Enable C4018 globally for the GN build, and disable per-target.

Enable C4018 globally for the GN build, and disable per-target.

This matches what the GYP build does.

BUG=398202
TEST=none

Committed: https://crrev.com/d408e1965f8e03629738039289ab7779c42a318c
Cr-Commit-Position: refs/heads/master@{#340590}

[Peter Kasting, 2015-07-22 00:43:38]

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

[Peter Kasting, 2015-07-22 00:43:38]

"gn gen" seemed to work, so I presumably didn't botch the syntax too badly? 
Have a look anyway.

[brettw, 2015-07-22 04:30:25]

the win8 bot is actually GN in case you didn't realize (confusing name).

LGTM

https://codereview.chromium.org/1246103007/diff/1/third_party/leveldatabase/B...
File third_party/leveldatabase/BUILD.gn (right):

https://codereview.chromium.org/1246103007/diff/1/third_party/leveldatabase/B...
third_party/leveldatabase/BUILD.gn:121: cflags = [ "/wd4018" ]  #
Signed/unsigned mismatch in comparison.
You'll want to put this above in the "leveldatabase" target above.. Right here
it's actually outside of any target,

[Peter Kasting, 2015-07-22 05:58:08]

https://codereview.chromium.org/1246103007/diff/1/third_party/leveldatabase/B...
File third_party/leveldatabase/BUILD.gn (right):

https://codereview.chromium.org/1246103007/diff/1/third_party/leveldatabase/B...
third_party/leveldatabase/BUILD.gn:121: cflags = [ "/wd4018" ]  #
Signed/unsigned mismatch in comparison.
On 2015/07/22 at 04:30:25, brettw wrote:
> You'll want to put this above in the "leveldatabase" target above.. Right here
it's actually outside of any target,

That would explain the bot failure...

[Peter Kasting, 2015-07-23 00:06:49]

The CQ bit was checked by pkasting@chromium.org

[Peter Kasting, 2015-07-23 00:06:50]

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/1246103007/#ps20001
(title: "Fix leveldatabase")

[commit-bot: I haz the power, 2015-07-23 00:07:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1246103007/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1246103007/20001

[commit-bot: I haz the power, 2015-07-23 01:19:09]

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

[commit-bot: I haz the power, 2015-07-23 01:19:10]

Try jobs failed on following builders:
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)

[Peter Kasting, 2015-07-27 20:19:18]

pkasting@chromium.org changed reviewers:
+ erg@chromium.org

[Peter Kasting, 2015-07-27 20:19:19]

+erg for components/filesystem/file_impl.cc -- is this warning fix correct?  I'm
not sure what INT was in the old code or why it compiled correctly.

[Elliot Glaysher, 2015-07-27 20:35:29]

On 2015/07/27 20:19:19, Peter Kasting wrote:
> +erg for components/filesystem/file_impl.cc -- is this warning fix correct? 
I'm
> not sure what INT was in the old code or why it compiled correctly.

lgtm...probably.

(Trying to reconstruct what happened; recollection may be wrong.) ssize_t was
not available for windows, but we want to have the same behaviour across all
platforms, so I used what I thought was the comparable data type on windows. I
vaguely remember that I copied INT from another piece of Windows specific code
which was doing a similar cast.

[Peter Kasting, 2015-07-27 20:41:58]

On 2015/07/27 20:35:29, Elliot Glaysher wrote:
> On 2015/07/27 20:19:19, Peter Kasting wrote:
> > +erg for components/filesystem/file_impl.cc -- is this warning fix correct? 
> I'm
> > not sure what INT was in the old code or why it compiled correctly.
> 
> lgtm...probably.
> 
> (Trying to reconstruct what happened; recollection may be wrong.) ssize_t was
> not available for windows, but we want to have the same behaviour across all
> platforms, so I used what I thought was the comparable data type on windows. I
> vaguely remember that I copied INT from another piece of Windows specific code
> which was doing a similar cast.

WinDef.h #defines INT as int, so I think I'm not changing the code's meaning.

Honestly, I think the code here should use int for all platforms. 
WriteAtCurrentPos() returns an int on all platforms, so I think your use of
ssize_t for non-Windows is incorrect.  I'm going to leave that to you to look
into further.

[Peter Kasting, 2015-07-27 20:42:00]

The CQ bit was checked by pkasting@chromium.org

[Peter Kasting, 2015-07-27 20:42:01]

The patchset sent to the CQ was uploaded after l-g-t-m from brettw@chromium.org
Link to the patchset: https://codereview.webrtc.org/1246103007/#ps60001 (title:
"Fix compile")

[commit-bot: I haz the power, 2015-07-27 20:42:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1246103007/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1246103007/60001

[commit-bot: I haz the power, 2015-07-27 22:54:57]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-07-27 22:55:55]

Patchset 4 (id:??) landed as
https://crrev.com/d408e1965f8e03629738039289ab7779c42a318c
Cr-Commit-Position: refs/heads/master@{#340590}

[ccameron, 2015-07-28 01:04:51]

A revert of this CL (patchset #4 id:60001) has been created in
https://codereview.chromium.org/1257583007/ by ccameron@chromium.org.

The reason for reverting is: Speculative revert based on windows gn bots failing
with

http://build.chromium.org/p/chromium.win/builders/Win8%20GN%20%28dbg%29/build...

FAILED: ninja -t msvc -e environment.x64 -- E:\b\build\goma/gomacc.exe
"E:\b\depot_tools\win_toolchain\vs2013_files\VC\bin\amd64/cl.exe" /nologo
/showIncludes /FC @obj/base/base/precompile.c.obj.rsp /c
../../build/precompile.cc /Foobj/base/base/precompile.c.obj
/Fdobj/base/base_c.pdb
e:\b\build\slave\win_x64_gn__dbg_\build\src\build\precompile.cc(8) : error
C2220: warning treated as error - no 'object' file generated
e:\b\build\slave\win_x64_gn__dbg_\build\src\build\precompile.cc(8) : warning
C4206: nonstandard extension used : translation unit is empty
ninja: build stopped: subcommand failed.

In theory this warning shouldn't have affected other warnings, but there are no
other clear candidates..