Re: [OpenOCD-user] beaglebone + gdb segfault

[Jorge V. <jor...@gm...>]

Paul,
I tried your fisrt suggestion to 6Mhz, 1Mhz and 200 kHz with no success.

The second suggestion doesn't make sense because beaglebone use the ftdi
chipset
through a usb and the jtag is built in.

I will post something at devel mailing list.

Thank you.




On Sat, Jun 28, 2014 at 12:05 AM, Paul Fertser <fer...@gm...> wrote:

> Hi Jorge,
>
> You've omitted the mailing list from CC, I assume by mistake, readding
> it back; let's keep the communication public so more people would be
> able to help you.
>
> And please avoid topposting, I'm really allergic to it.
>
> On Fri, Jun 27, 2014 at 11:33:03PM -0500, Jorge Ventura wrote:
> > I didn't understand your last instruction about gdb --args ...,
>
> I mean to fix the segfault itself it's better to have the backtrace of
> the openocd itself, so you can start openocd from inside host gdb to
> obtain that.
>
> > Debug: 329 6 arm_adi_v5.c:827 ahbap_debugport_init():
> > Debug: 330 7 arm_adi_v5.c:938 ahbap_debugport_init(): MEM_AP Packed
> Transfers: enabled
> > Debug: 331 7 arm_adi_v5.c:949 ahbap_debugport_init(): MEM_AP CFG: large
> data 0 long address 0, big-endian 0
> > Debug: 332 7 arm_adi_v5.c:1011 dap_find_ap(): Found APB-AP at AP index:
> 1 (IDR=0x04770002)
> > Debug: 333 7 arm_adi_v5.c:1011 dap_find_ap(): Found AHB-AP at AP index:
> 0 (IDR=0x14770001)
> > Debug: 334 7 adi_v5_jtag.c:271 jtagdp_transaction_endcheck(): jtag-dp:
> CTRL/STAT error, 0xf0000021
> > Debug: 335 7 adi_v5_jtag.c:291 jtagdp_transaction_endcheck(): MEM-AP
> Cached values: ap_bank 0x10, ap_csw 0xa2000002, ap_tar 0x80001d00
> > Error: 336 7 adi_v5_jtag.c:298 jtagdp_transaction_endcheck(): JTAG-DP
> STICKY ERROR
> > Debug: 337 7 adi_v5_jtag.c:315 jtagdp_transaction_endcheck(): jtag-dp:
> CTRL/STAT 0xf0000001
> > Error: 338 7 adi_v5_jtag.c:331 jtagdp_transaction_endcheck(): MEM_AP_CSW
> 0x800000c2, MEM_AP_TAR 0x80001d00
> > Debug: 339 7 openocd.c:147 handle_init_command(): target examination
> failed
>
> Here is the reason of the later segfault. Target couldn't be examined
> and as the result you can't work with it later.
>
> Possible things to try:
>
> 1. Lower JTAG clock, 16MHz might be really pushing it, depending on
> how exactly you connect the target;
>
> 2. Check SRST and TRST signals, your config says you're using them
> both, make sure they're properly connected;
>
> 3. Write to the devel mailing list (my own opinion is that the user
> list should better be merged with the devel; ignoring threads is easy
> in any decent MUA).
>
>
> --
> Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software!
> mailto:fer...@gm...
>