On Friday 16 April 2010, Koeller, T. wrote: > Hi, > > I am using the Amontec JTAGkey2 USB JTAG adapter > in conjunction with a Spectrum Digital TMS320DM365 > eval board.
There are two basic issues I'm asware of with that board when used with OpenOCD -- once it's set up correctly. * JTAG resets don't work .... at least on the original boards, the CPLD programming ignores the SRST signal from the JTAG connector. That's presumably fixable by reprogramming the CPLD, but longer term I thnk the best fix wouldinvolve information from TI about how to use ICEpick-C to reset chips. (It would help other DaVinci chips, OMAPs, andthe new TMD6570 devices (dual core Cortex-R). The's a not-very-good-workaround in place though .... not-very-good because it can't work when you most need it. Like, when the chip is in WFI mode (core clock gated) and thus is not responsive (which is just when you most need to reset it). (The workaround involves using the watchdog to force a reset. But if you can't talk to the core, it can't talk to the watchdog for you, hence no reset.) The other basic issue is that the "reset-init" method isn't yet able to set up the clock tree. Each generation of DaVinci chips tweaked the clock setup differently. Net rsult, DM365 memory controllers aren't set up, unless the NAND boot loaders run for long enough to do that. (The earlier DM355 EVM boards work better with OpenOCD. No CPLD bug, and the clock tree setup works well enough.) > I built openocd-0.4.0 under windows / > cygwin, using the 'CDM 2.06.00 WHQL Certified' > driver distribution from FTDI. Openocd basically > works, using the supplied configuration files, which I > wrapped in openocd.conf like this: > > $ cat .openocd/openocd.cfg > set CS0MODE NAND > set JRC_TAPID 0x8b83e02f > source [find interface/jtagkey2.cfg] > source [find target/davinci.cfg] > source [find board/dm365evm.cfg] > > However, a number of rather nasty problems remains: > > 1. The most annoying thing is that the JTAGkey2 easily > gets locked up and enters a state where it apparently no > longer communicates with openocd. This can be forced by > simply terminating and restarting openocd: If you power cycle and the board reboots from its NAND boot loader(s) it will come up happier But you'll likely have a fw other issues then. > > tkoel...@gerd0022 ~ > $ openocd > Open On-Chip Debugger 0.4.0 (2010-04-13-19:21) > Licensed under GNU GPL v2 > For bug reports, read > http://openocd.berlios.de/doc/doxygen/bugs.html > RCLK - adaptive > fast memory access is enabled > dcc downloads are enabled > trst_only separate trst_push_pull > CS0 NAND > Info : device: 6 "2232H" > Info : deviceID: 67358712 > Info : SerialNumber: 53SZGA1LA > Info : Description: Amontec JTAGkey-2 A > Info : max TCK change to: 30000 kHz > Info : RCLK (adaptive clock speed) > Info : JTAG tap: dm365.jrc tap/device found: 0x8b83e02f (mfg: 0x017, part: > 0xb83e, ver: 0x8) > Info : JTAG tap: dm365.etb enabled > Info : JTAG tap: dm365.arm enabled > Info : Embedded ICE version 6 > Info : dm365.arm: hardware has 2 breakpoint/watchpoint units > Info : ETM v1.3 > ^C > > tkoel...@gerd0022 ~ > $ openocd > Open On-Chip Debugger 0.4.0 (2010-04-13-19:21) > Licensed under GNU GPL v2 > For bug reports, read > http://openocd.berlios.de/doc/doxygen/bugs.html > RCLK - adaptive > fast memory access is enabled > dcc downloads are enabled > trst_only separate trst_push_pull > CS0 NAND > Info : device: 6 "2232H" > Info : deviceID: 67358712 > Info : SerialNumber: 53SZGA1LA > Info : Description: Amontec JTAGkey-2 A > Info : max TCK change to: 30000 kHz > Info : RCLK (adaptive clock speed) > Error: couldn't read enough bytes from FT2232 device (0 < 81) > Error: couldn't read from FT2232 > > The only way to recover from this is by disconnecting > the JTAGkey2 from the PC and power-cycling the EVM board. > If only the EVM board is power-cycled without disconnecting > the JTAKkey2, I get weird results when restarting > openocd: > > tkoel...@gerd0022 ~ > $ openocd > Open On-Chip Debugger 0.4.0 (2010-04-13-19:21) > Licensed under GNU GPL v2 > For bug reports, read > http://openocd.berlios.de/doc/doxygen/bugs.html > RCLK - adaptive > fast memory access is enabled > dcc downloads are enabled > trst_only separate trst_push_pull > CS0 NAND > Info : device: 6 "2232H" > Info : deviceID: 67358712 > Info : SerialNumber: 53SZGA1LA > Info : Description: Amontec JTAGkey-2 A > Info : max TCK change to: 30000 kHz > Info : RCLK (adaptive clock speed) > Info : JTAG tap: dm365.jrc tap/device found: 0x2b900f0f (mfg: 0x787, part: > 0xb900, ver: 0x2) > Warn : JTAG tap: dm365.jrc UNEXPECTED: 0x2b900f0f (mfg: 0x787, part: > 0xb900, ver: 0x2) > Error: JTAG tap: dm365.jrc expected 1 of 1: 0x8b83e02f (mfg: 0x017, part: > 0xb83e, ver: 0x8) > Warn : Unexpected idcode after end of chain: 32 0x0792602f > Warn : Unexpected idcode after end of chain: 64 0x8b83e02f I'm suspecting you didn't set up the EMU0/EMU1 jumpers correctly ... I've not fired up a DM365 in a long time, but that easy stuff was working fine. > Error: double-check your JTAG setup (interface, speed, missing TAPs, ...) > Info : JTAG tap: dm365.jrc tap/device found: 0x2b900f0f (mfg: 0x787, part: > 0xb900, ver: 0x2) > Warn : JTAG tap: dm365.jrc UNEXPECTED: 0x2b900f0f (mfg: 0x787, part: > 0xb900, ver: 0x2) > Error: JTAG tap: dm365.jrc expected 1 of 1: 0x8b83e02f (mfg: 0x017, part: > 0xb83e, ver: 0x8) > Warn : Unexpected idcode after end of chain: 32 0x0792602f > Warn : Unexpected idcode after end of chain: 64 0x8b83e02f > Error: double-check your JTAG setup (interface, speed, missing TAPs, ...) > Command handler execution failed > Warn : jtag initialization failed; try 'jtag init' again. > > While in this state, I tried to re-animate the adapter, but > this failed: > > Open On-Chip Debugger > > targets > TargetName Type Endian TapName State > -- ------------------ ---------- ------ ------------------ ------------ > 0* dm365.arm arm926ejs little dm365.arm tap-disabled > > jtag tapenable dm365.arm > JTAG tap: dm365.arm enabled > 1 > > targets > TargetName Type Endian TapName State > -- ------------------ ---------- ------ ------------------ ------------ > 0* dm365.arm arm926ejs little dm365.arm unknown > > jtag init > 'jtag init' has already been called > > I should also mention that I tried to increase the USB > latency value to 10ms using the ft2232_latency command, > to no avail. > > 2. NAND flash support does not work for me: > > > halt > target state: halted > target halted in ARM state due to debug-request, current mode: Supervisor > cpsr: 0x60000013 pc: 0xc00a7190 > MMU: enabled, D-Cache: enabled, I-Cache: enabled > > nand list > #0: not probed > #1: not probed > > nand probe 0 > memory read caused data abort (address: 0x01d10060, size: 0x4, count: 0x1) > chip address 02000000 not NAND-enabled? > BUG: unknown controller initialization failure > probing failed for NAND flash device > > 3. The 'reset' command does not work: > > > reset run > RCLK - adaptive > couldn't read enough bytes from FT2232 device (0 < 81) > couldn't read from FT2232 > error: -104 > Command handler execution failed > in procedure 'reset' called at file "command.c", line 650 > called at file "command.c", line 361 > > Again, the JTAGkey2 is locked up at this point. > > > > Any help with the problems described above would be greatly appreciated! > > > _______________________________ > > Thomas Köller, Software Developer > > Basler Vision Technologies > An der Strusbek 60-62 > 22926 Ahrensburg > Germany > > Tel. +49 (0) 4102 463-390 > Fax +49 (0) 4102 463-46390 > > mailto:thomas.koel...@baslerweb.com > http://www.baslerweb.com > > Vorstand: Dr.-Ing. Dietmar Ley (Vorsitzender) · John P. Jennings · > Aufsichtsratsvorsitzender: Norbert Basler > Basler AG · Amtsgericht Ahrensburg HRB 4090 · Ust-IdNr.: DE 135 098 121 · > Steuer-Nr.: 30 292 04497 · WEEE-Reg.-Nr. DE 83888045 > > _______________________________ > > > > _______________________________________________ > Openocd-development mailing list > Openocd-development@lists.berlios.de > https://lists.berlios.de/mailman/listinfo/openocd-development > _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
