Closed Bug 1222487 Opened 9 years ago Closed 9 years ago

Thunderbird crash in mozilla::gfx::UserData::Add when printing

Categories

(Core :: Graphics, defect)

Product:
Core
Component:
Graphics
Version:
38 Branch
Platform:
x86
Windows NT
Type:
defect
Priority:
Not set
Severity:
critical

Tracking

()

Status:
RESOLVED WORKSFORME
Tracking Flags:
Tracking Status
thunderbird_esr38 --- wontfix

People

(Reporter: wsmwk, Assigned: sotaro)

References

Details

(4 keywords, Whiteboard: [regression:TB38.3.0][gfx-noted][tbird topcrash])

Crash Data

Not a new crash [1], but has risen significantly in rankings. #8 for version 38.3.0. Was #15 in 38.2.0. All have printing on the stack and printing is often mentioned in user comments. bp-89d71f90-04e9-4a24-afd5-8a97a2151101. 0 xul.dll mozilla::gfx::UserData::Add(mozilla::gfx::UserDataKey*, void*, void (*)(void*)) gfx/2d/UserData.h 1 xul.dll nsDeviceContext::CreateRenderingContext() gfx/src/nsDeviceContext.cpp 2 xul.dll nsSimplePageSequenceFrame::PrintNextPage() layout/generic/nsSimplePageSequenceFrame.cpp 3 xul.dll nsPrintEngine::PrintPage(nsPrintObject*, bool&) layout/printing/nsPrintEngine.cpp 4 xul.dll nsPagePrintTimer::Run() layout/printing/nsPagePrintTimer.cpp Other examples, all with user comments and email addresses: bp-e09683d4-d99a-4994-96a6-475a42151101 bp-0f2137b0-8177-4c1a-b531-ea59b2151105 bp-b299b005-d630-4168-98aa-8a6432151104 bp-21b81ce3-395e-414f-b69a-d749b2151105 bp-1effa082-6b12-44ae-a43a-8215d2151104 [1] crash exists in version 31 https://crash-stats.mozilla.com/search/?signature=~mozilla%3A%3Agfx%3A%3AUserData%3A%3AAdd&product=Thunderbird&date=%3E2015-05-01&date=%3C2015-08-01&_facets=signature&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform#crash-reports Note mozilla::gfx::UserData::Add(mozilla::gfx::UserDataKey*, void*, void (*)(void*)) exists in TB 31.* and 38.2.0 but virtually gone in TB 38.3.0 - probably from changes in signature creation for crash-stats. But that should not quite so dramatic a rise in ranking. So I'm suggesting a regression is involved.
However, for firefox the crash pretty much does not exist in version 41 and newer https://crash-stats.mozilla.com/report/list?signature=mozilla%3A%3Agfx%3A%3AUserData%3A%3AAdd If it matters, note that hardware acceleration is disabled starting in TB38.2.0
status-thunderbird_esr38: --- → affected
tracking-thunderbird_esr38: --- → ?
Component: General → Graphics
Product: Thunderbird → Core
Version: unspecified → 38 Branch
Whiteboard: [regression:TB38.3.0] → [regression:TB38.3.0][gfx-noted]
We fail to create a DrawTarget here http://hg.mozilla.org/releases/mozilla-esr38/annotate/fc916052746/gfx/src/nsDeviceContext.cpp#l410 Since we are in the middle of printing I assume we need to do something more than null-check and early-return, because printing will be broken so we should at least warn the user or something. Jeff, do you know if there's a mechanism for handling errors during printing?
Flags: needinfo?(jmuizelaar)
Assignee: nobody → sotaro.ikeda.g
Flags: needinfo?(jmuizelaar)
Whiteboard: [regression:TB38.3.0][gfx-noted] → [regression:TB38.3.0][gfx-noted][tbird topcrash]
The call stack of the crash did not specific to the print. It was just normal rendering update. printing just seemed to trigger oom.
(In reply to Sotaro Ikeda [:sotaro] from comment #4) > The call stack of the crash did not specific to the print. It was just > normal rendering update. printing just seemed to trigger oom. The above was for the stacks in comment 3. Call stacks in comment 0 was different than comment 3. comment 0 was printing related crash as in comment 2. comment 3 is a different crash that might not be related to printing.
Blocks: 1257112
Bug 1257112 is created for crashes of comment 3.
No longer blocks: 1257112
See Also: → 1257112
Bug 1019063 already addressed the cause of the crash. It handles oom failure in nsDeviceContext::CreateRenderingContext(). Uplifting Bug 1019063 should address the crash.
Depends on: 1019063
From Bug 1019063 Comment 78, we could not uplift the fix to esr38.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → WONTFIX
I believe this signature is no longer relevant to Thunderbird 45. All 4 of those crashes are startup, and have different signatures than the THunderbird "printing" crashes. Either the crash is gone for Thunderbird or the signature changed (but I find not one TB45 crash so far whose comments mentions printing). Similarly, the rate for Thunderbird printing bug 1063723 has changed in 45, but not entirely gone.
status-thunderbird_esr38: affected → wontfix
tracking-thunderbird_esr38: ? → ---
Resolution: WONTFIX → WORKSFORME
Summary: crash in mozilla::gfx::UserData::Add when printing → Thunderbird crash in mozilla::gfx::UserData::Add when printing
You need to log in before you can comment on or make changes to this bug.