[PATCH v2 0/7] spi: zynq-qspi: Clarify and fix the chip selection

[PATCH v2 0/7] spi: zynq-qspi: Clarify and fix the chip selection

[Miquel Raynal]

Hello,

While working on SPI-NOR support I figured the chip select handling of
the Zynq 7000 QSPI driver was not behaving as I would have
expected. While cheking out what was wrong I decided to clarify things
around so I did a bit of cleaning. The step-by-step changes are the
reason for patches [1-6]. This way, the last patch actually adding
support for both CS is much more understandable.

Thanks,
Miquèl

Changes since v1:
* Rebased on top of v5.4-rc6 and fixed all the conflicts
  (minors). Compile tested only (actual testing done on a v5.1).

Miquel Raynal (7):
  spi: zynq-qspi: Anything else than CS0 is not supported yet
  spi: zynq-qspi: Keep the naming consistent across the driver
  spi: zynq-qspi: Keep the bitfields naming consistent
  spi: zynq-qspi: Enhance the Linear CFG bit definitions
  spi: zynq-qspi: Clarify the select chip function
  spi: zynq-qspi: Do the actual hardware initialization later in the
    probe
  spi: zynq-qspi: Support two chip selects

 drivers/spi/spi-zynq-qspi.c | 83 ++++++++++++++++++++++---------------
 1 file changed, 50 insertions(+), 33 deletions(-)

-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 1/7] spi: zynq-qspi: Anything else than CS0 is not supported yet

[Miquel Raynal]

Unlike what the driver is currently advertizing, CS0 only can be used,
CS1 is not supported at all. Prevent people to use CS1.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 5cf6993ddce5..713f9a03a40a 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -681,10 +681,14 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 
 	ret = of_property_read_u32(np, "num-cs",
 				   &num_cs);
-	if (ret < 0)
+	if (ret < 0) {
 		ctlr->num_chipselect = ZYNQ_QSPI_DEFAULT_NUM_CS;
-	else
+	} else if (num_cs > ZYNQ_QSPI_DEFAULT_NUM_CS) {
+		dev_err(&pdev->dev, "anything but CS0 is not yet supported\n");
+		goto remove_master;
+	} else {
 		ctlr->num_chipselect = num_cs;
+	}
 
 	ctlr->mode_bits =  SPI_RX_DUAL | SPI_RX_QUAD |
 			    SPI_TX_DUAL | SPI_TX_QUAD;
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Anything else than CS0 is not supported yet" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Anything else than CS0 is not supported yet

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From 087622d09472f96f1f5d6ced36ca75c92e86af21 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:38 +0100
Subject: [PATCH] spi: zynq-qspi: Anything else than CS0 is not supported yet

Unlike what the driver is currently advertizing, CS0 only can be used,
CS1 is not supported at all. Prevent people to use CS1.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-2-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index b1c56e9d7c94..9f53ea08adf7 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -680,10 +680,14 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 
 	ret = of_property_read_u32(np, "num-cs",
 				   &num_cs);
-	if (ret < 0)
+	if (ret < 0) {
 		ctlr->num_chipselect = ZYNQ_QSPI_DEFAULT_NUM_CS;
-	else
+	} else if (num_cs > ZYNQ_QSPI_DEFAULT_NUM_CS) {
+		dev_err(&pdev->dev, "anything but CS0 is not yet supported\n");
+		goto remove_master;
+	} else {
 		ctlr->num_chipselect = num_cs;
+	}
 
 	ctlr->mode_bits =  SPI_RX_DUAL | SPI_RX_QUAD |
 			    SPI_TX_DUAL | SPI_TX_QUAD;
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 2/7] spi: zynq-qspi: Keep the naming consistent across the driver

[Miquel Raynal]

