Closed
Bug 1122994
Opened 11 years ago
Closed 10 years ago
Intermittent browser_dbg_watch-expressions-02.js | application crashed [@ mozilla::image::SurfaceCacheImpl::StopTracking(mozilla::image::CachedSurface*)] after Assertion failure: mAvailableCost <= mMaxCost (More available cost than we started with)
Categories
(Core :: Graphics, defect)
Tracking
()
RESOLVED
WORKSFORME
People
(Reporter: philor, Unassigned, NeedInfo)
Details
(Keywords: assertion, intermittent-failure, Whiteboard: [gfx-noted])
No description provided.
| Comment hidden (Legacy TBPL/Treeherder Robot) |
Updated•11 years ago
|
Whiteboard: [gfx-noted]
Comment 3•11 years ago
|
||
I do. It means there's a bug in the accounting for ImageLib's SurfaceCache. We are probably somehow managing to remove the same surface from the cache twice or something similar.
Comment 4•11 years ago
|
||
For posterity, the crashing stack:
14:45:25 INFO - Assertion failure: mAvailableCost <= mMaxCost (More available cost than we started with), at /builds/slave/m-in-osx64-d-00000000000000000/build/src/image/src/SurfaceCache.cpp:427
14:45:25 INFO - #01: nsBaseHashtable<nsGenericHashKey<mozilla::image::SurfaceKey>, nsRefPtr<mozilla::image::CachedSurface>, mozilla::image::CachedSurface*>::s_EnumReadStub(PLDHashTable*, PLDHashEntryHdr*, unsigned int, void*) [xpcom/glue/nsBaseHashtable.h:391]
14:45:25 INFO - #02: PLDHashTable::Enumerate(PLDHashOperator (*)(PLDHashTable*, PLDHashEntryHdr*, unsigned int, void*), void*) [xpcom/glue/pldhash.cpp:727]
14:45:25 INFO - #03: mozilla::image::SurfaceCacheImpl::RemoveImage(mozilla::image::Image*) [xpcom/glue/nsTHashtable.h:172]
14:45:25 INFO - #04: mozilla::image::SurfaceCache::RemoveImage(mozilla::image::Image*) [xpcom/glue/Mutex.h:169]
14:45:25 INFO - #05: mozilla::image::RasterImage::~RasterImage() [xpcom/base/nsAutoPtr.h:74]
14:45:25 INFO - #06: mozilla::image::RasterImage::~RasterImage() [memory/mozalloc/mozalloc.h:233]
14:45:25 INFO - #07: mozilla::image::RasterImage::Release() [image/src/RasterImage.cpp:255]
14:45:25 INFO - #08: imgRequest::~imgRequest() [xpcom/base/nsRefPtr.h:59]
14:45:25 INFO - #09: imgRequest::~imgRequest() [memory/mozalloc/mozalloc.h:233]
14:45:25 INFO - #10: imgRequest::Release() [image/src/imgRequest.cpp:58]
14:45:25 INFO - #11: imgCacheEntry::Release() [memory/mozalloc/mozalloc.h:233]
14:45:25 INFO - #12: nsTArray_Impl<nsRefPtr<imgCacheEntry>, nsTArrayInfallibleAllocator>::RemoveElementsAt(unsigned long, unsigned long) [xpcom/glue/nsTArray.h:1700]
14:45:25 INFO - #13: imgLoader::EvictEntries(nsRefPtrHashtable<nsCStringHashKey, imgCacheEntry>&) [xpcom/glue/nsTArray-inl.h:21]
14:45:25 INFO - #14: imgLoader::~imgLoader() [image/src/imgLoader.cpp:1263]
14:45:25 INFO - #15: imgLoader::~imgLoader() [memory/mozalloc/mozalloc.h:233]
14:45:25 INFO - #16: imgLoader::Release() [image/src/imgLoader.cpp:1001]
14:45:25 INFO - #17: imgLoader::Shutdown() [image/src/imgLoader.cpp:1252]
14:45:25 INFO - #18: mozilla::image::ShutdownModule() [image/build/nsImageModule.cpp:109]
14:45:25 INFO - #19: LayoutModuleDtor [layout/build/nsLayoutModule.cpp:1312]
14:45:25 INFO - #20: nsTArray_Impl<nsAutoPtr<nsComponentManagerImpl::KnownModule>, nsTArrayInfallibleAllocator>::RemoveElementsAt(unsigned long, unsigned long) [xpcom/glue/nsCOMPtr.h:389]
14:45:25 INFO - #21: nsComponentManagerImpl::Shutdown() [xpcom/components/nsComponentManager.cpp:900]
14:45:25 INFO - #22: mozilla::ShutdownXPCOM(nsIServiceManager*) [xpcom/build/XPCOMInit.cpp:978]
14:45:25 INFO - #23: ScopedXPCOMStartup::~ScopedXPCOMStartup() [toolkit/xre/nsAppRunner.cpp:1332]
14:45:25 INFO - #24: XREMain::XRE_main(int, char**, nsXREAppData const*) [memory/mozalloc/mozalloc.h:233]
14:45:25 INFO - #25: XRE_main [toolkit/xre/nsAppRunner.cpp:4441]
14:45:25 INFO - #26: main [browser/app/nsBrowserApp.cpp:294]
14:45:29 INFO - TEST-INFO | Main app process: killed by SIGHUP
Looks like yet another shutdown-related crash.
Comment 5•10 years ago
|
||
Inactive; closing (see bug 1180138).
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → WORKSFORME
You need to log in
before you can comment on or make changes to this bug.
Description
•