add scalar stubs for neon functions to give arm guys a place to work

add scalar stubs for neon functions to give arm guys a place to work

BUG=

Committed: http://code.google.com/p/skia/source/detail?r=10431

[humper, 2013-07-26 19:41:37]

This passed android trybot tests; shouldn't affect android behavior at all.

[reed1, 2013-07-29 13:31:49]

How will this get plumbed in, and is it a clean/scalable technique?

I see a (growing) list of extern functions declared in SkBitmapProcState.cpp,
and that doesn't feel like the right long-term technique. Can we create
something more hidden, perhaps pass more info to our "PlatformProc" kind of
hook, so that the decision tree and available procs is more hidden from the
caller?

[humper, 2013-07-29 22:41:06]

On 2013/07/29 13:31:49, reed1 wrote:
> How will this get plumbed in, and is it a clean/scalable technique?
> 
> I see a (growing) list of extern functions declared in SkBitmapProcState.cpp,
> and that doesn't feel like the right long-term technique. Can we create
> something more hidden, perhaps pass more info to our "PlatformProc" kind of
> hook, so that the decision tree and available procs is more hidden from the
> caller?

I see what you're talking about now.

This particular change didn't add anything to that list, and is already plumbed.

There's definitely some ugliness going on with that list of externs, but I think
it's orthogonal to this CL.

[reed1, 2013-07-30 12:24:15]

On 2013/07/29 22:41:06, humper wrote:
> On 2013/07/29 13:31:49, reed1 wrote:
> > How will this get plumbed in, and is it a clean/scalable technique?
> > 
> > I see a (growing) list of extern functions declared in
SkBitmapProcState.cpp,
> > and that doesn't feel like the right long-term technique. Can we create
> > something more hidden, perhaps pass more info to our "PlatformProc" kind of
> > hook, so that the decision tree and available procs is more hidden from the
> > caller?
> 
> I see what you're talking about now.
> 
> This particular change didn't add anything to that list, and is already
plumbed.
> 
> There's definitely some ugliness going on with that list of externs, but I
think
> it's orthogonal to this CL.

Agreed (ortho), but lets make it a priority to switch to a more
structure/scalable architecture.

[reed1, 2013-07-30 12:24:24]

lgtm

[commit-bot: I haz the power, 2013-07-30 13:04:22]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/humper@google.com/20749003/1

[commit-bot: I haz the power, 2013-07-30 13:16:20]

Change committed as 10431