In this driver (and also in a lot of other drivers in drivers/spi/),
the spi_controller structure is sometimes referred as 'ctlr' and
sometimes as 'ctrl'. Grepping there shows that 'ctlr' seems to be more
common so keep the naming consistent in this driver and s/ctrl/ctlr/.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 713f9a03a40a..641691f5dedf 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -284,8 +284,8 @@ static void zynq_qspi_txfifo_op(struct zynq_qspi *xqspi, unsigned int size)
  */
 static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 {
-	struct spi_controller *ctrl = spi->master;
-	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctrl);
+	struct spi_controller *ctlr = spi->master;
+	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctlr);
 	u32 config_reg;
 
 	config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_CONFIG_OFFSET);
@@ -365,10 +365,10 @@ static int zynq_qspi_config_op(struct zynq_qspi *xqspi, struct spi_device *spi)
  */
 static int zynq_qspi_setup_op(struct spi_device *spi)
 {
-	struct spi_controller *ctrl = spi->master;
-	struct zynq_qspi *qspi = spi_controller_get_devdata(ctrl);
+	struct spi_controller *ctlr = spi->master;
+	struct zynq_qspi *qspi = spi_controller_get_devdata(ctlr);
 
-	if (ctrl->busy)
+	if (ctlr->busy)
 		return -EBUSY;
 
 	clk_enable(qspi->refclk);
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 3/7] spi: zynq-qspi: Keep the bitfields naming consistent

[Miquel Raynal]

Most of the bits/bitfields #define'd in this driver are composed with:
1/ the driver prefix
2/ the name of the register they apply to

Keep the naming consistent by applying this rule to the CONFIG register
internals. These definitions will be used in a following change set.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 641691f5dedf..78711fe955f4 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -61,9 +61,9 @@
  * These are the values used in the calculation of baud rate divisor and
  * setting the slave select.
  */
-#define ZYNQ_QSPI_BAUD_DIV_MAX		GENMASK(2, 0) /* Baud rate maximum */
-#define ZYNQ_QSPI_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift in CR */
-#define ZYNQ_QSPI_SS_SHIFT		10 /* Slave Select field shift in CR */
+#define ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX	GENMASK(2, 0) /* Baud rate maximum */
+#define ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift */
+#define ZYNQ_QSPI_CONFIG_PCS		10 /* Peripheral Chip Select */
 
 /*
  * QSPI Interrupt Registers bit Masks
@@ -293,7 +293,7 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 		/* Select the slave */
 		config_reg &= ~ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
 		config_reg |= (((~(BIT(spi->chip_select))) <<
-				ZYNQ_QSPI_SS_SHIFT) &
+				ZYNQ_QSPI_CONFIG_PCS) &
 				ZYNQ_QSPI_CONFIG_SSCTRL_MASK);
 	} else {
 		config_reg |= ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
@@ -332,7 +332,7 @@ static int zynq_qspi_config_op(struct zynq_qspi *xqspi, struct spi_device *spi)
 	 *      ----------------
 	 *      111 - divide by 256
 	 */
-	while ((baud_rate_val < ZYNQ_QSPI_BAUD_DIV_MAX)  &&
+	while ((baud_rate_val < ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX)  &&
 	       (clk_get_rate(xqspi->refclk) / (2 << baud_rate_val)) >
 		spi->max_speed_hz)
 		baud_rate_val++;
@@ -348,7 +348,7 @@ static int zynq_qspi_config_op(struct zynq_qspi *xqspi, struct spi_device *spi)
 		config_reg |= ZYNQ_QSPI_CONFIG_CPOL_MASK;
 
 	config_reg &= ~ZYNQ_QSPI_CONFIG_BDRATE_MASK;
-	config_reg |= (baud_rate_val << ZYNQ_QSPI_BAUD_DIV_SHIFT);
+	config_reg |= (baud_rate_val << ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT);
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_CONFIG_OFFSET, config_reg);
 
 	return 0;
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Keep the bitfields naming consistent" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Keep the bitfields naming consistent

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From 941be723735113028361c4664cd10134d3d27006 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:40 +0100
Subject: [PATCH] spi: zynq-qspi: Keep the bitfields naming consistent

