On Tue, Nov 01, 2016 at 12:33:45PM +0100, Meindert Sprang wrote: > Hello all, > > I am the developer of the ShipModul (CustomWare) MiniPlex products where > we use FTDI chips for USB communications. > I see that even the latest PID we use (FD4B) is already in the > ftdi_sio.c driver and I'm glad to see the Linux community picks this up > to fast. > > There is one thing however that I'd like to see added to the driver and > that is fixed baud rates for our products. Since PID FD49 (the second > PID we assigned) we use fixed baudrates, so the user of our products > doesn't have to set the right baud rate in his application. > > When browsing through the source code I found here > (http://lxr.free-electrons.com/source/drivers/usb/serial/ftdi_sio.c), I > found function > > static __u32 get_ftdi_divisor(struct tty_struct *tty, struct > usb_serial_port *port) > > where the devisor for a specific baud rate is set. > > In the case at line 1323, where the FT232BM and similar chips are > handled, I see an if statement which obviously checks a few PID's to set > a fixed baud rate. That's not exactly what is being done here; rather a nasty one-off hack is mapping a specific baudrate to a higher non-standard rate for some PIDs. There is now a standard way of setting such rates (BOTHER), which should have been used for this. > So it appears that it would be fairly easy to add a few other exceptions > for our PID's. Am I correct? I'm afraid this is not something I think we should do. Limiting maximum baudrate etc based on chip types is one thing, but forcing a fixed rate based on a particular application a chip is used in is something quite different (e.g. could change after a firmware update). Johan -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
