cc: Fix tile manager queues perftests to use number of layers.

cc: Fix tile manager queues perftests to use number of layers.

I made a bit of a blunder when writing the tile manager queues perftests
and made them use a variable number of tilings, instead of layers.

The important thing for us to test here is how the system behaves with
multiple layers. Multiple tilings case is tested at the tiling level
iterator level.

This patch fixes up all of this.

NOTE TO PERF SHERIFFS
=====================

Please disregard any performance changes in the tile manager perftests
as a result of this patch. This changes fundamentally what the
perftests are measuring.

R=reveman

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=287104

[vmpstr, 2014-07-31 18:59:07]

PTAL. Here are the results:

$ ./out/Release/cc_perftests --single-process-tests
--gtest_filter="TileManager*Queue*"

Note: Google Test filter = TileManager*Queue*
[==========] Running 4 tests from 1 test case.
[----------] Global test environment set-up.
[----------] 4 tests from TileManagerPerfTest
[ RUN      ] TileManagerPerfTest.RasterTileQueueConstruct
*RESULT tile_manager_raster_tile_queue_construct: 2= 1273727.5 runs/s
*RESULT tile_manager_raster_tile_queue_construct: 10= 266219.59375 runs/s
*RESULT tile_manager_raster_tile_queue_construct: 50= 33431.20703125 runs/s
[       OK ] TileManagerPerfTest.RasterTileQueueConstruct (6025 ms)
[ RUN      ] TileManagerPerfTest.RasterTileQueueConstructAndIterate
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 2_16=
98692.6328125 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 2_32=
48960.83984375 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 2_64=
23855.3359375 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 2_128=
10862.8603515625 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 10_16= 62418.78125
runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 10_32=
34529.7578125 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 10_64=
18466.333984375 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 10_128=
8956.009765625 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 50_16=
21422.150390625 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 50_32=
16150.130859375 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 50_64=
10996.42578125 runs/s
*RESULT tile_manager_raster_tile_queue_construct_and_iterate: 50_128=
6341.48388671875 runs/s
[       OK ] TileManagerPerfTest.RasterTileQueueConstructAndIterate (24191 ms)
[ RUN      ] TileManagerPerfTest.EvictionTileQueueConstruct
*RESULT tile_manager_eviction_tile_queue_construct: 2= 338110.4375 runs/s
*RESULT tile_manager_eviction_tile_queue_construct: 10= 83365.203125 runs/s
[ RUN      ] TileManagerPerfTest.EvictionTileQueueConstructAndIterate
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 2_16= 62443.9375
runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 2_32=
29872.580078125 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 2_64=
14589.4677734375 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 2_128=
7002.51416015625 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 10_16=
31774.4765625 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 10_32=
15053.46484375 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 10_64=
7702.64306640625 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 10_128=
3893.92529296875 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 50_16=
9212.0703125 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 50_32=
4567.61572265625 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 50_64=
2491.5244140625 runs/s
*RESULT tile_manager_eviction_tile_queue_construct_and_iterate: 50_128=
1250.7861328125 runs/s
[       OK ] TileManagerPerfTest.EvictionTileQueueConstructAndIterate (24179 ms)
[----------] 4 tests from TileManagerPerfTest (60420 ms total)

[----------] Global test environment tear-down
[==========] 4 tests from 1 test case ran. (60421 ms total)
[  PASSED  ] 4 tests.

[reveman, 2014-07-31 19:07:37]

lgtm with nit

https://codereview.chromium.org/434713002/diff/1/cc/resources/tile_manager_pe...
File cc/resources/tile_manager_perftest.cc (right):

https://codereview.chromium.org/434713002/diff/1/cc/resources/tile_manager_pe...
cc/resources/tile_manager_perftest.cc:442: RunRasterQueueConstructTest("2", 2);
nit: maybe include the one layer test case? more generally, should the layer
counts here be consistent with the ManageTiles test?

[vmpstr, 2014-07-31 20:11:25]

https://codereview.chromium.org/434713002/diff/1/cc/resources/tile_manager_pe...
File cc/resources/tile_manager_perftest.cc (right):

https://codereview.chromium.org/434713002/diff/1/cc/resources/tile_manager_pe...
cc/resources/tile_manager_perftest.cc:442: RunRasterQueueConstructTest("2", 2);
On 2014/07/31 19:07:37, reveman wrote:
> nit: maybe include the one layer test case? more generally, should the layer
> counts here be consistent with the ManageTiles test?

Done. I actually changed the ManageTiles to be consistent with this.

[vmpstr, 2014-07-31 20:11:31]

The CQ bit was checked by vmpstr@chromium.org

[commit-bot: I haz the power, 2014-07-31 20:15:00]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/vmpstr@chromium.org/434713002/20001

[commit-bot: I haz the power, 2014-08-01 00:22:10]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_chromium_x64_rel on tryserver.chromium.win
(http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[commit-bot: I haz the power, 2014-08-01 13:05:55]

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

[commit-bot: I haz the power, 2014-08-01 13:05:56]

Try jobs failed on following builders:
  win_chromium_x64_rel on tryserver.chromium.win
(http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[vmpstr, 2014-08-01 16:10:19]

The CQ bit was checked by vmpstr@chromium.org

[commit-bot: I haz the power, 2014-08-01 16:14:48]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/vmpstr@chromium.org/434713002/20001

[commit-bot: I haz the power, 2014-08-01 22:18:07]

Change committed as 287104