On Sat, May 26, 2012 at 11:26 PM, Tomek CEDRO <tomek.ce...@gmail.com> wrote: > It is important to clarify nomenclature for libswd and my approach: > 1. Bitbang can set/read port state - one usb transfer per operation.
There is a penalty associated with this. This is a short transfer in terms of data load and yet the USB round-trip time (USB frame 1ms or micro-frame 125us) will apply. That is probably why Simon mentioned that USB latency time is more critical to SWD. And this part should really be in the MCU firmware for intelligent HW debugger. > 2. Transfer can read/write bistream from char array (one bit up to dozen > bytes) on miso mosi lines with clock - one usb transfer per operation. This part can benefit from the libusb-1.0 async API and probably some work should be invested to abstract this part. > These two functions when implemented can make any interface do any > kind of transport :-) i would place them in interface, Yes I agree. Interface will handle this kind of low level stuff. > other stuff should go to transport and target... But I think some HW debugger may handle part of the transport and target functions inside FW. For them, maybe some part of those supporting functions can be inside Interface. -- Xiaofan ------------------------------------------------------------------------------ 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 OpenOCD-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openocd-devel
