Always define the `gcc_version' variable to some value.

Always define the `gcc_version' variable to some value.

gcc_version previously existed only under certain conditions; currently,
os_posix==1 and OS!="mac" and OS!="ios".

These same checks had to be repeated everywhere gcc_version was used, and
they often got out of sync (many places still do not check for OS!="ios",
for example).

We now always gcc_version. Its default value is 0, and it is then changed to
something meaningful under that condition mentioned above. This allows us to
remove the duplicated checks currently spread across the .gyp files in the
tree.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=196910

[Raphael Kubo da Costa (rakuco), 2013-04-26 12:06:47]

Most of the checks are actually inside Blink; since I can't change both
repositories at once, I need to get the chromium/ changes first for the bots to
test everything correctly.

[Raphael Kubo da Costa (rakuco), 2013-04-26 12:13:39]

On 2013/04/26 12:06:47, Raphael Kubo da Costa (rakuco) wrote:
> Most of the checks are actually inside Blink; since I can't change both
> repositories at once, I need to get the chromium/ changes first for the bots
to
> test everything correctly.

BTW, it'd be good if someone with enough karma could trigger some trybots for us
to see if I haven't done anything wrong.

[Paweł Hajdan Jr., 2013-04-26 17:02:07]

Raphael, thank you for the patch. Nice to see a general Linux theme in your
patches I've seen so far. :)

https://codereview.chromium.org/13912027/diff/8001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/13912027/diff/8001/build/common.gypi#newcode901
build/common.gypi:901: 'gcc_version%': 0,
We already have a block like:

          ['clang==0 and asan==0 and tsan==0 and msan==0', {
            # This will set gcc_version to XY if you are running gcc X.Y.*.
            # This is used to tweak build flags for gcc 4.5.
            'gcc_version%': '<!(python <(DEPTH)/build/compiler_version.py)',
          }, {
            'gcc_version%': 0,
          }],

in this file. Please extract it and update the condition. This will make
gcc_version defined in only one place, making it easier to reason about.

https://codereview.chromium.org/13912027/diff/8001/build/common.gypi#newcode4187
build/common.gypi:4187: # Don't warn about the "typedef 'foo' locally defined
but not used"
Why is this code being moved?

[Raphael Kubo da Costa (rakuco), 2013-04-26 17:48:52]

On 2013/04/26 17:02:07, Paweł Hajdan Jr. wrote:
> Raphael, thank you for the patch. Nice to see a general Linux theme in your
> patches I've seen so far. :)

Being a FreeBSD packager and CMake fanboy myself, I can't help being attracted
to this kind of thing :-)

> in this file. Please extract it and update the condition. This will make
> gcc_version defined in only one place, making it easier to reason about.

I'll have to send a patch v3 anyway, for some reason the original value I set at
the top isn't being picked up so I guess there's something I haven't understood
about gyp yet.

>
https://codereview.chromium.org/13912027/diff/8001/build/common.gypi#newcode4187
> build/common.gypi:4187: # Don't warn about the "typedef 'foo' locally defined
> but not used"
> Why is this code being moved?

Because it was inside a 'os_posix==1 and OS!="mac" and OS!="ios"' block before,
and it can be moved out of there now.

[Raphael Kubo da Costa (rakuco), 2013-04-26 18:23:26]

Patch v3 is up. I ended up setting 'gcc_version' without the '%' at the end. For
one, it doesn't make much sense to override this on the command-line; and we
were also having some problems with its value because of some subprojects (such
as v8) appear to have borrowed older versions of common.gypi in the past and set
it to, say, "unknown" or some other value (they don't seem to use the value
themselves, but override it and cause problems in the top-level
build/common.gypi).

[Paweł Hajdan Jr., 2013-04-26 19:06:19]

LGTM

[commit-bot: I haz the power, 2013-04-26 19:14:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/139120...

[commit-bot: I haz the power, 2013-04-26 19:41:01]

Sorry for I got bad news for ya.
Compile failed with a clobber build on android_dbg.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_db...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-04-26 20:32:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/139120...

[commit-bot: I haz the power, 2013-04-26 20:47:00]

Sorry for I got bad news for ya.
Compile failed with a clobber build on android_dbg.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_db...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-04-26 21:17:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/raphael.kubo.da.costa@intel.com/139120...

[commit-bot: I haz the power, 2013-04-27 01:53:22]

Change committed as 196910

[Nico, 2013-04-30 00:50:08]

This seems to cause issues on the bots:
http://build.chromium.org/p/tryserver.chromium/builders/linux_rel/builds/1220...

[Raphael Kubo da Costa (rakuco), 2013-04-30 08:19:52]

On 2013/04/30 00:50:08, Nico wrote:
> This seems to cause issues on the bots:
>
http://build.chromium.org/p/tryserver.chromium/builders/linux_rel/builds/1220...

This is weird, especially since the subsequent linux_rel builds seem to have
worked fine.

And since this is a Linux bot, compiler_version.py would have been called
without this change as well (since os_posix==1, OS!="mac", OS!="ios" and
clang!=1).

Do you have any idea of what could be causing this?

[Nico, 2013-04-30 18:01:29]

On Tue, Apr 30, 2013 at 1:19 AM, <raphael.kubo.da.costa@intel.com> wrote:

> On 2013/04/30 00:50:08, Nico wrote:
>
>> This seems to cause issues on the bots:
>>
>
> http://build.chromium.org/p/**tryserver.chromium/builders/**
>
linux_rel/builds/122024/steps/**runhooks/logs/stdio<http://build.chromium.org/p/tryserver.chromium/builders/linux_rel/builds/122024/steps/runhooks/logs/stdio>
>
> This is weird, especially since the subsequent linux_rel builds seem to
> have
> worked fine.
>
> And since this is a Linux bot, compiler_version.py would have been called
> without this change as well (since os_posix==1, OS!="mac", OS!="ios" and
> clang!=1).
>
> Do you have any idea of what could be causing this?
>

Apparently there was some unrelated infrastructure issue that could've
caused this. Until it happens again, I think it's all good.


>
>
https://codereview.chromium.**org/13912027/<https://codereview.chromium.org/1...
>