Skip to contentSkip to conversation

i2c: adding driver for pxa27x, pxa3xx, pxa25x

PATCH New Andrew Ruder 4 participants 9 messages Aug 12, 2015 – Aug 21, 2015 I2C

lore
Share & tools
lore

Andrew Ruder

author
#0 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Cc Marek Vasut <[email protected]>
Cc Heiko Schocher <[email protected]>
Signed-off-by Andrew Ruder <[email protected]>

---

This driver was written before the driver model stuff was really around (or at least based on a U-Boot version that only had very preliminary support) so there might be some older conventions in use here. I have tested on a PXA270 board running upstream master with this driver, however - so to some extent this all still works fine. Let me know if things need to change to be accepted, and I'll try to find some time to update!

 drivers/i2c/Makefile  |   1 +
 drivers/i2c/pxa_i2c.c | 796 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 797 insertions(+)
 create mode 100644 drivers/i2c/pxa_i2c.c
drivers/i2c/Makefile +1 -0
index 9b45248..cf2c8a8 100644
@@ -30,6 +30,7 @@ obj-$(CONFIG_SYS_I2C_MXS) += mxs_i2c.o
 obj-$(CONFIG_SYS_I2C_OMAP24XX) += omap24xx_i2c.o
 obj-$(CONFIG_SYS_I2C_OMAP34XX) += omap24xx_i2c.o
 obj-$(CONFIG_SYS_I2C_PPC4XX) += ppc4xx_i2c.o
+obj-$(CONFIG_SYS_I2C_PXA) += pxa_i2c.o
 obj-$(CONFIG_SYS_I2C_RCAR) += rcar_i2c.o
 obj-$(CONFIG_SYS_I2C_S3C24X0) += s3c24x0_i2c.o
 obj-$(CONFIG_SYS_I2C_SANDBOX) += sandbox_i2c.o i2c-emul-uclass.o
drivers/i2c/pxa_i2c.c new file +796 -0

796 lines not on this page · raw message

Signature · 1 line
2.1.4
#1 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

On Wednesday, August 12, 2015 at 08:43:52 PM, Andrew Ruder wrote:

Commit message is missing :'-(

Quoted 57 lines
Cc Marek Vasut <[email protected]>
Cc Heiko Schocher <[email protected]>
Signed-off-by Andrew Ruder <[email protected]>

---

This driver was written before the driver model stuff was really around (or at least based on a U-Boot version that only had very preliminary support) so there might be some older conventions in use here. I have tested on a PXA270 board running upstream master with this driver, however - so to some extent this all still works fine. Let me know if things need to change to be accepted, and I'll try to find some time to update!

 drivers/i2c/Makefile  |   1 +
 drivers/i2c/pxa_i2c.c | 796
++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 797
insertions(+)
 create mode 100644 drivers/i2c/pxa_i2c.c
drivers/i2c/Makefile +1 -0
index 9b45248..cf2c8a8 100644
@@ -30,6 +30,7 @@ obj-$(CONFIG_SYS_I2C_MXS) += mxs_i2c.o
 obj-$(CONFIG_SYS_I2C_OMAP24XX) += omap24xx_i2c.o
 obj-$(CONFIG_SYS_I2C_OMAP34XX) += omap24xx_i2c.o
 obj-$(CONFIG_SYS_I2C_PPC4XX) += ppc4xx_i2c.o
+obj-$(CONFIG_SYS_I2C_PXA) += pxa_i2c.o
 obj-$(CONFIG_SYS_I2C_RCAR) += rcar_i2c.o
 obj-$(CONFIG_SYS_I2C_S3C24X0) += s3c24x0_i2c.o
 obj-$(CONFIG_SYS_I2C_SANDBOX) += sandbox_i2c.o i2c-emul-uclass.o
drivers/i2c/pxa_i2c.c new file +14 -0
new file mode 100644
index 0000000..859d6cf
@@ -0,0 +1,796 @@
+/*
+ *  pxa_i2c.c
+ *
+ *  I2C adapter for the PXA I2C bus access.
+ *
+ *  Copyright (C) 2002 Intrinsyc Software Inc.
+ *  Copyright (C) 2004-2005 Deep Blue Solutions Ltd.
+ *  Copyright (C) 2014 Andrew Ruder
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ *  This is a heavily modified version of the PXA I2C driver from the

