Add missing dependency in base_win64.

Add missing dependency in base_win64.

BUG=477851
Committed: https://crrev.com/cff5a0c6ca6dcaf284b43cd6a8984e6bdffb6aa4
Cr-Commit-Position: refs/heads/master@{#325708}

[Georges Khalil, 2015-04-17 14:20:39]

Patchset #1 (id:1) has been deleted

[Georges Khalil, 2015-04-17 14:22:35]

georgesak@chromium.org changed reviewers:
+ thestig@chromium.org

[Georges Khalil, 2015-04-17 14:22:35]

PTAL.

[Lei Zhang, 2015-04-17 17:06:36]

lgtm

[Georges Khalil, 2015-04-17 17:07:06]

The CQ bit was checked by georgesak@chromium.org

[commit-bot: I haz the power, 2015-04-17 17:07:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1090223002/20001

[wtc, 2015-04-17 18:11:56]

wtc@chromium.org changed reviewers:
+ wtc@chromium.org

[wtc, 2015-04-17 18:11:57]

Patch set 1 LGTM. Thanks for writing the CL.

I think a better fix would be something like the following, which adds the
dependency near the file that requires this dependency:

diff --git a/base/base.gyp b/base/base.gyp
index 1d6ba34..e99b0d6 100644
--- a/base/base.gyp
+++ b/base/base.gyp
@@ -190,9 +190,6 @@
               ],
             },
           ],
-          'dependencies': [
-           'trace_event/etw_manifest/etw_manifest.gyp:etw_manifest',
-          ],
         }],
         ['OS == "mac" or (OS == "ios" and _toolset == "host")', {
           'link_settings': {
diff --git a/base/base.gypi b/base/base.gypi
index 23f1c8a..06d94f9 100644
--- a/base/base.gypi
+++ b/base/base.gypi
@@ -974,6 +974,11 @@
           ['OS=="mac" or OS=="ios" or <(chromeos)==1 or <(chromecast)==1', {
             'defines': ['SYSTEM_NATIVE_UTF8'],
           }],
+          ['OS == "win"', {
+            'dependencies': [
+              'trace_event/etw_manifest/etw_manifest.gyp:etw_manifest',
+            ],
+          }],
         ],
       }],
       ['base_i18n_target==1', {

The file that needs this dependency is
base/trace_event/trace_event_etw_export_win.cc, which is in
'trace_event_sources'.

'<@(trace_event_sources)' is added to 'sources' if 'base_target==1'. Therefore,
it would be better add the dependency at that place.

[Georges Khalil, 2015-04-17 20:28:53]

That was my first reflex. However, I don't think you can have dependencies
in base.gypi (a quick search will yield 0 dependencies declared in this file).

If there's a workaround, I would be more than happy to change it.

On 2015/04/17 18:11:57, wtc wrote:
> Patch set 1 LGTM. Thanks for writing the CL.
> 
> I think a better fix would be something like the following, which adds the
> dependency near the file that requires this dependency:
> 
> diff --git a/base/base.gyp b/base/base.gyp
> index 1d6ba34..e99b0d6 100644
> --- a/base/base.gyp
> +++ b/base/base.gyp
> @@ -190,9 +190,6 @@
>                ],
>              },
>            ],
> -          'dependencies': [
> -           'trace_event/etw_manifest/etw_manifest.gyp:etw_manifest',
> -          ],
>          }],
>          ['OS == "mac" or (OS == "ios" and _toolset == "host")', {
>            'link_settings': {
> diff --git a/base/base.gypi b/base/base.gypi
> index 23f1c8a..06d94f9 100644
> --- a/base/base.gypi
> +++ b/base/base.gypi
> @@ -974,6 +974,11 @@
>            ['OS=="mac" or OS=="ios" or <(chromeos)==1 or <(chromecast)==1', {
>              'defines': ['SYSTEM_NATIVE_UTF8'],
>            }],
> +          ['OS == "win"', {
> +            'dependencies': [
> +              'trace_event/etw_manifest/etw_manifest.gyp:etw_manifest',
> +            ],
> +          }],
>          ],
>        }],
>        ['base_i18n_target==1', {
> 
> The file that needs this dependency is
> base/trace_event/trace_event_etw_export_win.cc, which is in
> 'trace_event_sources'.
> 
> '<@(trace_event_sources)' is added to 'sources' if 'base_target==1'.
Therefore,
> it would be better add the dependency at that place.

[commit-bot: I haz the power, 2015-04-17 20:53:05]

Committed patchset #1 (id:20001)

[commit-bot: I haz the power, 2015-04-17 20:54:09]

Patchset 1 (id:??) landed as
https://crrev.com/cff5a0c6ca6dcaf284b43cd6a8984e6bdffb6aa4
Cr-Commit-Position: refs/heads/master@{#325708}

[wtc, 2015-04-20 16:22:48]

Patch set 1 LGTM. Thanks.

On 2015/04/17 20:28:53, Georges Khalil wrote:
> That was my first reflex. However, I don't think you can have dependencies
> in base.gypi (a quick search will yield 0 dependencies declared in this file).
> 
> If there's a workaround, I would be more than happy to change it.

Hi Georges,

I asked Mark Mentovai about this. He said what I suggested won't work. So
we need to go with your fix.

[Georges Khalil, 2015-04-21 09:58:44]

On 2015/04/20 16:22:48, wtc wrote:
> Patch set 1 LGTM. Thanks.
> 
> On 2015/04/17 20:28:53, Georges Khalil wrote:
> > That was my first reflex. However, I don't think you can have dependencies
> > in base.gypi (a quick search will yield 0 dependencies declared in this
file).
> > 
> > If there's a workaround, I would be more than happy to change it.
> 
> Hi Georges,
> 
> I asked Mark Mentovai about this. He said what I suggested won't work. So
> we need to go with your fix.

Sounds good. Let me know if you need something else from me.

[Jeffrey Yasskin, 2015-04-25 03:28:15]

On 2015/04/21 09:58:44, Georges Khalil wrote:
> On 2015/04/20 16:22:48, wtc wrote:
> > Patch set 1 LGTM. Thanks.
> > 
> > On 2015/04/17 20:28:53, Georges Khalil wrote:
> > > That was my first reflex. However, I don't think you can have dependencies
> > > in base.gypi (a quick search will yield 0 dependencies declared in this
> file).
> > > 
> > > If there's a workaround, I would be more than happy to change it.
> > 
> > Hi Georges,
> > 
> > I asked Mark Mentovai about this. He said what I suggested won't work. So
> > we need to go with your fix.
> 
> Sounds good. Let me know if you need something else from me.

The build seems to still be flaky due to this set of patches:
http://build.chromium.org/p/chromium/builders/Win/builds/31260/steps/compile/...
http://build.chromium.org/p/chromium/builders/Win/builds/31272/steps/compile/...