. Then Launch Review can decide whether this is worth doing.

[brettw, 2016-11-11 17:51:17]

For my information, how does the even viewer know to load this DLL? I don't see
any mention of it. Or is that waiting on resolution for how to ship it?

[pastarmovj, 2016-11-11 23:15:34]

On 2016/11/11 17:51:17, brettw (ping on IM after 24h) wrote:
> For my information, how does the even viewer know to load this DLL? I don't
see
> any mention of it. Or is that waiting on resolution for how to ship it?

Thanks for the review, Brett!

The registration requires listing the dll in the registry
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\eventlog\Application which
can only happen with admin rights. Therefore the plan is to register this new
dll only with system wide installs. We are still discussion if this should be
part of the setup package or not but the prevailing opinion so far is that it
should be.

[brettw, 2016-11-12 05:02:12]

On 2016/11/11 23:15:34, pastarmovj wrote:
> On 2016/11/11 17:51:17, brettw (ping on IM after 24h) wrote:
> > For my information, how does the even viewer know to load this DLL? I don't
> see
> > any mention of it. Or is that waiting on resolution for how to ship it?
> 
> Thanks for the review, Brett!
> 
> The registration requires listing the dll in the registry
> HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\eventlog\Application
which
> can only happen with admin rights. Therefore the plan is to register this new
> dll only with system wide installs. We are still discussion if this should be
> part of the setup package or not but the prevailing opinion so far is that it
> should be.

Thanks!

[pastarmovj, 2016-11-14 08:31:36]

The CQ bit was checked by pastarmovj@chromium.org

[commit-bot: I haz the power, 2016-11-14 08:31:47]

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, 2016-11-14 09:17:25]

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

[commit-bot: I haz the power, 2016-11-14 09:17:26]

Try jobs failed on following builders:
  win_clang on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_clang/builds/...)

[pastarmovj, 2016-11-14 12:49:38]

The CQ bit was checked by pastarmovj@chromium.org

[pastarmovj, 2016-11-14 12:49:38]

The patchset sent to the CQ was uploaded after l-g-t-m from brettw@chromium.org
Link to the patchset: https://codereview.chromium.org/2476353003/#ps20001
(title: "Make the mc file unicode encoded.")

[commit-bot: I haz the power, 2016-11-14 12:49:51]

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, 2016-11-14 13:50:18]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-11-14 13:52:48]

Description was changed from

==========
Adds eventlog provider dll to describe the message types of SYSLOG.

Creates a dll file that can be registered with the system event-logging
mechanisms on Windows to make the events generated when using the SYSLOG
macro nicer to read in the Event Viewer snapin.

BUG=642115
TEST=manual
==========

to

==========
Adds eventlog provider dll to describe the message types of SYSLOG.

Creates a dll file that can be registered with the system event-logging
mechanisms on Windows to make the events generated when using the SYSLOG
macro nicer to read in the Event Viewer snapin.

BUG=642115
TEST=manual

Committed: https://crrev.com/cee58011177e2cf4bb8cfd08e64b0cd48629b0ff
Cr-Commit-Position: refs/heads/master@{#431872}
==========

[commit-bot: I haz the power, 2016-11-14 13:52:50]

Patchset 2 (id:??) landed as
https://crrev.com/cee58011177e2cf4bb8cfd08e64b0cd48629b0ff
Cr-Commit-Position: refs/heads/master@{#431872}

[pdr., 2016-11-14 18:04:32]

On 2016/11/14 at 13:52:50, commit-bot wrote:
> Patchset 2 (id:??) landed as
https://crrev.com/cee58011177e2cf4bb8cfd08e64b0cd48629b0ff
> Cr-Commit-Position: refs/heads/master@{#431872}

I think this change is causing windows compile errors:
https://uberchromegw.corp.google.com/i/chromium/builders/Win/builds/49108/ste...

[1/1] LINK(DLL) eventlog_provider.dll eventlog_provider.dll.lib
eventlog_provider.dll.pdb
ninja explain: output eventlog_provider.dll.lib doesn't exist
ninja explain: eventlog_provider.dll.lib is dirty
<Thread(Thread-1, started 6936)> ProcessRead: proc.stdout finished.
<Thread(Thread-1, started 6936)> ProcessRead: cleaning up.
<Thread(Thread-2, started daemon 6460)> TimedFlush: Finished
<Thread(Thread-1, started 6936)> ProcessRead: finished.
Failing build because ninja reported work to do.
This means that after completing a compile, another was run and
it resulted in still having work to do (that is, a no-op build
wasn't a no-op). Consult the first "ninja explain:" line for a
likely culprit.

[brettw, 2016-11-14 18:12:50]

This error is because Visual Studio doesn't write a .lib file if there are no
exports from the DLL (in some cases, I'm not sure why it doesn't always error).
Our toolchain declares that this link step produces a .lib file, so Ninja is
getting confused that the expected outputs were not generated.

I don't have a good idea how to fix. It's not possible to express that some
outputs only happen sometimes. In the case where I ran into it before, I think
the answer was to make sure there's always a symbol so VS produces a lib file.
Since this DLL has no code, that's probably not a good solution.

[pdr., 2016-11-14 18:14:46]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/2502643003/ by pdr@chromium.org.

The reason for reverting is: Caused windows compile errors:
https://uberchromegw.corp.google.com/i/chromium/builders/Win/builds/49108/ste...

[1/1] LINK(DLL) eventlog_provider.dll eventlog_provider.dll.lib
eventlog_provider.dll.pdb
ninja explain: output eventlog_provider.dll.lib doesn't exist
ninja explain: eventlog_provider.dll.lib is dirty.