Linux + * kernel 3.15.10.
+ */

drivers/i2c/pxa_i2c.c new file +5 -0
+
+#include <common.h>
+#include <compiler.h>
+#include <i2c.h>
+#include <asm/arch/pxa-regs.h>

drivers/i2c/pxa_i2c.c line 31

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+
+#include <asm/errno.h>
+#include <asm/io.h>
+
+#define MAX_TRANSFER_MS 5000
+#define WAIT_MASTER_MS 64
+#define ABORT_LENGTH_MS 50
+
+/* Set up some defaults for various configurable pieces of this driver
+ * so that most of the time we don't need extra things defined

Multiline comment should be of the form:

/*
* foo
* bar
*/

Quoted 6 lines

+ */

drivers/i2c/pxa_i2c.c new file +5 -0
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SPEED)
+#define CONFIG_SYS_PXA_STD_I2C_SPEED 100000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SLAVE)

drivers/i2c/pxa_i2c.c line 47

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+#define CONFIG_SYS_PXA_STD_I2C_SLAVE 1
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SPEED)
+#define CONFIG_SYS_PXA_PWR_I2C_SPEED 40000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SLAVE)
+#define CONFIG_SYS_PXA_PWR_I2C_SLAVE 1
+#endif

Please remove all this stuff, it's only going to hide bugs.

Quoted 4 lines
drivers/i2c/pxa_i2c.c new file +4 -0
+#if defined(CONFIG_CPU_MONAHANS) && defined(CONFIG_PXA_PWR_I2C)
+#warning Monahans + PWRI2C is unlikely to work without additional testing
+#endif
+

[...]

drivers/i2c/pxa_i2c.c line 142

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+#define ISR_ITE		(1 << 6)	   /* tx buffer empty */
+#define ISR_IRF		(1 << 7)	   /* rx buffer full */
+#define ISR_GCAD	(1 << 8)	   /* general call address detected */
+#define ISR_SAD		(1 << 9)	   /* slave address detected */
+#define ISR_BED		(1 << 10)	   /* bus error no ACK/NAK */
+
+struct pxa_i2c_msg {
+	u16 addr;
+	u16 flags;
+#define I2C_M_TEN		0x0010	/* this is a ten bit chip address */

Be consistent about using 1 << n or 0x0n00 please. Just use the BIT(n) macro.

drivers/i2c/pxa_i2c.c

Quoted 6 lines
drivers/i2c/pxa_i2c.c new file +6 -0
+#define I2C_M_RD		0x0001	/* read data, from slave to master */
+#define I2C_M_STOP		0x8000	/* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_NOSTART		0x4000	/* if I2C_FUNC_NOSTART */
+#define I2C_M_REV_DIR_ADDR	0x2000	/* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_IGNORE_NAK	0x1000	/* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_NO_RD_ACK		0x0800	/* if I2C_FUNC_PROTOCOL_MANGLING

