Added directwrite warmup for PPAPI processes.

Added directwrite warmup for PPAPI processes.
This patch warms up the directwrite skia font code when running a PPAPI
process under win32k lockdown. This allows users of the Skia (such as the
PPAPI itself) to bypass GDI calls when dealing with fonts.

BUG=523278
Committed: https://crrev.com/30ce0db09000893a6a44e9ade4912bbd3d45eccc
Cr-Commit-Position: refs/heads/master@{#350191}

[forshaw, 2015-09-03 00:03:58]

forshaw@chromium.org changed reviewers:
+ palmer@chromium.org, wfh@chromium.org

[forshaw, 2015-09-03 00:03:59]

Please take a look.

[palmer, 2015-09-15 21:43:43]

https://codereview.chromium.org/1325843002/diff/20001/content/browser/ppapi_p...
File content/browser/ppapi_plugin_process_host.cc (right):

https://codereview.chromium.org/1325843002/diff/20001/content/browser/ppapi_p...
content/browser/ppapi_plugin_process_host.cc:13: #include
"base/strings/string_number_conversions.h"
You don't use this.

https://codereview.chromium.org/1325843002/diff/20001/content/ppapi_plugin/pp...
File content/ppapi_plugin/ppapi_thread.cc (right):

https://codereview.chromium.org/1325843002/diff/20001/content/ppapi_plugin/pp...
content/ppapi_plugin/ppapi_thread.cc:103: // Get win32k lockdown policy using a
lazy instance so it's only initialized
This is copied and pasted from another CL, right? Rather than duplicating it,
put it in a shared location.

[forshaw, 2015-09-15 21:59:32]

https://codereview.chromium.org/1325843002/diff/20001/content/ppapi_plugin/pp...
File content/ppapi_plugin/ppapi_thread.cc (right):

https://codereview.chromium.org/1325843002/diff/20001/content/ppapi_plugin/pp...
content/ppapi_plugin/ppapi_thread.cc:103: // Get win32k lockdown policy using a
lazy instance so it's only initialized
On 2015/09/15 21:43:43, palmer wrote:
> This is copied and pasted from another CL, right? Rather than duplicating it,
> put it in a shared location.

I'd love to. Initially I had it in base but wfh@ recommended that would be a
hassle getting it into that from a maintainers perspective. So instead I moved
it to sandbox. But it turns out the skia code can't access sandbox code and I
felt it was probably unwise to change dependencies. So currently at a loss as
where best to put it. I could probably access the one in skia/ext from content,
but that seems a bit sketchy. If you've got a better idea for a place to dump it
I'm all for it.

[palmer, 2015-09-15 22:01:12]

> I'd love to. Initially I had it in base but wfh@ recommended that would be a
> hassle getting it into that from a maintainers perspective. So instead I moved
> it to sandbox. But it turns out the skia code can't access sandbox code and I
> felt it was probably unwise to change dependencies. So currently at a loss as
> where best to put it. I could probably access the one in skia/ext from
content,
> but that seems a bit sketchy. If you've got a better idea for a place to dump
it
> I'm all for it.

Hmm, base seems the natural place for it. wfh: What's the hassle?

[forshaw, 2015-09-16 12:43:25]

Patchset #3 (id:40001) has been deleted

[forshaw, 2015-09-16 12:46:57]

https://codereview.chromium.org/1325843002/diff/20001/content/browser/ppapi_p...
File content/browser/ppapi_plugin_process_host.cc (right):

https://codereview.chromium.org/1325843002/diff/20001/content/browser/ppapi_p...
content/browser/ppapi_plugin_process_host.cc:13: #include
"base/strings/string_number_conversions.h"
On 2015/09/15 21:43:43, palmer wrote:
> You don't use this.

Used for DoubleToString below.

[palmer, 2015-09-17 20:34:17]

> Used for DoubleToString below.

Ohh, OK. LGTM.

[jam, 2015-09-17 21:48:48]

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

[jam, 2015-09-17 21:48:48]

https://codereview.chromium.org/1325843002/diff/60001/content/ppapi_plugin/pp...
File content/ppapi_plugin/ppapi_plugin_main.cc (right):

https://codereview.chromium.org/1325843002/diff/60001/content/ppapi_plugin/pp...
content/ppapi_plugin/ppapi_plugin_main.cc:71: void WarmupDirectWrite() {
we shouldn't duplicate that code. it should be shared in content/child.

[forshaw, 2015-09-18 13:07:11]

PTAL. I've moved the shared code into a single location which any content
process can access.

[forshaw, 2015-09-18 13:45:43]

Patchset #3 (id:60001) has been deleted

[jam, 2015-09-21 21:26:26]

https://codereview.chromium.org/1325843002/diff/80001/content/common/font_war...
File content/common/font_warmup_win.cc (right):

https://codereview.chromium.org/1325843002/diff/80001/content/common/font_war...
content/common/font_warmup_win.cc:216: warmup_initialized = true;
I don't see why these first four lines are needed? we know that it's only called
once.

https://codereview.chromium.org/1325843002/diff/80001/content/ppapi_plugin/pp...
File content/ppapi_plugin/ppapi_plugin_main.cc (right):

https://codereview.chromium.org/1325843002/diff/80001/content/ppapi_plugin/pp...
content/ppapi_plugin/ppapi_plugin_main.cc:66: 
undo

[forshaw, 2015-09-21 21:34:01]

Patchset #4 (id:100001) has been deleted

[jam, 2015-09-21 21:40:46]

lgtm

[forshaw, 2015-09-21 22:03:29]

On 2015/09/21 21:40:46, jam wrote:
> lgtm

Thanks very much jam@ for the review.

[Will Harris, 2015-09-22 16:47:29]

content/common/sandbox_win.cc lgtm

[forshaw, 2015-09-22 17:13:14]

The CQ bit was checked by forshaw@chromium.org

[forshaw, 2015-09-22 17:13:15]

The patchset sent to the CQ was uploaded after l-g-t-m from palmer@chromium.org
Link to the patchset: https://codereview.chromium.org/1325843002/#ps120001
(title: "Removed init check.")

[commit-bot: I haz the power, 2015-09-22 17:13:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1325843002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1325843002/120001

[commit-bot: I haz the power, 2015-09-22 18:09:06]

Committed patchset #4 (id:120001)

[commit-bot: I haz the power, 2015-09-22 18:44:10]

Patchset 4 (id:??) landed as
https://crrev.com/30ce0db09000893a6a44e9ade4912bbd3d45eccc
Cr-Commit-Position: refs/heads/master@{#350191}