Extract Linear and Radial Gradients.

[djg]

In support of #3385, extract gradients into separate interned stores.

---

This change is 

[gw3583]

Reviewed 8 of 8 files at r1.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @djg)

---

webrender/src/prim_store/mod.rs, line 591 at r1 (raw file):

}

/// A hashable point for using as a key during primitive interning.

PointKey is (or will be) used by many types, so let's keep it in mod.rs.

---

webrender/src/prim_store/mod.rs, line 3796 at r1 (raw file):

    assert_eq!(mem::size_of::<PrimitiveInstance>(), 128, "PrimitiveInstance size changed");
    assert_eq!(mem::size_of::<PrimitiveInstanceKind>(), 40, "PrimitiveInstanceKind size changed");
    assert_eq!(mem::size_of::<PrimitiveTemplate>(), 144, "PrimitiveTemplate size changed");

Woohoo, this should hopefully be a small-ish talos win.

[gw3583]

Just one minor change, and we should do a try / talos run to verify, but otherwise looks good!

@bors-servo delegate+

[bors-servo]

✌️ @djg can now approve this pull request

[djg]

Try + talos: https://treeherder.mozilla.org/#/jobs?repo=try&revision=fa45ad0eadabe152a7eaf7e00d7202638c737b93

[djg]

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @gw3583)

---

webrender/src/prim_store/mod.rs, line 591 at r1 (raw file):

Previously, gw3583 (Glenn Watson) wrote…

PointKey is (or will be) used by many types, so let's keep it in mod.rs.

Done.

[gw3583]

Try looks good - a couple of expected failures from the recent blur shader change, and a couple of unrelated known intermittent failures. Talos looks like a small win (no regression at least).

@bors-servo r+

[bors-servo]

📌 Commit 2ca99dd has been approved by gw3583

[bors-servo]

⌛ Testing commit 2ca99dd with merge 74b9312...

[bors-servo]

☀️ Test successful - status-appveyor, status-taskcluster
Approved by: gw3583
Pushing 74b9312 to master...