Skip to contentSkip to conversation

spi_mpc83xx: fix prescale modulus calculation

PATCH In review Anton Vorontsov 2 participants 3 messages Aug 6, 2007 2 messages not in the archive SPI

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 spi_mpc83xx: fix prescale modulus calculation

Long ago I've noticed (but didn't pay much attention) that spi_mpc83xx using PM calculations that differs from what specs describe. I.e.

u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 4);

While specs says: "The SPI baud rate generator clock source (either system clock or system clock divided by 16, depending on DIV16 bit) is divided by 4 * ([PM] + 1), a range from 4 to 64.".

Thus " - 1" is missing in the spi_mpc83xx's formula.

Why nobody noticed that bug? Probably because sysclk usually less then user expects, e.g. you expect 200 MHz, but real clock is 198 MHz, and integer rounding helps when this formula is used.

Suppose it's SPI in QE, SYSCLK at 198 MHz, thus SPIBRG at 99MHz, 25 MHz requested.

PM = (99MHz / ( 25 MHz * 4 )), PM == 0, output SPICLK will be 24.75 MHz

At lower frequencies this bug is more noticeable, though.

And this bug shows itself in all its beauty if SYSCLK is equal or a bit more than you expect (200 MHz SYSCLK, 100 MHz SPIBRG):
PM = (100MHz / ( 25 MHz * 4 )), PM == 1, output SPICLK will be 12.625 MHz!

---
 drivers/spi/spi_mpc83xx.c |   10 +++++++++-
 1 files changed, 9 insertions(+), 1 deletions(-)
drivers/spi/spi_mpc83xx.c +9 -1
index 446b624..042704b 100644
@@ -170,7 +170,8 @@ static void mpc83xx_spi_chipselect(struct spi_device *spi, int value)
 		regval |= SPMODE_LEN(len);
 
 		if ((mpc83xx_spi->spibrg / spi->max_speed_hz) >= 64) {
-			u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 64);
+			u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 64) - 1;
+
 			if (pm > 0x0f) {
 				dev_warn(&spi->dev, "Requested speed is too "
 					"low: %d Hz. Will use %d Hz instead.\n",
@@ -180,6 +181,13 @@ static void mpc83xx_spi_chipselect(struct spi_device *spi, int value)
 			regval |= SPMODE_PM(pm) | SPMODE_DIV16;
 		} else {
 			u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 4);
+
+			if (pm)
+				pm--;
+			else /* this floods dmesg if using mmc_spi, so dbg */
+				dev_dbg(&spi->dev, "Requested speed is too "
+					"high: %d Hz. Will use %d Hz instead.\n",
+					spi->max_speed_hz, mpc83xx_spi->spibrg / 4);
 			regval |= SPMODE_PM(pm);
 		}
 
Signature · 7 lines
1.5.0.6

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

A message that is not in the archive (sent off-list or not yet received): 20070806131014.GA6913-bi+AKbBUZKY6gyzm1THtWbp2dZbC/[email protected]

#2 lore raw
Quoted 7 lines · Anton Vorontsov

+
+                       if (pm)
+                               pm--;
+                       else /* this floods dmesg if using mmc_spi, so dbg */ +                               dev_dbg(&spi->dev, "Requested speed is too " +                                       "high: %d Hz. Will use %d Hz instead.\n", +                                       spi->max_speed_hz, mpc83xx_spi->spibrg / 4);

Except that's not worthy of any message whatsoever, so it shouldn't even be a debug message: user says "give me at most X", driver says "fine", end of story.

I'm more bothered by the earlier message, which gives a rate higher than X instead of cleanly failing.

- Dave

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/

A message that is not in the archive (sent off-list or not yet received): 200708060844.40310.david-b-yBeKhBN/[email protected]

#4 lore raw

Re: [PATCH] [SPI][POWERPC] spi_mpc83xx: fix prescale modulus?calculation

Quoted 15 lines · David Brownell
Quoted 7 lines · Anton Vorontsov

+
+                       if (pm)
+                               pm--;
+                       else /* this floods dmesg if using mmc_spi, so dbg */ +                               dev_dbg(&spi->dev, "Requested speed is too " +                                       "high: %d Hz. Will use %d Hz instead.\n", +                                       spi->max_speed_hz, mpc83xx_spi->spibrg / 4);

Except that's not worthy of any message whatsoever, so it shouldn't even be a debug message: user says "give me at most X", driver says "fine", end of story.

I'm more bothered by the earlier message, which gives a rate higher than X instead of cleanly failing.

[...]

Quoted 1 line

So the patch should be more like the appended ...

Thanks! Reposting here.

- - - -
From: Anton Vorontsov <[email protected]>
Subject: spi_mpc83xx: fix prescale modulus calculation

Long ago I've noticed (but didn't pay much attention) that spi_mpc83xx using PM calculations that differs from what specs describe. I.e.

u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 4);

While specs says: "The SPI baud rate generator clock source (either system clock or system clock divided by 16, depending on DIV16 bit) is divided by 4 * ([PM] + 1), a range from 4 to 64.".

Thus " - 1" is missing in the spi_mpc83xx's formula.

Why nobody noticed that bug? Probably because sysclk usually less then user expects, e.g. you expect 200 MHz, but real clock is 198 MHz, and integer rounding helps when this formula is used.

Suppose it's SPI in QE, SYSCLK at 198 MHz, thus SPIBRG at 99MHz, 25 MHz requested.

PM = (99MHz / ( 25 MHz * 4 )), PM == 0, output SPICLK will be 24.75 MHz

At lower frequencies this bug is more noticeable, though.

And this bug shows itself in all its beauty if SYSCLK is equal or a bit more than you expect (200 MHz SYSCLK, 100 MHz SPIBRG):
PM = (100MHz / ( 25 MHz * 4 )), PM == 1, output SPICLK will be 12.625 MHz!

---
 drivers/spi/spi_mpc83xx.c |    8 ++++++--
 1 files changed, 6 insertions(+), 2 deletions(-)
drivers/spi/spi_mpc83xx.c +6 -2
index fe69e94..2adf856 100644
@@ -148,6 +148,8 @@ static void mpc83xx_spi_chipselect(struct spi_device *spi, int value)
 	if (value == BITBANG_CS_ACTIVE) {
 		u32 regval = mpc83xx_spi_read_reg(&mpc83xx_spi->base->mode);
 		u32 len = spi->bits_per_word;
+		u8 pm;
+
 		if (len == 32)
 			len = 0;
 		else
@@ -170,7 +172,7 @@ static void mpc83xx_spi_chipselect(struct spi_device *spi, int value)
 		regval |= SPMODE_LEN(len);
 
 		if ((mpc83xx_spi->spibrg / spi->max_speed_hz) >= 64) {
-			u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 64);
+			pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 64) - 1;
 			if (pm > 0x0f) {
 				dev_err(&spi->dev, "Requested speed is too "
 					"low: %d Hz. Will use %d Hz instead.\n",
@@ -180,7 +182,9 @@ static void mpc83xx_spi_chipselect(struct spi_device *spi, int value)
 			}
 			regval |= SPMODE_PM(pm) | SPMODE_DIV16;
 		} else {
-			u8 pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 4);
+			pm = mpc83xx_spi->spibrg / (spi->max_speed_hz * 4);
+			if (pm)
+				pm--;
 			regval |= SPMODE_PM(pm);
 		}
 
Signature · 12 lines
1.5.0.6


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
spi-devel-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spi-devel-general