It looks like this pull request may not have tests. Please make sure to add tests before merging. While there are exceptions to this rule, if this patch modifies code it is probably not an exception.

Reviewers: Read the Tree Hygine page and make sure this patch meets those guidelines before LGTMing.

/cc @dnfield

Why not a 10x10 or a 1x1 surface instead?

Why not a 10x10 or a 1x1 surface instead?

The draw calls have to be onscreen. We might be able to drop below 100x100 by tweaking the draw calls, but at 100x100 our read pixels is 4ms. I don't know if we'll be able to drop much below that. (Ideally we don't even read pixels but I've given up on that for now).

I realized that "adding a test to verify the number of shaders being warmed up is as expected" could be a little complicated, so I'm Ok to just have it manually checked for this PR, and maybe add it as a test in another dedicated PR.

On the other hand, add a test to measure the total time of shader warm-up so we can track it in our dashboard (instead of doing an observatory tracing manually every time) shouldn't be very difficult. For example, you can create MeasureDefaultShaderWarmUp that extends DefaultShaderWarmUp and overrides the execute like this:

class MeasureDefaultShaderWarmUp extends DefaultShaderWarmUp {
  Duration measuredTime;

  @override
  Future<void> execute() async {
    final DateTime start = DateTime.now();
    await super.execute();
    measuredTime = DateTime.now() - start;
  }
}

You can then write a device lab driver test to report the measurement: https://github.com/flutter/flutter/tree/master/dev/devicelab

Please let me know if you have any questions about the device lab test. Ideally, maybe we can land the driver test before the merge of this PR so we can get an official number of how much time is saved on our test devices 😄

You can then write a device lab driver test to report the measurement: https://github.com/flutter/flutter/tree/master/dev/devicelab

Hey Yuqian, I looked into this and it doesn't seem trivial. Whatever tests covered this code previously should cover this PR. I'm not sure if continual tracking of shader warmup is something we want to do, considering the effort to do so. This is demonstrably faster with profiling. I agree that the benchmark would be nice to have, but I'm failing to see how it is a good return on investment. Let's chat tomorrow.

The 2 outstanding tests have passed, they just aren't reporting so here for some reason.

I'm investigating the historic slowdowns in the complex_layout_scroll_perf__memory benchmark. It looks like we were at around 5000 in late July and since then we've been stuck around 9000+.

It appears that this commit is responsible for that change. Here are 3 runs of the benchmark on a Moto G4 first running on commit 0d0af31 (which is just before this commit) and then on this commit 9b150f1

Stats for hash 0d0af3

On G4:

  "data": {
    "start-min": 34210,
    "start-max": 40838,
    "start-median": 39220,
    "end-min": 44052,
    "end-max": 46190,
    "end-median": 45034,
    "diff-min": 4464,
    "diff-max": 11980,
    "diff-median": 5033
  },
  "data": {
    "start-min": 35164,
    "start-max": 41233,
    "start-median": 40473,
    "end-min": 44973,
    "end-max": 45653,
    "end-median": 45321,
    "diff-min": 4125,
    "diff-max": 10322,
    "diff-median": 4487
  },
  "data": {
    "start-min": 32907,
    "start-max": 40356,
    "start-median": 39693,
    "end-min": 44393,
    "end-max": 48066,
    "end-median": 44879,
    "diff-min": 4099,
    "diff-max": 15159,
    "diff-median": 4907
  },

Stats for hash 9b150f:

On G4:

  "data": {
    "start-min": 31106,
    "start-max": 33471,
    "start-median": 31184,
    "end-min": 40494,
    "end-max": 45019,
    "end-median": 40980,
    "diff-min": 7793,
    "diff-max": 11548,
    "diff-median": 9319
  },
  "data": {
    "start-min": 31144,
    "start-max": 32886,
    "start-median": 31240,
    "end-min": 40506,
    "end-max": 41104,
    "end-median": 40906,
    "diff-min": 7932,
    "diff-max": 9798,
    "diff-median": 9690
  },
  "data": {
    "start-min": 30958,
    "start-max": 32838,
    "start-median": 31365,
    "end-min": 40332,
    "end-max": 41224,
    "end-median": 40892,
    "diff-min": 8054,
    "diff-max": 9874,
    "diff-median": 9501
  },

See #40406