Réf : Re: Réf : Re: RTEMS LEON2 EDAC Trap Management

Leonard Bise leonard.bise at syderal.ch
Thu Dec 3 07:22:16 UTC 2009


Thanks for your hints guys.

I managed to fix my problem, it was indeed a problem of trap management, 
instead of being treated as a synchronous trap, my double EDAC trap would 
be managed as an asynchronous one.

It was simply fixed by using the following RTEMS call:

set_vector((OS_FUNCPTR)BGD_trap_DE_handler, SPARC_SYNCHRONOUS_TRAP(0x09), 
1);

Thanks a lot Jamie Bowman and Neil MacEwen for your help !

Best Regards,
Léonard.




Aleix Conchillo Flaqué <aconchillo at ice.csic.es> 
Envoyé par : aconchillo at gmail.com
02.12.2009 08:44

A
Leonard Bise <leonard.bise at syderal.ch>
cc
rtems-users at rtems.org
Objet
Re: Réf : Re: RTEMS LEON2 EDAC Trap Management






This makes use of rtems_interrupt_catch (as you passed 1),so the default 
ISR dispatcher is used. The default ISR detects whether the interrupt was 
synchronous or not and modifies %l1 and %l2 accordingly if so.

Did you run this in a simulator and did a step-by-step? May be clearing 
the failing status register is not enough and the IU error in the Error 
status register is also set and you need to clear it as well (in test 
mode).

2009/12/2 Leonard Bise <leonard.bise at syderal.ch>

Thank you Aleix! But yeah I'm working on uncorrectable EDAC errors. 

Also I was wondering, the trap table is installed using this function, 
maybe this could be the cause of my problem? 

void OS_install_handler(OS_FUNCPTR handler, uint32 irq) { 
    { 
        /*#[ operation OS_install_handler(OS_FUNCPTR,uint32) */ 
        set_vector ((const rtems_isr_entry) handler, irq, 1); 
        /*#]*/ 
    } 
} 




Aleix Conchillo Flaqué <aconchillo at ice.csic.es> 
Envoyé par : aconchillo at gmail.com 
01.12.2009 19:43 


A
Leonard Bise <leonard.bise at syderal.ch> 
cc
rtems-users at rtems.org 
Objet
Re: RTEMS LEON2 EDAC Trap Management








Sorry Leonard, I just read your mail too fast. May be my answer helps
someone else with single errors. :-)

Cheers,

Aleix

On Tue, Dec 1, 2009 at 17:24, Leonard Bise <leonard.bise at syderal.ch> 
wrote:
>
> Hi all,
>
> I'm working on an RTEMS application running on a LEON2 processor.
> I have some issues in regard to the Double EDAC error trap (0x09)
> management.
>
> I'm trying to validate that my application correctly manages EDAC errors 
by
> launching my application, breaking then generating an EDAC error and 
then
> resuming.
> My application then detects that an error has hapened and launches the
> correct trap handler.
>
> After the trap has been handled, we resume normal execution by executing 
the
> rett instruction (return from trap), with the address of the next
> instruction after the one that has been trapped.
> Only that instead of continuing executing after the rett, the same
> instruction that triggered the first error triggers another error (0x11)
> which should be a correctable EDAC error.
>
> This behavior then loops forever (Double EDAC than Single EDAC etc...) 
and
> after some time my application resets (I have approximatively 32'000 
trap
> triggered for each one before it resets).
>
> I've been looking around for help in the LEON "community" but I could 
not
> find much help. I know it is not necessarily RTEMS related but a 
colleague
> of mine which does the exact same thing on another project but only in C 
has
> no problem so I'm wondering what might be causing this.
>
> For info here is my double edac trap handler.
>
> static void BGD_trap_DE_handler(void) {
>     {
>         /*#[ operation BGD_trap_DE_handler() */
>         volatile uint32*          failing_address;
>         boolean accepted = FALSE;
>
>         /* read Failure Address Register */
>         asm("TEST_EDAC_DOUBLE_START:nop;");
>         failing_address = *FAILAR;
>         FIFO_edac_address [FIFO_edac_next] = (uint32)failing_address;
>         FIFO_edac_double [FIFO_edac_next] = 1;
>         FIFO_edac_nb = (FIFO_edac_nb & 0xF) + 1;
>         /* Note: in case more than 16 errors occur during 1 second, the 
16
> oldest will be lost. */
>         FIFO_edac_next = (FIFO_edac_next+1) & 0xF;
>
>         /* SYD MOD */
>         if ((failing_address < START_EEPROM) || (failing_address >=
> 0x80000000))
>         {
>             BGD_process_d_edac_sram();
>         }
>         else
>         {
>             BGD_process_d_edac_eeprom();
>         }
>         /* reset fail status register */
>         *FAILSR = 0;
>         *FAILAR = 0;
>
>         //BDT_abort_request (&accepted);
>         /*#]*/
>                 asm("TEST_EDAC_DOUBLE_END:nop;");
>     }
> }
>
> Here is a disassembly of the instruction that triggers the  Double EDAC
> Trap, which is correct:
>  1046115464  40001400  cmp  %i0, 15                [00000ff1]
>  1046115466  40001404  bgu  0x40001414             [00000000]
>  1046115467  40001408  nop                         [00000000]
>  1046115468  40001414  add  %fp, -20, %i5          [401e5f34]
>  1046115473            ahb read,  mst=0, size=2    [401e5f34 40053854]
>  1046115474  40001418  ld  [%i5], %i2              [40053854]
>  1046115475  4000141c  add  %fp, -24, %i4          [401e5f30]
>  1046115480            ahb read,  mst=0, size=2    [401e5f30 40100000]
>  1046115481  40001420  ld  [%i4], %i1              [40100000]
>  1046115482  40001424  mov  %i2, %i3               [40053854]
>  1046115483  40001428  mov  %i1, %i0               [40100000]
>  1046115491            ahb read,  mst=0, size=2    [40100000 a6102003]
>  1046115492  4000142c  ld  [%i0], %i0              [trapped]
>
> Here is the second trap triggered, which should not happen:
>  1046117903  4002bdc8  mov  %l0, %psr              [000000c4]
>  1046117904  4002bdcc  nop                         [00000000]
>  1046117905  4002bdd0  nop                         [00000000]
>  1046117906  4002bdd4  nop                         [00000000]
>  1046117911            ahb read,  mst=0, size=2    [401e5e84 00000000]
>  1046117912  4002bdd8  ld  [%g1 + 0x6c], %g1       [00000000]
>  1046117913  4002bddc  jmp  %l1                    [4002bddc]
>  1046117914  4002bde0  rett  %l2                   [40001430]
>  1046117916  4000142c  ld  [%i0], %i0              [trapped]
>
> I hope someone can help ;)
>
> Léonard.
>
> _______________________________________________
> rtems-users mailing list
> rtems-users at rtems.org
> http://www.rtems.org/mailman/listinfo/rtems-users
>
>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20091203/217792cb/attachment-0001.html>


More information about the users mailing list