All of lore.kernel.org
 help / color / mirror / Atom feed
From: Johan Hovold <johan@kernel.org>
To: Michael Hanselmann <public@hansmi.ch>
Cc: linux-usb@vger.kernel.org, Johan Hovold <johan@kernel.org>,
	Michael Dreher <michael@5dot1.de>,
	Jonathan Olds <jontio@i4free.co.nz>
Subject: Re: [PATCH v2 3/6] USB: serial: ch341: Limit prescaler on quirky chips
Date: Thu, 14 May 2020 16:17:43 +0200	[thread overview]
Message-ID: <20200514141743.GE25962@localhost> (raw)
In-Reply-To: <ee0499594a75018c0e31da4af9feb8a322ff448d.1585697281.git.public@hansmi.ch>

On Tue, Mar 31, 2020 at 11:37:19PM +0000, Michael Hanselmann wrote:
> A subset of all CH341 devices stop responding to bulk transfers, usually
> after the third byte, when the highest prescaler bit (0b100) is set.
> There is one exception, namely a prescaler of exactly 0b111 (fact=1,
> ps=3). Limit these to the lower prescaler values at the cost of timing
> precision.
> 
> Rates above 46875 baud use the same timings as the normal algorithm.
> Below that the worst difference between desired and actual baud rate is
> 2.17 percentage points. The worst difference is 1.06 p.p. when only
> looking at divisors differing from the normal algorithm.
> 
> Signed-off-by: Michael Hanselmann <public@hansmi.ch>
> ---
>  drivers/usb/serial/ch341.c | 70 ++++++++++++++++++++++++++++----------
>  1 file changed, 52 insertions(+), 18 deletions(-)
> 
> diff --git a/drivers/usb/serial/ch341.c b/drivers/usb/serial/ch341.c
> index 9c839f67c3d4..67a5d4c3df42 100644
> --- a/drivers/usb/serial/ch341.c
> +++ b/drivers/usb/serial/ch341.c
> @@ -24,6 +24,8 @@
>  #define DEFAULT_BAUD_RATE 9600
>  #define DEFAULT_TIMEOUT   1000
>  
> +#define CH341_QUIRK_LIMITED_PRESCALER 0x01
> +
>  /* flags for IO-Bits */
>  #define CH341_BIT_RTS (1 << 6)
>  #define CH341_BIT_DTR (1 << 5)
> @@ -195,35 +197,67 @@ static int ch341_calc_divisor(speed_t speed, unsigned int ps,
>   *		2 <= div <= 256 if fact = 0, or
>   *		9 <= div <= 256 if fact = 1
>   */
> -static int ch341_get_divisor(speed_t speed)
> +static int ch341_get_divisor(struct ch341_private *priv)
>  {
> +	speed_t speed;
>  	unsigned int fact, div, clk_div;
>  	int ps;
>  
> +	speed = priv->baud_rate;
> +
>  	/*
>  	 * Clamp to supported range, this makes the (ps < 0) and (div < 2)
>  	 * sanity checks below redundant.
>  	 */
>  	speed = clamp(speed, 46U, 3000000U);
>  
> -	/*
> -	 * Start with highest possible base clock and find a divisor for the
> -	 * requested baud rate.
> -	 */
> -	for (ps = 3; ps >= 0; --ps) {
> -		if (ch341_calc_divisor(speed, ps, 1U, &div, &clk_div) == 0) {
> -			fact = 1;
> -			break;
> -		}
> -
> +	if (priv->quirks & CH341_QUIRK_LIMITED_PRESCALER) {
>  		/*
> -		 * Prefer half base clock (fact = 0) before trying lower
> -		 * prescaler values. This makes the receiver more tolerant to
> -		 * errors.
> +		 * A subset of all CH34x devices stop responding to bulk
> +		 * transfers when configured with certain prescaler values.
> +		 *
> +		 * fact=0, ps=0..3: Works
> +		 * fact=1, ps=0..2: Unreliable
> +		 * fact=1, ps=3: Works
> +		 *
> +		 * Limit these devices to working prescaler values at the cost
> +		 * of precision for speeds up to 46875 baud above which
> +		 * fact = 1 with ps = 3 is used.
>  		 */
> -		if (ch341_calc_divisor(speed, ps, 0U, &div, &clk_div) == 0) {
> +		if (ch341_calc_divisor(speed, 3U, 1U, &div, &clk_div) == 0) {
> +			ps = 3U;
> +			fact = 1U;
> +		} else {
>  			fact = 0;
> -			break;
> +
> +			for (ps = 3; ps >= 0; --ps) {
> +				if (ch341_calc_divisor(speed, ps, fact,
> +						       &div, &clk_div) == 0)
> +					break;
> +			}
> +		}
> +	} else {
> +		/*
> +		 * Start with highest possible base clock and find a divisor
> +		 * for the requested baud rate.
> +		 */
> +		for (ps = 3; ps >= 0; --ps) {
> +			if (ch341_calc_divisor(speed, ps, 1U,
> +					       &div, &clk_div) == 0) {
> +				fact = 1;
> +				break;
> +			}
> +
> +			/*
> +			 * Prefer half base clock (fact = 0) before trying
> +			 * lower prescaler values. This makes the receiver more
> +			 * tolerant to errors.
> +			 */
> +			if (ch341_calc_divisor(speed, ps, 0U,
> +					       &div, &clk_div) == 0) {
> +				fact = 0;
> +				break;
> +			}
>  		}
>  	}
>  
> @@ -250,7 +284,7 @@ static int ch341_set_baudrate_lcr(struct usb_device *dev,
>  	if (!priv->baud_rate)
>  		return -EINVAL;
>  
> -	val = ch341_get_divisor(priv->baud_rate);
> +	val = ch341_get_divisor(priv);
>  	if (val < 0)
>  		return -EINVAL;
>  
> @@ -353,7 +387,7 @@ static int ch341_detect_quirks(struct usb_device *dev)
>  	if (r == -EPIPE) {
>  		dev_dbg(&dev->dev, "%s - reading break condition register"
>  			" failed (%d)\n", __func__, r);
> -		r = 0;
> +		r = CH341_QUIRK_LIMITED_PRESCALER;
>  		goto out;
>  	}

