http://forums.leaflabs.com/topic.php?id=17 A place to share, learn, and grow... en-US Fri, 22 Jan 2016 00:10:29 +0000 http://bbpress.org/?v=1.0.2 q http://forums.leaflabs.com/search.php http://forums.leaflabs.com/topic.php?id=17#post-82 Sat, 03 Apr 2010 14:54:36 +0000 importer 82@http://forums.leaflabs.com/ <p>[on Tue 12 of Jan., 2010 16:32 EST, by poslathian]</p> <p>Were fooling with bootloader based fixed now, although Im not certain that there isnt a Mac-end fix also.</p> <p>Unfortunately, the chip boots into appIDLE mode, not dfuIDLE, and exposes the application descriptors by default. Also, unfortunately, you would need a JTAG to flash a new bootloader on. You might get down and dirty and cut the traces on the boot0/1 pins and try and use the embedded serial bootloader that is fixed on the chip. By the way, were going make Maple V2 compatible with the serial bootloader for just this reason.</p> <p>The bootloader-side fix were trying is this:<br /> DFU spec supports both host side bus reset and client side bus reset to enter dfu mode. Right now it looks something like this:</p> <p>Maple in appIDLE<br /> Host-&gt;Maple DFU Request<br /> Maple in appDETACH<br /> Host-&gt;Maple usb bus reset<br /> Maple in dfuIDLE</p> <p>but, on mac, dfu-util is disallowed from issuing the usb bus reset. However, the following is also a valid dfu startup sequence (with the appropriate modifications to the usb functional descriptor)</p> <p>Maple in appIDLE<br /> Host-&gt;Maple DFU Request<br /> Maple-&gt;Host usb bus reset<br /> Maple in dfuIDLE</p> <p>well see if this works. It wont fix anything without a bootloader upgrade though. Again, im still unconvinced there isnt a software solution </p> http://forums.leaflabs.com/topic.php?id=17#post-81 Sat, 03 Apr 2010 14:53:46 +0000 importer 81@http://forums.leaflabs.com/ <p>[on Fri 08 of Jan., 2010 10:52 EST, by rsbohn]</p> <p>Seems like the bootloader trouble on Mac OS could be solved if there was some way to force the bootloader to remain active instead of launching an application. By this I mean that the board remains in dfuIDLE mode. Suppose the bootloader would check a pin, perhaps one on the extension port, and remain resident if that pin is grounded. I think this would allow you to flash the board from Mac OS.</p> <p>Alternatively, if you could clear out the application in Flash it looks like the bootloader would go to an endless loop (main.c:165). If you can stand to run your programs from RAM every time that might work.</p> <p>Do you need JTAG to re-install the bootloader, or is there some way to patch it? </p>