*/

Quoted 1 line

+#define I2C_M_RECV_LEN 0x0400 /* length will be first received

byte */

Quoted 33 lines
drivers/i2c/pxa_i2c.c new file +32 -0
+#define I2C_M_FIRST		0x0800	/* First message in a chain */
+	u16 len;
+	u8 *buf;
+};
+
+struct pxa_i2c_state {
+	struct pxa_i2c_msg	*msg;
+	unsigned int		msg_num;
+	unsigned int		msg_ptr;
+};
+
+#define STD_I2C_BASE 0x40301680
+#if defined(CONFIG_CPU_PXA27X) || defined(CONFIG_CPU_PXA25X)
+#define PWR_I2C_BASE	0x40f00180
+#define IBMR_OFFSET	0x00
+#define IDBR_OFFSET	0x08
+#define ICR_OFFSET	0x10
+#define ISR_OFFSET	0x18
+#define ISAR_OFFSET	0x20
+#elif defined(CONFIG_CPU_MONAHANS)
+#define PWR_I2C_BASE	0x40f500c0
+#define IBMR_OFFSET	0x00
+#define IDBR_OFFSET	0x04
+#define ICR_OFFSET	0x08
+#define ISR_OFFSET	0x0c
+#define ISAR_OFFSET	0x10
+#endif
+
+static inline void __iomem *
+pxa_i2c_base(const struct i2c_adapter *adap)
+{
+	return (adap->hwadapnr == 0) ? (void *)STD_I2C_BASE : (void

*)PWR_I2C_BASE; +}

drivers/i2c/pxa_i2c.c line 187

Quoted 5 lines
drivers/i2c/pxa_i2c.c new file +5 -0
+
+static inline void __iomem *
+pxa_i2c_ibmr(const struct i2c_adapter *adap)
+{
+	return (void *)((char *)pxa_i2c_base(adap) + IBMR_OFFSET);

No need for these hideous casts, just use pxa_i2c_base() + OFFSET in the code.

Quoted 1 line

+}

[...]

drivers/i2c/pxa_i2c.c line 329

Quoted 8 lines
drivers/i2c/pxa_i2c.c new file +8 -0
+static void
+i2c_pxa_abort(struct i2c_adapter *adap)
+{
+	unsigned long long end;
+
+	end = get_ticks() + (get_tbclk() / 1000) * ABORT_LENGTH_MS;
+
+	while ((get_ticks() < end) &&

Please use get_timer() to implement the delay loop.

time = get_timer(0);

while (true) {
	reg = readl();
	if (reg & bit)
		break;

	if (get_timer(time) > timeout)
		return -ETIMEDOUT;
}

Also, the lpc32xx patches recently implemented a wait_for_bit() function. If you make this generic, that'd be really cool.

Quoted 5 lines
drivers/i2c/pxa_i2c.c new file +5 -0
+	       (readl(_IBMR(adap)) & 0x1) == 0) {
+		unsigned long icr = readl(_ICR(adap));
+
+		icr &= ~ICR_START;
+		icr |= ICR_ACKNAK | ICR_STOP | ICR_TB;

drivers/i2c/pxa_i2c.c line 344

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+
+		writel(icr, _ICR(adap));
+
+		show_state(adap);
+
+		mdelay(1);
+	}
+
+	writel(readl(_ICR(adap)) & ~(ICR_MA | ICR_START | ICR_STOP),
+	       _ICR(adap));

Please use clrsetbits_le32();

Quoted 1 line

+}

[...]

author
#2 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Quoted 7 lines · Marek Vasut

On Wednesday, August 12, 2015 at 08:43:52 PM, Andrew Ruder wrote:

Commit message is missing :'-(

I actually was going with subject being sufficient on this one. I'll fill in some more info on this as far as origin and what it has been tested on and send in again.

Quoted 213 lines
Quoted 58 lines
Cc Marek Vasut <[email protected]>
Cc Heiko Schocher <[email protected]>
Signed-off-by Andrew Ruder <[email protected]>

---

This driver was written before the driver model stuff was really around (or at least based on a U-Boot version that only had very preliminary support) so there might be some older conventions in use here. I have tested on a PXA270 board running upstream master with this driver, however - so to some extent this all still works fine. Let me know if things need to change to be accepted, and I'll try to find some time to update!

