Opened 9 years ago

Last modified 2 years ago

#5386 assigned bug

Another SegFault in Tracker

Reported by: jahaiku Owned by: nobody
Priority: normal Milestone: R1
Component: Applications/Tracker Version: R1/Development
Keywords: Cc:
Blocked By: Blocking:
Has a Patch: no Platform: All

Description

This is a gcc4 build of hrev35431. I got this while shutting down haiku.

86: DEBUGGER: _numBlocks > 0
debug_server: Thread 86 entered the debugger: Debugger call: `_numBlocks > 0'
stack trace, current PC 0xffff0114  :
  (0x7ffee8e8)  0x5e8f8b  __assert_fail + 0x46
  (0x7ffee908)  0x6536db  _ZN8BPrivate11processHeap4freeEPv + 0x89
  (0x7ffee958)  0x6550d2  free + 0x154
  (0x7ffee978)  0x37639e  _ZN9entry_refD1Ev + 0x1e
  (0x7ffee998)  0x50ac7b  _ZN8BPrivate5ModelD1Ev + 0x77
  (0x7ffee9c8)  0x4c48ed  _ZN11BObjectListIN8BPrivate5ModelEE9MakeEmptyEb + 0x61
  (0x7ffee9f8)  0x4c49b0  _ZN11BObjectListIN8BPrivate5ModelEED1Ev + 0x34
  (0x7ffeea28)  0x513601  _ZN8BPrivate13NodePreloaderD0Ev + 0x59
  (0x7ffeea58)  0x56381f  _ZN8BPrivate8TTracker4QuitEv + 0x81
  (0x7ffeea98)  0x2a8e90  _ZN7BLooper14_QuitRequestedEP8BMessage + 0xec
  (0x7ffeeb18)  0x2a8ee2  _ZN7BLooper15DispatchMessageEP8BMessageP8BHandler + 0x46
  (0x7ffeeb38)  0x29fc00  _ZN12BApplication15DispatchMessageEP8BMessageP8BHandler + 0x416
  (0x7ffeed58)  0x2aaa3b  _ZN7BLooper11task_looperEv + 0x205
  (0x7ffeed98)  0x2a00e2  _ZN12BApplication3RunEv + 0x64
  (0x7ffeedc8)  0x200922  main + 0x36
  (0x7ffeef68)  0x2007cd  _start + 0x51
  (0x7ffeef9c)  0x105367  runtime_loader + 0x12f

Change History (4)

comment:1 Changed 9 years ago by anevilyak

Version: R1/alpha1R1/Development

Is this reproducible with a clean rebuild? I haven't seen this with any recent revision.

comment:2 in reply to:  1 Changed 9 years ago by jahaiku

Replying to anevilyak:

Is this reproducible with a clean rebuild? I haven't seen this with any recent revision.

No, I tried to reproduce another shutdown bug with media_addons_server and I got this instead of the other.

comment:3 Changed 9 years ago by mmlr

Not every debugger invokation is a segfault btw, this is actually an assert failing in the heap due to misuse.

I could imagine that this one is related/caused by the one reliably reproducible on quit when running Tracker with libroot_debug.

comment:4 Changed 2 years ago by axeld

Owner: changed from axeld to nobody
Status: newassigned
Note: See TracTickets for help on using tickets.