Hi,
Sorry if this has asked and answered, but I have not been able to find this
issue in the archives.
I'm running a trivial multithreaded program on a Redhat 5.1 linux system,
kernel 2.6.18, and I'm seeing some unexpected behavior with gdbserver. Here's
the source:
#include <stdio.h>
#include <stdlib.h>
#include <pthread.h>
void *print_message_function( void *ptr );
main()
{
pthread_t thread1, thread2;
char *message1 = "Thread 1";
char *message2 = "Thread 2";
int iret1, iret2;
/* Create independent threads each of which will execute function */
iret1 = pthread_create( &thread1, NULL, print_message_function, (void*)
message1);
iret2 = pthread_create( &thread2, NULL, print_message_function, (void*)
message2);
/* Wait till threads are complete before main continues. Unless we */
/* wait we run the risk of executing an exit which will terminate */
/* the process and all threads before the threads have completed. */
pthread_join( thread1, NULL);
pthread_join( thread2, NULL);
printf("Thread 1 returns: %d\n",iret1);
printf("Thread 2 returns: %d\n",iret2);
exit(0);
}
void *print_message_function( void *ptr )
{
char *message;
message = (char *) ptr;
printf("%s \n", message);
}
Under gdb 6.8 it all seems to work fine-in particular new threads are reported
as they are created and "info threads" works:
[jbradsha-linux](jbradsha)294$ /export/jbradsha/workdir/gdb/gdb-6.8/gdb/gdb
pthread
GNU gdb 6.8
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-pc-linux-gnu"...
(gdb) b pthread.c:10
Breakpoint 1 at 0x80484a5: file pthread.c, line 10.
(gdb) n
The program is not being run.
(gdb) c
The program is not being run.
(gdb) r
Starting program: /home/rtrsvcs/jbradsha/sandbox/pthread
[Thread debugging using libthread_db enabled]
[New Thread 0xb7fd26c0 (LWP 13381)]
[Switching to Thread 0xb7fd26c0 (LWP 13381)]
Breakpoint 1, main () at pthread.c:10
10 char *message1 = "Thread 1";
(gdb) info threads
* 1 Thread 0xb7fd26c0 (LWP 13381) main () at pthread.c:10
(gdb) n
11 char *message2 = "Thread 2";
(gdb) n
16 iret1 = pthread_create( &thread1, NULL, print_message_function,
(void*) message1);
(gdb) info threads
* 1 Thread 0xb7fd26c0 (LWP 13381) main () at pthread.c:16
(gdb) n
[New Thread 0xb7fd1b90 (LWP 13386)]
17 iret2 = pthread_create( &thread2, NULL, print_message_function,
(void*) message2);
(gdb) info threads
2 Thread 0xb7fd1b90 (LWP 13386) 0x0081afc8 in clone () from /lib/libc.so.6
* 1 Thread 0xb7fd26c0 (LWP 13381) main () at pthread.c:17
(gdb) c
Continuing.
Thread 1
[New Thread 0xb75d0b90 (LWP 13400)]
[Thread 0xb7fd1b90 (LWP 13386) exited]
Thread 2
[Thread 0xb75d0b90 (LWP 13400) exited]
Thread 1 returns: 0
Thread 2 returns: 0
Program exited normally.
(gdb)
Now when I run the program under gdbserver and connect with gdb (both on the
localhost) I find that threads don't seem to be reported correctly to gdb:
(gdb) target remote localhost:4500
`system-supplied DSO at 0xcb0000' has disappeared; keeping its symbols.
Remote debugging using localhost:4500
[New Thread 13460]
0x0072c810 in _start () from /lib/ld-linux.so.2
(gdb) c
Continuing.
Breakpoint 1, main () at pthread.c:10
10 char *message1 = "Thread 1";
(gdb) info threads
* 1 Thread 13460 main () at pthread.c:10
(gdb) n
11 char *message2 = "Thread 2";
(gdb) n
16 iret1 = pthread_create( &thread1, NULL, print_message_function,
(void*) message1);
(gdb) n
17 iret2 = pthread_create( &thread2, NULL, print_message_function,
(void*) message2);
(gdb) n
23 pthread_join( thread1, NULL);
(gdb) info threads
* 1 Thread 13460 main () at pthread.c:23
(gdb) c
Continuing.
Program exited normally.
(gdb)
However, the console output clearly indicates that the program runs correctly:
[jbradsha-linux](jbradsha)257$
/export/jbradsha/workdir/gdb/gdb-6.8/gdb/gdbserver/gdbserver :4500 pthread
Process pthread created; pid = 13460
Listening on port 4500
Remote debugging from host 127.0.0.1
Thread 1
Thread 2
Thread 1 returns: 0
Thread 2 returns: 0
Child exited with retcode = 0
Child exited with status 0
GDBserver exiting
[jbradsha-linux](jbradsha)258$
Here are the shared libraries I'm using with gdb/gdbserver:
[jbradsha-linux](jbradsha)295$ ldd /export/jbradsha/workdir/gdb/gdb-6.8/gdb/gdb
linux-gate.so.1 => (0x00c55000)
libncurses.so.5 => /usr/lib/libncurses.so.5 (0x02db3000)
libm.so.6 => /lib/libm.so.6 (0x0088f000)
libexpat.so.0 => /lib/libexpat.so.0 (0x00a92000)
libdl.so.2 => /lib/libdl.so.2 (0x008b8000)
libc.so.6 => /lib/libc.so.6 (0x0074d000)
/lib/ld-linux.so.2 (0x0072c000)
[jbradsha-linux](jbradsha)259$ ldd
/export/jbradsha/workdir/gdb/gdb-6.8/gdb/gdbserver/gdbserver
linux-gate.so.1 => (0x00126000)
libthread_db.so.1 => /lib/libthread_db.so.1 (0x0088f000)
libc.so.6 => /lib/libc.so.6 (0x0074d000)
/lib/ld-linux.so.2 (0x0072c000)
Can someone tell me if this is a gdbserver bug or just an error on my part,
and, if so, how to fix it so that gdbserver correctly reports thread
information to gdb?
Thanks very much.
Ted Bradshaw
