Closed Bug 1440088 Opened 7 years ago Closed 7 years ago

Categories

(Core :: Graphics: WebRender, defect, P1)

defect

Tracking

()

RESOLVED DUPLICATE of bug 1454187
Tracking Status
firefox-esr52 --- unaffected
firefox-esr60 --- unaffected
firefox58 --- unaffected
firefox59 --- unaffected
firefox60 --- disabled
firefox61 --- disabled
firefox62 --- disabled

People

(Reporter: Tobbi, Assigned: jrmuizel)

References

(Depends on 1 open bug, Blocks 1 open bug, )

Details

(4 keywords)

Crash Data

Attachments

(2 files)

Steps to reproduce: 1. Go to http://www.adfc-tourenportal.de/viewtrackobject.php?trackObjectID=23944 2. Scroll in and out of the map. (The resulting rendering issue has been reported as bug 1440087). 3. I got a crash while doing so repeatedly in a short period of time. Crash report: https://crash-stats.mozilla.com/report/index/007e7c63-6f18-414b-8772-24f850180221
See Also: → 1440087
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → DUPLICATE
Blocks: 1391255
Status: RESOLVED → REOPENED
Resolution: DUPLICATE → ---
Attached video Crash.mp4
Attached video Memory Leak.mp4
Severity: major → normal
OS: Mac OS X → All
Hardware: x86 → All
It looks like bug 1391255 could have multiple causes. This one is easy to reproduce for me.
Severity: normal → critical
> Memory Leak.mp4 OBS was dying. In reality, zooming stutters only at the end.
Failed to allocate a surface due to invalid size (DTD) Size(12290,12290) preceding the assert/panic.
Assignee: nobody → kvark
Priority: -- → P1
Try completely zooming out and then zooming into the center of Moscow. I had no luck with other cities like London. (lol)
(In reply to Jan Andre Ikenmeyer [:darkspirit] from comment #7) > Try completely zooming out and then zooming into the center of Moscow. mozregression --launch 9cd8e03d9e472f07041be3cd80cc95a8194f91a5 --pref gfx.webrender.all:true startup.homepage_welcome_url:"https://www.adfc-tourenportal.de/map.php?&minimap=1&maptypecontrol=1&largenavigation=1&poi=1&viewbox=50.5964,6.08269,54.8396,10.0339&selectId=16782716385aac1ef0225d2&sid=tjr78raeoc0j55vunekjr60gd2&TrackobjectId=23944" -B debug > 0:43.72 INFO: [Child 3475, ImageIO] WARNING: Appending an extra chunk for SourceBuffer: file /builds/worker/workspace/build/src/image/SourceBuffer.cpp, line 148 > 0:43.72 INFO: [Child 3475, ImageIO] WARNING: Appending an extra chunk for SourceBuffer: file /builds/worker/workspace/build/src/image/SourceBuffer.cpp, line 148 > 0:43.79 INFO: [Child 3475, ImageIO] WARNING: Appending an extra chunk for SourceBuffer: file /builds/worker/workspace/build > [...] > 0:56.11 INFO: [GFX3-]: Surface size too large (exceeds allocation limit)! > 0:56.11 INFO: [GFX3-]: Surface size too large (exceeds extent limit)! > 0:56.11 INFO: [GFX1-]: Failed to allocate a surface due to invalid size (DTD) Size(13122,9863) > 0:56.11 INFO: thread 'WRWorker#1' panicked at 'Moz2D replay problem', gfx/webrender_bindings/src/moz2d_renderer.rs:187:21 Behavior is the same: Sometimes Nightly just freezes with a broken UI and uses more and more memory (until my desktop freezes) or I get a crash report like this one: bp-7e13d0b5-775f-487a-bdd1-4478f0180316 [@ mozalloc_abort | abort | rayon_core::job::{{impl}}::execute<T> ]
Blocks: 1446286
