On Mon, Apr 4, 2011 at 4:26 PM, Øyvind Harboe <oyvind.har...@zylin.com> wrote: > Is #3 a hack or do we know why flushing the buffer > is necessary? > > Any other objections out there?
I can not confirm this only from experience - not flushing the buffer systematically leads to overfill of the ft2232 buffer in my case, which later results in error (EJTAG miniprogram is not executed correctly). I hope that some MIPS experts out there can provide us few insights. If it does not bother the performance I'd opt for implementing this change, otherwise FASTDATA is useless from my point of view - I have not found other way to make it work and I did a lot of experiments. Of course, I can be wrong and the reason for this might be somewhere else... BR, Drasko _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
