Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

ace dumps core in mutex_lock

39 views
Skip to first unread message

pcaf...@iel.ie

unread,
Sep 1, 1999, 3:00:00 AM9/1/99
to
Hello

I'm looking at using ACE for the first time.

I'm trying to build on solaris 2.5 and 2.7 and am getting core dumps
when running the tests ie ./Basic_Types_Test
segv

compiler: g++ 2.95.1
linker /usr/ccs/bin/ld

libACE.so builds ok but when I try any of the tests they all core dump

The stack trace suggests that it's crashing in mutex_lock because
the parameter is invalid ie

#0 0xfef3a540 in mutex_lock () from /usr/lib/libthread.so.1
#1 0xff2c7cd8 in ACE_OS::mutex_lock (m=0x8)
at /export/home/pcaffrey/ACE_wrappers/ace/OS.i:1928
#2 0xff2c7760 in ACE_OS::thread_mutex_lock (m=0x8)
at /export/home/pcaffrey/ACE_wrappers/ace/OS.i:2222
#3 0xff2df0b4 in ACE_Thread_Mutex::acquire (this=0x8)
at /export/home/pcaffrey/ACE_wrappers/ace/Synch.i:568
#4 0xff200f68 in ACE_Guard<ACE_Thread_Mutex>::acquire (this=0xffbefac0)
at /export/home/pcaffrey/ACE_wrappers/ace/Synch_T.i:39
#5 0xff200fa4 in ACE_Guard<ACE_Thread_Mutex>::ACE_Guard
(this=0xffbefac0,
l=@0x8) at /export/home/pcaffrey/ACE_wrappers/ace/Synch_T.i:12
#6 0xff1f1e74 in ACE_Log_Msg::instance () at Log_Msg.cpp:178
#7 0x1ca70 in main () at Basic_Types_Test.cpp:69

Does this problem look familiar?

Thanks

Paul


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

Steve Huston

unread,
Sep 1, 1999, 3:00:00 AM9/1/99
to
Hi Paul,

> I'm looking at using ACE for the first time.

Welcome! For future reference, when you run into problems, please use
the PROBLEM-REPORT-FORM in the top-level ACE directory so we know
everything needed to help.

> I'm trying to build on solaris 2.5 and 2.7 and am getting core dumps
> when running the tests ie ./Basic_Types_Test
> segv
>
> compiler: g++ 2.95.1
> linker /usr/ccs/bin/ld

Ok... ACE 5.0 was not tested w/ gcc 2.95 - you might want to grab the
latest beta (http://www.cs.wustl.edu/~schmidt/ACE-obtain.html) and try
that - some work has been done with gcc 2.95 after ACE 5.0 was released.

> libACE.so builds ok but when I try any of the tests they all core dump
>
> The stack trace suggests that it's crashing in mutex_lock because
> the parameter is invalid ie
>
> #0 0xfef3a540 in mutex_lock () from /usr/lib/libthread.so.1
> #1 0xff2c7cd8 in ACE_OS::mutex_lock (m=0x8)
> at /export/home/pcaffrey/ACE_wrappers/ace/OS.i:1928
> #2 0xff2c7760 in ACE_OS::thread_mutex_lock (m=0x8)
> at /export/home/pcaffrey/ACE_wrappers/ace/OS.i:2222
> #3 0xff2df0b4 in ACE_Thread_Mutex::acquire (this=0x8)
> at /export/home/pcaffrey/ACE_wrappers/ace/Synch.i:568
> #4 0xff200f68 in ACE_Guard<ACE_Thread_Mutex>::acquire (this=0xffbefac0)
> at /export/home/pcaffrey/ACE_wrappers/ace/Synch_T.i:39
> #5 0xff200fa4 in ACE_Guard<ACE_Thread_Mutex>::ACE_Guard
> (this=0xffbefac0,
> l=@0x8) at /export/home/pcaffrey/ACE_wrappers/ace/Synch_T.i:12
> #6 0xff1f1e74 in ACE_Log_Msg::instance () at Log_Msg.cpp:178
> #7 0x1ca70 in main () at Basic_Types_Test.cpp:69
>
> Does this problem look familiar?

It looks like ACE wasn't initialized properly... please fill out the
whole PROBLEM-REPORT-FORM and then we'll know what we're looking at
here.

Regards,

-Steve

--
Steve Huston Riverace Corporation
Email: shu...@riverace.com http://www.riverace.com
Specializing in TCP/IP, CORBA, ACE (508) 541-9183, FAX 541-9185
Expertise to help your projects succeed We support ACE!

Exams Buzz

unread,
Oct 7, 2020, 6:47:38 AM10/7/20
to
Your post is too amazing. I have found with ease what I was looking for. Moreover, the content quality is awesome. Thanks for the nudge! www.ace4sure.com/ACE-questions.html
0 new messages