This can be implemented in a more compact way which makes the algorithm
easier to follow and doesn't depend on your patch 1/6 either. With a
smaller fix we can get this backported to stable as well.

I've verified that the below works here on top of your (slightly
modified) 2/6. Would you mind giving it a try on your end as well?

Johan


From ea23730616b6406101d4efeb12e1ae2312dd20c9 Mon Sep 17 00:00:00 2001
From: Johan Hovold <johan@kernel.org>
Date: Thu, 14 May 2020 11:36:45 +0200
Subject: [PATCH] USB: serial: ch341: fix lockup of devices with limited
 prescaler

Michael Hanselmann reports that

	[a] subset of all CH341 devices stop responding to bulk
	transfers, usually after the third byte, when the highest
	prescaler bit (0b100) is set. There is one exception, namely a
	prescaler of exactly 0b111 (fact=1, ps=3).

Fix this by forcing a lower base clock (fact = 0) whenever needed.

This specifically makes the standard rates 110, 134 and 200 bps work
again with these devices.

Fixes: 35714565089e ("USB: serial: ch341: reimplement line-speed handling")
Cc: stable <stable@vger.kernel.org>	# 5.5
Reported-by: Michael Hanselmann <public@hansmi.ch>
Signed-off-by: Johan Hovold <johan@kernel.org>
---
 drivers/usb/serial/ch341.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/serial/ch341.c b/drivers/usb/serial/ch341.c
index f2b93f4554a7..89675ee29645 100644
--- a/drivers/usb/serial/ch341.c
+++ b/drivers/usb/serial/ch341.c
@@ -73,6 +73,8 @@
 #define CH341_LCR_CS6          0x01
 #define CH341_LCR_CS5          0x00
 
