So because I am so late into the party, I needed to update a few things to get the openocd script to work.
First, I had an issue where the syntax of the "target create" command no longer accepts the chain-position parameter, I rather had to create a dap separately.
I have followed the guide here:
https://visualgdb.com/support/chainposition/The resulting script is:
source [find interface/jlink.cfg]
reset_config trst_and_srst
if { [info exists CHIPNAME] } {
set _CHIPNAME $CHIPNAME
} else {
set _CHIPNAME fpgasoc
}
if { [info exists DAP_TAPID] } {
set _DAP_TAPID $DAP_TAPID
} else {
set _DAP_TAPID 0x4ba00477
}
jtag newtap $_CHIPNAME cpu -irlen 4 -ircapture 0x01 -irmask 0x0f -expected-id $_DAP_TAPID
adapter_khz 4000
dap create $_CHIPNAME.dap -chain-position $_CHIPNAME.cpu
set _TARGETNAME1 $_CHIPNAME.cpu.0
target create $_TARGETNAME1 cortex_a -dap $_CHIPNAME.dap -endian little -coreid 0 -dbgbase 0x80110000
init
cortex_a dbginit
If anyone is more familiar with OpenOCD than I am, and can check that I did this right, feel free to take a look.
I am now downloading the flash using the rtb_nand_dump.pl script (thanks ElectronMan for the fantastic background work in post #1, really couldn't have done it without you).
I am using a clone JLink V9 (this is what I had in stock...). It had an internal VTref of 3.3V, and I didn't have an exposed 2.5V on the connector of the scope, so I switched the AMS1117 3.3V LDO that creates that reference inside the probe to a 2.5V one (the scope uses 2.5V signals). I don't seem to have issues reading the data, and I checked the signals with a voltmeter and all is at 2.5V.
As to the speed, here is a quick measurement:
$ ls -l RTB_NAND_IMAGE.img && sleep 60 && ls -l RTB_NAND_IMAGE.img
-rw-rw-r-- 1 x x 16191568 Nov 5 13:20 RTB_NAND_IMAGE.img
-rw-rw-r-- 1 x x 17897344 Nov 5 13:21 RTB_NAND_IMAGE.img
1705776 bytes per minute. Image is 512MB. 512*1024*1024/1705776=314 minutes, so around 5 hours if all goes well (no interruptions - hopefully my probe is stable)
Stay tuned!
[EDIT] Image file seems to contain valid data so far. I can see the bootloader that was discussed previously at the beginning of the image, with legible text etc.