Re: [OpenOCD-user] Advice on Snapdragon 820

[Jiri K. <cz1...@gm...>]

snapdragon 820 is afaik kryo based, so there should be some differencies from aarch64
cores.

anyway, even you can't handle core, when you telnet to 4444, you should be able to
list using 'dap info' all components visible via apb-ap.

there would be probably ~3 components with invalid CID (modem related debug components).

and maybe, like in previous generation 'krait' cores, all components tied to core
will have same partnumber (PMU, CTI, debug unit).

and just guessing - can you try cortex_a instead of aarch64 and paste somewhere output?

On Thu, Jan 05, 2017 at 11:21:36PM +0100, Sergei Antonov wrote:
> On 5 January 2017 at 18:59, Sergei Antonov <sa...@gm...> wrote:
> > On 5 January 2017 at 10:11, Matthias Welwarsky <mat...@we...> wrote:
> >> Hello Sergei,
> >>
> >> On Thursday 05 January 2017 01:25:35 Sergei Antonov wrote:
> >>> Hello!
> >>> I'd like to use OpenOCD with a development board containing a Qualcomm
> >>> Snapdragon 820 (APQ8096) processor, its architecture is AArch64. Here
> >>> is the scanned JTAG chain:
> >>>
> >>> Info : JTAG tap: auto0.tap tap/device found: 0x5ba00477 (mfg: 0x23b
> >>> (ARM Ltd.), part: 0xba00, ver: 0x5)
> >>> Info : JTAG tap: auto1.tap tap/device found: 0x000016dd (mfg: 0x36e
> >>> (Certicom Corporation), part: 0x0001, ver: 0x0)
> >>> Info : JTAG tap: auto2.tap tap/device found: 0x4003e0e1 (mfg: 0x070
> >>> (Qualcomm), part: 0x003e, ver: 0x4)
> >>>
> >>> I suppose, I can debug device 0x5ba00477 as aarch64 target using
> >>> development source code. What repo/branch of OpenOCD should I use?
> >>> On Gerrit there are lots of patches for aarch64, I'm confused about
> >>> what exactly is state-of-the-art or has best chances to work on my
> >>> hardware.
> >>
> >> If you checkout http://openocd.zylin.com/#/c/3832, you get a pretty good base
> >> for aarch64 debugging. You will also likely need this patch:
> >>
> >> http://openocd.zylin.com/#/c/3710 on top, you should be able to cherry-pick
> >> it.
> >
> > Thanks, Matthias! I took c/3832, cherry-picked changes from c/3710,
> > and added -ignore-syspwrupack to my .cfg. Unfortunately, target
> > examination fails. Details are below.
> >
> > == openocd.cfg ==
> > source [find interface/altera-usb-blaster.cfg]
> > jtag newtap auto0 tap -irlen 4 -expected-id 0x5ba00477 -ignore-syspwrupack
> > jtag newtap auto1 tap -irlen 5 -expected-id 0x000016dd
> > jtag newtap auto2 tap -irlen 11 -expected-id 0x4003e0e1
> > target create target0 aarch64 -endian little -chain-position auto0.tap
> > init
> >
> > == Relevant part of debug output ==
> > Debug: 193 21 core.c:1399 jtag_init_inner(): Init JTAG chain
> > Debug: 194 21 core.c:327 jtag_call_event_callbacks(): jtag event: TAP reset
> > Debug: 195 21 core.c:1060 jtag_examine_chain(): DR scan interrogation
> > for IDCODE/BYPASS
> > Debug: 196 21 core.c:327 jtag_call_event_callbacks(): jtag event: TAP reset
> > Info : 197 29 core.c:959 jtag_examine_chain_display(): JTAG tap:
> > auto0.tap tap/device found: 0x5ba00477 (mfg: 0x23b (ARM Ltd.), part:
> > 0xba00, ver: 0x5)
> > Info : 198 29 core.c:959 jtag_examine_chain_display(): JTAG tap:
> > auto1.tap tap/device found: 0x000016dd (mfg: 0x36e (Certicom
> > Corporation), part: 0x0001, ver: 0x0)
> > Info : 199 29 core.c:959 jtag_examine_chain_display(): JTAG tap:
> > auto2.tap tap/device found: 0x4003e0e1 (mfg: 0x070 (Qualcomm), part:
> > 0x003e, ver: 0x4)
> > Debug: 200 29 core.c:1190 jtag_validate_ircapture(): IR capture validation scan
> > Debug: 201 34 core.c:1248 jtag_validate_ircapture(): auto0.tap: IR capture 0x01
> > Debug: 202 34 core.c:1248 jtag_validate_ircapture(): auto1.tap: IR capture 0x01
> > Debug: 203 34 core.c:1248 jtag_validate_ircapture(): auto2.tap: IR capture 0x1
> > Debug: 204 34 openocd.c:148 handle_init_command(): Examining targets...
> > Debug: 205 34 target.c:1513 target_call_event_callbacks(): target
> > event 21 (examine-start)
> > Debug: 206 34 arm_adi_v5.c:603 dap_dp_init():
> > Debug: 207 34 arm_adi_v5.c:636 dap_dp_init(): DAP: wait CDBGPWRUPACK
> > Debug: 208 34 arm_adi_v5.h:432 dap_dp_poll_register(): DAP: poll 4,
> > mask 0x20000000, value 0x20000000
> > Debug: 209 135 arm_adi_v5.c:789 dap_find_ap(): Found APB-AP at AP
> > index: 1 (IDR=0x44770002)
> > Debug: 210 173 adi_v5_jtag.c:572 jtagdp_transaction_endcheck():
> > jtag-dp: CTRL/STAT 0x70000021
> > Error: 211 173 adi_v5_jtag.c:584 jtagdp_transaction_endcheck():
> > JTAG-DP STICKY ERROR
> > Error: 212 178 aarch64.c:1825 aarch64_examine_first(): Could not
> > initialize the APB-AP
> > Debug: 213 178 openocd.c:150 handle_init_command(): target examination failed
> 
> The first time I try to connect after rebooting the device, behavior
> is different (though examination fails anyway):
> 
> Debug: 206 35 arm_adi_v5.c:603 dap_dp_init():
> Debug: 207 35 arm_adi_v5.c:636 dap_dp_init(): DAP: wait CDBGPWRUPACK
> Debug: 208 35 arm_adi_v5.h:432 dap_dp_poll_register(): DAP: poll 4,
> mask 0x20000000, value 0x20000000
> Debug: 209 136 arm_adi_v5.c:789 dap_find_ap(): Found APB-AP at AP
> index: 1 (IDR=0x44770002)
> Debug: 210 174 arm_adi_v5.c:716 mem_ap_init(): MEM_AP Packed Transfers: disabled
> Debug: 211 174 arm_adi_v5.c:727 mem_ap_init(): MEM_AP CFG: large data
> 0, long address 0, big-endian 0
> Error: 212 3761 adi_v5_jtag.c:450 jtagdp_overrun_check(): Timeout
> during WAIT recovery
> Error: 213 3780 arm_adi_v5.c:845 dap_lookup_cs_component(): Can't read
> component with base address 0x80030000, the corresponding core might
> be turned off
> Debug: 214 3780 openocd.c:150 handle_init_command(): target examination failed
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most 
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> _______________________________________________
> OpenOCD-user mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openocd-user