Start enforcing strict checks for {virtual,override,final} on Linux

Start enforcing strict checks for {virtual,override,final} on Linux

BUG=417463
Committed: https://crrev.com/e8d2cde55fe3fcc12f557e413c7afaef574cb4b4
Cr-Commit-Position: refs/heads/master@{#309799}

[dcheng, 2014-12-30 23:27:05]

dcheng@chromium.org changed reviewers:
+ cjhopman@chromium.org

[dcheng, 2014-12-30 23:27:05]

This works on my local machine. Hopefully it works on the bots...

[Nico, 2014-12-30 23:48:56]

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

[Nico, 2014-12-30 23:48:56]

lgtm

[Nico, 2014-12-30 23:49:08]

(do you want to do this in gn too?)

[dcheng, 2014-12-30 23:50:49]

On 2014/12/30 at 23:49:08, thakis wrote:
> (do you want to do this in gn too?)

Eventually. I've only been testing in gyp, so I'm sure gn will turn out some new
fun. I'd like to get the flag enabled at all somewhere for starters =)

[cjhopman, 2014-12-30 23:55:30]

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi#newcod...
build/common.gypi:2146: 'clang_chrome_plugins_flags': [
gyp list merging is done by appending and I believe conditions in variables are
applied after other variables are evaluated, and so you may be able to drop the
nested variables block and the else condition here (and remove
<@(clang_chrome_plugins_flags) from this condition).

[cjhopman, 2014-12-30 23:55:53]

also, lgtm either way

[dcheng, 2014-12-30 23:57:03]

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi#newcod...
build/common.gypi:2146: 'clang_chrome_plugins_flags': [
On 2014/12/30 23:55:30, cjhopman wrote:
> gyp list merging is done by appending and I believe conditions in variables
are
> applied after other variables are evaluated, and so you may be able to drop
the
> nested variables block and the else condition here (and remove
> <@(clang_chrome_plugins_flags) from this condition).

thakis@ mentioned this as well.

I took an old patch that removed the last platform-dependent plugin flags and
applied it in reverse, because gyp is scary and I wanted something that is
99.99% going to work.

As it turns out, the Linux GPU bots don't build the same bits I've been testing
with, so there's a bunch of new failures to fix there. So I'll look at
simplifying the gyp a bit.

[dcheng, 2014-12-31 01:33:54]

Looks much simpler now. For GYP, anyway.

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/831863003/diff/20001/build/common.gypi#newcod...
build/common.gypi:2146: 'clang_chrome_plugins_flags': [
On 2014/12/30 at 23:55:30, cjhopman wrote:
> gyp list merging is done by appending and I believe conditions in variables
are applied after other variables are evaluated, and so you may be able to drop
the nested variables block and the else condition here (and remove
<@(clang_chrome_plugins_flags) from this condition).

Done.

[Nico, 2014-12-31 01:35:56]

lgtm++

https://codereview.chromium.org/831863003/diff/40001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/831863003/diff/40001/build/common.gypi#newcod...
build/common.gypi:2143: ['OS=="linux" and chromeos==0', {
(if this ends up breaking the official bots, you could land this with an `and
buildtype!="Official"` clause added)

[dcheng, 2014-12-31 01:39:27]

On 2014/12/31 at 01:35:56, thakis wrote:
> lgtm++
> 
> https://codereview.chromium.org/831863003/diff/40001/build/common.gypi
> File build/common.gypi (right):
> 
>
https://codereview.chromium.org/831863003/diff/40001/build/common.gypi#newcod...
> build/common.gypi:2143: ['OS=="linux" and chromeos==0', {
> (if this ends up breaking the official bots, you could land this with an `and
buildtype!="Official"` clause added)

I'll try landing it without for now; hopefully it won't be too many failures (if
any).

[dcheng, 2014-12-31 01:39:39]

(Also, I think I did an official build as well locally and it was fine)

[dcheng, 2014-12-31 01:48:15]

The CQ bit was checked by dcheng@chromium.org

[commit-bot: I haz the power, 2014-12-31 01:48:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/831863003/40001

[commit-bot: I haz the power, 2014-12-31 03:00:33]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2014-12-31 03:01:26]

Patchset 3 (id:??) landed as
https://crrev.com/e8d2cde55fe3fcc12f557e413c7afaef574cb4b4
Cr-Commit-Position: refs/heads/master@{#309799}