Initialize our quads to zero before we gl_->BufferData() with them.

Initialize our quads to zero before we gl_->BufferData() with them.

The data was never actually used, but msan was complaining.

Fixed build

BUG=

[awoloszyn, 2015-02-27 21:26:21]

awoloszyn@chromium.org changed reviewers:
+ enne@chromium.org

[awoloszyn, 2015-02-27 21:35:43]

awoloszyn@chromium.org changed reviewers:
+ danakj@chromium.org

[enne (OOO), 2015-02-27 21:36:01]

enne@chromium.org changed reviewers:
- danakj@chromium.org

[enne (OOO), 2015-02-27 21:36:01]

https://codereview.chromium.org/959403002/diff/1/cc/output/dynamic_geometry_b...
File cc/output/dynamic_geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/1/cc/output/dynamic_geometry_b...
cc/output/dynamic_geometry_binding.cc:15: GeometryBindingQuad quads[1];
Can you initialize this in the class itself rather than here? If somebody else
ever instantiates it, they'll probably forget as well.

[awoloszyn, 2015-02-27 22:08:37]

https://codereview.chromium.org/959403002/diff/1/cc/output/dynamic_geometry_b...
File cc/output/dynamic_geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/1/cc/output/dynamic_geometry_b...
cc/output/dynamic_geometry_binding.cc:15: GeometryBindingQuad quads[1];
On 2015/02/27 21:36:01, enne wrote:
> Can you initialize this in the class itself rather than here? If somebody else
> ever instantiates it, they'll probably forget as well.

Done.

[danakj, 2015-02-27 22:10:53]

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

[danakj, 2015-02-27 22:10:54]

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
File cc/output/geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
cc/output/geometry_binding.cc:46: memset(&v0, 0x0,
sizeof(GeometryBindingVertex));
(0x0 is a very verbose way of just saying 0)

[enne (OOO), 2015-02-27 22:12:07]

enne@chromium.org changed reviewers:
- danakj@chromium.org

[enne (OOO), 2015-02-27 22:12:07]

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
File cc/output/geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
cc/output/geometry_binding.cc:46: memset(&v0, 0x0,
sizeof(GeometryBindingVertex));
Please no memset.  Just initialize these directly.

https://codereview.chromium.org/959403002/diff/20001/cc/output/static_geometr...
File cc/output/static_geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/20001/cc/output/static_geometr...
cc/output/static_geometry_binding.cc:43: GeometryBindingQuadIndex y =
{static_cast<uint16>(0 + 4 * i),
Use direct or copy initialization, not list initialization.

[awoloszyn, 2015-02-27 22:13:54]

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
File cc/output/geometry_binding.cc (right):

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
cc/output/geometry_binding.cc:46: memset(&v0, 0x0,
sizeof(GeometryBindingVertex));
On 2015/02/27 22:10:54, danakj wrote:
> (0x0 is a very verbose way of just saying 0)

Its an old habbit will fix. On that note, the CL this was being created for is
getting reverted anyway, so I will move this over to there.

https://codereview.chromium.org/959403002/diff/20001/cc/output/geometry_bindi...
cc/output/geometry_binding.cc:46: memset(&v0, 0x0,
sizeof(GeometryBindingVertex));
On 2015/02/27 22:12:07, enne wrote:
> Please no memset.  Just initialize these directly.

Will do. My other patch is getting reverted for other reasons (strangely added a
static initializer, want to figure out why). 
I will move these changes into that patch and update when it is done.

[enne (OOO), 2015-03-09 17:52:04]

I'll close this patch then, if these changes are getting rolled into a different
patch.