Most of the bits/bitfields #define'd in this driver are composed with:
1/ the driver prefix
2/ the name of the register they apply to

Keep the naming consistent by applying this rule to the CONFIG register
internals. These definitions will be used in a following change set.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-4-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 9f53ea08adf7..530533b6c974 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -60,9 +60,9 @@
  * These are the values used in the calculation of baud rate divisor and
  * setting the slave select.
  */
-#define ZYNQ_QSPI_BAUD_DIV_MAX		GENMASK(2, 0) /* Baud rate maximum */
-#define ZYNQ_QSPI_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift in CR */
-#define ZYNQ_QSPI_SS_SHIFT		10 /* Slave Select field shift in CR */
+#define ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX	GENMASK(2, 0) /* Baud rate maximum */
+#define ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift */
+#define ZYNQ_QSPI_CONFIG_PCS		10 /* Peripheral Chip Select */
 
 /*
  * QSPI Interrupt Registers bit Masks
@@ -292,7 +292,7 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 		/* Select the slave */
 		config_reg &= ~ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
 		config_reg |= (((~(BIT(spi->chip_select))) <<
-				ZYNQ_QSPI_SS_SHIFT) &
+				ZYNQ_QSPI_CONFIG_PCS) &
 				ZYNQ_QSPI_CONFIG_SSCTRL_MASK);
 	} else {
 		config_reg |= ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
@@ -331,7 +331,7 @@ static int zynq_qspi_config_op(struct zynq_qspi *xqspi, struct spi_device *spi)
 	 *      ----------------
 	 *      111 - divide by 256
 	 */
-	while ((baud_rate_val < ZYNQ_QSPI_BAUD_DIV_MAX)  &&
+	while ((baud_rate_val < ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX)  &&
 	       (clk_get_rate(xqspi->refclk) / (2 << baud_rate_val)) >
 		spi->max_speed_hz)
 		baud_rate_val++;
@@ -347,7 +347,7 @@ static int zynq_qspi_config_op(struct zynq_qspi *xqspi, struct spi_device *spi)
 		config_reg |= ZYNQ_QSPI_CONFIG_CPOL_MASK;
 
 	config_reg &= ~ZYNQ_QSPI_CONFIG_BDRATE_MASK;
-	config_reg |= (baud_rate_val << ZYNQ_QSPI_BAUD_DIV_SHIFT);
+	config_reg |= (baud_rate_val << ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT);
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_CONFIG_OFFSET, config_reg);
 
 	return 0;
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 4/7] spi: zynq-qspi: Enhance the Linear CFG bit definitions

[Miquel Raynal]

Using masks makes sense when manipulating fields of several bits. When
only one bit is involved, it is usual to just use the BIT() macro but
in this case using the term mask is abusive. Fix the #define macros
and their comments.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 78711fe955f4..1151443ec830 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -99,9 +99,9 @@
  * It is named Linear Configuration but it controls other modes when not in
  * linear mode also.
  */
-#define ZYNQ_QSPI_LCFG_TWO_MEM_MASK	BIT(30) /* LQSPI Two memories Mask */
-#define ZYNQ_QSPI_LCFG_SEP_BUS_MASK	BIT(29) /* LQSPI Separate bus Mask */
-#define ZYNQ_QSPI_LCFG_U_PAGE_MASK	BIT(28) /* LQSPI Upper Page Mask */
+#define ZYNQ_QSPI_LCFG_TWO_MEM		BIT(30) /* LQSPI Two memories */
+#define ZYNQ_QSPI_LCFG_SEP_BUS		BIT(29) /* LQSPI Separate bus */
+#define ZYNQ_QSPI_LCFG_U_PAGE		BIT(28) /* LQSPI Upper Page */
 
 #define ZYNQ_QSPI_LCFG_DUMMY_SHIFT	8
 
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Enhance the Linear CFG bit definitions" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Enhance the Linear CFG bit definitions

