If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

Intermittent B2G emulator object-fit-scale-down-svg-002e.html | image comparison (==), max difference: 255, number of differing pixels: 32

NEW
Unassigned

Status

()

Core
Layout
P3
normal
3 years ago
a year ago

People

(Reporter: RyanVM, Unassigned)

Tracking

({intermittent-failure})

Trunk
ARM
Gonk (Firefox OS)
intermittent-failure
Points:
---

Firefox Tracking Flags

(Not tracked)

Details

(Whiteboard: [test disabled on B2G])

Attachments

(2 attachments)

(Reporter)

Description

3 years ago
Created attachment 8534421 [details]
test screenshot

07:25:20 INFO - REFTEST TEST-START | http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002e.html
07:25:20 INFO - REFTEST TEST-LOAD | http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002e.html | 313 / 522 (59%)
07:25:20 INFO - REFTEST TEST-LOAD | http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002-ref.html | 313 / 522 (59%)
07:25:20 INFO - REFTEST TEST-UNEXPECTED-FAIL | http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002e.html | image comparison (==), max difference: 255, number of differing pixels: 32
07:25:20 INFO - REFTEST IMAGE 1 (TEST): 
07:25:56 INFO - REFTEST IMAGE 2 (REFERENCE): 
07:25:56 INFO - REFTEST INFO | Saved log: START http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002e.html
07:25:56 INFO - REFTEST INFO | Saved log: [CONTENT] OnDocumentLoad triggering AfterOnLoadScripts
07:25:56 INFO - REFTEST INFO | Saved log: Initializing canvas snapshot
07:25:56 INFO - REFTEST INFO | Saved log: DoDrawWindow 0,0,800,1000
07:25:56 INFO - REFTEST INFO | Saved log: [CONTENT] RecordResult fired
07:25:56 INFO - REFTEST INFO | Saved log: RecordResult fired
07:25:56 INFO - REFTEST INFO | Saved log: START http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002-ref.html
07:25:56 INFO - REFTEST INFO | Saved log: [CONTENT] OnDocumentLoad triggering AfterOnLoadScripts
07:25:56 INFO - REFTEST INFO | Saved log: Initializing canvas snapshot
07:25:56 INFO - REFTEST INFO | Saved log: DoDrawWindow 0,0,800,1000
07:25:56 INFO - REFTEST INFO | Saved log: [CONTENT] RecordResult fired
07:25:56 INFO - REFTEST INFO | Saved log: RecordResult fired
07:25:56 INFO - REFTEST INFO | Loading a blank page
07:25:56 INFO - REFTEST TEST-END | http://10.0.2.2:8888/tests/layout/reftests/w3c-css/submitted/images3/object-fit-scale-down-svg-002e.html
(Reporter)

Comment 1

3 years ago
Created attachment 8534423 [details]
reference screenshot
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(Reporter)

Comment 5

3 years ago
Bleh, this is hitting lots of tests within this directory.
(Reporter)

Comment 6

3 years ago
Looks like the first instances of this go back to late November in bug 1104098. I'll try to bisect.
(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #6)
> Looks like the first instances of this go back to late November in bug
> 1104098. I'll try to bisect.

Don't bother -- they were added around that time. I think it's just a low-frequency failure, and they've failed since forever.

Hoping to look at it in more detail at some point. I've been justifying holding off w/ the excuse of "gathering more data". :)
(This seems to be a bug in nsSubDocumentFrame.)

needinfo=me to mark these tests as random-if(b2g), per discussion w/ RyanVM.
Flags: needinfo?(dholbert)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(In reply to Daniel Holbert [:dholbert] from comment #8)
> needinfo=me to mark these tests as random-if(b2g), per discussion w/ RyanVM.

Actually, while I was prepping to do that, I discovered & filed bug 1110950, which I suspect might fix this bug.

So, I'd rather not mark these as random just yet, because I'd like to fix bug 1110950 and discover whether that bug's fix also fixes these failures (which would be great).
Flags: needinfo?(dholbert)
[Tentatively marking as "depends-on" bug 1110950, since per previous comment, it seems possible that bug's fix would fix this class of intermittent-failures.]
Depends on: 1110950
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(Reporter)

Comment 19

3 years ago
(In reply to Daniel Holbert [:dholbert] from comment #12)
> [Tentatively marking as "depends-on" bug 1110950, since per previous
> comment, it seems possible that bug's fix would fix this class of
> intermittent-failures.]

Any news here? :)
Flags: needinfo?(dholbert)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #19)
> (In reply to Daniel Holbert [:dholbert] from comment #12)
> > [Tentatively marking as "depends-on" bug 1110950
[...] 
> Any news here? :)

Yeah, sorry -- I thought I'd get to fixing that bug sooner.  It's still high on my list of todo items, but it's fighting against my review-queue -- so to avoid causing sheriffs further pain, I'll go ahead and push some annotations to silence these test failures for now.
Flags: needinfo?(dholbert)
I temporarily marked this test as random-if(B2G), as noted in bug 1110950 comment 5.
Whiteboard: [test marked as random-if(B2G)]
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(Note: comment 25, 26, and 27 are from different tests from the one that this bug was originally filed on -- tests that I didn't mark as random in comment 24. They're hitting the same basic issue, though, so they're appropriate here; I'm just explaining why we have failure-reports here even after a random-if annotation was added)

Anyway -- I've now removed the random-if annotation, after fixing bug 1110950 (as noted in bug 1110950 comment 38), because hopefully bug 1110950's fix should fix this randomorange.

I'm leaving this open for now, but let's close this in a week or so if we don't get any more reports from m-c & related branches (m-i, fx-team).
Whiteboard: [test marked as random-if(B2G)]
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(Reporter)

Comment 33

3 years ago
Any idea why we're still seeing these, Dan?
Flags: needinfo?(dholbert)
No. :-/ I'm glad the rate has at least gone down a lot (from what I've seen), with bug 1110950's fix, though.

Realistically, I probably won't have cycles to look into these soon, particularly given that the frequency is so low (which makes them harder to debug/investigate).

To avoid wasting sheriff time & spamming random-oranges until I *do* have time to investigate, I'm tempted to mark these as random-if(b2g) for now, with a tracking bug filed for digging more.
Flags: needinfo?(dholbert)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Summary: Intermittent object-fit-scale-down-svg-002e.html | image comparison (==), max difference: 255, number of differing pixels: 32 → Intermittent B2G emulator object-fit-scale-down-svg-002e.html | image comparison (==), max difference: 255, number of differing pixels: 32
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
This seems to have gotten worse in the last ~24 hours... (since comment 51)
I probably won't have a chance to a deep-dive into what's happening here, but I'm happy to rubber-stamp marking this test & any similarly-affected friends as random-if(B2G) for the time being, if this stays bad.

(I may push an annotation along those lines tomorrow, but I'm happy with someone else making that change sooner if they like.)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
BTW: I annotated this bug's test as "skip-if(B2G)" in bug 1148588 comment 2.
Whiteboard: [test disabled on B2G]
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)

Comment 84

a year ago
Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.
Priority: -- → P3
You need to log in before you can comment on or make changes to this bug.