1706846 - https://codepen.io/thebabydino/pen/OJRoapw renders as a full disc (it should render as a crescent)

Status: VERIFIED FIXED

(Core :: Graphics: WebRender, defect, P3)

Description

[Mayank Bansal]

Repros with both sw-wr and hw-wr

go to https://codepen.io/thebabydino/pen/OJRoapw

ER: Crescent shape
AR: Full round disc

Regression points to bug 1682195 , (though Im not 100% certain its correct as there was a period during the regression range where night would crash on the testcase.)

2021-04-22T08:59:10.707000: INFO : Narrowed integration regression window from [d12aec06, f3b8d237] (4 builds) to [7534ee1e, f3b8d237] (2 builds) (~1 steps left)
2021-04-22T08:59:10.720000: DEBUG : Starting merge handling...
2021-04-22T08:59:10.721000: DEBUG : Using url: https://hg.mozilla.org/integration/autoland/json-pushes?changeset=f3b8d237e29af262b2042809e13ce6043b75c922&full=1
2021-04-22T08:59:10.721000: DEBUG : redo: attempt 1/3
2021-04-22T08:59:10.721000: DEBUG : redo: retry: calling _default_get with args: ('https://hg.mozilla.org/integration/autoland/json-pushes?changeset=f3b8d237e29af262b2042809e13ce6043b75c922&full=1',), kwargs: {}, attempt #1
2021-04-22T08:59:10.725000: DEBUG : urllib3.connectionpool: Resetting dropped connection: hg.mozilla.org
2021-04-22T08:59:12.236000: DEBUG : urllib3.connectionpool: https://hg.mozilla.org:443 "GET /integration/autoland/json-pushes?changeset=f3b8d237e29af262b2042809e13ce6043b75c922&full=1 HTTP/1.1" 200 None
2021-04-22T08:59:12.299000: DEBUG : Found commit message:
Bug 1682195 - Ensure that area outside image mask is cleared if it is first clip. r=gw

This fixes a previous bug in the image mask code that the tile rect caused cs_clip_image
to use discard on a first clip, the outside area would be left to whatever the initial value
of the texture from texture cache was. Since we need to support trimming down to the tile
rect for tiled image masks, this opts to just clear the background of the task in tricky
cases.

Differential Revision: https://phabricator.services.mozilla.com/D111375

2021-04-22T08:59:12.299000: DEBUG : Did not find a branch, checking all integration branches
2021-04-22T08:59:12.299000: INFO : The bisection is done.
2021-04-22T08:59:12.299000: INFO : Stopped

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

Set release status flags based on info from the regressing bug 1682195

Comment 2

[Lee Salzman [:lsalzman]]

It looks like I just need to modify my patch from bug 1682195 to also detect if the clip transform is axis aligned, not just the prim transform.

Comment 3

[Lee Salzman [:lsalzman]]

Attachment: Bug 1706846 - Verify that total clip transform is axis-aligned rather than just the primitive transform. r?nical

It's possible that the overall clip transform isn't axis-aligned even if the primitive transform is. So we rather
need to verify that this overall transform is axis-aligned.

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/9057a10b4b9c
Verify that total clip transform is axis-aligned rather than just the primitive transform. r=gfx-reviewers,aosmond

Comment 5

[Iulian Moraru]

https://hg.mozilla.org/mozilla-central/rev/9057a10b4b9c

Comment 6

[Mayank Bansal]

This is fixed in the latest nightly https://hg.mozilla.org/mozilla-central/rev/9b7bac5af873863628d90e89299a813228ddbb83

Comment 7

[Lee Salzman [:lsalzman]]

Comment on attachment 9217901 [details]
Bug 1706846 - Verify that total clip transform is axis-aligned rather than just the primitive transform. r?nical

Beta/Release Uplift Approval Request

• User impact if declined: Potential rendering artifacts with HW and SW WR when CSS 3D transforms are used.
• Is this code covered by automated tests?: Unknown
• Has the fix been verified in Nightly?: Yes
• Needs manual test from QE?: No
• If yes, steps to reproduce:
• List of other uplifts needed: None
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): Just adds some more checks around some rendering code. Verified fix.
• String changes made/needed:

Comment 8

[Pascal Chevrel:pascalc]

Comment on attachment 9217901 [details]
Bug 1706846 - Verify that total clip transform is axis-aligned rather than just the primitive transform. r?nical

Fix for an 89 regression. On nightly for 2 weeks, uplift approved for 89 beta 9, thanks

Comment 9

[Pascal Chevrel:pascalc]

https://hg.mozilla.org/releases/mozilla-beta/rev/cd1b4179392f