Should fix https://github.com/flutter/flutter/issues/159520 ~~but I still need to check locally.~~ Seems to do the trick.
All cmd bindings were copying the shader metadata, which meant allocating/de-allocating a lot of strings per draw.
This is a dumb performance optimization. because we only use the DescriptorType enum to represent vk descriptor types, lets just make the enum values match. Then we can static cast instead of switch.
I do see this function showing up in profiles, though a very small slice.
Avoid Sprintf'ing labels for textures/cmd buffers/render passes. For the render target textures, defer combined label construction until we know the label will be used.
Many of our labeling APIs accepted a std::string. When provided with a string literal (const char*), this would force the allocation of a new std::string object and a copy of the original char*. std::string_view is a view of either a std::string, or a const char* (or other), which allows an API that doesn not need to store the string to accept both without a copy. Of course, we can't store a std::string_view, as that doesn't provide memory ownership - so many of the GLES implementations will still copy the string_view into a string locally.
Also updates several of the debug labelling functions to no-op in release mode and removes some expensive interpolations that aren't super necessary.
Vulkan and GLES backends support triangle fans, so only "un-fan" these geometries when necessary. Now that dispatch gives the vertices geometry as a shared_ptr, we can hold onto that directly without performing another copy.
Measures as substantially faster on windows. See the bufferSubData docs:
> When replacing the entire data store, consider using glBufferSubData rather than completely recreating the data store with glBufferData. This avoids the cost of reallocating the data store.
Tracks invalidated ranges and only updates those ranges using the Flush API.
Fixes https://github.com/flutter/flutter/issues/104447
It seems like on some ANdroid devices we can actually get 4 pending frames even with the 3 frame wait. I thought about this a lot but too hard. Lets bump it up one.
Also minor comment fixes in ahb swapchain.
This is similar to our handling of degrees and radians or really any of the stuff in chrono.
Storage is always in bytes. Before being displayed, it can be converted into any of unit that makes the most sense. Conversions are non-truncating and everything is constexpr safe.
Match DL/Skia sampling defaults. This does not change the default SamplerDescriptor value, but that is irrelevant as any draws from the framework will pass through DL conversion.
The SamplerDescriptor default of nearest is generally a good idea.
Fixes https://github.com/flutter/flutter/issues/148253
Resolves https://github.com/flutter/flutter/issues/134748.
This was a really fun experiment. I learned a lot from it, and it
genuinely helped me solve some coverage-related problems, but the
reality is it was too little too late -- by the time we had this
capture system, we had already solved most of the problems that would
have benefitted from this.
It's been a few months since I've used or extended the capabilities of
this capture system for something, and I don't have the spare
time/energy to give it the love it needs to realize the vision I had
for it. I still almost exclusively use a combination of native frame
captures and print debugging to solve problems.
RIP in peace.
This reverts commit 104eb98e62. (https://github.com/flutter/engine/pull/52680)
part of https://github.com/flutter/flutter/issues/138798
Works around https://github.com/flutter/flutter/issues/144498 for the glyph atlas.
Adds BlitPass::AddCopy implementation for GLES, which is mostly a copy of Texture::SetContents. Updates the glyph atlas to use a blit pass and device private textures instead of set contents (which is unsafely on Metal).
This also removes DeviceBuffer::AsTexture, which isn't actually used anywhere - and creates a linear texture on iOS (and fails with an unsupported API on simulators).
Note that in general, we don't actually have support for hostVisible textures on GLES _or_ Vulkan. Instead, VMA is falling back to device private textures. We may want to, separately, remove the concept of host visible from textures - and similarly remove the concept of transient from buffers.
Resolves https://github.com/flutter/flutter/issues/134748.
This was a really fun experiment. I learned a lot from it, and it genuinely helped me solve some coverage-related problems, but the reality is it was too little too late -- by the time we had this capture system, we had already solved most of the problems that would have benefitted from this.
It's been a few months since I've used or extended the capabilities of this capture system for something, and I don't have the spare time/energy to give it the love it needs to realize the vision I had for it. I still almost exclusively use a combination of native frame captures and print debugging to solve problems.
RIP in peace.
Part of https://github.com/flutter/flutter/issues/143077
Only allocate into reused arenas instead of allocating a new vector of data. Fixes https://github.com/flutter/flutter/issues/133348
Also moves tessellation logic into the c/tessellator and out of the impeller tessellator. This was necessary to fix a compilation error. introduced by including host_buffer -> allocator -> fml mapping -> window.h include which has a function definition that conflicts with the c tessellator definition.
Fixes missing points in case a filled path is not explicitly closed.
Reverts: flutter/engine#52131
Initiated by: jonahwilliams
Reason for reverting: breaking flutter logo rendering
Original PR Author: jonahwilliams
Reviewed By: {chinmaygarde}
This change reverts the following previous change:
Part of https://github.com/flutter/flutter/issues/143077
Only allocate into reused arenas instead of allocating a new vector of data. Fixes https://github.com/flutter/flutter/issues/133348
Also moves tessellation logic into the c/tessellator and out of the impeller tessellator. This was necessary to fix a compilation error. introduced by including host_buffer -> allocator -> fml mapping -> window.h include which has a function definition that conflicts with the c tessellator definition.
Part of https://github.com/flutter/flutter/issues/143077
Only allocate into reused arenas instead of allocating a new vector of data. Fixes https://github.com/flutter/flutter/issues/133348
Also moves tessellation logic into the c/tessellator and out of the impeller tessellator. This was necessary to fix a compilation error. introduced by including host_buffer -> allocator -> fml mapping -> window.h include which has a function definition that conflicts with the c tessellator definition.
Relands https://github.com/flutter/engine/pull/51589
The fix is in 74397bc171c74d2bfb24e82b47f2aa29d70c1711. I couldn't
figure out how to get a test in the engine to cover it. The test is in
the devicelab.
Here's what I attempted:
```c++
TEST_P(AiksTest, BlendModePlusAlphaColorFilterAlphaWideGamut) {
if (GetParam() != PlaygroundBackend::kMetal) {
GTEST_SKIP_("This backend doesn't yet support wide gamut.");
}
EXPECT_EQ(GetContext()->GetCapabilities()->GetDefaultColorFormat(),
PixelFormat::kR16G16B16A16Float);
Canvas canvas;
canvas.Scale(GetContentScale());
canvas.DrawPaint({.color = Color(0.1, 0.2, 0.1, 0.5)});
canvas.SaveLayer({
.color_filter = ColorFilter::MakeBlend(BlendMode::kPlus,
Color(Vector4{1, 0, 0, 0.5})),
});
Paint paint;
paint.color = Color(1, 0, 0, 0.5);
canvas.DrawRect(Rect::MakeXYWH(100, 100, 400, 400), paint);
paint.color = Color::White();
canvas.Restore();
ASSERT_TRUE(OpenPlaygroundHere(canvas.EndRecordingAsPicture()));
}
```
## Pre-launch Checklist
- [x] I read the [Contributor Guide] and followed the process outlined
there for submitting PRs.
- [x] I read the [Tree Hygiene] wiki page, which explains my
responsibilities.
- [x] I read and followed the [Flutter Style Guide] and the [C++,
Objective-C, Java style guides].
- [x] I listed at least one issue that this PR fixes in the description
above.
- [x] I added new tests to check the change I am making or feature I am
adding, or the PR is [test-exempt]. See [testing the engine] for
instructions on writing and running engine tests.
- [x] I updated/added relevant documentation (doc comments with `///`).
- [x] I signed the [CLA].
- [x] All existing and new tests are passing.
If you need help, consider asking for advice on the #hackers-new channel
on [Discord].
<!-- Links -->
[Contributor Guide]:
https://github.com/flutter/flutter/wiki/Tree-hygiene#overview
[Tree Hygiene]: https://github.com/flutter/flutter/wiki/Tree-hygiene
[test-exempt]:
https://github.com/flutter/flutter/wiki/Tree-hygiene#tests
[Flutter Style Guide]:
https://github.com/flutter/flutter/wiki/Style-guide-for-Flutter-repo
[C++, Objective-C, Java style guides]:
https://github.com/flutter/engine/blob/main/CONTRIBUTING.md#style
[testing the engine]:
https://github.com/flutter/flutter/wiki/Testing-the-engine
[CLA]: https://cla.developers.google.com/
[flutter/tests]: https://github.com/flutter/tests
[breaking change policy]:
https://github.com/flutter/flutter/wiki/Tree-hygiene#handling-breaking-changes
[Discord]: https://github.com/flutter/flutter/wiki/Chat
Uses the utility added in https://github.com/flutter/engine/pull/51361
I counted the removal of 58 static casts. There was one addition made to the original utility however. Vulkan HPP was promoting all enums to its own mask type. This in itself is problematic but we got away with it because there was no one else doing this kind of promotion. Till we added our own utility. To avoid polluting the namespace with methods that may cause ambiguity, enums that are masks must explicitly be marked as maskable with `IMPELLER_ENUM_IS_MASK` in the `impeller` namespace.
No change in functionality.
Part of https://github.com/flutter/flutter/issues/144617
Adds MemoryBudgetUsageMB which includes the MB of VMA allocated GPU and host memory, approximately per frame. This will be recorded in the devicelab and used to track how much memory pressure we're creating.
Split out from https://github.com/flutter/engine/pull/51187 since that was reverted (and doing big changes is a bad idea anyway).
Various cleanups to Vulkan allocator implementation:
1. Fixes https://github.com/flutter/flutter/issues/137454
2. Fold device transient cap check into allocator.
3. adds debug tracking for total memory usage in MB (a followup change needs to be made to driver to plumb it through)
4. Small cleanups to mock vulkan so an allocator can be created from it.
5. depth/stencil shouldn't be input attachments.
Part of https://github.com/flutter/flutter/issues/144617
