cc: Do the math for a tile's content rect in layer space once.

cc: Do the math for a tile's content rect in layer space once.

We currently do this math in CoversRect(), GatherPixelRefs() and each
time a tile is rastered to compute damage. We also do this multiple
times in the process of creating a tile.

We can do this once at the time we create the tile and pass it around
and store it on the Tile instead. Less math!

R=vmpstr
BUG=342848
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/d11f79b35cf3958bec49a8271c8a9059a983828f
Cr-Commit-Position: refs/heads/master@{#346491}

[vmpstr, 2015-08-28 21:04:17]

lgtm with something to consider. I don't know if the code looks cleaner with i,
j function or TileMapKey functions, so up to you.

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
File cc/tiles/picture_layer_tiling.cc (right):

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
cc/tiles/picture_layer_tiling.cc:106: const int i = info.tiling_i_index;
While here and doing this, could you just use info.tiling_index here (type being
TileMapKey or std::pair<int, int>)

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
cc/tiles/picture_layer_tiling.cc:137: Tile::CreateInfo info =
CreateInfoForTile(key.index_x, key.index_y);
... Then don't expand key here, just pass a key.

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
cc/tiles/picture_layer_tiling.cc:348: gfx::Rect tile_rect =
tiling_data_.TileBoundsWithBorder(i, j);
... And expand here? Wdyt?

[vmpstr, 2015-08-28 21:05:30]

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc
File cc/tiles/tile.cc (right):

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc#newcode20
cc/tiles/tile.cc:20: const CreateInfo& info,
Also, why not just store the info?

[danakj, 2015-08-31 21:12:12]

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
File cc/tiles/picture_layer_tiling.cc (right):

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/picture_layer_tili...
cc/tiles/picture_layer_tiling.cc:106: const int i = info.tiling_i_index;
On 2015/08/28 21:04:17, vmpstr wrote:
> While here and doing this, could you just use info.tiling_index here (type
being
> TileMapKey or std::pair<int, int>)

As per offline convo.. nope. :D This is kinda weird cuz PLT knows the TileMapKey
is a struct, but Tile doesn't..

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc
File cc/tiles/tile.cc (right):

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc#newcode20
cc/tiles/tile.cc:20: const CreateInfo& info,
On 2015/08/28 21:05:30, vmpstr wrote:
> Also, why not just store the info?

Well, that would add a lot of indirections/typing in the code, and I don't see a
strong benefit. It would save a few lines in the constructor basically. I like
it a lot for plumbing, but less so as a storage type basically.

[danakj, 2015-08-31 21:12:15]

The CQ bit was checked by danakj@chromium.org

[commit-bot: I haz the power, 2015-08-31 21:13:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1318733006/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1318733006/1

[vmpstr, 2015-08-31 21:15:10]

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc
File cc/tiles/tile.cc (right):

https://codereview.chromium.org/1318733006/diff/1/cc/tiles/tile.cc#newcode20
cc/tiles/tile.cc:20: const CreateInfo& info,
On 2015/08/31 21:12:12, danakj wrote:
> On 2015/08/28 21:05:30, vmpstr wrote:
> > Also, why not just store the info?
> 
> Well, that would add a lot of indirections/typing in the code, and I don't see
a
> strong benefit. It would save a few lines in the constructor basically. I like
> it a lot for plumbing, but less so as a storage type basically.

Makes sense.

[commit-bot: I haz the power, 2015-08-31 21:49:28]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-31 21:49:29]

Try jobs failed on following builders:
  android_chromium_gn_compile_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_chromiu...)

[danakj, 2015-08-31 21:50:19]

The CQ bit was checked by danakj@chromium.org

[commit-bot: I haz the power, 2015-08-31 21:51:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1318733006/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1318733006/1

[commit-bot: I haz the power, 2015-08-31 22:22:14]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-08-31 22:22:51]

Patchset 1 (id:??) landed as
https://crrev.com/d11f79b35cf3958bec49a8271c8a9059a983828f
Cr-Commit-Position: refs/heads/master@{#346491}