Øyvind Harboe wrote: > >>> For the OpenOCD, we know there are some optimizations do be done on > >>> the software to accelerate the USB JTAG cable. > >> > >> The bottleneck is stupid adapters. > > I don't know that "stupid" is derogatory when we talk about > hardware. KISS often does it...
I am a firm believer in having relevant expert knowledge in the hardware. Yes, moving that to a PC makes for cheaper solutions, but performance and all sorts of hell will follow. And really, how much cheaper than a $6 USB Cortex-M3 microcontroller running at 72MHz does JTAG need? > > Maybe as the first step, OpenOCD can be modified to make use of > > the async API provided by libusb-1.0 and libftdi-1.0. Very good point, Xiaofan. > OpenOCD will create an entire read/write sequence before sending > *anything* to the USB adapter. I'm not sure if it helps, because USB bus time may still not be used optimally when using the sync libusb APIs. libusb-1.0 improves sync performance somewhat over 0.1 (also improvement when using libusb-1.0 and libusb-compat-0.1 to get the old API) but to schedule USB operations the most efficiently it's neccessary to use the async API. Maybe less important for a high-speed device, but more important for full-speed devices. I'm happy that UrJTAG got as much as a 1000% speed increase! //Peter _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