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From 044ac826200f08adebe3207a618d829279e9f3b1 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:41 +0100
Subject: [PATCH] spi: zynq-qspi: Enhance the Linear CFG bit definitions

Using masks makes sense when manipulating fields of several bits. When
only one bit is involved, it is usual to just use the BIT() macro but
in this case using the term mask is abusive. Fix the #define macros
and their comments.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-5-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 530533b6c974..70ecefd817f7 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -98,9 +98,9 @@
  * It is named Linear Configuration but it controls other modes when not in
  * linear mode also.
  */
-#define ZYNQ_QSPI_LCFG_TWO_MEM_MASK	BIT(30) /* LQSPI Two memories Mask */
-#define ZYNQ_QSPI_LCFG_SEP_BUS_MASK	BIT(29) /* LQSPI Separate bus Mask */
-#define ZYNQ_QSPI_LCFG_U_PAGE_MASK	BIT(28) /* LQSPI Upper Page Mask */
+#define ZYNQ_QSPI_LCFG_TWO_MEM		BIT(30) /* LQSPI Two memories */
+#define ZYNQ_QSPI_LCFG_SEP_BUS		BIT(29) /* LQSPI Separate bus */
+#define ZYNQ_QSPI_LCFG_U_PAGE		BIT(28) /* LQSPI Upper Page */
 
 #define ZYNQ_QSPI_LCFG_DUMMY_SHIFT	8
 
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 5/7] spi: zynq-qspi: Clarify the select chip function

[Miquel Raynal]

The code used to assert and de-assert a chip select line is very
complicated for no reason. Simplify the logic by either setting or
resetting the concerned bit, which actually only changes an electrical
state.

Update the comment to reflect that there is no possibility to actually
choose a CS as the default (CS0) will be driven in any case.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 1151443ec830..9adfcc66b02b 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -51,7 +51,6 @@
 #define ZYNQ_QSPI_CONFIG_BDRATE_MASK	GENMASK(5, 3) /* Baud Rate Mask */
 #define ZYNQ_QSPI_CONFIG_CPHA_MASK	BIT(2) /* Clock Phase Control */
 #define ZYNQ_QSPI_CONFIG_CPOL_MASK	BIT(1) /* Clock Polarity Control */
-#define ZYNQ_QSPI_CONFIG_SSCTRL_MASK	BIT(10) /* Slave Select Mask */
 #define ZYNQ_QSPI_CONFIG_FWIDTH_MASK	GENMASK(7, 6) /* FIFO width */
 #define ZYNQ_QSPI_CONFIG_MSTREN_MASK	BIT(0) /* Master Mode */
 
@@ -63,7 +62,7 @@
  */
 #define ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX	GENMASK(2, 0) /* Baud rate maximum */
 #define ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift */
-#define ZYNQ_QSPI_CONFIG_PCS		10 /* Peripheral Chip Select */
+#define ZYNQ_QSPI_CONFIG_PCS		BIT(10) /* Peripheral Chip Select */
 
 /*
  * QSPI Interrupt Registers bit Masks
@@ -288,16 +287,12 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctlr);
 	u32 config_reg;
 
+	/* Ground the line to assert the CS */
 	config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_CONFIG_OFFSET);
-	if (assert) {
-		/* Select the slave */
-		config_reg &= ~ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
-		config_reg |= (((~(BIT(spi->chip_select))) <<
-				ZYNQ_QSPI_CONFIG_PCS) &
-				ZYNQ_QSPI_CONFIG_SSCTRL_MASK);
-	} else {
-		config_reg |= ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
-	}
+	if (assert)
+		config_reg &= ~ZYNQ_QSPI_CONFIG_PCS;
+	else
+		config_reg |= ZYNQ_QSPI_CONFIG_PCS;
 
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_CONFIG_OFFSET, config_reg);
 }
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Clarify the select chip function" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Clarify the select chip function

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From dffaf7439979e6dbb820633bb4f44dcadcac847e Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:42 +0100
Subject: [PATCH] spi: zynq-qspi: Clarify the select chip function

The code used to assert and de-assert a chip select line is very
complicated for no reason. Simplify the logic by either setting or
resetting the concerned bit, which actually only changes an electrical
state.

