Bug 1061463 - [abrt] openocd: ioutil_Jim_Command_peek(): openocd killed by SIGSEGV
Summary: [abrt] openocd: ioutil_Jim_Command_peek(): openocd killed by SIGSEGV
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: openocd
Version: 20
Hardware: x86_64
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Dean Glazeski
QA Contact: Fedora Extras Quality Assurance
URL: https://retrace.fedoraproject.org/faf...
Whiteboard: abrt_hash:a812c3416e07637bae479376b46...
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-02-04 21:13 UTC by Josep
Modified: 2014-02-05 12:35 UTC (History)
4 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2014-02-05 10:14:56 UTC
Type: ---
Embargoed:

Attachments (Terms of Use)
File: backtrace (10.28 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: cgroup (172 bytes, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: core_backtrace (5.20 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: dso_list (1.29 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: environ (2.82 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: exploitable (82 bytes, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: limits (1.29 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: maps (7.59 KB, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: open_fds (639 bytes, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: proc_pid_status (934 bytes, text/plain)
2014-02-04 21:13 UTC, Josep 		no flags Details
File: var_log_messages (324 bytes, text/plain)
2014-02-04 21:14 UTC, Josep 		no flags Details
View All

Description Josep 2014-02-04 21:13:36 UTC 
Description of problem:
I was connected to the opencd server via telnet to por 4444 and I issued the command:
> peek 0x0

Version-Release number of selected component:
openocd-0.7.0-3.fc20

Additional info:
reporter:       libreport-2.1.12
backtrace_rating: 4
cmdline:        openocd -s /usr/share/openocd/scripts -f ../daemon.cfg -f interface/olimex-arm-usb-ocd-h.cfg -f board/olimex_stm32_h103.cfg
crash_function: ioutil_Jim_Command_peek
executable:     /usr/bin/openocd
kernel:         3.12.9-301.fc20.x86_64
runlevel:       N 5
type:           CCpp
uid:            1000

Truncated backtrace:
Thread no. 1 (10 frames)
 #0 ioutil_Jim_Command_peek at ioutil.c:421
 #1 JimInvokeCommand at jim.c:10074
 #2 Jim_EvalObj at jim.c:10519
 #3 Jim_EvalSource at jim.c:10838
 #4 command_run_line at command.c:659
 #5 telnet_input at telnet_server.c:391
 #6 server_loop at server.c:465
 #7 openocd_thread at openocd.c:295
 #8 openocd_main at openocd.c:328
 #10 _start
Comment 1 Josep 2014-02-04 21:13:41 UTC 
Created attachment 859374 [details]
File: backtrace
Comment 2 Josep 2014-02-04 21:13:43 UTC 
Created attachment 859375 [details]
File: cgroup
Comment 3 Josep 2014-02-04 21:13:45 UTC 
Created attachment 859376 [details]
File: core_backtrace
Comment 4 Josep 2014-02-04 21:13:47 UTC 
Created attachment 859377 [details]
File: dso_list
Comment 5 Josep 2014-02-04 21:13:49 UTC 
Created attachment 859378 [details]
File: environ
Comment 6 Josep 2014-02-04 21:13:51 UTC 
Created attachment 859379 [details]
File: exploitable
Comment 7 Josep 2014-02-04 21:13:53 UTC 
Created attachment 859380 [details]
File: limits
Comment 8 Josep 2014-02-04 21:13:55 UTC 
Created attachment 859381 [details]
File: maps
Comment 9 Josep 2014-02-04 21:13:57 UTC 
Created attachment 859382 [details]
File: open_fds
Comment 10 Josep 2014-02-04 21:13:58 UTC 
Created attachment 859383 [details]
File: proc_pid_status
Comment 11 Josep 2014-02-04 21:14:00 UTC 
Created attachment 859384 [details]
File: var_log_messages
Comment 12 Markus Mayer 2014-02-05 10:14:56 UTC 
Hi Josep,

I'm not that familiar with openocd, so take my comment with a grain of salt.

'peek' reads from the memory address given by the first parameter. It reads the memory of the machine openocd is running (that's your PC). It doesn't read the memory of the debugged device.

So 'peek 0x0' will read/dereference a null pointer, which will cause your x64-cpu to emit a segmentation fault. That's by design!

I'm closing this as 'not a bug', but if you disagree fell free to reopen it.


regards
   Markus
Comment 13 Josep 2014-02-05 12:35:24 UTC 
Hi Markus,
I didn't know peek looked at the memory on the machine where openocd was running, that seems an odd feature, but why not...
Closing seems the right thing to do, thanks anyway.
Note You need to log in before you can comment on or make changes to this bug.