On Monday 08 February 2010, Øyvind Harboe wrote: > On Mon, Feb 8, 2010 at 11:31 PM, David Brownell <davi...@pacbell.net> wrote: > > On Monday 08 February 2010, Ųyvind Harboe wrote: > >> The reason why jtag_add_runtest() is so much faster than > >> pathmove is that jtag_add_runtest() can be queued in > >> a hardware fifo. > > > > But does this guarantee the same path transition? > > Ah.... that's the other half of the problem, isn't it?
That's why pathmove() is used... > > The ARM11 debug TAP cares about specific paths when > > the host is using the ITR mechanism ... > > I think this patch (and the branch I created tonight) was helpful > in pointing out where the performance problems are. > > Something to tinker with post 0.4.... > > 100kBytes/s @ 8MHz is ~1:3 of arm7/9 performance, so it isn't > disastrously bad. > Is that using the DCC? _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
