Conditional data watchpoint doesn't work in ARM GDB - c

My intent is to be able to catch when a global variable has some exact value. GDB has data watchpoints according to which this can be implemented.
Consider this simple program written for x86 Linux:
int myVar = 0;
void debug_watchpoints() {
for(int i=0; i < 2000; i++) {
myVar++;
}
}
int main() {
debug_watchpoints();
return 0;
}
Compiling the program with
gcc -o main -ggdb3 -Og main.c
And starting debugging with GDB:
max#PC-LT-23:~/stackoverflow$ gdb ./main
GNU gdb (Ubuntu 8.3-0ubuntu1) 8.3
Copyright (C) 2019 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 "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./main...
(gdb) b main
Breakpoint 1 at 0x1146: file main.c, line 9.
(gdb) start
Temporary breakpoint 2 at 0x1146: file main.c, line 9.
Starting program: /home/max/stackoverflow/main
Breakpoint 1, main () at main.c:9
9 int main() {
(gdb) watch myVar if myVar==1337
Hardware watchpoint 3: myVar
(gdb) continue
Continuing.
Hardware watchpoint 3: myVar
Old value = 1336
New value = 1337
debug_watchpoints () at main.c:4
4 for(int i=0; i < 2000; i++) {
(gdb)
As you can see it halted the program at exactly the point in time when the variable was set to 1337.
Consider the exact same program, compiled with arm-none-eabi-gcc for a STM32L476RG microcontroller which has a Cortex-M4F core. The used IDE here is System Workbench for STM32 (aka Eclipse) with a project generated by STM32CubeMX.
Now launching openocd gives
Open On-Chip Debugger 0.10.0+dev-00021-g524e8c8 (2019-04-12-08:33)
Licensed under GNU GPL v2
For bug reports, read
http://openocd.org/doc/doxygen/bugs.html
srst_only separate srst_nogate srst_open_drain connect_assert_srst
Info : The selected transport took over low-level target control. The results might differ compared to plain JTAG/SWD
padded zone erase set to 1
adapter speed: 8000 kHz
adapter_nsrst_delay: 100
Info : Listening on port 6666 for tcl connections
Info : Listening on port 4444 for telnet connections
Info : clock speed 8000 kHz
Info : STLINK v2.1 JTAG v34 API v2 M25 VID 0x0483 PID 0x374B
Info : using stlink api v2
Info : Target voltage: 0.011074
Error: target voltage may be too low for reliable debugging
Info : Unable to match requested speed 8000 kHz, using 4000 kHz
Info : Stlink adapter speed set to 4000 kHz
Info : STM32L476RGTx.cpu: hardware has 6 breakpoints, 4 watchpoints
Info : Listening on port 3333 for gdb connections
Info : accepting 'gdb' connection on tcp/3333
target halted due to debug-request, current mode: Thread
xPSR: 0x01000000 pc: 0x08001340 msp: 0x20018000
A breakpoint is set in the main function and then the watchpoint is setup exactly like before. Also, a breakpoint is is set after the execution of the debug_watchpoints() function.
GNU gdb (GNU Tools for Arm Embedded Processors 9-2019-q4-major) 8.3.0.20190709-git
Copyright (C) 2019 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 "--host=x86_64-linux-gnu --target=arm-none-eabi".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
(gdb) Reset_Handler () at ../startup/startup_stm32l476xx.s:63
63 ldr sp, =_estack /* Set stack pointer */
Temporary breakpoint 3, main () at ../Src/main.c:65
65 {
(gdb) watch myVar if myVar==1337
Hardware watchpoint 4: myVar
(gdb) info breakpoints
Num Type Disp Enb Address What
2 breakpoint keep y 0x08000f46 in main at ../Src/main.c:70
4 hw watchpoint keep y myVar
stop only if myVar==1337
(gdb)
When continuing the program, it now stops at every modification of the variable with a SIGTRAP, regardless of whether the condition was met or not.
Program received signal SIGTRAP, Trace/breakpoint trap.
0x08000ec2 in debug_watchpoints () at ../Src/main.c:54
54 for(int i=0; i < 2000; i++) {
(gdb) continue
Continuing.
Program received signal SIGTRAP, Trace/breakpoint trap.
0x08000ec2 in debug_watchpoints () at ../Src/main.c:54
54 for(int i=0; i < 2000; i++) {
(gdb) continue
Continuing.
Program received signal SIGTRAP, Trace/breakpoint trap.
0x08000ec2 in debug_watchpoints () at ../Src/main.c:54
54 for(int i=0; i < 2000; i++) {
(gdb) info breakpoint
Num Type Disp Enb Address What
2 breakpoint keep y 0x08000f46 in main at ../Src/main.c:70
4 hw watchpoint keep y myVar
stop only if myVar==1337
(gdb) print myVar
$2 = 3
I can continue for as many times as I want, it just breaks each and every time the variable is changed.
In my scenario of "debugging a memory corruption on the stack", I really need GDB to evaluate the condition correctly, otherwise the program stops a thousand times or more (every time a variable which happens to be at this memory location is changed) and not at only the specific time a specific value is written into it to catch a bug.
Why does arm-none-eabi-gdb differ here in its behavior from normal gdb? Could the error lie in the Cortex-M4 hardware debugging capabilities, arm-none-eabi-gdb, or openocd as the GDB server?