+#define CH341_QUIRK_LIMITED_PRESCALER	BIT(0)
+
 static const struct usb_device_id id_table[] = {
 	{ USB_DEVICE(0x4348, 0x5523) },
 	{ USB_DEVICE(0x1a86, 0x7523) },
@@ -160,9 +162,11 @@ static const speed_t ch341_min_rates[] = {
  *		2 <= div <= 256 if fact = 0, or
  *		9 <= div <= 256 if fact = 1
  */
-static int ch341_get_divisor(speed_t speed)
+static int ch341_get_divisor(struct ch341_private *priv)
 {
 	unsigned int fact, div, clk_div;
+	speed_t speed = priv->baud_rate;
+	bool force_fact0 = false;
 	int ps;
 
 	/*
@@ -188,8 +192,12 @@ static int ch341_get_divisor(speed_t speed)
 	clk_div = CH341_CLK_DIV(ps, fact);
 	div = CH341_CLKRATE / (clk_div * speed);
 
+	/* Some devices require a lower base clock if ps < 3. */
+	if (ps < 3 && (priv->quirks & CH341_QUIRK_LIMITED_PRESCALER))
+		force_fact0 = true;
+
 	/* Halve base clock (fact = 0) if required. */
-	if (div < 9 || div > 255) {
+	if (div < 9 || div > 255 || force_fact0) {
 		div /= 2;
 		clk_div *= 2;
 		fact = 0;
@@ -228,7 +236,7 @@ static int ch341_set_baudrate_lcr(struct usb_device *dev,
 	if (!priv->baud_rate)
 		return -EINVAL;
 
-	val = ch341_get_divisor(priv->baud_rate);
+	val = ch341_get_divisor(priv);
 	if (val < 0)
 		return -EINVAL;
 
@@ -333,6 +341,7 @@ static int ch341_detect_quirks(struct usb_serial_port *port)
 			    CH341_REG_BREAK, 0, buffer, size, DEFAULT_TIMEOUT);
 	if (r == -EPIPE) {
 		dev_dbg(&port->dev, "break control not supported\n");
+		quirks = CH341_QUIRK_LIMITED_PRESCALER;
 		r = 0;
 		goto out;
 	}
-- 
2.26.2



  reply	other threads:[~2020-05-14 14:17 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-31 23:37 [PATCH v2 0/6] USB: serial: ch341: Add support for limited chips (was: Add support for HL340 devices) Michael Hanselmann
2020-03-31 23:37 ` [PATCH v2 1/6] USB: serial: ch341: Reduce special cases in clock calculation Michael Hanselmann
2020-03-31 23:37 ` [PATCH v2 2/6] USB: serial: ch341: Add basis for quirk detection Michael Hanselmann
2020-05-14 14:09   ` Johan Hovold
2020-03-31 23:37 ` [PATCH v2 3/6] USB: serial: ch341: Limit prescaler on quirky chips Michael Hanselmann
2020-05-14 14:17   ` Johan Hovold [this message]
2020-05-27 13:16     ` Johan Hovold
2020-05-27 15:41       ` Michael Hanselmann
2020-05-29  7:15         ` Johan Hovold
2020-03-31 23:37 ` [PATCH v2 4/6] USB: serial: ch341: Name prescaler, divisor registers Michael Hanselmann
2020-05-14 14:24   ` Johan Hovold
2020-05-27 20:59     ` Michael Hanselmann
2020-06-29  9:51       ` Johan Hovold
2020-03-31 23:37 ` [PATCH v2 5/6] USB: serial: ch341: Compute minimum baud rate Michael Hanselmann
2020-05-27 22:19   ` Michael Hanselmann
2020-06-30  9:57     ` Johan Hovold
2020-03-31 23:37 ` [PATCH v2 6/6] USB: serial: ch341: Simulate break condition if not supported Michael Hanselmann
2020-05-14 14:47   ` Johan Hovold
2020-05-27 22:21     ` Michael Hanselmann
2020-06-30 11:39       ` Johan Hovold
2020-07-04 18:25         ` Michael Hanselmann
2020-07-06  9:31           ` Johan Hovold
2020-05-14 14:02 ` [PATCH v2 0/6] USB: serial: ch341: Add support for limited chips (was: Add support for HL340 devices) Johan Hovold

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200514141743.GE25962@localhost \
    --to=johan@kernel.org \
    --cc=jontio@i4free.co.nz \
    --cc=linux-usb@vger.kernel.org \
    --cc=michael@5dot1.de \
    --cc=public@hansmi.ch \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.