|
f75e654d
|
2019-01-05T16:28:34
|
|
Add DrawElements on WGL perf test.
Bug: angleproject:2995
Change-Id: I3ed814935ba587d260502b664976322a57f32d98
Reviewed-on: https://chromium-review.googlesource.com/c/1396502
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
4627b370
|
2019-01-02T18:38:32
|
|
Add gl null driver tests to DrawElements.
Requires a couple tweaks to the back-end to be compatible with the
null driver.
Bug: angleproject:2966
Change-Id: Ia83651aabb0dd14b7d6f64152c276463a8fb28fc
Reviewed-on: https://chromium-review.googlesource.com/c/1392392
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
828c5946
|
2018-12-13T23:09:05
|
|
Split angle_perftests into white and black box.
The black box version tests traditional GL API features. The white box
version is designed to test internal classes and uses a static link
against libANGLE and libGLESv2. The black box version will also be
extendable to test native drivers directly instead of going through
ANGLE.
Bug: angleproject:2995
Change-Id: I0b672db37ed38f82157c7a1db70080ff539337ce
Reviewed-on: https://chromium-review.googlesource.com/c/1359515
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
f3acb8c1
|
2018-11-14T16:24:25
|
|
Retool perf test running.
This change does a few things involving the way we run the perf tests:
- Perf test runner can capture different metrics
Useful for capturing the new "nanoSecPerIteration" metric.
- Removes the "score" metric
We'll move to the new time-based metrics. These new metrics are scaled
correctly with iteration counts.
- Runs three trials per perf test
This gives more measurements per test. Each trial is approximately one
second. First the perf tests set a fixed number of iterations after
calibrating the number of steps that we can run in one second. After
that the three trials are run. This should give more stable results.
- Apply more CPU stabilization on Windows
Use SetPriorityClass to apply more CPU priority. Also upgrade
SetThreadPriority to the highest level.
- Always build the Vulkan command buffer test
This catches build regressions more easily. We still skip the test on
non-Android platforms.
Bug: angleproject:2923
Change-Id: I7da234c5af07775ba4a232bb8d65e0138ee7073f
Reviewed-on: https://chromium-review.googlesource.com/c/1330262
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
|
6bf09284
|
2018-10-31T13:45:32
|
|
Scale down a slow index buffer perf test.
This test was slow enough that running 20k iterations might have been
causing a timeout on the Nexus 5.
Bug: angleproject:2923
Change-Id: I03f6e95135d8c521e6ef3e81bc9a4b0dfd7f19df
Reviewed-on: https://chromium-review.googlesource.com/c/1310693
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
3402d523
|
2018-10-30T15:14:52
|
|
Try to reduce variance in angle_perftests.
This change does a few things:
- make perf test runner script print % variation instead of stddev
This makes it a bit more clear how much variance there is.
- stabilize CPU in the render perf tests
Setting a thread affinity and priority should stop from switching cores
during the run. Hopefully can prevent background noise from changing
the test results.
- warm up the benchmark with a few iterations
This should hopefully make the test results a bit more stable.
- output a new normalized perf result value
The new result is normalized against the number of iterations. So it
should hopefully be stable even if the number of iterations is changed.
- increases the iteration count in the draw call perf tests.
These tests were completely dominated by SwapBuffers time. Increasing
the iterations per step means we actually are bottlenecked on CPU time
instead.
Bug: angleproject:2923
Change-Id: I5ee347cf93df239ac33b83dc5effe4c21e066736
Reviewed-on: https://chromium-review.googlesource.com/c/1303679
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
|
0ea96210
|
2018-10-30T15:14:51
|
|
Vulkan: Enable more perf tests.
Also fixes the check for the correct UINT index extension that wasn't
available on Vulkan. Also includes a workaround for the mock ICD not
implementing buffer state for index ranges.
Bug: angleproject:2923
Change-Id: Iab35809d15f890525a9e658d4148272c46cf1320
Reviewed-on: https://chromium-review.googlesource.com/c/1308733
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
be0e2c0c
|
2017-11-20T11:38:16
|
|
Fix type conversion warnings
Warnings about implicit type conversions were showing up as errors on
MSVS 2017.
TEST=build with MSVS 2017
Change-Id: I5f173a6838da16e96fad134756d31f56890c147b
Reviewed-on: https://chromium-review.googlesource.com/778841
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
8957b6c0
|
2017-11-14T12:40:38
|
|
Add perf test for ushort DrawElements.
The code path in the D3D11 back-end is sigificantly different for
unsigned short indices vs unsigned int. Because of the workaround we
have for the primitive restart index in D3D11, ushort rendering can
be slower.
BUG=angleproject:2229
Change-Id: I303dcc55b0314ec45508044995ba47b250cbb87d
Reviewed-on: https://chromium-review.googlesource.com/767149
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
b8d2664f
|
2017-10-27T18:14:14
|
|
Run angle_perftests on GLES backend
Adds OPENGLES_NULL configuration
and selects between OPENGL and OPENGLES based on whether building for Android.
Also 2 small changes to get the newly enabled tests to pass on N5X:
1. Require GL_EXT_texture_storage in TexSubImage test
2. Limit numVertexUniforms and numFragmentUniforms to 64 in MatrixUniforms test
BUG=675997
Change-Id: I5439e5fb7e93b3a928f12594761115d56f60d81b
Reviewed-on: https://chromium-review.googlesource.com/748522
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
7478ce7f
|
2017-10-13T13:19:07
|
|
D3D11: Minor Optimization to DrawCallNeedsTranslation
Remove IsStreamingIndexData from DrawCallNeedsTranslation since
IsStreamingIndexData is only needed for drawElementsIndirect for fast path.
This change gives about a 4% improvement on the drawElements benchmark for
the D3D11 backend with the null driver.
BUG=angleproject:1155
Change-Id: Ife2a9748f6b6fe9bc1f3a67c96672ab941a89d44
Reviewed-on: https://chromium-review.googlesource.com/717946
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
c2493e38
|
2017-06-20T17:16:25
|
|
Add DrawElementsPerf test
This is a follow-up CL for
https://chromium-review.googlesource.com/c/544634/.
In this commit, we add drawElements perf test for d3d9, d3d11 and
OpenGL. Specially, we test DrawElements with unchanged index buffer
every frame and changed index buffer every frame. This is also a
preparation for further remove IndexRange in validation.
Paste a set of data (repeated 30) for reference on Intel skylake
Win10 desktop.
DrawElementsPerfBenchmark.Run/d3d11:
mean: 13186.1666667, stddev: 200.479605502,
truncated mean: 13192.5416667, stddev: 156.284190704
DrawElementsPerfBenchmark.Run/d3d11_index_buffer_changed:
mean: 43.6, stddev: 4.02160830183,
truncated mean: 43.5833333333, stddev: 2.97092391166
BUG=angleproject:1393
TEST=angle_perftests
Change-Id: I94bd181aa2b4515df1affc79f591f5db399feac3
Reviewed-on: https://chromium-review.googlesource.com/585981
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|