As said, there's no working implementation in openocd, the debugger bridge you are using.
You can set by sw. Here a small snippet. COMP2, FUNCTION2, COMP1 and FUNCTION1 are DWT registers. Please note that you need to use two comparators linked together. Not all Cortex-m implementations support these features and not all comparators are linkable, it depends on your silicon. Also it seems that the capture is asynchonous. Generally my PC stops a few istruction after the trigger.
DWT->COMP2 = <address to compare>;
DWT->FUNCTION2 = 0;
DWT->COMP1 = <word to compare>;
DWT->FUNCTION1 = 0x20B06;

None of the software I know does it out of the box. When |I use the DWT usually I set it in my software. I have not found any way to program it via any variant of gdb (except manual setting the registers - but it is too annoying)

Related

examine function argument using GDB

I am new to GDB, and need to examine content of function using gdb. The program need to be debugged is perf and function name is ___fprintf_chk() , and tried some thing like:
perf stat -I 1000 -e branch-misses
time counts unit events
1.000222090 1746 branch-misses
2.000486444 1986 branch-misses
3.000712797 1783 branch-misses
sudo ./gdb --pid $(pidof perf)
GNU gdb (GDB) 10.2
Copyright (C) 2021 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 "aarch64-none-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<https://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
Attaching to process 38977
Reading symbols from /usr/bin/perf...
warning: Unable to determine the number of hardware watchpoints available.
warning: Unable to determine the number of hardware breakpoints available.
Reading symbols from /lib/aarch64-linux-gnu/libpthread.so.0...
(No debugging symbols found in /lib/aarch64-linux-gnu/libpthread.so.0)
Error while reading shared library symbols for /lib/aarch64-linux-gnu/libpthread.so.0:
Cannot find user-level thread for LWP 38977: generic error
Reading symbols from /lib/aarch64-linux-gnu/librt.so.1...
(No debugging symbols found in /lib/aarch64-linux-gnu/librt.so.1)
Reading symbols from /lib/aarch64-linux-gnu/libm.so.6...
(No debugging symbols found in /lib/aarch64-linux-gnu/libm.so.6)
Reading symbols from /lib/aarch64-linux-gnu/libdl.so.2...
(No debugging symbols found in /lib/aarch64-linux-gnu/libdl.so.2)
Reading symbols from /lib/libopencsd_c_api.so.1...
--Type <RET> for more, q to quit, c to continue without paging--q
Quit
(gdb) b __fprintf_chk
Breakpoint 1 at 0xaaaae55b2614
(gdb) c
Continuing.
Breakpoint 1, 0x0000aaaae55b2614 in __fprintf_chk#plt ()
(gdb) info args
No symbol table info available.
Now, need to know when break point is put at __fprintf_chk, does it on __fprintf_chk that is part of perf binary or from other shared objects (.so files, as I see Reading symbols from /lib/aarch64-linux-gnu/libpthread.so.0...)
Also, info args gives nothing, does it mean perf need to be compiled with debugging info (how)?

