On 19/07/2012 11:22 AM, Andrew Leech wrote: > On 19/07/12 7:29 AM, Freddie Chopin wrote:
>> Anyone on the list willing to do some testing with this new piece of >> code (already merged - I can provide binaries for Windows) and >> Flyswatter and/or luminary-icdi adapters, which supposedly don't work? >> > I can try it on icdi, if I can grab windows binaries from you it'd be > quicker to get going. > Is it set up to use the winusb driver backend for libusb? > Ok, I've compiled it myself in the mean time, although it's compiled against my copy of the libusb-pbatard-K branch that's got support for libusbk driver. I loaded the WinUSB backend driver to the A port on my Luminary Micro ICDI Board. I've previously used official ftdi drivers rather than libusb, maybe I'll change for good now. Where are the updated interface scripts supposed to come from? I've seen an interface/ftdi folder referenced in older messages, but I don't a ftdi folder in my freshly git master pulled folder openocd/tcl/interface nor can I see the folder in http://openocd.git.sourceforge.net/git/gitweb.cgi?p=openocd/openocd;a=tree;f=tcl/interface Regardless, I found an updated luminary.cfg attached in a past message (Andreas on 10/05/12) on this list and modified it. I had to change the desc and vid/pid to icdi but more importantly, I had to change the ftdi_layout_signal nSRST command to be -noe rather than -data. I think the reset was just inverted and changing this works for me, with -data as soon as I tried to reset the board with openocd my srst line was held low permanently. The icdi board's srst has a tri state buffer whose input is grounded, and the (inverted) enable line driven by ADBUS5, so the -noe makes sense to me as the correct option. Should there be a -data or something else as well? There's a schematic available for the icdi in various development kit pdf's on ti site, I can find a link to one if anyone wants a copy for reference. I can now load my program with gdb and debug it no worries. ########### script start ################ interface ftdi ftdi_device_desc "Luminary Micro ICDI Board" ftdi_vid_pid 0x0403 0xbcda ftdi_layout_init 0x00a8 0x00eb ftdi_layout_signal nSRST -noe 0x0020 adapter_khz 3000 reset_config srst_only srst_pulls_trst source [find target/lpc3131.cfg] # FPGA Tap: A3PN060 A3P060 jtag newtap fpga tap -irlen 8 -expected-id 0x1b9521cf -expected-id 0x129121cf -expected-id 0x1b9121cf init #scan_chain ########### script finish ################ for reference in my lpc3131.cfg I've got arm7_9 fast_memory_access enabled, this still work fine. I should post my changes to the list, I've fixed a couple of minor things in it since I first submitted it. I also tried it with the libusbk driver, and that worked great too (thanks to the special build of libusb). It does seem a lot faster too, although this is based on this icdi (my personal one) being set to a much higher adapter_khz than the jtag programmer I usually use here at work. Andrew ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ OpenOCD-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openocd-devel