Summary: Crash in [@ mozalloc_abort | abort | webrender::resource_cache::ResourceCache::block_until_all_resources_added::h37f200fba556c4c6 ] with webrender on http://www.adfc-tourenportal.de/viewtrackobject.php?trackObjectID=23944 → Crash on http://www.adfc-tourenportal.de/viewtrackobject.php?trackObjectID=23944
Nightly 61 x64 20180403100105 de_DE 4a3275936ddf871103b53e00608e2b8d5aee7e69 @ Debian Testing, KDE, Radeon RX480, 2560x1440 fresh profile [@ mozalloc_abort | abort | libxul.so@0x3d69b08 | libxul.so@0x3d69af8 | libxul.so@0x3d59650 | rayon_core::job::{{impl}}::execute<T> ] bp-5af27b2a-6477-4fc0-83e3-d88680180403 03.04.18 21:35 gfx.webrender.all + gfx.webrender.blob.invalidation bp-ff9fa5c0-7a23-455d-8256-74d0e0180403 03.04.18 21:34 gfx.webrender.all > MOZ_CRASH Reason Moz2D replay problem > GraphicsCriticalError |[0][GFX1-]: Failed to allocate a surface due to invalid size (DTD) Size(15170,9507) (t=31.865) fastly zooming in: memory leak zooming in not too fast, but also not too slow: crash Slightly different crash signature, otherwise unchanged.
Assignee: kvark → jmuizelaar
Crash Signature: [@ mozalloc_abort | abort | libxul.so@0x3d69b08 | libxul.so@0x3d69af8 | libxul.so@0x3d59650 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3d6b258 | libxul.so@0x3d6b248 | libxul.so@0x3d5ada0 | rayon_core::job::{{impl}}:…
Crash Signature: rayon_core::job::{{impl}}::execute<T> ] → rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3d08258 | libxul.so@0x3d08248 | libxul.so@0x3cf7da0 | rayon_core::job::{{impl}}::execute<T> ]
Blocks: 1454187
Crash Signature: rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3d08258 | libxul.so@0x3d08248 | libxul.so@0x3cf7da0 | rayon_core::job::{{impl}}::execute<T> ] → rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3d08258 | libxul.so@0x3d08248 | libxul.so@0x3cf7da0 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3d031f8 | libxul.so@0x3d031e8 | libxul…
gfx.webrender.all + gfx.webrender.async-scene-build;2 = bug 1455743 bp-775d7165-52dc-4fab-9288-b6def0180424 gfx.webrender.all + gfx.webrender.async-scene-build;0 = same as before. bp-c1322c32-f3d3-4142-8254-fd6000180424
Crash Signature: libxul.so@0x3cf2d40 | rayon_core::job::{{impl}}::execute<T> ] → libxul.so@0x3cf2d40 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3cdfb28 | libxul.so@0x3cdfb18 | libxul.so@0x3ccf670 | rayon_core::job::{{impl}}::execute<T> ]
Depends on: 1454660
Debian Testing, KDE, Radeon RX480 bad = a crash. This issue behaves the same in the past. mozregression --good 2018-01-15 --bad 2018-03-06 --pref layers.acceleration.force-enabled:true gfx.webrender.enabled:true image.mem.shared:false gfx.webrender.blob-images:true gfx.webrender.hit-test:true gfx.webrender.async-scene-build:0 gfx.webrender.blob.invalidation:false startup.homepage_welcome_url:'https://www.adfc-tourenportal.de/map.php?&minimap=1&maptypecontrol=1&largenavigation=1&poi=1&viewbox=50.5964,6.08269,54.8396,10.0339&selectId=16782716385aac1ef0225d2&sid=tjr78raeoc0j55vunekjr60gd2&TrackobjectId=23944' > 25:40.50 INFO: Last good revision: 0561afeb9ca153847505cb3857ebda9d380205a2 > 25:40.50 INFO: First bad revision: b2804c6bec4c61bffd17cd1526877db79508f454 > 25:40.50 INFO: Pushlog: > https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=0561afeb9ca153847505cb3857ebda9d380205a2&tochange=b2804c6bec4c61bffd17cd1526877db79508f454 2018-02-06 > b2804c6bec4c Alexis Beingessner — Bug 1362115 - turn on blob-images