sparse_strips: Add more new/refactored code into `vello_common` by LaurenzV · Pull Request #1745 · linebender/vello

LaurenzV · 2026-07-14T15:29:23Z

This PR is based on #1744 and introduces some actual changes, once again in preparation for the Vello Hybrid rewrite.

The first commit adds some utility types like Offset and Size, which will layer on be used by Vello Hybrid (they are currently partially duplicated there, but I will just remove them later on). In theory they could just live in Vello Hybrid, but since we already have RectU16 in vello_common I figured this is the best place.

The second commit resolved a TODO and creates a new generic method that allows generating fill/alpha-fill sequences from a slice of strips. We also use this for the existing strip_bbox function now. I'm aware that this will result in a slight slowdown since we will do the clipping even though it isn't necessary to clip anything since we are just calculating the bbox, but I think it's better to accept this cost for less code duplication.

The third commit extracts all of the transforms we currently have into a custom struct, so that, once again, they can more easily be used by Vello Hybrid.

Fourth commit introduces a new ViewportState abstraction, that abstracts all of the complexity of handling different viewport so that Vello Hybrid can use this as well.

Fifth commit adds a way of querying whether an encoded paint has any opacities directly.

PR train

sparse_strips: Move some code from vello_cpu to vello_common #1744 — Move some code from vello_cpu to vello_common
sparse_strips: Add more new/refactored code into vello_common #1745 — Add more new/refactored code into vello_common 👈
vello_common: Generalize the command recorder abstraction and add more metadata #1746 — Generalize the command recorder abstraction and add more metadata
sparse_strips: Introduce Padding type and streamline clip handling #1751 — Introduce Padding and streamline clip handling

LaurenzV · 2026-07-14T15:30:44Z

+// TODO: Remove these types once we've completely moved to u16 everywhere in Vello Hybrid.
+
+/// An offset represented by two 32-bit unsigned integers.
+#[repr(C)]


These types are likely temporary and we will use u16 where possible. Unfortunately, right now we kind of convert between u16 and u32 in all kinds of different places even though it could be more unified, but I want to leave this cleanup to the future.

LaurenzV · 2026-07-16T08:46:27Z

+        match self {
+            Self::Gradient(gradient) => !gradient.may_have_transparency,
+            Self::Image(image) => !image.may_have_transparency,
+            Self::ExternalTexture(_) => false,


TODO: Need to implement this for external textures as well.

Could we have a single method here that indicates whether opaque paint optimizations can be used in cpu/hybrid?

What do you mean? Isn't it the same across CPU/Hybrid? I just need to make sure to read the field from the external texture.

Ah, I presume you meant something like this? 😄 : 90705e6

grebmeg · 2026-07-17T04:31:37Z

+#[derive(Copy, Clone, Debug, Pod, Zeroable, PartialEq, Eq)]
+pub struct SizeU16(pub [u16; 2]);
+
+impl SizeU16 {


As I understand it, the U32 types below will be removed later? If so, do you think it's still worth introducing generic Size<T> and Rect<T> types?

I also considered this but probably not. For now it's easier to just keep it this way. 😄

grebmeg · 2026-07-17T04:35:30Z

    /// Return this rectangle relative to `origin`, clamping negative coordinates to zero.
    #[inline(always)]
-    pub const fn relative_to_origin(self, origin: (u16, u16)) -> Self {
+    pub fn relative_to_origin(self, origin: (u16, u16)) -> Self {


Could we add a couple of small tests for this?

Will add two tests.

grebmeg · 2026-07-17T04:38:08Z

+    type Output = Self;
+
+    fn add(self, rhs: Self) -> Self::Output {
+        Self::from_wh(self.width() + rhs.width(), self.height() + rhs.height())


I don't think this should ever overflow in our cases, but in theory it could. Maybe it's worth adding a short comment noting that it wraps on overflow in release

I'll turn it into a checked add.

grebmeg · 2026-07-17T05:05:44Z

+        match self {
+            Self::Gradient(gradient) => !gradient.may_have_transparency,
+            Self::Image(image) => !image.may_have_transparency,
+            Self::ExternalTexture(_) => false,


Could we have a single method here that indicates whether opaque paint optimizations can be used in cpu/hybrid?

grebmeg · 2026-07-17T06:30:21Z

+    }
+
+    /// Return the currently active root transform.
+    pub fn root_transform(&self) -> Affine {


I think cumulative_transform would be a better name, wouldn't it? When I read root_transform, I think of a single fixed top-level transform. In reality, this represents the cumulative transform, so root_transform feels a bit misleading from that point of view. Would the Cumulative prefix make the intent clearer for this type?

Hmm, I can change, but I think root_transform makes sense. It represents the root transform for the currently active filter viewport that should be used as the basis for any subsequent path/paint transforms.

grebmeg · 2026-07-17T06:59:53Z

Thanks for your work! Everything looks much tidier now 🤩 I left a few comments, but they're mostly minor.

LaurenzV commented Jul 14, 2026

View reviewed changes

LaurenzV mentioned this pull request Jul 14, 2026

vello_common: Generalize the command recorder abstraction and add more metadata #1746

Open

LaurenzV requested a review from grebmeg July 14, 2026 16:04

laurenz-canva force-pushed the laurenz/vello_hybrid_shift_part1 branch from 8806e12 to a93be43 Compare July 15, 2026 11:12

LaurenzV added 2 commits July 15, 2026 13:12

Add more geometry types

85f8273

Share strip fill segment iteration with Vello CPU

89e5866

laurenz-canva force-pushed the laurenz/vello_hybrid_shift_part2 branch from 86cb568 to d10a407 Compare July 15, 2026 11:12

LaurenzV added 3 commits July 16, 2026 10:13

Share transform state with Vello CPU

0540b3e

Extract a ViewportState from Vello CPU

f8cd052

Add opaque paint queries

3f61c40

laurenz-canva force-pushed the laurenz/vello_hybrid_shift_part2 branch from d10a407 to 3f61c40 Compare July 16, 2026 08:14

LaurenzV commented Jul 16, 2026

View reviewed changes

This was referenced Jul 16, 2026

sparse_strips: Move some code from vello_cpu to vello_common #1744

Open

sparse_strips: Introduce Padding type and streamline clip handling #1751

Open

grebmeg reviewed Jul 17, 2026

View reviewed changes

LaurenzV added 9 commits July 17, 2026 13:07

add two tests

974c329

use checked add

ee61987

construct rect directly

a6abcc9

Rename method and var

60247e4

extract filter transform

825ec4f

Extract debug assertions

310ec0b

scene transform and scene_paint_transform

60fd08d

add tests

bea285e

unify may_have_transparenc

90705e6

Uh oh!

Conversation

LaurenzV commented Jul 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR train

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

grebmeg commented Jul 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

LaurenzV commented Jul 14, 2026 •

edited

Loading