Update the comment to reflect that there is no possibility to actually
choose a CS as the default (CS0) will be driven in any case.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-6-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 70ecefd817f7..80e51c894eaa 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -50,7 +50,6 @@
 #define ZYNQ_QSPI_CONFIG_BDRATE_MASK	GENMASK(5, 3) /* Baud Rate Mask */
 #define ZYNQ_QSPI_CONFIG_CPHA_MASK	BIT(2) /* Clock Phase Control */
 #define ZYNQ_QSPI_CONFIG_CPOL_MASK	BIT(1) /* Clock Polarity Control */
-#define ZYNQ_QSPI_CONFIG_SSCTRL_MASK	BIT(10) /* Slave Select Mask */
 #define ZYNQ_QSPI_CONFIG_FWIDTH_MASK	GENMASK(7, 6) /* FIFO width */
 #define ZYNQ_QSPI_CONFIG_MSTREN_MASK	BIT(0) /* Master Mode */
 
@@ -62,7 +61,7 @@
  */
 #define ZYNQ_QSPI_CONFIG_BAUD_DIV_MAX	GENMASK(2, 0) /* Baud rate maximum */
 #define ZYNQ_QSPI_CONFIG_BAUD_DIV_SHIFT	3 /* Baud rate divisor shift */
-#define ZYNQ_QSPI_CONFIG_PCS		10 /* Peripheral Chip Select */
+#define ZYNQ_QSPI_CONFIG_PCS		BIT(10) /* Peripheral Chip Select */
 
 /*
  * QSPI Interrupt Registers bit Masks
@@ -287,16 +286,12 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctlr);
 	u32 config_reg;
 
+	/* Ground the line to assert the CS */
 	config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_CONFIG_OFFSET);
-	if (assert) {
-		/* Select the slave */
-		config_reg &= ~ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
-		config_reg |= (((~(BIT(spi->chip_select))) <<
-				ZYNQ_QSPI_CONFIG_PCS) &
-				ZYNQ_QSPI_CONFIG_SSCTRL_MASK);
-	} else {
-		config_reg |= ZYNQ_QSPI_CONFIG_SSCTRL_MASK;
-	}
+	if (assert)
+		config_reg &= ~ZYNQ_QSPI_CONFIG_PCS;
+	else
+		config_reg |= ZYNQ_QSPI_CONFIG_PCS;
 
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_CONFIG_OFFSET, config_reg);
 }
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 6/7] spi: zynq-qspi: Do the actual hardware initialization later in the probe

[Miquel Raynal]

Supporting more than one CS will need some tweaking of the linear
configuration register which is (rightfully) initialized in the
hardware initialization helper. The extra initialization needs the
knowledge of the actual number of CS, which is retrieved by reading
the value of the num-cs DT property.

As the initialization helper is called pretty early and might be
called much later in the probe without side effect, let's delay it a
bit so that the number of CS will be available when running this
helper. This way, adding support for multiple CS lines in a next patch
will be eased.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 9adfcc66b02b..07d361501aa0 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -658,9 +658,6 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 		goto clk_dis_pclk;
 	}
 
-	/* QSPI controller initializations */
-	zynq_qspi_init_hw(xqspi);
-
 	xqspi->irq = platform_get_irq(pdev, 0);
 	if (xqspi->irq <= 0) {
 		ret = -ENXIO;
@@ -691,6 +688,10 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ctlr->setup = zynq_qspi_setup_op;
 	ctlr->max_speed_hz = clk_get_rate(xqspi->refclk) / 2;
 	ctlr->dev.of_node = np;
+
+	/* QSPI controller initializations */
+	zynq_qspi_init_hw(xqspi);
+
 	ret = devm_spi_register_controller(&pdev->dev, ctlr);
 	if (ret) {
 		dev_err(&pdev->dev, "spi_register_master failed\n");
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Do the actual hardware initialization later in the probe" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Do the actual hardware initialization later in the probe