drivers/i2c/Makefile | 1 +
drivers/i2c/pxa_i2c.c | 796 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 797 insertions(+)
create mode 100644 drivers/i2c/pxa_i2c.c

drivers/i2c/Makefile +0 -0
index 9b45248..cf2c8a8 100644
@@ -30,6 +30,7 @@ obj-$(CONFIG_SYS_I2C_MXS) += mxs_i2c.o

obj-$(CONFIG_SYS_I2C_OMAP24XX) += omap24xx_i2c.o obj-$(CONFIG_SYS_I2C_OMAP34XX) += omap24xx_i2c.o obj-$(CONFIG_SYS_I2C_PPC4XX) += ppc4xx_i2c.o +obj-$(CONFIG_SYS_I2C_PXA) += pxa_i2c.o obj-$(CONFIG_SYS_I2C_RCAR) += rcar_i2c.o obj-$(CONFIG_SYS_I2C_S3C24X0) += s3c24x0_i2c.o obj-$(CONFIG_SYS_I2C_SANDBOX) += sandbox_i2c.o i2c-emul-uclass.o

drivers/i2c/pxa_i2c.c new file +14 -0
new file mode 100644
index 0000000..859d6cf
@@ -0,0 +1,796 @@
+/*
+ *  pxa_i2c.c
+ *
+ *  I2C adapter for the PXA I2C bus access.
+ *
+ *  Copyright (C) 2002 Intrinsyc Software Inc.
+ *  Copyright (C) 2004-2005 Deep Blue Solutions Ltd.
+ *  Copyright (C) 2014 Andrew Ruder
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ *  This is a heavily modified version of the PXA I2C driver from the

Linux + * kernel 3.15.10.
+ */

drivers/i2c/pxa_i2c.c new file +5 -0
+
+#include <common.h>
+#include <compiler.h>
+#include <i2c.h>
+#include <asm/arch/pxa-regs.h>

drivers/i2c/pxa_i2c.c line 31

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+
+#include <asm/errno.h>
+#include <asm/io.h>
+
+#define MAX_TRANSFER_MS 5000
+#define WAIT_MASTER_MS 64
+#define ABORT_LENGTH_MS 50
+
+/* Set up some defaults for various configurable pieces of this driver
+ * so that most of the time we don't need extra things defined

Multiline comment should be of the form:

/*
* foo
* bar
*/

Quoted 6 lines

+ */

drivers/i2c/pxa_i2c.c new file +5 -0
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SPEED)
+#define CONFIG_SYS_PXA_STD_I2C_SPEED 100000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SLAVE)

drivers/i2c/pxa_i2c.c line 47

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+#define CONFIG_SYS_PXA_STD_I2C_SLAVE 1
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SPEED)
+#define CONFIG_SYS_PXA_PWR_I2C_SPEED 40000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SLAVE)
+#define CONFIG_SYS_PXA_PWR_I2C_SLAVE 1
+#endif

Please remove all this stuff, it's only going to hide bugs.

Quoted 4 lines
drivers/i2c/pxa_i2c.c new file +4 -0
+#if defined(CONFIG_CPU_MONAHANS) && defined(CONFIG_PXA_PWR_I2C)
+#warning Monahans + PWRI2C is unlikely to work without additional testing
+#endif
+

[...]

drivers/i2c/pxa_i2c.c line 142

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+#define ISR_ITE        (1 << 6)       /* tx buffer empty */
+#define ISR_IRF        (1 << 7)       /* rx buffer full */
+#define ISR_GCAD    (1 << 8)       /* general call address detected */
+#define ISR_SAD        (1 << 9)       /* slave address detected */
+#define ISR_BED        (1 << 10)       /* bus error no ACK/NAK */
+
+struct pxa_i2c_msg {
+    u16 addr;
+    u16 flags;
+#define I2C_M_TEN        0x0010    /* this is a ten bit chip address */

Be consistent about using 1 << n or 0x0n00 please. Just use the BIT(n) macro.

drivers/i2c/pxa_i2c.c

Quoted 6 lines
drivers/i2c/pxa_i2c.c new file +6 -0
+#define I2C_M_RD        0x0001    /* read data, from slave to master */
+#define I2C_M_STOP        0x8000    /* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_NOSTART        0x4000    /* if I2C_FUNC_NOSTART */
+#define I2C_M_REV_DIR_ADDR    0x2000    /* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_IGNORE_NAK    0x1000    /* if I2C_FUNC_PROTOCOL_MANGLING */
+#define I2C_M_NO_RD_ACK        0x0800    /* if I2C_FUNC_PROTOCOL_MANGLING

*/

Quoted 1 line

+#define I2C_M_RECV_LEN 0x0400 /* length will be first received

byte */

Quoted 33 lines
drivers/i2c/pxa_i2c.c new file +32 -0
+#define I2C_M_FIRST        0x0800    /* First message in a chain */
+    u16 len;
+    u8 *buf;
+};
+
+struct pxa_i2c_state {
+    struct pxa_i2c_msg    *msg;
+    unsigned int        msg_num;
+    unsigned int        msg_ptr;
+};
+
+#define STD_I2C_BASE 0x40301680
+#if defined(CONFIG_CPU_PXA27X) || defined(CONFIG_CPU_PXA25X)
+#define PWR_I2C_BASE    0x40f00180
+#define IBMR_OFFSET    0x00
+#define IDBR_OFFSET    0x08
+#define ICR_OFFSET    0x10
+#define ISR_OFFSET    0x18
+#define ISAR_OFFSET    0x20
+#elif defined(CONFIG_CPU_MONAHANS)
+#define PWR_I2C_BASE    0x40f500c0
+#define IBMR_OFFSET    0x00
+#define IDBR_OFFSET    0x04
+#define ICR_OFFSET    0x08
+#define ISR_OFFSET    0x0c
+#define ISAR_OFFSET    0x10
+#endif
+
+static inline void __iomem *
+pxa_i2c_base(const struct i2c_adapter *adap)
+{
+    return (adap->hwadapnr == 0) ? (void *)STD_I2C_BASE : (void

*)PWR_I2C_BASE; +}

drivers/i2c/pxa_i2c.c line 187

Quoted 5 lines
drivers/i2c/pxa_i2c.c new file +5 -0
+
+static inline void __iomem *
+pxa_i2c_ibmr(const struct i2c_adapter *adap)
+{
+    return (void *)((char *)pxa_i2c_base(adap) + IBMR_OFFSET);

No need for these hideous casts, just use pxa_i2c_base() + OFFSET in the code.

Quoted 1 line

+}

[...]

drivers/i2c/pxa_i2c.c line 329

Quoted 8 lines
drivers/i2c/pxa_i2c.c new file +8 -0
+static void
+i2c_pxa_abort(struct i2c_adapter *adap)
+{
+    unsigned long long end;
+
+    end = get_ticks() + (get_tbclk() / 1000) * ABORT_LENGTH_MS;
+
+    while ((get_ticks() < end) &&

Please use get_timer() to implement the delay loop.

time = get_timer(0);

while (true) {
   reg = readl();
   if (reg & bit)
       break;

   if (get_timer(time) > timeout)
       return -ETIMEDOUT;
}

Also, the lpc32xx patches recently implemented a wait_for_bit() function. If you make this generic, that'd be really cool.

Quoted 5 lines
drivers/i2c/pxa_i2c.c new file +5 -0
+           (readl(_IBMR(adap)) & 0x1) == 0) {
+        unsigned long icr = readl(_ICR(adap));
+
+        icr &= ~ICR_START;
+        icr |= ICR_ACKNAK | ICR_STOP | ICR_TB;

drivers/i2c/pxa_i2c.c line 344

Quoted 10 lines
drivers/i2c/pxa_i2c.c new file +10 -0
+
+        writel(icr, _ICR(adap));
+
+        show_state(adap);
+
+        mdelay(1);
+    }
+
+    writel(readl(_ICR(adap)) & ~(ICR_MA | ICR_START | ICR_STOP),
+           _ICR(adap));

Please use clrsetbits_le32();

Quoted 1 line

+}

[...]

#3 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Quoted 9 lines · Andrew Ruder
Quoted 5 lines · Marek Vasut

On Wednesday, August 12, 2015 at 08:43:52 PM, Andrew Ruder wrote:

Commit message is missing :'-(

I actually was going with subject being sufficient on this one. I'll fill in some more info on this as far as origin and what it has been tested on and send in again.

It's really a minor thing, but a commit message should always be present.

Thanks!

Best regards,
Marek Vasut

#4 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

In reply to Marek Vasut
Quoted 10 lines · Marek Vasut
Quoted 9 lines · Andrew Ruder
drivers/i2c/pxa_i2c.c new file +9 -0
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SPEED)
+#define CONFIG_SYS_PXA_STD_I2C_SPEED 100000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_STD_I2C_SLAVE)
+#define CONFIG_SYS_PXA_STD_I2C_SLAVE 1
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SPEED)
Quoted 8 lines

drivers/i2c/pxa_i2c.c line 47

Quoted 6 lines
drivers/i2c/pxa_i2c.c new file +6 -0
+#define CONFIG_SYS_PXA_PWR_I2C_SPEED 40000
+#endif
+
+#if !defined(CONFIG_SYS_PXA_PWR_I2C_SLAVE)
+#define CONFIG_SYS_PXA_PWR_I2C_SLAVE 1
+#endif

Please remove all this stuff, it's only going to hide bugs.

Hide in what way? Just require that ports define them when enabling the i2c driver? ... or don't make it externally configurable?

Thanks,
Andrew

#5 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

In reply to Andrew Ruder
Quoted 5 lines · Andrew Ruder
Cc Marek Vasut <[email protected]>
Cc Heiko Schocher <[email protected]>
Signed-off-by Andrew Ruder <[email protected]>

---

Quoted 11 lines

This driver was written before the driver model stuff was really around (or at
least based on a U-Boot version that only had very preliminary support) so
there might be some older conventions in use here. I have tested on a PXA270
board running upstream master with this driver, however - so to some extent
this all still works fine. Let me know if things need to change to be
accepted, and I'll try to find some time to update!

Apart from the missing commit message.

Signed-off-by Marcel Ziswiler <[email protected]>
Tested-by Marcel Ziswiler <[email protected]>

Tested-on: Colibri PXA270 V1.2C, V1.2D, V2.2B, V2.4A on Colibri Evaluation Board V3.1A
Note: While power I2C works on the older V1.2C version it does not seem to work on any of the newer ones but this might just be a pin muxing resp. missing pull up issue.

#6 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Quoted 18 lines · Marcel Ziswiler
Quoted 5 lines · Andrew Ruder
Cc Marek Vasut <[email protected]>
Cc Heiko Schocher <[email protected]>
Signed-off-by Andrew Ruder <[email protected]>

---

Quoted 11 lines

This driver was written before the driver model stuff was really around (or at
least based on a U-Boot version that only had very preliminary support) so
there might be some older conventions in use here. I have tested on a PXA270
board running upstream master with this driver, however - so to some extent
this all still works fine. Let me know if things need to change to be
accepted, and I'll try to find some time to update!

Quoted 3 lines

Apart from the missing commit message.

Signed-off-by Marcel Ziswiler <[email protected]>

I think you're misunderstanding the meaning of a SoB line here, the SoB line is a certificate of origin. You didn't write the driver and/or you didn't make changes to it, so you don't place your certificate on it :)

Quoted 1 line
Tested-by Marcel Ziswiler <[email protected]>

This one is of course good.

Quoted 1 line

Tested-on: Colibri PXA270 V1.2C, V1.2D, V2.2B, V2.4A on Colibri

Is this some kind of a newly invented tag ? :-)

Quoted 4 lines

Evaluation Board V3.1A
Note: While power I2C works on the older V1.2C version it does not seem to work on any of the newer ones but this might just be a pin muxing resp. missing pull up issue.