GDB quits when running the 'finish' command

I downloaded mingw-w64 so I can work on writing programs and debugging them from my home laptop, but for whatever reason, when ever I'm debugging a program with gdb and I wish to exit function using 'finish' command, gdb simply runs until it returns from function and simply exits and I return to the shell's prompt. I have no idea what is causing this bug, and it is especially annoying and make debugging very hard because it forces me to go through every line in a function before exiting (God forbids there's a loop). Has someone encountered this bug before? Is there a solution?
PS C:\Users\gamef\OneDrive\Desktop\CPE453\program4_tinyfs> gdb .\a.exe
GNU gdb (GDB) 8.1
Copyright (C) 2018 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 "x86_64-w64-mingw32".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from .\a.exe...done.
(gdb) br writeData
Breakpoint 1 at 0x402e44: file TinyFS.c, line 299.
(gdb) run
Starting program: C:\Users\gamef\OneDrive\Desktop\CPE453\program4_tinyfs\a.exe
[New Thread 40396.0x1cc8]
[New Thread 40396.0x6e30]
[New Thread 40396.0x3090]
[New Thread 40396.0x9e38]
Thread 1 hit Breakpoint 1, writeData (begBlock=0, data=0x61fab2 "\001_", 'ÿ' <repeats 31 times>, size=254,
type=DIRECTORY) at TinyFS.c:299
299 Byte buff[BLOCKSIZE] = {0};
(gdb) s
300 buff[MAGIC_OFFSET] = MAGIC_NUMBER;
(gdb) finish
Run till exit from #0 writeData (begBlock=0, data=0x61fab2 "\001_", 'ÿ' <repeats 31 times>, size=254, type=DIRECTORY)
at TinyFS.c:300
PS C:\Users\gamef\OneDrive\Desktop\CPE453\program4_tinyfs>
It seems that this problem is a caused by a bug inside the 8.1 version of GDB. So I reverted back to GDB 8.0 which seems to work perfectly.

Debug when thread is stuck in syscall 32bit program on 64 bit host

I have a problem where one or more threads lock each other. I dont know what going on there. The debugger cannot break (thread 1), breaks but cannot get a backtrace (thread 2+5) or shows the backtrace (thread 3)
Gdb native shows the same.
I learned that this is case because libc imlements this in assembler an gdb cannot walt the stack correctly. Sometimes (i dont know when), i can do a few steps in the assembly, then i see the backtrace.
I just tried a x64 program and it works.
See my sample code:
#include <time.h>
int main()
{
while(1)
{
struct timespec ts;
ts.tv_sec = 1;
ts.tv_nsec = 0;
clock_nanosleep(CLOCK_MONOTONIC, 0, &ts, 0);
}
return 1;
}
gdb output 32 bit:
vagrant#PC41388-spvm-4650:/tmp$ gdb main32
GNU gdb (Ubuntu
7.7.1-0ubuntu5~14.04.2) 7.7.1 Copyright (C) 2014 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 "x86_64-linux-gnu". Type "show
configuration" for configuration details. For bug reporting
instructions, please see: http://www.gnu.org/software/gdb/bugs/.
Find the GDB manual and other documentation resources online at:
http://www.gnu.org/software/gdb/documentation/. For help, type
"help". Type "apropos word" to search for commands related to
"word"... Reading symbols from main32...(no debugging symbols
found)...done.
(gdb) r Starting program: /tmp/main32 [Thread
debugging using libthread_db enabled] Using host libthread_db library
"/lib/x86_64-linux-gnu/libthread_db.so.1". ^C Program received signal
SIGINT, Interrupt. 0x55579cd9 in ?? () (gdb) bt
#0 0x55579cd9 in ?? ()
#1 0x555b0af3 in __libc_start_main (main=0x80484dd , argc=1,
argv=0xffffcee4, init=0x8048520 <__libc_csu_init>,
fini=0x8048590 <__libc_csu_fini>, rtld_fini=0x55564160 <_dl_fini>,
stack_end=0xffffcedc) at libc-start.c:287
#2 0x08048401 in _start () (gdb)
gdb output 64 bit:
vagrant#PC41388-spvm-4650:/tmp$ gdb main64
GNU gdb (Ubuntu
7.7.1-0ubuntu5~14.04.2) 7.7.1 Copyright (C) 2014 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 "x86_64-linux-gnu". Type "show
configuration" for configuration details. For bug reporting
instructions, please see: http://www.gnu.org/software/gdb/bugs/.
Find the GDB manual and other documentation resources online at:
http://www.gnu.org/software/gdb/documentation/. For help, type
"help". Type "apropos word" to search for commands related to
"word"... Reading symbols from main64...(no debugging symbols
found)...done.
(gdb) r Starting program: /tmp/main64 [Thread
debugging using libthread_db enabled] Using host libthread_db library
"/lib/x86_64-linux-gnu/libthread_db.so.1". b ^C Program received
signal SIGINT, Interrupt. 0x00002aaaaafe092a in __clock_nanosleep
(clock_id=1, flags=0,
req=0x7fffffffdc10, rem=0x2aaaaafe092a <__clock_nanosleep+58>)
at ../sysdeps/unix/sysv/linux/clock_nanosleep.c:41 41 ../sysdeps/unix/sysv/linux/clock_nanosleep.c: No such file or
directory.
(gdb) bt
#0 0x00002aaaaafe092a in __clock_nanosleep (clock_id=1, flags=0,
req=0x7fffffffdc10, rem=0x2aaaaafe092a <__clock_nanosleep+58>)
at ../sysdeps/unix/sysv/linux/clock_nanosleep.c:41
#1 0x0000000000400630 in main () (gdb)
set architecture i386 does not help either.
More news: info proc mapp shows the x32 app is in [vvar] whereas the x64 app is at libc. This would explain why gdb cant find the backtrace.
So my question is: Is there a different version of the libc, where this works? I am using ubuntu14.04.
I updated to a newer gdb version (currently the latest, 7.12.1). This fixed the problem.
Note that gbd:i386 did not work either on lubuntu x64, whereas it worked fine under lubuntu x32. Also note that both main32 and libc are binary identical on lubuntu x64 and x32.