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From 8f16292d8b492ca6b0d58ac0769de1c1a7bbb544 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:43 +0100
Subject: [PATCH] spi: zynq-qspi: Do the actual hardware initialization later
 in the probe

Supporting more than one CS will need some tweaking of the linear
configuration register which is (rightfully) initialized in the
hardware initialization helper. The extra initialization needs the
knowledge of the actual number of CS, which is retrieved by reading
the value of the num-cs DT property.

As the initialization helper is called pretty early and might be
called much later in the probe without side effect, let's delay it a
bit so that the number of CS will be available when running this
helper. This way, adding support for multiple CS lines in a next patch
will be eased.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-7-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 80e51c894eaa..e76f9c9738f0 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -657,9 +657,6 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 		goto clk_dis_pclk;
 	}
 
-	/* QSPI controller initializations */
-	zynq_qspi_init_hw(xqspi);
-
 	xqspi->irq = platform_get_irq(pdev, 0);
 	if (xqspi->irq <= 0) {
 		ret = -ENXIO;
@@ -690,6 +687,10 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ctlr->setup = zynq_qspi_setup_op;
 	ctlr->max_speed_hz = clk_get_rate(xqspi->refclk) / 2;
 	ctlr->dev.of_node = np;
+
+	/* QSPI controller initializations */
+	zynq_qspi_init_hw(xqspi);
+
 	ret = devm_spi_register_controller(&pdev->dev, ctlr);
 	if (ret) {
 		dev_err(&pdev->dev, "spi_register_master failed\n");
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v2 7/7] spi: zynq-qspi: Support two chip selects

[Miquel Raynal]

The Zynq QSPI controller features 2 CS. When the num-cs DT property
is set to 2, the hardware will be initialized to support having two
devices connected over each CS.

In this case, both CS lines are driven by the state of the U_PAGE
(upper page) bit. When unset, the lower page (CS0) is selected,
otherwise it is the upper page (CS1).

Change tested on a custom design featuring two SPI-NORs with different
CS on the Zynq-7000 QSPI bus.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/spi/spi-zynq-qspi.c | 33 +++++++++++++++++++++++++--------
 1 file changed, 25 insertions(+), 8 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index 07d361501aa0..a041ac774331 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -115,8 +115,8 @@
  */
 #define ZYNQ_QSPI_MODEBITS			(SPI_CPOL | SPI_CPHA)
 
-/* Default number of chip selects */
-#define ZYNQ_QSPI_DEFAULT_NUM_CS	1
+/* Maximum number of chip selects */
+#define ZYNQ_QSPI_MAX_NUM_CS		2
 
 /**
  * struct zynq_qspi - Defines qspi driver instance
@@ -160,6 +160,7 @@ static inline void zynq_qspi_write(struct zynq_qspi *xqspi, u32 offset,
 /**
  * zynq_qspi_init_hw - Initialize the hardware
  * @xqspi:	Pointer to the zynq_qspi structure
+ * @num_cs:	Number of connected CS (to enable dual memories if needed)
  *
  * The default settings of the QSPI controller's configurable parameters on
  * reset are
@@ -177,7 +178,7 @@ static inline void zynq_qspi_write(struct zynq_qspi *xqspi, u32 offset,
  *	- Set the little endian mode of TX FIFO and
  *	- Enable the QSPI controller
  */
-static void zynq_qspi_init_hw(struct zynq_qspi *xqspi)
+static void zynq_qspi_init_hw(struct zynq_qspi *xqspi, unsigned int num_cs)
 {
 	u32 config_reg;
 
@@ -185,7 +186,12 @@ static void zynq_qspi_init_hw(struct zynq_qspi *xqspi)
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_IDIS_OFFSET, ZYNQ_QSPI_IXR_ALL_MASK);
 
