Always set using_map_image when using software renderer.

Always set using_map_image when using software renderer.

This saves an extra memory copy, so we'll always want this enabled when using software.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=262784

[danakj, 2014-03-13 22:02:52]

Can we remove the setting from LayerTreeSettings then, and instead change the
value of context_caps.gpu.map_image based on the command line flag?


https://code.google.com/p/chromium/codesearch#chromium/src/cc/output/gl_rende...

[jbauman, 2014-03-13 22:10:13]

Possibly. I'm not certain why we have this command-line flag to begin with.

On 2014/03/13 22:02:52, danakj wrote:
> Can we remove the setting from LayerTreeSettings then, and instead change the
> value of context_caps.gpu.map_image based on the command line flag?
> 
> 
>
https://code.google.com/p/chromium/codesearch#chromium/src/cc/output/gl_rende...

[reveman, 2014-03-13 22:17:11]

On 2014/03/13 22:10:13, jbauman wrote:
> Possibly. I'm not certain why we have this command-line flag to begin with.

Are we talking about the --enable/disable-map-image flags? These flags turn
on/off 0-copy texture updates. On some platforms this falls back to a shared
memory implementation for testing purposes, on some platforms it uses real
0-copy but is experimental.

> 
> On 2014/03/13 22:02:52, danakj wrote:
> > Can we remove the setting from LayerTreeSettings then, and instead change
the
> > value of context_caps.gpu.map_image based on the command line flag?
> > 
> > 
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/cc/output/gl_rende...

[jbauman, 2014-03-21 21:10:48]

On 2014/03/13 22:02:52, danakj wrote:
> Can we remove the setting from LayerTreeSettings then, and instead change the
> value of context_caps.gpu.map_image based on the command line flag?

It sounds like that's a non-starter.
> 
> 
>
https://code.google.com/p/chromium/codesearch#chromium/src/cc/output/gl_rende...

[jbauman, 2014-03-21 21:24:55]

On 2014/03/13 22:17:11, reveman wrote:
> On 2014/03/13 22:10:13, jbauman wrote:
> > Possibly. I'm not certain why we have this command-line flag to begin with.
> 
> Are we talking about the --enable/disable-map-image flags? These flags turn
> on/off 0-copy texture updates. On some platforms this falls back to a shared
> memory implementation for testing purposes, on some platforms it uses real
> 0-copy but is experimental.

So would it be ok to ignore the flag for software bitmaps (considering that
0-copy texture update is always the best option there)?
> 
> > 
> > On 2014/03/13 22:02:52, danakj wrote:
> > > Can we remove the setting from LayerTreeSettings then, and instead change
> the
> > > value of context_caps.gpu.map_image based on the command line flag?
> > > 
> > > 
> > >
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/cc/output/gl_rende...

[danakj, 2014-04-07 15:59:38]

On Fri, Mar 21, 2014 at 5:24 PM, <jbauman@chromium.org> wrote:

> On 2014/03/13 22:17:11, reveman wrote:
>
>> On 2014/03/13 22:10:13, jbauman wrote:
>> > Possibly. I'm not certain why we have this command-line flag to begin
>> with.
>>
>
>  Are we talking about the --enable/disable-map-image flags? These flags
>> turn
>> on/off 0-copy texture updates. On some platforms this falls back to a
>> shared
>> memory implementation for testing purposes, on some platforms it uses real
>> 0-copy but is experimental.
>>
>
> So would it be ok to ignore the flag for software bitmaps (considering that
> 0-copy texture update is always the best option there)?


Did this get answered? +reveman


>
>
>  >
>> > On 2014/03/13 22:02:52, danakj wrote:
>> > > Can we remove the setting from LayerTreeSettings then, and instead
>> change
>> the
>> > > value of context_caps.gpu.map_image based on the command line flag?
>> > >
>> > >
>> > >
>> >
>>
>
> https://code.google.com/p/chromium/codesearch#chromium/
> src/cc/output/gl_renderer.cc&l=239
>
>
>
> https://codereview.chromium.org/195733010/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[reveman, 2014-04-07 17:05:40]

On 2014/04/07 15:59:38, danakj wrote:
> On Fri, Mar 21, 2014 at 5:24 PM, <mailto:jbauman@chromium.org> wrote:
> 
> > On 2014/03/13 22:17:11, reveman wrote:
> >
> >> On 2014/03/13 22:10:13, jbauman wrote:
> >> > Possibly. I'm not certain why we have this command-line flag to begin
> >> with.
> >>
> >
> >  Are we talking about the --enable/disable-map-image flags? These flags
> >> turn
> >> on/off 0-copy texture updates. On some platforms this falls back to a
> >> shared
> >> memory implementation for testing purposes, on some platforms it uses real
> >> 0-copy but is experimental.
> >>
> >
> > So would it be ok to ignore the flag for software bitmaps (considering that
> > 0-copy texture update is always the best option there)?
> 
> 
> Did this get answered? +reveman

It's ok to ignore this setting for software bitmaps and always use 0-copy path
in that case. You could even remove the support in ResourceProvider for the
alternative code path too.


> 
> 
> >
> >
> >  >
> >> > On 2014/03/13 22:02:52, danakj wrote:
> >> > > Can we remove the setting from LayerTreeSettings then, and instead
> >> change
> >> the
> >> > > value of context_caps.gpu.map_image based on the command line flag?
> >> > >
> >> > >
> >> > >
> >> >
> >>
> >
> > https://code.google.com/p/chromium/codesearch#chromium/
> > src/cc/output/gl_renderer.cc&l=239
> >
> >
> >
> > https://codereview.chromium.org/195733010/
> >
> 
> To unsubscribe from this group and stop receiving emails from it, send an
email
> to mailto:chromium-reviews+unsubscribe@chromium.org.

[jbauman, 2014-04-08 23:52:50]

Ok, then could I get an lgtm on this patch?

[danakj, 2014-04-09 17:01:53]

LGTM

[jbauman, 2014-04-09 18:01:00]

The CQ bit was checked by jbauman@chromium.org

[commit-bot: I haz the power, 2014-04-09 18:01:38]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jbauman@chromium.org/195733010/1

[commit-bot: I haz the power, 2014-04-09 20:02:42]

Change committed as 262784