Re: [Patch/resend]drivers/usb/serial/ftdi_sio.c: For the FT232H FTDI_SIO_SET_BAUDRATE_REQUEST, index needs to be shifted too
From: Sergei Shtylyov <hidden>
Date: 2011-07-22 12:21:22
Hello. On 22-07-2011 12:41, Uwe Bonnes wrote:
the recent addition of the FT232H showed that baudrate was set wrong. See gmane.linux.usb.general: "[ftdi_sio] FT232H support". With the old code, the MSB of the 4 encoded fractional divider bits and more important the clock predivider bits got lost. Adding the FT232H to the code patch were these bits are shifted solves the problem. I verified baud rates with a scope now.
I suspect, that the BM device probably needs these bits shifted too. But there is no predivider bit, so this is not obvious, and a missing MSB of the encoded fractional divider only shifts the resulting baudrate minimal. The AM has only 3 bits of encoded fractional divider, so it is not impacted.
I have no BM device to test, so I only added a comment and left the code for the BM untouched.
This time with checkpatch happy.
Bye
-- Uwe Bonnes bon@elektron.ikp.physik.tu-darmstadt.de
Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- From 892f1ac2b0b47ffc7adf70ecbe5d7d2f31c79232 Mon Sep 17 00:00:00 2001 From: Uwe Bonnes <redacted> Date: Tue, 12 Jul 2011 18:47:08 +0200 Subject: ftdi_sio: For the FT232H FTDI_SIO_SET_BAUDRATE_REQUEST, index needs to be shifted too
--- drivers/usb/serial/ftdi_sio.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-)
quoted hunk ↗ jump to hunk
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 2e06b90..0692a36 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c@@ -1205,7 +1205,10 @@ static int change_speed(struct tty_struct *tty, struct usb_serial_port *port) urb_index_value = get_ftdi_divisor(tty, port); urb_value = (__u16)urb_index_value; urb_index = (__u16)(urb_index_value >> 16); - if (priv->interface) { /* FT2232C */ + if ((priv->chip_type == FT2232C) || (priv->chip_type == FT2232H) || + (priv->chip_type == FT4232H) || (priv->chip_type == FT232H)) {
This rather asks for a *switch* statement.
+ /* Probably the BM type needs the MSB of the encoded fractional + * divider also moved like for the chips above. Any infos? */
The preferred comment style:
/*
* bla
* bla
*/
urb_index = (__u16)((urb_index << 8) | priv->interface); } -- 1.7.3.4 Signed-off-by: Uwe Bonnes <redacted>
Signoff should precede the patch, not follow it.
WBR, Sergei