by default with webrender. r=kats > cae458b810a8 Alexis Beingessner — Bug 1362115 - properly handle empty slices in FFI bindings. r=kats > 87b04696a381 Alexis Beingessner — Bug 1362115 - don't emit a blob-image for contentless videos. r=kats (Note that mozregression always had gfx.webrender.blob-images:true.) ----- mozregression --repo autoland --launch b2804c6bec4c61bffd17cd1526877db79508f454 -B debug --pref layers.acceleration.force-enabled:true gfx.webrender.enabled:true image.mem.shared:false gfx.webrender.blob-images:true gfx.webrender.hit-test:true gfx.webrender.async-scene-build:0 gfx.webrender.blob.invalidation:false startup.homepage_welcome_url:'https://www.adfc-tourenportal.de/map.php?&minimap=1&maptypecontrol=1&largenavigation=1&poi=1&viewbox=50.5964,6.08269,54.8396,10.0339&selectId=16782716385aac1ef0225d2&sid=tjr78raeoc0j55vunekjr60gd2&TrackobjectId=23944' > 0:51.62 INFO: [GFX3-]: Surface size too large (exceeds allocation limit)! > 0:51.62 INFO: [GFX3-]: Surface size too large (exceeds extent limit)! > 0:51.62 INFO: [GFX1-]: Failed to allocate a surface due to invalid size (DTD) Size(13330,11452) > 0:51.62 INFO: thread 'WRRenderBackend#1' panicked at 'Vector image error Unknown error', /builds/worker/workspace/build/src/gfx/webrender/src/resource_cache.rs:868:28 is the same reason reason as bp-5e7875ff-8e57-43a7-9d9d-2179b0180306 from comment 2. (Same with gfx.webrender.hit-test:false.) Please don't slap me. xD
Blocks: 1362115
Has Regression Range: --- → yes
Has STR: --- → yes
Keywords: regression
Summary: Crash on http://www.adfc-tourenportal.de/viewtrackobject.php?trackObjectID=23944 → blob-images: Crash on http://www.adfc-tourenportal.de/viewtrackobject.php?trackObjectID=23944
*crash reason
First bad: RUST_BACKTRACE=1 mozregression --repo autoland --launch b2804c6bec4c61bffd17cd1526877db79508f454 -B debug --pref layers.acceleration.force-enabled:true gfx.webrender.enabled:true image.mem.shared:false gfx.webrender.blob-images:true gfx.webrender.hit-test:false gfx.webrender.async-scene-build:0 gfx.webrender.blob.invalidation:false startup.homepage_welcome_url:'https://www.adfc-tourenportal.de/map.php?&minimap=1&maptypecontrol=1&largenavigation=1&poi=1&viewbox=50.5964,6.08269,54.8396,10.0339&selectId=16782716385aac1ef0225d2&sid=tjr78raeoc0j55vunekjr60gd2&TrackobjectId=23944' > 0:50.79 INFO: [GFX3-]: Surface size too large (exceeds allocation limit)! > 0:50.79 INFO: [GFX3-]: Surface size too large (exceeds extent limit)! > 0:50.79 INFO: [GFX1-]: Failed to allocate a surface due to invalid size (DTD) Size(14210,8906) > 0:50.79 INFO: thread 'WRRenderBackend#1' panicked at 'Vector image error Unknown error', /builds/worker/workspace/build/src/gfx/webrender/src/resource_cache.rs:868:28 > 0:51.51 INFO: --DOMWINDOW == 15 (0x7f3a1db29c00) [pid = 15564] [serial = 2] [outer = (nil)] [url = about:blank] > 0:51.51 INFO: --DOMWINDOW == 14 (0x7f3a07b7d800) [pid = 15564] [serial = 12] [outer = (nil)] [url = about:blank] > 0:51.51 INFO: --DOMWINDOW == 13 (0x7f3a1af0a800) [pid = 15564] [serial = 11] [outer = (nil)] [url = about:blank] > 0:51.51 INFO: --DOMWINDOW == 12 (0x7f3a07b82800) [pid = 15564] [serial = 15] [outer = (nil)] [url = about:blank] > 0:52.46 INFO: stack backtrace: > 0:52.46 INFO: 0: std::sys::imp::backtrace::tracing::imp::unwind_backtrace > 0:52.46 INFO: 1: