HANG (1.5.0.1 openoffice.org)

[derekbruening]

From [email protected] on January 24, 2010 18:17:38

launch openoffice: bin64/drdeploy -ops "-pause_on_error" -debug -client
samples/bin64/libempty.so 0x1 "" openoffice.org
<Starting application sh (7513)>
<Starting application sh (7522)>
<Starting application stat (7524)>
<Stopping application stat (7524)>
<Starting application uname (7526)>
<Stopping application uname (7526)>
<Starting application uname (7528)>
<Stopping application uname (7528)>
<Stopping application sh (7530)>
Segmentation fault (core dumped)
<Starting application dirname (7532)>
<Stopping application dirname (7532)>
<Stopping application sh (7534)>
Segmentation fault
<Starting application basename (7535)>
<Stopping application basename (7535)>
<Starting application javaldx (7537)>
<Stopping application javaldx (7537)>
<Starting application pagein (7549)>
<Stopping application pagein (7549)>
<Starting application oosplash.bin (7551)>

after 40 minutes of cpu time, here's a backtrace of oosplash.bin (pid
7557):

---

#0 0x000000007116e89c in syscall_ready ()
from /home/david/Applications/Source/DynamoRIO-
#1 0x0000000077532680 in ?? ()
#2 0x0000000071178e45 in thread_yield ()
at /work/dr/tot/opensource/core/linux/os.c:2024
#3 0x000000007117ac35 in mutex_wait_contended_lock (lock=0x775325a8)
at /work/dr/tot/opensource/core/linux/os.c:6256
#4 0x00000000710c9fa3 in mutex_lock (lock=0x775325a8)
at /work/dr/tot/opensource/core/utils.c:876
#5 0x000000007118a749 in signal_thread_exit (dcontext=0x77527700)
at /work/dr/tot/opensource/core/linux/signal.c:1260
#6 0x000000007117f2a0 in os_thread_exit (dcontext=0x77527700)
at /work/dr/tot/opensource/core/linux/os.c:1491
#7 0x000000007104197c in dynamo_thread_exit_common (dcontext=0x77527700,
id=7551, other_thread=1) at /work/dr/tot/opensource/core/dynamo.c:2140
#8 0x0000000071041d5b in dynamo_other_thread_exit (tr=0x774fa168)
at /work/dr/tot/opensource/core/dynamo.c:2242
#9 0x0000000071043178 in dynamorio_fork_init (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dynamo.c:731
#10 0x000000007117e0dd in post_system_call (dcontext=0x77532680)
at /work/dr/tot/opensource/core/linux/os.c:4566
#11 0x000000007109654b in handle_post_system_call (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:1832
#12 0x000000007109a1a0 in dispatch_enter_dynamorio (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:708
#13 0x000000007109a31e in dispatch (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:140
#14 0x000000007750ed59 in ?? ()
#15 0x0000000000000000 in ?? ()

and the second one:

---

#0 0x000000007116e89c in syscall_ready ()
from /home/david/Applications/Source/DynamoRIO-
#1 0x0000000077532680 in ?? ()
#2 0x0000000071178e45 in thread_yield ()
at /work/dr/tot/opensource/core/linux/os.c:2024
#3 0x000000007117ac35 in mutex_wait_contended_lock (lock=0x775325a8)
at /work/dr/tot/opensource/core/linux/os.c:6256
#4 0x00000000710c9fa3 in mutex_lock (lock=0x775325a8)
at /work/dr/tot/opensource/core/utils.c:876
#5 0x000000007118a749 in signal_thread_exit (dcontext=0x77527700)
at /work/dr/tot/opensource/core/linux/signal.c:1260
#6 0x000000007117f2a0 in os_thread_exit (dcontext=0x77527700)
at /work/dr/tot/opensource/core/linux/os.c:1491
#7 0x000000007104197c in dynamo_thread_exit_common (dcontext=0x77527700,
id=7551, other_thread=1) at /work/dr/tot/opensource/core/dynamo.c:2140
#8 0x0000000071041d5b in dynamo_other_thread_exit (tr=0x774fa168)
at /work/dr/tot/opensource/core/dynamo.c:2242
#9 0x0000000071043178 in dynamorio_fork_init (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dynamo.c:731
#10 0x000000007117e0dd in post_system_call (dcontext=0x77532680)
at /work/dr/tot/opensource/core/linux/os.c:4566
#11 0x000000007109654b in handle_post_system_call (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:1832
#12 0x000000007109a1a0 in dispatch_enter_dynamorio (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:708
#13 0x000000007109a31e in dispatch (dcontext=0x77532680)
at /work/dr/tot/opensource/core/dispatch.c:140
#14 0x000000007750ed59 in ?? ()
#15 0x0000000000000000 in ?? ()

i tried to kill the process with Ctrl+C:
^C<Stopping application oosplash.bin (7551)>
<Stopping application sh (7522)>
<Stopping application sh (7513)>
oosplash.bin kept on running (it has pid 7557) and only a "kill -9" would
bring it down.

tested this on 2 systems running the same debian linux. see the other
issue for more details: issue #252

Original issue: http://code.google.com/p/dynamorio/issues/detail?id=253

[derekbruening]

From [email protected] on January 26, 2010 08:45:17

My call back trace is a little bit different from the posted.
The first one is the same, which seems the DR is busy waiting on a lock.
While the second one is different. My gdb tells me that it stops at the system call
in the code cache.

[derekbruening]

From [email protected] on February 21, 2010 12:57:26

looks like issue #239 to me based on quick glance
are the locks held by threads in the parent at fork time?