Debugger in C::B. Can't open cygwin.S

Hi I just discovered a quite weird behaviour of the debugger when declaring a simple two-dimensional array. It looks like it can't open a file cygwin.S in the library.
Cannot open file: ../../../../../src/gcc-4.8.1/libgcc/config/i386/cygwin.S
At ../../../../../src/gcc-4.8.1/libgcc/config/i386/cygwin.S:169
An execution without debugging works fine. There is an example of the code:
#include <stdio.h>
#include <stdlib.h>
int main()
{
const int strNumTries = 15;
const int strLength = 98;
char strName[strLength][strNumTries];
printf("Hello world!\n");
return 0;
}
Debugger stops on the char array declaration using 'step in' method. So what's the problem it may be?
I suspect your seeing something like this:
GNU gdb (Ubuntu 7.7.1-0ubuntu5~14.04.2) 7.7.1
Copyright (C) 2014 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 "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from untitled...done.
(gdb) br main
Breakpoint 1 at 0x4005f1: file untitled.c, line 6.
(gdb) r
Starting program: /home/xxxx/untitled
.
.
.
Breakpoint 1, main () at untitled.c:6
6 const int strNumTries = 15;
(gdb) s
7 const int strLength = 98;
(gdb)
8 char strName[strLength][strNumTries];
(gdb)
10 printf("Hello world!\n");
(gdb)
_IO_puts (str=0x400794 "Hello world!") at ioputs.c:34
34 ioputs.c: No such file or directory.
(gdb)
36 in ioputs.c
(gdb)
strlen () at ../sysdeps/x86_64/strlen.S:66
66 ../sysdeps/x86_64/strlen.S: No such file or directory.
(gdb)
67 in ../sysdeps/x86_64/strlen.S
(gdb)
68 in ../sysdeps/x86_64/strlen.S
where that last line repeats many times before moving on to any other sub functions.
This is not an error, but rather part of the call to printf()
the following is the only gdb output that has anything to do with the array declaration:
8 char strName[strLength][strNumTries];
Not sure if this will help your situation, but ...
I had exactly the same errors from GDB, and GDB admitted it had failed and had a problem (see GDB listing below). I could reproduce this with GCC/gFortran 6.4 and 7.1 100% (running C::B under Win, with MingW).
Of course, I had not ever installed cygwin, and I only use seh and sjlj variants of the MingW64 compilers. Also, I don't have, and never had any of the dir's that GDB is complaining about (e.g. "/../src/gcc-7.1.0/") so it made no sense ... and the problem arose only after adding one more ostensibly similar s/r to my lib of thousands of s/r's (none of which ever complained about this).
... to make a long story short, the problem turned out to be "out of stack space", as I had declared some (Automatic) arrays with large size. Changing either the array sizes (smaller) or changing the Automatic arrays to Allocatable arrays (the former goes on the stack, the later on the heap) fixed the problem.
... so, GDB seems to have a bug, and its complaints had nothing to do with the actual error.
---- for completeness, here is the relevant portion of my GDB listing:
[debug][New Thread 740.0x9fc]
[debug]172 ../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S: No such file or directory.
[debug]Thread 1 received signal SIGSEGV, Segmentation fault.
[debug]__chkstk_ms () at ../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S:172
[debug]>>>>>>cb_gdb:
[debug]> info frame
[debug]Stack level 0, frame at 0x136f00:
[debug] eip = 0x664974bb in __chkstk_ms (../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S:172); saved eip = 0x664958b3
[debug] called by frame at 0x136f14
[debug] source language asm.
[debug] Arglist at 0x136ef8, args:
[debug] Locals at 0x136ef8, Previous frame's sp is 0x136f00
[debug] Saved registers:
[debug] eax at 0x136ef4, ecx at 0x136ef8, eip at 0x136efc
[debug]>>>>>>cb_gdb:
Cannot open file: ../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S
At ../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S:172
[debug]> info locals
[debug]No locals.
[debug]>>>>>>cb_gdb:
[debug]> info args
[debug]No symbol table info available.
[debug]>>>>>>cb_gdb:
[debug]> bt 30
[debug]../../../../src/gdb-7.11.1/gdb/dwarf2loc.c:364: internal-error: dwarf_expr_frame_base: Assertion `framefunc != NULL' failed.
[debug]A problem internal to GDB has been detected,
[debug]further debugging may prove unreliable.
[debug]This is a bug, please report it. For instructions, see:
[debug]<http://www.gnu.org/software/gdb/bugs/>.
[debug]This application has requested the Runtime to terminate it in an unusual way.
[debug]Please contact the application's support team for more information.
[debug]#0 __chkstk_ms () at ../../../../../src/gcc-7.1.0/libgcc/config/i386/cygwin.S:172
[debug]#1 0x664958b3 in fadcern_sixtrack_xl (
Debugger finished with status 1
I received this error when i was trying to debug using gdb in vscode. Vscode don't offer redirected input especially for C. So i was using freopen(). Everything worked fine until it stepped into the declaration part of array.
int arr[n]
So I replaced it with a pointer and allocated memory dynamically.
int *arr=(int *) malloc(sizeof(int)*n);
and this worked.
I think gdb is unable to handle those arr[variable] declaration. But I may be completely wrong. Hope this helps.