std::sys_common::backtrace::_print > 0:52.46 INFO: 2: std::panicking::default_hook::{{closure}} > 0:52.46 INFO: 3: std::panicking::default_hook > 0:52.46 INFO: 4: std::panicking::rust_panic_with_hook > 0:52.46 INFO: 5: std::panicking::begin_panic > 0:52.46 INFO: 6: std::panicking::begin_panic_fmt > 0:52.46 INFO: 7: <unknown> > 0:52.46 INFO: Redirecting call to abort() to mozalloc_abort > 0:52.46 INFO: > 0:52.46 INFO: Hit MOZ_CRASH() at /builds/worker/workspace/build/src/memory/mozalloc/mozalloc_abort.cpp:33 Although this has the same crash reason as bp-5e7875ff-8e57-43a7-9d9d-2179b0180306, the crash stack looks different. Today: RUST_BACKTRACE=1 mozregression --launch 2018-04-23 -B debug --pref layers.acceleration.force-enabled:true gfx.webrender.enabled:true image.mem.shared:false gfx.webrender.blob-images:true gfx.webrender.hit-test:true gfx.webrender.async-scene-build:0 gfx.webrender.blob.invalidation:false startup.homepage_welcome_url:'https://www.adfc-tourenportal.de/map.php?&minimap=1&maptypecontrol=1&largenavigation=1&poi=1&viewbox=50.5964,6.08269,54.8396,10.0339&selectId=16782716385aac1ef0225d2&sid=tjr78raeoc0j55vunekjr60gd2&TrackobjectId=23944' > 1:01.92 INFO: [GFX3-]: Surface size too large (exceeds allocation limit)! > 1:01.92 INFO: [GFX3-]: Surface size too large (exceeds extent limit)! > 1:01.92 INFO: [GFX1-]: Failed to allocate a surface due to invalid size (DTD) Size(15554,9749) > 1:01.92 INFO: thread 'WRWorker#5' panicked at 'Moz2D replay problem', gfx/webrender_bindings/src/moz2d_renderer.rs:454:21 > 1:02.91 INFO: --DOMWINDOW == 6 (0x7fdff4d11400) [pid = 15161] [serial = 3] [outer = (nil)] [url = about:blank] > 1:03.30 INFO: stack backtrace: > 1:03.30 INFO: 0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace > 1:03.30 INFO: 1: std::sys_common::backtrace::_print > 1:03.30 INFO: 2: std::panicking::default_hook::{{closure}} > 1:03.30 INFO: 3: std::panicking::default_hook > 1:03.30 INFO: 4: std::panicking::rust_panic_with_hook > 1:03.30 INFO: 5: <unknown> > 1:03.30 INFO: Redirecting call to abort() to mozalloc_abort > 1:03.30 INFO: > 1:03.30 INFO: Hit MOZ_CRASH() at /builds/worker/workspace/build/src/memory/mozalloc/mozalloc_abort.cpp:34
Crash Signature: libxul.so@0x3cf2d40 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3cdfb28 | libxul.so@0x3cdfb18 | libxul.so@0x3ccf670 | rayon_core::job::{{impl}}::execute<T> ] → libxul.so@0x3cf2d40 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3cdfb28 | libxul.so@0x3cdfb18 | libxul.so@0x3ccf670 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | libxul.so@0x3cdf308 | libxul…
Crash Signature: libxul.so@0x3cdf2f8 | libxul.so@0x3ccee50 | rayon_core::job::{{impl}}::execute<T> ] → libxul.so@0x3cdf2f8 | libxul.so@0x3ccee50 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | rayon_core::job::{{impl}}::execute<T> ]
Crash Signature: libxul.so@0x3cdf2f8 | libxul.so@0x3ccee50 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | rayon_core::job::{{impl}}::execute<T> ] → libxul.so@0x3cdf2f8 | libxul.so@0x3ccee50 | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | rayon_core::job::{{impl}}::execute<T> ] [@ mozalloc_abort | abort | _$LT$rayon_core..job..HeapJob$LT$BODY$GT$$u20$as$u20$rayon_core..job..Jo…
No longer blocks: 1454187
Status: REOPENED → RESOLVED
Closed: 7 years ago7 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: