Added a command-line flag to enable h/w accelerated rendering of content layers.

Added a command-line flag to enable h/w accelerated rendering of content layers.

R=tony@chromium.org,darin@chromium.org
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=80822

[alokp, 2011-03-22 19:57:48]

I will commit this only after webkit side changes have landed.

[vangelis, 2011-03-22 20:10:08]

I would recommend using a different name as eventually we'll want to be drawing
other layers via the accelerated path too.  Maybe:
enable-accelerated-layer-drawing ?  -enable-accelerated-layer-content-drawing?
...

[alokp, 2011-03-22 20:46:13]

On 2011/03/22 20:10:08, vangelis wrote:
> I would recommend using a different name as eventually we'll want to be
drawing
> other layers via the accelerated path too.  Maybe:

which other layers? I thought only content and root was left - we already have
different flags for video, webgl, canvas2d, and plugins. I think jamesr wanted
to have a fine-grained control on which types of layers to accelerate. Are you
suggesting to use this flag for all types of layers except the ones that already
have a flag?

I was following the naming convention of --enable-accelerated-2d-canvas. But I
like enable-accelerated-layer-drawing.

[vangelis, 2011-03-22 20:58:54]

You're right, it's the root layer and the content layers.  I guess the term
content layer doesn't mean much to the outside world.  Let's go with
enable-accelerated-layer-drawing..

On 2011/03/22 20:46:13, Alok Priyadarshi wrote:
> On 2011/03/22 20:10:08, vangelis wrote:
> > I would recommend using a different name as eventually we'll want to be
> drawing
> > other layers via the accelerated path too.  Maybe:
> 
> which other layers? I thought only content and root was left - we already have
> different flags for video, webgl, canvas2d, and plugins. I think jamesr wanted
> to have a fine-grained control on which types of layers to accelerate. Are you
> suggesting to use this flag for all types of layers except the ones that
already
> have a flag?
> 
> I was following the naming convention of --enable-accelerated-2d-canvas. But I
> like enable-accelerated-layer-drawing.

[jamesr, 2011-03-22 21:01:39]

Apple's calling the prefs acceleratedDrawingEnabled /
canvasUsesAcceleratedDrawing, FWIW

[vangelis, 2011-03-22 21:04:18]

enable-accelerated-drawing sounds good too.  Removing the word "layer" from
the flag makes it a bit less confusing since there will be instances where
we'll be doing accelerated drawing but have no layers (other than the root).



On Tue, Mar 22, 2011 at 2:01 PM, <jamesr@chromium.org> wrote:

> Apple's calling the prefs acceleratedDrawingEnabled /
> canvasUsesAcceleratedDrawing, FWIW
>
>
> http://codereview.chromium.org/6726005/
>

[alokp, 2011-03-22 22:41:05]

On 2011/03/22 21:04:18, vangelis wrote:
> enable-accelerated-drawing sounds good too.  Removing the word "layer" from
> the flag makes it a bit less confusing since there will be instances where
> we'll be doing accelerated drawing but have no layers (other than the root).
> 
> 
> 
> On Tue, Mar 22, 2011 at 2:01 PM, <mailto:jamesr@chromium.org> wrote:
> 
> > Apple's calling the prefs acceleratedDrawingEnabled /
> > canvasUsesAcceleratedDrawing, FWIW
> >

Used acceleratedDrawingEnabled.
I wonder why we added accelerated2dCanvas when canvasUsesAcceleratedDrawing was
available.

[tony, 2011-03-22 22:53:18]

LGTM if it's OK with vangelis.

[vangelis, 2011-03-23 07:19:25]

On 2011/03/22 22:53:18, tony wrote:
> LGTM if it's OK with vangelis.

LGTM for me too.

[alokp, 2011-04-07 16:49:15]

On 2011/03/23 07:19:25, vangelis wrote:
> On 2011/03/22 22:53:18, tony wrote:
> > LGTM if it's OK with vangelis.
> 
> LGTM for me too.

Darin: OWNERS approval required.

[jam, 2011-04-07 17:46:00]

lgtm with the one change

http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
File chrome/common/chrome_switches.h (right):

http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
chrome/common/chrome_switches.h:124: extern const char
kEnableAcceleratedDrawing[];
this belongs in content/common/content_switches.cc

[alokp, 2011-04-07 18:09:58]

http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
File chrome/common/chrome_switches.h (right):

http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
chrome/common/chrome_switches.h:124: extern const char
kEnableAcceleratedDrawing[];
On 2011/04/07 17:46:00, John Abd-El-Malek wrote:
> this belongs in content/common/content_switches.cc

Done.

[darin (slow to review), 2011-04-07 18:22:03]

LGTM

http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
File content/common/content_switches.cc (right):

http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
content/common/content_switches.cc:76: const char kEnableAcceleratedDrawing[] =
"enable-accelerated-drawing";
nit: please note the indentation of the "=" sign.  it should line up in a column
with the others.

[vangelis, 2011-04-07 18:35:04]

On 2011/04/07 17:46:00, John Abd-El-Malek wrote:
> lgtm with the one change
> 
>
http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
> File chrome/common/chrome_switches.h (right):
> 
>
http://codereview.chromium.org/6726005/diff/14001/chrome/common/chrome_switch...
> chrome/common/chrome_switches.h:124: extern const char
> kEnableAcceleratedDrawing[];
> this belongs in content/common/content_switches.cc
John, should we move the rest of the gpu/compositor related flags to
content_switches as well?

[alokp, 2011-04-07 18:59:19]

http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
File content/common/content_switches.cc (right):

http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
content/common/content_switches.cc:76: const char kEnableAcceleratedDrawing[] =
"enable-accelerated-drawing";
On 2011/04/07 18:22:03, darin wrote:
> nit: please note the indentation of the "=" sign.  it should line up in a
column
> with the others.

Done. Does it really improve readability?

[darin (slow to review), 2011-04-07 19:04:59]

On Thu, Apr 7, 2011 at 11:59 AM, <alokp@chromium.org> wrote:

>
>
>
http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
> File content/common/content_switches.cc (right):
>
>
>
http://codereview.chromium.org/6726005/diff/18003/content/common/content_swit...
> content/common/content_switches.cc:76: const char
> kEnableAcceleratedDrawing[] = "enable-accelerated-drawing";
> On 2011/04/07 18:22:03, darin wrote:
>
>> nit: please note the indentation of the "=" sign.  it should line up
>>
> in a column
>
>> with the others.
>>
>
> Done. Does it really improve readability?
>
>
heh.... it is good to be consistent.
-darin



>
> http://codereview.chromium.org/6726005/
>