call to ffi_call fails even though arguments look right

Consider this gist. I have checked and double checked this piece of code for defects and can't find any apparent flaws in the code. It also compiles fine when I use g++ -g -std=c++11 -Wall dynlibtest.cc -ldl -lffi -lstdc++ -odynlibtest && ./dynlibtest (the -ldl and -lffi switches are for the dynamic loading and FFI libraries, respectively).
However, when the highlighted line (l.96) executes it segfaults.
I have also tried pulling it through gdb, and after installing the libc debugging symbols it spits this message out when the ./dynlibtest bin segfaults:
(gdb) next
Program received signal SIGSEGV, Segmentation fault.
__memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
157 ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S: No such file or directory.
Who can help me understand why this segfaults? Is it a bug of some kind or am I using one of the API's wrong?
For reference: the first part of the code calls gettimeofday directly to show that the code can indeed find it, and that even the args are correct when it is called directly.
EDIT: I have added the gdb output when the code segfaults with the output of bt also attached:
$ gdb ./dynlibtest
GNU gdb (Ubuntu 7.7.1-0ubuntu5~14.04.2) 7.7.1
Copyright (C) 2014 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 "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
http://www.gnu.org/software/gdb/bugs/.
Find the GDB manual and other documentation resources online at:
http://www.gnu.org/software/gdb/documentation/.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./dynlibtest...done.
(gdb) break 96
Breakpoint 1 at 0x401032: file dynlibtest.cc, line 96.
(gdb) run
Starting program: /home/j/dev/elisp-ffi/dynlibtest
Test started...
Got main program handle
pre-alloc: tv.tv_sec = 140737340592552
Sleeping for 1 second
post-alloc: tv.tv_sec = 1432058412
Sleeping for 1 second
Fn ptr call: tv.tv_sec = 1432058413
FFI CIF preparation is OK
Sleeping for 1 second
Breakpoint 1, main () at dynlibtest.cc:96
96 ffi_call(&cif, FFI_FN(gettimeofday), &result, args);
(gdb) next
Program received signal SIGSEGV, Segmentation fault.
__memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
157 ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S: No such file or directory.
(gdb) bt
#0 __memcpy_sse2_unaligned () at ../sysdeps/x86_64/multiarch/memcpy-sse2-unaligned.S:157
#1 0x00007ffff79d34c2 in memcpy (__len=8, __src=0x0, __dest=0x7fffffffda48) at /usr/include/x86_64-linux-gnu/bits/string3.h:51
#2 ffi_call (cif=0x7fffffffdca0, fn=0x400ab0 , rvalue=0x7fffffffdc40, avalue=0x7fffffffdc00) at ../src/x86/ffi64.c:504
#3 0x000000000040104e in main () at dynlibtest.cc:96
(gdb)

Resources