STM32F4 failure on semaphore release in termios.

[Karel Gardas]

On 08/30/13 08:58 AM, Sebastian Huber wrote:
> On 2013-08-28 23:30, Karel Gardas wrote:
>> _Internal_error_Occurred (the_source=RTEMS_FATAL_SOURCE_EXCEPTION,
>> is_internal=false, the_error=536875260)
>
> If you get an exception like this, then you can use in GDB to figure out
> the context (the actual program counter of the exception is not the
> direct value of register_pc):
>
> Breakpoint 6, _Internal_error_Occurred
> (the_source=RTEMS_FATAL_SOURCE_EXCEPTION, is_internal=false,
> the_error=1049524) at
> ../../../../../../rtems/c/src/../../cpukit/score/src/interr.c:35
> 35 _User_extensions_Fatal( the_source, is_internal, the_error );
> (gdb) p/x *(CPU_Exception_frame *)the_error
> $1 = {

Thanks for this nice trick!

(gdb) p/x *(CPU_Exception_frame *)the_error
$2 = {register_r0 = 0x20006acc, register_r1 = 0x1a010004, register_r2 = 
0x0, register_r3 = 0x1a010003, register_r4 = 0x20006ab8, register_r5 = 
0x20001794, register_r6 = 0x20006d80, register_r7 = 0x20003c70, 
register_r8 = 0x20003e7c, register_r9 = 0x20001440, register_r10 = 0x0, 
register_r11 = 0x18fc0, register_r12 = 0x0, register_sp = 0x0, 
register_lr = 0xe57b, register_pc = 0xe584, register_xpsr = 0x21000000, 
vector = 0x3, vfp_context = 0x1}
(gdb) dis 0xe584
warning: bad breakpoint number at or near '0xe584'
(gdb) disas 0xe584
Dump of assembler code for function _Chain_Get_first_unprotected:
    0x0000e56c <+0>:     push	{r7, lr}
    0x0000e56e <+2>:     sub	sp, #24
    0x0000e570 <+4>:     add	r7, sp, #0
    0x0000e572 <+6>:     str	r0, [r7, #4]
    0x0000e574 <+8>:     ldr	r0, [r7, #4]
    0x0000e576 <+10>:    bl	0xe4d8 <_Chain_Head>
    0x0000e57a <+14>:    str	r0, [r7, #20]
    0x0000e57c <+16>:    ldr	r3, [r7, #20]
    0x0000e57e <+18>:    ldr	r3, [r3, #0]
    0x0000e580 <+20>:    str	r3, [r7, #16]
    0x0000e582 <+22>:    ldr	r3, [r7, #16]
    0x0000e584 <+24>:    ldr	r3, [r3, #0]
    0x0000e586 <+26>:    str	r3, [r7, #12]
    0x0000e588 <+28>:    ldr	r3, [r7, #20]
    0x0000e58a <+30>:    ldr	r2, [r7, #12]
    0x0000e58c <+32>:    str	r2, [r3, #0]
    0x0000e58e <+34>:    ldr	r3, [r7, #12]
    0x0000e590 <+36>:    ldr	r2, [r7, #20]
    0x0000e592 <+38>:    str	r2, [r3, #4]
    0x0000e594 <+40>:    ldr	r3, [r7, #16]
    0x0000e596 <+42>:    mov	r0, r3
    0x0000e598 <+44>:    add.w	r7, r7, #24
    0x0000e59c <+48>:    mov	sp, r7
    0x0000e59e <+50>:    pop	{r7, pc}
End of assembler dump.

So it looks like I've debugged well and got up to the culprit. Now the 
question is why this ldr fails...

Thanks,
Karel