Fix race between DPI and window size

Fix race between DPI and window size

Because the display scale factor and the window size are sent in two different
IPCs, and because the window size (ViewMsg_Resize) may be delayed (to avoid
spamming the renderer with resize messages while dragging), the renderer can
think it's using a different resolution than it is.

This merges the ViewMsg_ScreenInfoChanged and ViewMsg_Resize messages,
so that the renderer cannot not have a consistent view of the window size and
DPI.

BUG=239624
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=202814

[ccameron, 2013-05-25 00:25:24]

This is the final part to fix the DPI-change-related flashes on
Mac (and presumably any hi-DPI platforms).

There is still a flash, but it looks to be separate (it flashes garbage
data ... as though we did a swapbuffers without doing a new draw).

[piman, 2013-05-25 00:37:25]

Thanks for this, it's definitely a win to have a single message, and no stale
transient state.
I just have one thing, and after that it's good to go.

https://codereview.chromium.org/16048003/diff/1/content/browser/renderer_host...
File content/browser/renderer_host/render_widget_host_impl.cc (right):

https://codereview.chromium.org/16048003/diff/1/content/browser/renderer_host...
content/browser/renderer_host/render_widget_host_impl.cc:548:
GetWebScreenInfo(&params.screen_info);
This could be a bit expensive, like round-tripping to X on linux, but done on
every resize. Could we still query it in NotifyScreenInfoChanged which should
only happen rarely, and cache it as a field?

[ccameron, 2013-05-25 01:53:36]

Thanks!!

https://codereview.chromium.org/16048003/diff/1/content/browser/renderer_host...
File content/browser/renderer_host/render_widget_host_impl.cc (right):

https://codereview.chromium.org/16048003/diff/1/content/browser/renderer_host...
content/browser/renderer_host/render_widget_host_impl.cc:548:
GetWebScreenInfo(&params.screen_info);
On 2013/05/25 00:37:25, piman wrote:
> This could be a bit expensive, like round-tripping to X on linux, but done on
> every resize. Could we still query it in NotifyScreenInfoChanged which should
> only happen rarely, and cache it as a field?

Yes, I was on the fence about this. I didn't initially put it in because of the
include order issue. I've updated this to have a scoped_ptr<> which is reset
when it becomes invalid.

[piman, 2013-05-25 01:59:16]

OK, that works. LGTM!

[commit-bot: I haz the power, 2013-05-27 08:14:20]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/7001

[commit-bot: I haz the power, 2013-05-27 08:20:59]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[ccameron, 2013-05-27 08:49:24]

Adding cevans for security IPC review.

[ccameron, 2013-05-28 20:50:07]

Ping cevans/cdn for security IPC review.

[Cris Neckar, 2013-05-28 21:02:30]

LGTM for security review

[ccameron, 2013-05-28 21:06:21]

Thanks!!

[commit-bot: I haz the power, 2013-05-28 21:06:37]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/7001

[commit-bot: I haz the power, 2013-05-28 21:29:31]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[commit-bot: I haz the power, 2013-05-28 21:31:33]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/7001

[commit-bot: I haz the power, 2013-05-28 22:10:44]

Sorry for I got bad news for ya.
Compile failed with a clobber build on android_clang_dbg.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_cl...
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.

[ccameron, 2013-05-28 23:47:24]

This broke some tests because screen_info_ assumed that the renderer didn't have
correct initial screen info -- this isn't the case.

I've added a bit to explicitly track if NotifyScreenInfoChanged has been called,
and only force a resize message if this has happened.

Throwing back in the CQ -- lmk if you want to look again.

[commit-bot: I haz the power, 2013-05-28 23:49:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/37002

[commit-bot: I haz the power, 2013-05-29 00:11:29]

Sorry for I got bad news for ya.
Compile failed with a clobber build on linux_chromeos.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...
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, 2013-05-29 00:58:58]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/57001

[commit-bot: I haz the power, 2013-05-29 01:30:28]

Sorry for I got bad news for ya.
Compile failed with a clobber build on linux_chromeos_clang.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...
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, 2013-05-29 01:37:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ccameron@chromium.org/16048003/3002

[commit-bot: I haz the power, 2013-05-29 08:05:57]

Change committed as 202814