CrashKill/2014-02-03

Socorro

• Beta counting (b99) is buggy. selenamarie is working on untangling it (no bug yet, but coming)
• Shipped 72 last week - notable because it was shipped by devs (lars)
• Shipped again on Thursday (see email)
• Moving to having a shipping window every morning, and tagging once a week
• Why don't we calculate topcrashes across all versions?

Flash

• bug 774281 - F1398665248, doubled in volume in last few days. Related to weekend spike across branches?
  • See https://bugzilla.mozilla.org/show_bug.cgi?id=774281#c21
  • many comments about Facebook games crashing
  • As seen in Fx29 list of explosive reports

Desktop

• bug 530074 (German top crash) Some users responded to our mails and this appears to be caused by a program called Gutscheinfilter/G-Filter

Beta

• crash rate for 27 dropped below release on Jan 30. (related to b99 counting?)
• bug 966543
  • OMTC not on for 27
    • For OSX 10.7+ it was actually shipped on Fx25 and 10.6 is targeting Fx27 below are the tracking bugs
      • OSX :
        • Bug 756601 - Finish the OMTC implementation on OS X (meta bug)
        • Bug 894470 - Disable OMTC for Firefox 24 (Was enabled for Fx25 and up)
        • Bug 890997 - Enable OMTC on OSX 10.6 (Firefox 27)
      • Windows (WIP)
        • bug 899785
        • bug 899787

Aurora

• gfxContext::PushClipsToDT(mozilla::gfx::DrawTarget*) remains our #1 top crash, 2.6 3-day explosiveness rating
• __delayLoadHelper2 | ffi_closure_STDCALL is #2, 3-day explosiveness is 1.2

Trunk

• bug 965921 - #3 topcrash, reproducible, has been going up in the past few days.
  • Looks like it can be reproduced with AVG toolbar installation.

Mobile

• bug 958706 - Crash on JNI exceptions - should only happen on non-release builds
• bug 959237 - Define GetJNIEnv and GetJNIForThread as infallible

B2G

• Summary : need symbols from ZTE for OpenC on 1.3, Alcatel for OT-995, flatfish; dev who build their own builds also show up without symbols.
• Symbol missing on 1.4 : dev builds that hasn't had symbols pushed. Chances are these are builds that dev made themselves and flashed with drivers that may be old. ie they haven't repulled the OEM blob out from the latest blob.
  • https://crash-stats.mozilla.com/report/list?signature=android_atomic_add
  • https://crash-stats.mozilla.com/report/list?signature=libxul.so%400x5ba316%20%7C%20libxul.so%400x5ae375%20%7C%20libxul.so%400x174a172%20%7C%20libxul.so%400x5ae483%20%7C%20libxul.so%400x153a923%20%7C%20libxul.so%400x138419c%20%7C%20libnss3.so%400xfa3fe
  • LGE AOSP on Mako, WonderMedia WM-8880-MID
• Symbols missing on 1.3 : inari, buri, ZTE OPENC, geeksphone, Alcatel OT-995, flatfish, rockchip rk30sdk, , Sony R800i, Xony Xperia S,
• not enough data for 1.2 and below.

1.4

• bug 925819 - crash in js::jit::Assembler::bind(js::jit::Label*, js::jit::BufferOffset)
• bug 964328 - crash in mozilla::layers::DeprecatedContentClientRemoteBuffer::CreateAndAllocateDeprecatedTextureClient(mozilla::RefPtr<mozilla::layers::DeprecatedTextureClient>&)

1.3

• bug 871574 - crash in mozilla::dom::indexedDB::PIndexedDBRequestChild::OnMessageReceived
  • happens on 1.4 as well
• bug 967075 - crash in GrDrawTarget::getClip() const