1468294 - panicked at 'animate should only be used for interpolating or accumulating transforms'

Status: RESOLVED FIXED

(Core :: DOM: Animation, defect)

Description

[Jesse Schwartzentruber (:truber)]

Attachment: testcase.html

The attached testcase causes a panic in m-c 20180611-62ab31ea0ec7

thread '<unnamed>' panicked at 'animate should only be used for interpolating or accumulating transforms', style-2f3f2642796fb364/out/properties.rs:88389:9
...
#5: std::panicking::begin_panic
        at src/libstd/panicking.rs:538
#6: _<style..properties..animated_properties..Quaternion as style..values..animated..Animate>::animate
        at /out/properties.rs:88389
#7: _<style..properties..animated_properties..MatrixDecomposed3D as style..values..animated..Animate>::animate
        at /out/properties.rs:88341
#8: style::properties::animated_properties::_<impl style..values..animated..Animate for style..values..generics..transform..Matrix3D<f32>>::animate
        at /out/properties.rs:88127
#9: style::properties::animated_properties::_<impl style..values..animated..Animate for style..values..generics..transform..TransformOperation<style..values..computed..angle..Angle, f32, style..values..computed..length..CSSPixelLeng
th, i32, style..values..computed..length..LengthOrPercentage>>::animate
        at /out/properties.rs:87711
#10: _<&'a mut I as core..iter..iterator..Iterator>::next
        at /out/properties.rs:89261
#11: _<alloc..vec..Vec<T> as alloc..vec..SpecExtend<T, I>>::from_iter
        at src/liballoc/vec.rs:1801
#12: _<style..properties..animated_properties..AnimationValue as style..values..animated..Animate>::animate
        at src/liballoc/vec.rs:1713
#13: geckoservo::glue::compose_animation_segment
        at servo/ports/geckolib/glue.rs:591
#14: geckoservo::glue::Servo_ComposeAnimationSegment
        at servo/ports/geckolib/glue.rs:608
#15: mozilla::layers::AnimationHelper::SampleAnimationForEachNode(mozilla::TimeStamp, mozilla::TimeStamp, nsTArray<mozilla::layers::Animation>&, nsTArray<mozilla::layers::AnimData>&, RefPtr<RawServoAnimationValue>&, mozilla::layers:
:AnimatedValue const*)
        at gfx/layers/AnimationHelper.cpp:319

Comment 1

[Hiroyuki Ikezoe (:hiro)]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=b81f8bcd3a35bada0aa11483750f216277887754

Comment 2

[Hiroyuki Ikezoe (:hiro)]

Attachment: Bug 1468294 - Double f64::EPSILON for calculation error.

It's possible that both this_weight and other_weght have calculation errors
which are approximately equal to f64::EPSILON.

Review commit: https://reviewboard.mozilla.org/r/250732/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/250732/

Comment 3

[Brian Birtles (:birtles, Feb 26~ parental leave)]

Comment on attachment 8984997 [details]
Bug 1468294 - Double f64::EPSILON for calculation error.

https://reviewboard.mozilla.org/r/250732/#review257004

::: servo/components/style/properties/helpers/animated_properties.mako.rs:1741
(Diff revision 1)
>      fn animate(&self, other: &Self, procedure: Procedure) -> Result<Self, ()> {
>          use std::f64;
>  
>          let (this_weight, other_weight) = procedure.weights();
>          debug_assert!(
> -            (this_weight + other_weight - 1.0f64).abs() <= f64::EPSILON ||
> +            (this_weight + other_weight - 1.0f64).abs() <= f64::EPSILON * 2.0 ||

We should a comment to the code to explain why we are doubling the epsilon.

Comment 4

[Hiroyuki Ikezoe (:hiro)]

Comment on attachment 8984997 [details]
Bug 1468294 - Double f64::EPSILON for calculation error.

Review request updated; see interdiff: https://reviewboard.mozilla.org/r/250732/diff/1-2/

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/ee2f1c495bc1
Double f64::EPSILON for calculation error. r=birtles

Comment 6

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/ee2f1c495bc1