 	/* Disable linear mode as the boot loader may have used it */
-	zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, 0);
+	config_reg = 0;
+	/* At the same time, enable dual mode if more than 1 CS is available */
+	if (num_cs > 1)
+		config_reg |= ZYNQ_QSPI_LCFG_TWO_MEM;
+
+	zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, config_reg);
 
 	/* Clear the RX FIFO */
 	while (zynq_qspi_read(xqspi, ZYNQ_QSPI_STATUS_OFFSET) &
@@ -287,6 +293,17 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctlr);
 	u32 config_reg;
 
+	/* Select the lower (CS0) or upper (CS1) memory */
+	if (ctlr->num_chipselect > 1) {
+		config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET);
+		if (!spi->chip_select)
+			config_reg &= ~ZYNQ_QSPI_LCFG_U_PAGE;
+		else
+			config_reg |= ZYNQ_QSPI_LCFG_U_PAGE;
+
+		zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, config_reg);
+	}
+
 	/* Ground the line to assert the CS */
 	config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_CONFIG_OFFSET);
 	if (assert)
@@ -674,9 +691,9 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ret = of_property_read_u32(np, "num-cs",
 				   &num_cs);
 	if (ret < 0) {
-		ctlr->num_chipselect = ZYNQ_QSPI_DEFAULT_NUM_CS;
-	} else if (num_cs > ZYNQ_QSPI_DEFAULT_NUM_CS) {
-		dev_err(&pdev->dev, "anything but CS0 is not yet supported\n");
+		ctlr->num_chipselect = 1;
+	} else if (num_cs > ZYNQ_QSPI_MAX_NUM_CS) {
+		dev_err(&pdev->dev, "only 2 chip selects are available\n");
 		goto remove_master;
 	} else {
 		ctlr->num_chipselect = num_cs;
@@ -690,7 +707,7 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ctlr->dev.of_node = np;
 
 	/* QSPI controller initializations */
-	zynq_qspi_init_hw(xqspi);
+	zynq_qspi_init_hw(xqspi, ctlr->num_chipselect);
 
 	ret = devm_spi_register_controller(&pdev->dev, ctlr);
 	if (ret) {
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Applied "spi: zynq-qspi: Support two chip selects" to the spi tree

[Mark Brown]

The patch

   spi: zynq-qspi: Support two chip selects

has been applied to the spi tree at

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-5.5

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.  

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark

From d575c9b7c8b4c5ddfb1aa75ac91fdcc20ce328c4 Mon Sep 17 00:00:00 2001
From: Miquel Raynal <miquel.raynal@bootlin.com>
Date: Fri, 8 Nov 2019 15:07:44 +0100
Subject: [PATCH] spi: zynq-qspi: Support two chip selects

The Zynq QSPI controller features 2 CS. When the num-cs DT property
is set to 2, the hardware will be initialized to support having two
devices connected over each CS.

In this case, both CS lines are driven by the state of the U_PAGE
(upper page) bit. When unset, the lower page (CS0) is selected,
otherwise it is the upper page (CS1).

Change tested on a custom design featuring two SPI-NORs with different
CS on the Zynq-7000 QSPI bus.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20191108140744.1734-8-miquel.raynal@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
---
 drivers/spi/spi-zynq-qspi.c | 33 +++++++++++++++++++++++++--------
 1 file changed, 25 insertions(+), 8 deletions(-)

diff --git a/drivers/spi/spi-zynq-qspi.c b/drivers/spi/spi-zynq-qspi.c
index e76f9c9738f0..17641157354d 100644
--- a/drivers/spi/spi-zynq-qspi.c
+++ b/drivers/spi/spi-zynq-qspi.c
@@ -114,8 +114,8 @@
  */
 #define ZYNQ_QSPI_MODEBITS			(SPI_CPOL | SPI_CPHA)
 
