Closed Bug 946317 Opened 11 years ago Closed 10 years ago

hwaccel canvas overdraws shadows on windows when outside clip

Categories

(Core :: Graphics, defect)

7 Branch
x86_64
Windows 7
defect
Not set
normal

Tracking

()

RESOLVED FIXED
mozilla27
Tracking Status
firefox25 --- affected
firefox26 --- affected
firefox27 --- verified
firefox28 --- verified
firefox-esr17 --- affected
firefox-esr24 --- affected

People

(Reporter: kevin, Assigned: bas.schouten)

References

Details

(Keywords: regression, testcase)

Attachments

(3 files)

Attached file ffbug.html
User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.57 Safari/537.36

Steps to reproduce:

If you use drawImage to draw the contents of 1 canvas to another, the shadow on the source canvas can be drawn repeatedly to the destination if the destination is redrawn with a clip region that excludes the region where the source was placed. This bug only happens w/ acceleration is turned on.


Actual results:

Presumably the source canvas is serialized to a displaylist that is being replayed when the destination is redraw. The clip region is preserving the previous buffer, but not excluding the display list from being re-evaluated, and therefore the shadows compound to eventually produce a solid black color.


Expected results:

The display list should be excluded since it is outside the clip region, and the shadow should not accumulate.
Shows bug
Regression window(m-c)
Good:
http://hg.mozilla.org/mozilla-central/rev/450e4d9ea2d5
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0a1) Gecko/20110624 Firefox/7.0a1 ID:20110625025936
Bad:
http://hg.mozilla.org/mozilla-central/rev/48ad4ffc4230
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0a1) Gecko/20110625 Firefox/7.0a1 ID:20110625030821
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=450e4d9ea2d5&tochange=48ad4ffc4230

Regression window(m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/c65f1fb0449d
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0a1) Gecko/20110624 Firefox/7.0a1 ID:20110624094307
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/46c3d095bc75
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0a1) Gecko/20110624 Firefox/7.0a1 ID:20110624104626
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=c65f1fb0449d&tochange=46c3d095bc75

Regressed by: Bug 651858
Blocks: 651858
Status: UNCONFIRMED → NEW
Component: Untriaged → Graphics
Ever confirmed: true
Keywords: regression
Product: Firefox → Core
Version: 25 Branch → 7 Branch
Looks like Moz2D/D2D bug
Assignee: nobody → bas
Highly likely, yes. I already have some suspicion as to the cause.
This bug is closely related to bug 946475 and will be solved by the patches there. However in addition to this there's an additional issue that this shows that will not be fixed there, but that should be much more rare.
Depends on: 946475
This is WFM as per Bug 946475's checkin.

(In reply to Bas Schouten (:bas.schouten) from comment #5)

> However in addition to this there's an additional issue that this
> shows that will not be fixed there, but that should be much more rare.

Can this get resolved nevertheless? New bug for the other issue?
Flags: needinfo?(bas)
Keywords: testcase
Status: NEW → RESOLVED
Closed: 10 years ago
Flags: needinfo?(bas)
Resolution: --- → FIXED
Target Milestone: --- → mozilla27
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: