Seen while reviewing crash stat reports. https://crash-stats.mozilla.com/report/list?signature=SocketWrite - showing up more frequently on trunk lately but shows up on other branches as well. https://crash-stats.mozilla.com/report/index/b0f2c0d6-ebec-4e46-a51e-97b042110606 Frame Module Signature [Expand] Source 0 @0x88bcc30 1 nspr4.dll SocketWrite nsprpub/pr/src/io/prsocket.c:701 2 nspr4.dll PR_SetPollableEvent nsprpub/pr/src/io/prpolevt.c:232 3 xul.dll nsSocketTransportService::OnDispatchedEvent netwerk/base/src/nsSocketTransportService2.cpp:588 4 xul.dll nsThread::Dispatch xpcom/threads/nsThread.cpp:422 5 xul.dll nsSocketTransportService::Dispatch netwerk/base/src/nsSocketTransportService2.cpp:140 6 xul.dll nsHttpConnectionMgr::PostEvent netwerk/protocol/http/nsHttpConnectionMgr.cpp:207 7 xul.dll nsHttpConnectionMgr::AddTransaction netwerk/protocol/http/nsHttpConnectionMgr.cpp:275 8 xul.dll nsHttpChannel::Connect netwerk/protocol/http/nsHttpChannel.cpp:325 9 xul.dll nsHttpChannel::OnNormalCacheEntryAvailable netwerk/protocol/http/nsHttpChannel.cpp:2294 10 xul.dll nsHttpChannel::OnCacheEntryAvailable netwerk/protocol/http/nsHttpChannel.cpp:4711 11 xul.dll nsCacheListenerEvent::Run netwerk/cache/nsCacheService.cpp:1462 12 xul.dll nsThread::ProcessNextEvent xpcom/threads/nsThread.cpp:618 13 xul.dll mozilla::ipc::MessagePump::Run ipc/glue/MessagePump.cpp:110 14 xul.dll xul.dll@0xb65627 15 xul.dll MessageLoop::RunHandler ipc/chromium/src/base/message_loop.cc:202 16 xul.dll xul.dll@0x37209f 17 xul.dll MessageLoop::Run ipc/chromium/src/base/message_loop.cc:176 18 xul.dll gfxTextRun::SetPotentialLineBreaks gfx/thebes/gfxFont.cpp:3138 19 xul.dll nsBaseAppShell::Run widget/src/xpwidgets/nsBaseAppShell.cpp:189 20 @0x750fffff 21 userenv.dll DeleteLinkFileWorker 22 icm32.dll CheckOutputColorSpaceChk 23 pshook.dll pshook.dll@0x736d 24 advapi32.dll PcwCollectData 25 xul.dll combine_xor_ca obj-firefox/gfx/cairo/libpixman/src/pixman-combine.c.template:1884 26 @0x642e3232

83 crashes in the last week in this stack on the trunk.

SocketWrite seems to have been around for a while in general, but it rose up on the trunk with the 05/27 build id. Looking at the "Table" tab (which is per build id date) in https://crash-stats.mozilla.com/report/list?query_search=signature&query_type=contains&reason_type=contains&range_value=4&range_unit=weeks&hang_type=any&process_type=any&signature=SocketWrite and ignoring 2011042000 and 2011041300 (4.0.1 and 3.6.17), there's a sudden rise on 05/27 with having crashes for every build since then - that smells like a trunk regression.

It might be a dupe of bug 422044.

hope my firefox dosen't crash

This is a going concern - ~120 active crash-stats reports and they are pretty interesting. Each one is a calling PR_SetPollableEvent on the STS mThreadEvent on an unusual thread (i.e. its not the main thread or the socket thread) and is doing so after a timer expiration. At first I assumed shutdown of some sort, but the socket thread and main thread are clearly not going through shutdown - they're in the normal active states. ideas dragana or daniel? Seems awfully related to the other work on mThreadEvent we've been looking at - maybe this is more actionable - Even though this is hard to verify (there is only one crash before the beta audience) it seems worth working on. https://crash-stats.mozilla.com/signature/?signature=SocketWrite+|+PR_SetPollableEvent+|+nsSocketTransportService%3A%3AOnDispatchedEvent&_columns=date&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=reason&_columns=address&page=1#reports https://crash-stats.mozilla.com/report/index/5f591562-ecef-4bfe-812d-4274f2160205#allthreads https://crash-stats.mozilla.com/report/index/0985f0ad-a159-44b4-a705-396872160204#allthreads https://crash-stats.mozilla.com/report/index/bb7be556-2442-459a-bb92-50c402160207#allthreads

fixed by bug 698882