-/* Default number of chip selects */
-#define ZYNQ_QSPI_DEFAULT_NUM_CS	1
+/* Maximum number of chip selects */
+#define ZYNQ_QSPI_MAX_NUM_CS		2
 
 /**
  * struct zynq_qspi - Defines qspi driver instance
@@ -159,6 +159,7 @@ static inline void zynq_qspi_write(struct zynq_qspi *xqspi, u32 offset,
 /**
  * zynq_qspi_init_hw - Initialize the hardware
  * @xqspi:	Pointer to the zynq_qspi structure
+ * @num_cs:	Number of connected CS (to enable dual memories if needed)
  *
  * The default settings of the QSPI controller's configurable parameters on
  * reset are
@@ -176,7 +177,7 @@ static inline void zynq_qspi_write(struct zynq_qspi *xqspi, u32 offset,
  *	- Set the little endian mode of TX FIFO and
  *	- Enable the QSPI controller
  */
-static void zynq_qspi_init_hw(struct zynq_qspi *xqspi)
+static void zynq_qspi_init_hw(struct zynq_qspi *xqspi, unsigned int num_cs)
 {
 	u32 config_reg;
 
@@ -184,7 +185,12 @@ static void zynq_qspi_init_hw(struct zynq_qspi *xqspi)
 	zynq_qspi_write(xqspi, ZYNQ_QSPI_IDIS_OFFSET, ZYNQ_QSPI_IXR_ALL_MASK);
 
 	/* Disable linear mode as the boot loader may have used it */
-	zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, 0);
+	config_reg = 0;
+	/* At the same time, enable dual mode if more than 1 CS is available */
+	if (num_cs > 1)
+		config_reg |= ZYNQ_QSPI_LCFG_TWO_MEM;
+
+	zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, config_reg);
 
 	/* Clear the RX FIFO */
 	while (zynq_qspi_read(xqspi, ZYNQ_QSPI_STATUS_OFFSET) &
@@ -286,6 +292,17 @@ static void zynq_qspi_chipselect(struct spi_device *spi, bool assert)
 	struct zynq_qspi *xqspi = spi_controller_get_devdata(ctlr);
 	u32 config_reg;
 
+	/* Select the lower (CS0) or upper (CS1) memory */
+	if (ctlr->num_chipselect > 1) {
+		config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET);
+		if (!spi->chip_select)
+			config_reg &= ~ZYNQ_QSPI_LCFG_U_PAGE;
+		else
+			config_reg |= ZYNQ_QSPI_LCFG_U_PAGE;
+
+		zynq_qspi_write(xqspi, ZYNQ_QSPI_LINEAR_CFG_OFFSET, config_reg);
+	}
+
 	/* Ground the line to assert the CS */
 	config_reg = zynq_qspi_read(xqspi, ZYNQ_QSPI_CONFIG_OFFSET);
 	if (assert)
@@ -673,9 +690,9 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ret = of_property_read_u32(np, "num-cs",
 				   &num_cs);
 	if (ret < 0) {
-		ctlr->num_chipselect = ZYNQ_QSPI_DEFAULT_NUM_CS;
-	} else if (num_cs > ZYNQ_QSPI_DEFAULT_NUM_CS) {
-		dev_err(&pdev->dev, "anything but CS0 is not yet supported\n");
+		ctlr->num_chipselect = 1;
+	} else if (num_cs > ZYNQ_QSPI_MAX_NUM_CS) {
+		dev_err(&pdev->dev, "only 2 chip selects are available\n");
 		goto remove_master;
 	} else {
 		ctlr->num_chipselect = num_cs;
@@ -689,7 +706,7 @@ static int zynq_qspi_probe(struct platform_device *pdev)
 	ctlr->dev.of_node = np;
 
 	/* QSPI controller initializations */
-	zynq_qspi_init_hw(xqspi);
+	zynq_qspi_init_hw(xqspi, ctlr->num_chipselect);
 
 	ret = devm_spi_register_controller(&pdev->dev, ctlr);
 	if (ret) {
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel