Skip to contentSkip to conversation

net: phy: spi_ks8995: Do not overwrite SPI mode flags

PATCH In review Sven Van Asbroeck 3 participants 6 messages Nov 10, 2020 – Nov 11, 2020

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 net: phy: spi_ks8995: Do not overwrite SPI mode flags

From: Sven Van Asbroeck <[email protected]>

This driver makes sure the underlying SPI bus is set to "mode 0" by assigning SPI_MODE_0 to spi->mode. Which overwrites all other SPI mode flags.

In some circumstances, this can break the underlying SPI bus driver. For example, if SPI_CS_HIGH is set on the SPI bus, the driver will clear that flag, which results in a chip-select polarity issue.

Fix by changing only the SPI_MODE_N bits, i.e. SPI_CPHA and SPI_CPOL.

Fixes a8e510f682fe ("phy: Micrel KS8995MA 5-ports 10/100 managed Ethernet switch support added")
Fixes f3186dd87669 ("spi: Optionally use GPIO descriptors for CS GPIOs")

---

Tree: git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git # 989ef49bdf10

To be followed by a proposed spi helper function. Submit to net-next after net gets merged into net-next.

# large number of people, from get_maintainer.pl
To: Andrew Lunn <[email protected]>
To: Heiner Kallweit <[email protected]>
To: Jakub Kicinski <[email protected]>
Cc: Russell King <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc Mark Brown <[email protected]>
Cc Linus Walleij <[email protected]>
Cc Andy Shevchenko <[email protected]>
Cc Frederic LAMBERT <[email protected]>
Cc Gabor Juhos <[email protected]>
Cc Jonathan Cameron <[email protected]>

# Cc SPI group, suggested by Jakub Kicinski
Cc: [email protected]

 drivers/net/phy/spi_ks8995.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/phy/spi_ks8995.c +3 -1
index 4b198399bfa2..3c6c87a09b03 100644
@@ -491,7 +491,9 @@ static int ks8995_probe(struct spi_device *spi)
 
 	spi_set_drvdata(spi, ks);
 
-	spi->mode = SPI_MODE_0;
+	/* use SPI_MODE_0 without changing any other mode flags */
+	spi->mode &= ~(SPI_CPHA | SPI_CPOL);
+	spi->mode |= SPI_MODE_0;
 	spi->bits_per_word = 8;
 	err = spi_setup(spi);
 	if (err) {
Signature · 1 line
2.17.1
#1 lore raw
Quoted 7 lines · Sven Van Asbroeck

From: Sven Van Asbroeck <[email protected]>

This driver makes sure the underlying SPI bus is set to "mode 0" by assigning SPI_MODE_0 to spi->mode. Which overwrites all other SPI mode flags.

Quoted 5 lines

In some circumstances, this can break the underlying SPI bus driver. For example, if SPI_CS_HIGH is set on the SPI bus, the driver will clear that flag, which results in a chip-select polarity issue.

Fix by changing only the SPI_MODE_N bits, i.e. SPI_CPHA and SPI_CPOL.

I see that this is a fix for backporing, but maybe you can send a patches on top of this to:
1) introduce
#define SPI_MODE_MASK (SPI_CPHA | SPI_CPOL)

drivers/net/phy/spi_ks8995.c static int ks8995_probe(struct spi_device *spi) · line 495

Quoted 2 lines
drivers/net/phy/spi_ks8995.c +2 -0
+       /* use SPI_MODE_0 without changing any other mode flags */
+       spi->mode &= ~(SPI_CPHA | SPI_CPOL);

2)

       spi->mode &= ~SPI_MODE_MASK;
Quoted 1 line

+ spi->mode |= SPI_MODE_0;

?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#2 lore raw

Hi Andy, thank you for the feedback.

Quoted 7 lines · Andy Shevchenko

I see that this is a fix for backporing, but maybe you can send a patches on top of this to:
1) introduce

 #define SPI_MODE_MASK  (SPI_CPHA | SPI_CPOL)
       spi->mode &= ~SPI_MODE_MASK;

Andrew Lunn suggested that a spi helper function would probably fit the bill. I am planning to submit that to net-next after this patch is accepted in next (and next is merged into net-next).

I am learning that net is only for the most minimal of fixes.

See the previous discussion here:
https://patchwork.ozlabs.org/project/netdev/patch/[email protected]/

#3 lore raw
In reply to Andy Shevchenko

PING Jakub

Quoted 8 lines · Andy Shevchenko

I see that this is a fix for backporing, but maybe you can send a patches on top of this to:
1) introduce

 #define SPI_MODE_MASK  (SPI_CPHA | SPI_CPOL)
       spi->mode &= ~SPI_MODE_MASK;
Quoted 1 line

+ spi->mode |= SPI_MODE_0;

Jakub,

Is it possible to merge Andy's suggestion into net? Or should this go into net-next?

Thank you,
Sven

maintainer
#4 lore raw
Quoted 10 lines · Sven Van Asbroeck

PING Jakub

Quoted 6 lines · Andy Shevchenko

I see that this is a fix for backporing, but maybe you can send a patches on top of this to:
1) introduce

 #define SPI_MODE_MASK  (SPI_CPHA | SPI_CPOL)
       spi->mode &= ~SPI_MODE_MASK;
Quoted 6 lines
Quoted 2 lines
Quoted 1 line

+ spi->mode |= SPI_MODE_0;

Jakub,

Is it possible to merge Andy's suggestion into net? Or should this go into net-next?

I would keep with the minimal fix for the moment, it keeps the dependencies simple.

When you add a helper, it should really be somewhere in the SPI code, not the net code. So we need both the SPI and the net maintainers to cooperate to get the helper merged, and then this driver using the helper.

Andrew

#5 lore raw
In reply to Sven Van Asbroeck

A spi core fix has been accepted which makes this patch unnecessary.

https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git/commit/?id=766c6b63aa044e84b045803b40b14754d69a2a1d

Quoted 17 lines · Sven Van Asbroeck

This driver makes sure the underlying SPI bus is set to "mode 0" by assigning SPI_MODE_0 to spi->mode. Which overwrites all other SPI mode flags.

In some circumstances, this can break the underlying SPI bus driver. For example, if SPI_CS_HIGH is set on the SPI bus, the driver will clear that flag, which results in a chip-select polarity issue.

Fix by changing only the SPI_MODE_N bits, i.e. SPI_CPHA and SPI_CPOL.

Fixes a8e510f682fe ("phy: Micrel KS8995MA 5-ports 10/100 managed Ethernet switch support added")
Fixes f3186dd87669 ("spi: Optionally use GPIO descriptors for CS GPIOs")

---