In article <1310393182.8617.16.camel@AlexS-PC> you wrote: > Hello, > We are currently working with a development module which contains a > FT232H chip and communicates via UART with another chip at a baud rate > of exactly 115200. After the last patch of the ftdi_sio driver we saw > that, theoretically, the FT232H should be fully supported by the driver. > But, when we tried to use it, it didn't work. Argh: So my loop back test wasn't sufficent. > After digging further, we got to the 'get_ftdi_divisor()' function where > the FT232H chip was placed on the same branch as FT2232H and FT4232H > and, for a baud rate of 115200, 0xC068 was calculated as the divisor, > when the correct value should have been 0x1A. The latter being confirmed > by both the application note (AN120 from FTDI [1]) and the > 'ftdi_set_baudrate()' in libftdi. On the other side, the > 'ftdi_232bm_baud_to_divisor()', used for the FT232BM, FT2232C and > FT232RL appears to also return the correct divisor for FT232H. Up to 3.000.000 Baud, FT232BM, FT2232C, FT232RL, FT2232H, FT4232H and FT232H (should) calculate the baud rate the same way. > Basically, we have 2 issues that aren't really clear: > 1) Is the FT232H placed in the correct branch when calculating the > divisor? There are some issues withe the 2/3232H baud rate calculation, see the archive of the list > 2) What't the explanation behind the algorithm which calculates the > divisor in function "ftdi_2232h_baud_base_to_divisor()" ? For the FT232R, the data sheet states for baud rate calculation: Baud Rate Generator - The Baud Rate Generator provides a 16x clock input to the UART Controller from the 48MHz reference clock. It consists of a 14 bit pre-scaler and 3 register bits which provide fine tuning of the baud rate (used to divide by a number plus a fraction or a "sub-integer"). This determines the baud rate of the UART, which is programmable from 183 baud to 3 Mbaud. The FT232R supports all standard baud rates and non-standard baud rates from 183 Baud up to 3 Mbaud. Achievable non-standard baud rates are calculated as follows - Baud Rate = 3000000 / (n + x) where 'n' can be any integer between 2 and 16,384 ( = 2 ) and 'x' can be a sub-integer of the value 0, 14 0.125, 0.25, 0.375, 0.5, 0.625, 0.75, or 0.875. When n = 1, x = 0, i.e. baud rate divisors with values between 1 and 2 are not possible. The FT2/4232H has: Independent Baud Rate Generators - The Rate Generators provides a x16 or a x10 clock input to the UART'ss from a 120MHz reference clock and consists of a 14 bit pre-scaler and 4 register bits which provide fine tuning of the baud rate (used to divide by a number plus a fraction). This determines the Baud Rate of the UART which is programmable from 183 baud to 12 million baud. The FT2232H does not support the baud rates of 7 Mbaud 9 Mbaud, 10 Mbaud and 11 Mbaud. And the FT232H: Baud Rate Generators - The Baud Rate Generators provides a x16 or a x10 clock input to the UART's from a 120MHz reference clock and consists of a 14 bit pre-scaler and 4 register bits which provide fine tuning of the baud rate (used to divide by a number plus a fraction). This determines the Baud Rate of the UART which is programmable from 183 baud to 12 Mbaud. See FTDI application note AN_120 on the FTDI website for more details. This makes me confident with placing the FT232H into the FT2232H/FT4232H baud rate calculation path. I guess there is still something fishy with the Baud rate calculation... Bye -- Uwe Bonnes bon@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- -- 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