Very nice. Btw. how do you bootstrap those modules, do you use that OpenPXA OBM or OBM2 or something else entirely ?

Best regards,
Marek Vasut

#7 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Quoted 10 lines · Marek Vasut
Quoted 3 lines

Apart from the missing commit message.

Signed-off-by Marcel Ziswiler <[email protected]>

I think you're misunderstanding the meaning of a SoB line here, the SoB
line is a certificate of origin. You didn't write the driver and/or you
didn't make changes to it, so you don't place your certificate on it :)

Yeah, it's kind of late. Should have been an Acked-by of course, sorry.

Quoted 7 lines
Quoted 1 line
Tested-by Marcel Ziswiler <[email protected]>

This one is of course good.

Quoted 1 line

Tested-on: Colibri PXA270 V1.2C, V1.2D, V2.2B, V2.4A on Colibri

Is this some kind of a newly invented tag ? :-)

Well, others have been using that one on and off for a while. So at least I am not the inventor (;-p).

Quoted 10 lines
Quoted 2 lines

Evaluation Board V3.1A
Note: While power I2C works on the older V1.2C version it does not

seem

Quoted 1 line

to work on any of the newer ones but this might just be a pin

muxing

Quoted 1 line

resp. missing pull up issue.

Very nice. Btw. how do you bootstrap those modules, do you use that OpenPXA
OBM or OBM2 or something else entirely ?

No, I never used any of that crap. Just purely U-Boot even on the PXA3xx but the PXA270 being just a regular NOR beast is really nothing special. Just using ye olde BDI2000 if things go awry. Unfortunately my PXA3xx U-Boot NAND SPL stuff never made it mainline and you since dropped PXA3xx support entirely. If I ever get some precious time I might actually re-add those again especially now with Marvell having committed to keep selling us PXA320 chips to 2020.

#8 lore raw

PATCH 1/1 i2c: adding driver for pxa27x, pxa3xx, pxa25x

Quoted 11 lines · Marcel Ziswiler
Quoted 10 lines · Marek Vasut
Quoted 3 lines

Apart from the missing commit message.

Signed-off-by Marcel Ziswiler <[email protected]>

I think you're misunderstanding the meaning of a SoB line here, the SoB
line is a certificate of origin. You didn't write the driver and/or you
didn't make changes to it, so you don't place your certificate on it

Quoted 3 lines
Quoted 1 line

:)

Yeah, it's kind of late. Should have been an Acked-by of course, sorry.

I can understand that :)

Quoted 10 lines
Quoted 7 lines
Quoted 1 line
Tested-by Marcel Ziswiler <[email protected]>

This one is of course good.

Quoted 1 line

Tested-on: Colibri PXA270 V1.2C, V1.2D, V2.2B, V2.4A on Colibri

Is this some kind of a newly invented tag ? :-)

Well, others have been using that one on and off for a while. So at least I am not the inventor (;-p).

Heh :)

Quoted 9 lines
Quoted 9 lines
Quoted 2 lines

Evaluation Board V3.1A
Note: While power I2C works on the older V1.2C version it does not

seem

Quoted 1 line

to work on any of the newer ones but this might just be a pin

muxing

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

resp. missing pull up issue.

Very nice. Btw. how do you bootstrap those modules, do you use that OpenPXA
OBM or OBM2 or something else entirely ?

No, I never used any of that crap.

Well sorry it wasn't to your liking ... ;-)

Quoted 7 lines

Just purely U-Boot even on the
PXA3xx but the PXA270 being just a regular NOR beast is really nothing special. Just using ye olde BDI2000 if things go awry. Unfortunately my PXA3xx U-Boot NAND SPL stuff never made it mainline and you since dropped PXA3xx support entirely. If I ever get some precious time I might actually re-add those again especially now with Marvell having committed to keep selling us PXA320 chips to 2020.

There is some very rudimentary SPL support in the u-boot-pxa, but it's really old.

Best regards,
Marek Vasut