[PATCH v1 0/2] mtd: devices: add support for microchip 48l640 EERAM

[PATCH v1 0/2] mtd: devices: add support for microchip 48l640 EERAM

[Heiko Schocher]

add support for microchip 48l640 eeram connected via
SPI. Patch 1 introduces devicetree documentation
and patch 2 the driver.



Heiko Schocher (2):
  mtd: devices: add devicetree documentation for microchip 48l640
  mtd: devices: add support for microchip 48l640 EERAM

 .../bindings/mtd/microchip,mchp48l640.yaml    |  40 ++
 drivers/mtd/devices/Kconfig                   |   6 +
 drivers/mtd/devices/Makefile                  |   1 +
 drivers/mtd/devices/mchp48l640.c              | 379 ++++++++++++++++++
 4 files changed, 426 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
 create mode 100644 drivers/mtd/devices/mchp48l640.c

-- 
2.30.2

[PATCH v1 1/2] mtd: devices: add devicetree documentation for microchip 48l640

[Heiko Schocher]

The Microchip 48l640 is a 8KByte EERAM connected via SPI.
Add devicetree bindings documentation.

Signed-off-by: Heiko Schocher <hs@denx.de>
---
I must admit that I created the file

Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml

by hand, looking into other yaml files...

Is there a way to check such a file to its correctness?

 .../bindings/mtd/microchip,mchp48l640.yaml    | 40 +++++++++++++++++++
 1 file changed, 40 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml

diff --git a/Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml b/Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
new file mode 100644
index 0000000000000..f0290d09d4b25
--- /dev/null
+++ b/Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+title: MTD SPI driver for Microchip 48l640 (and similar) serial EERAM
+
+maintainers:
+  - Heiko Schocher <hs@denx.de>
+
+description: |
+  The Microchip 48l640 is a 8KByte EERAM connected via SPI.
+
+  datasheet: http://ww1.microchip.com/downloads/en/DeviceDoc/20006055B.pdf
+
+properties:
+  compatible:
+    items:
+      - const: microchip,48l640
+
+  reg:
+    maxItems: 1
+
+  spi-max-frequency: true
+
+required:
+  - compatible
+  - reg
+
+examples:
+  - |
+    spi {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      spi-eeram@0 {
+        compatible = "microchip,48l640";
+        reg = <0>;
+        spi-max-frequency = <20000000>;
+      };
+    };
+...
-- 
2.30.2

Re: [PATCH v1 1/2] mtd: devices: add devicetree documentation for microchip 48l640

[Rob Herring]

On Fri, 28 May 2021 09:23:50 +0200, Heiko Schocher wrote:
> The Microchip 48l640 is a 8KByte EERAM connected via SPI.
> Add devicetree bindings documentation.
> 
> Signed-off-by: Heiko Schocher <hs@denx.de>
> ---
> I must admit that I created the file
> 
> Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
> 
> by hand, looking into other yaml files...
> 
> Is there a way to check such a file to its correctness?
> 
>  .../bindings/mtd/microchip,mchp48l640.yaml    | 40 +++++++++++++++++++
>  1 file changed, 40 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
> 

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check'
on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
Traceback (most recent call last):
  File "/usr/local/bin/dt-doc-validate", line 67, in <module>
    ret = check_doc(f)
  File "/usr/local/bin/dt-doc-validate", line 33, in check_doc
    for error in sorted(dtschema.DTValidator.iter_schema_errors(testtree), key=lambda e: e.linecol):
  File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 723, in iter_schema_errors
    meta_schema = cls.resolver.resolve_from_url(schema['$schema'])
KeyError: '$schema'
make[1]: *** Deleting file 'Documentation/devicetree/bindings/processed-schema-examples.json'
Traceback (most recent call last):
  File "/usr/local/bin/dt-mk-schema", line 38, in <module>
    schemas = dtschema.process_schemas(args.schemas, core_schema=(not args.useronly))
  File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 587, in process_schemas
    sch = process_schema(os.path.abspath(filename))
  File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 568, in process_schema
    DTValidator.check_schema(schema)
  File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 738, in check_schema
    meta_schema = cls.resolver.resolve_from_url(schema['$schema'])
KeyError: '$schema'
make[1]: *** [Documentation/devicetree/bindings/Makefile:62: Documentation/devicetree/bindings/processed-schema-examples.json] Error 1
make: *** [Makefile:1416: dt_binding_check] Error 2

See https://patchwork.ozlabs.org/patch/1485038

This check can fail if there are any dependencies. The base for a patch
series is generally the most recent rc1.

If you already ran 'make dt_binding_check' and didn't see the above
error(s), then make sure 'yamllint' is installed and dt-schema is up to
date:

pip3 install dtschema --upgrade

Please check and re-submit.

Re: [PATCH v1 1/2] mtd: devices: add devicetree documentation for microchip 48l640

[Heiko Schocher]

Hello Rob,

On 28.05.21 18:21, Rob Herring wrote:
> On Fri, 28 May 2021 09:23:50 +0200, Heiko Schocher wrote:
>> The Microchip 48l640 is a 8KByte EERAM connected via SPI.
>> Add devicetree bindings documentation.
>>
>> Signed-off-by: Heiko Schocher <hs@denx.de>
>> ---
>> I must admit that I created the file
>>
>> Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
>>
>> by hand, looking into other yaml files...
>>
>> Is there a way to check such a file to its correctness?
>>
>>  .../bindings/mtd/microchip,mchp48l640.yaml    | 40 +++++++++++++++++++
>>  1 file changed, 40 insertions(+)
>>  create mode 100644 Documentation/devicetree/bindings/mtd/microchip,mchp48l640.yaml
>>
> 
> My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check'
> on your patch (DT_CHECKER_FLAGS is new in v5.13):
> 
> yamllint warnings/errors:
> 
> dtschema/dtc warnings/errors:
> Traceback (most recent call last):
>   File "/usr/local/bin/dt-doc-validate", line 67, in <module>
>     ret = check_doc(f)
>   File "/usr/local/bin/dt-doc-validate", line 33, in check_doc
>     for error in sorted(dtschema.DTValidator.iter_schema_errors(testtree), key=lambda e: e.linecol):
>   File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 723, in iter_schema_errors
>     meta_schema = cls.resolver.resolve_from_url(schema['$schema'])
> KeyError: '$schema'
> make[1]: *** Deleting file 'Documentation/devicetree/bindings/processed-schema-examples.json'
> Traceback (most recent call last):
>   File "/usr/local/bin/dt-mk-schema", line 38, in <module>
>     schemas = dtschema.process_schemas(args.schemas, core_schema=(not args.useronly))
>   File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 587, in process_schemas
>     sch = process_schema(os.path.abspath(filename))
>   File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 568, in process_schema
>     DTValidator.check_schema(schema)
>   File "/usr/local/lib/python3.8/dist-packages/dtschema/lib.py", line 738, in check_schema
>     meta_schema = cls.resolver.resolve_from_url(schema['$schema'])
> KeyError: '$schema'
> make[1]: *** [Documentation/devicetree/bindings/Makefile:62: Documentation/devicetree/bindings/processed-schema-examples.json] Error 1
> make: *** [Makefile:1416: dt_binding_check] Error 2
> 
> See https://patchwork.ozlabs.org/patch/1485038
> 
> This check can fail if there are any dependencies. The base for a patch
> series is generally the most recent rc1.
> 
> If you already ran 'make dt_binding_check' and didn't see the above
> error(s), then make sure 'yamllint' is installed and dt-schema is up to
> date:
> 
> pip3 install dtschema --upgrade
> 
> Please check and re-submit.

I just setup this stuff, and have now a

'make DT_CHECKER_FLAGS=-m dt_binding_check'

running without warnings from my new yaml file.

Thanks for your help!

Will send a v2 soon...

bye,
Heiko
-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-52   Fax: +49-8142-66989-80   Email: hs@denx.de

[PATCH v1 2/2] mtd: devices: add support for microchip 48l640 EERAM

[Heiko Schocher]

The Microchip 48l640 is a 8KByte EERAM connected via SPI.

Signed-off-by: Heiko Schocher <hs@denx.de>
---
I tried to use drivers/mtd/devices/mchp23k256.c but this
driver does not use any status register and there seems
slight differences in registers (no write enable register
for example), so I decided to make a new driver.

This driver sets the continuous mode bit in Status register,
which states you can write continuous ... but after writting
32 bytes the chip goes into an undefined state, so driver now
writes data in 32 byte chunks.

I also tried to use regmap, but it leads in a lot of more code,
and as this chip has only spi interface it makes no sense, or?

Tested  this driver on board imx8mp-phyboard-pollux-rdk
board, which is already in mainline.

Made some tbot tests, which write at random offset random
length bytes with dd and and random content. Reread the data
after a reboot and compare with the written data. Works fine.

 drivers/mtd/devices/Kconfig      |   6 +
 drivers/mtd/devices/Makefile     |   1 +
 drivers/mtd/devices/mchp48l640.c | 379 +++++++++++++++++++++++++++++++
 3 files changed, 386 insertions(+)
 create mode 100644 drivers/mtd/devices/mchp48l640.c

diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig
index 0f4c2d823de84..79cb981ececc9 100644
--- a/drivers/mtd/devices/Kconfig
+++ b/drivers/mtd/devices/Kconfig
@@ -89,6 +89,12 @@ config MTD_MCHP23K256
 	  platform data, or a device tree description if you want to
 	  specify device partitioning
 
+config MTD_MCHP48L640
+	tristate "Microchip 48L640 EERAM"
+	depends on SPI_MASTER
+	help
+	  This enables access to Microchip 48L640 EERAM chips, using SPI.
+
 config MTD_SPEAR_SMI
 	tristate "SPEAR MTD NOR Support through SMI controller"
 	depends on PLAT_SPEAR || COMPILE_TEST
diff --git a/drivers/mtd/devices/Makefile b/drivers/mtd/devices/Makefile
index 991c8d12c0160..0362cf6bdc67f 100644
--- a/drivers/mtd/devices/Makefile
+++ b/drivers/mtd/devices/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_MTD_LART)		+= lart.o
 obj-$(CONFIG_MTD_BLOCK2MTD)	+= block2mtd.o
 obj-$(CONFIG_MTD_DATAFLASH)	+= mtd_dataflash.o
 obj-$(CONFIG_MTD_MCHP23K256)	+= mchp23k256.o
+obj-$(CONFIG_MTD_MCHP48L640)	+= mchp48l640.o
 obj-$(CONFIG_MTD_SPEAR_SMI)	+= spear_smi.o
 obj-$(CONFIG_MTD_SST25L)	+= sst25l.o
 obj-$(CONFIG_MTD_BCM47XXSFLASH)	+= bcm47xxsflash.o
diff --git a/drivers/mtd/devices/mchp48l640.c b/drivers/mtd/devices/mchp48l640.c
new file mode 100644
index 0000000000000..e70a7cd2292bb
--- /dev/null
+++ b/drivers/mtd/devices/mchp48l640.c
@@ -0,0 +1,379 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Driver for Microchip 48L640 64 Kb SPI Serial EERAM
+ *
+ * Copyright Heiko Schocher <hs@denx.de>
+ *
+ * datasheet: http://ww1.microchip.com/downloads/en/DeviceDoc/20006055B.pdf
+ *
+ * we set continuous mode but reading/writing more bytes than
+ * pagesize seems to bring chip into state where readden values
+ * are wrong ... no idea why.
+ *
+ */
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/jiffies.h>
+#include <linux/module.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mutex.h>
+#include <linux/sched.h>
+#include <linux/sizes.h>
+#include <linux/spi/flash.h>
+#include <linux/spi/spi.h>
+#include <linux/of_device.h>
+
+struct mchp48_caps {
+	unsigned int size;
+	unsigned int page_size;
+};
+
+struct mchp48l640_flash {
+	struct spi_device	*spi;
+	struct mutex		lock;
+	struct mtd_info		mtd;
+	const struct mchp48_caps	*caps;
+};
+
+#define MCHP48L640_CMD_WREN		0x06
+#define MCHP48L640_CMD_WRDI		0x04
+#define MCHP48L640_CMD_WRITE		0x02
+#define MCHP48L640_CMD_READ		0x03
+#define MCHP48L640_CMD_WRSR		0x01
+#define MCHP48L640_CMD_RDSR		0x05
+
+#define MCHP48L640_STATUS_RDY		0x01
+#define MCHP48L640_STATUS_WEL		0x02
+#define MCHP48L640_STATUS_BP0		0x04
+#define MCHP48L640_STATUS_BP1		0x08
+#define MCHP48L640_STATUS_SWM		0x10
+#define MCHP48L640_STATUS_PRO		0x20
+#define MCHP48L640_STATUS_ASE		0x40
+
+#define MCHP48L640_TIMEOUT		100
+
+#define MAX_CMD_SIZE			0x10
+
+#define to_mchp48l640_flash(x) container_of(x, struct mchp48l640_flash, mtd)
+
+static int mchp48l640_mkcmd(struct mchp48l640_flash *flash, u8 cmd, loff_t addr, char *buf)
+{
+	int i;
+
+	buf[0] = cmd;
+	buf[1] = addr >> 8;
+	buf[2] = addr;
+
+	return 3;
+}
+
+static int mchp48l640_read_status(struct mchp48l640_flash *flash, int *status)
+{
+	unsigned char cmd[2];
+	int ret;
+
+	cmd[0] = MCHP48L640_CMD_RDSR;
+	cmd[1] = 0x00;
+	mutex_lock(&flash->lock);
+	ret = spi_write_then_read(flash->spi, &cmd[0], 1, &cmd[1], 1);
+	mutex_unlock(&flash->lock);
+	if (!ret)
+		*status = cmd[1];
+	dev_dbg(&flash->spi->dev, "read status ret: %d status: %x", ret, *status);
+
+	return ret;
+}
+
+static int mchp48l640_waitforbit(struct mchp48l640_flash *flash, int bit, bool set)
+{
+	int ret, status;
+	unsigned long deadline;
+
+	deadline = jiffies + msecs_to_jiffies(MCHP48L640_TIMEOUT);
+	do {
+		ret = mchp48l640_read_status(flash, &status);
+		dev_dbg(&flash->spi->dev, "read status ret: %d bit: %x %sset status: %x",
+			ret, bit, (set ? "" : "not"), status);
+		if (ret)
+			return ret;
+
+		if (set) {
+			if ((status & bit) == bit)
+				return 0;
+		} else {
+			if ((status & bit) == 0)
+				return 0;
+		}
+
+		usleep_range(1000, 2000);
+	} while (!time_after_eq(jiffies, deadline));
+
+	dev_err(&flash->spi->dev, "Timeout waiting for bit %x %s set in status register.",
+		bit, (set ? "" : "not"));
+	return -ETIMEDOUT;
+}
+
+static int mchp48l640_write_prepare(struct mchp48l640_flash *flash, bool enable)
+{
+	unsigned char cmd[2];
+	int ret;
+
+	if (enable)
+		cmd[0] = MCHP48L640_CMD_WREN;
+	else
+		cmd[0] = MCHP48L640_CMD_WRDI;
+
+	mutex_lock(&flash->lock);
+	ret = spi_write(flash->spi, cmd, 1);
+	mutex_unlock(&flash->lock);
+
+	if (ret)
+		dev_err(&flash->spi->dev, "write %sable failed ret: %d",
+			(enable ? "en" : "dis"), ret);
+
+	dev_dbg(&flash->spi->dev, "write %sable success ret: %d",
+		(enable ? "en" : "dis"), ret);
+	if (enable)
+		return mchp48l640_waitforbit(flash, MCHP48L640_STATUS_WEL, true);
+
+	return ret;
+}
+
+static int mchp48l640_set_mode(struct mchp48l640_flash *flash)
+{
+	unsigned char cmd[2];
+	int status;
+	int ret;
+
+	ret = mchp48l640_write_prepare(flash, true);
+	if (ret)
+		return ret;
+
+	cmd[0] = MCHP48L640_CMD_WRSR;
+	cmd[1] = MCHP48L640_STATUS_PRO;
+
+	mutex_lock(&flash->lock);
+	ret = spi_write(flash->spi, cmd, 2);
+	mutex_unlock(&flash->lock);
+	if (ret)
+		dev_err(&flash->spi->dev, "Could not set continuous mode ret: %d", ret);
+
+	return mchp48l640_waitforbit(flash, MCHP48L640_STATUS_PRO, true);
+}
+
+static int mchp48l640_wait_rdy(struct mchp48l640_flash *flash)
+{
+	return mchp48l640_waitforbit(flash, MCHP48L640_STATUS_RDY, false);
+};
+
+static int mchp48l640_write_page(struct mtd_info *mtd, loff_t to, size_t len,
+			    size_t *retlen, const unsigned char *buf)
+{
+	struct mchp48l640_flash *flash = to_mchp48l640_flash(mtd);
+	unsigned char *cmd;
+	int ret;
+	int cmdlen;
+
+	cmd = kmalloc((3 + len), GFP_KERNEL | GFP_DMA);
+	if (!cmd)
+		return -ENOMEM;
+
+	ret = mchp48l640_wait_rdy(flash);
+	if (ret)
+		goto fail;
+
+	ret = mchp48l640_write_prepare(flash, true);
+	if (ret)
+		goto fail;
+
+	mutex_lock(&flash->lock);
+	cmdlen = mchp48l640_mkcmd(flash, MCHP48L640_CMD_WRITE, to, cmd);
+	memcpy(&cmd[cmdlen], buf, len);
+	ret = spi_write(flash->spi, cmd, cmdlen + len);
+	mutex_unlock(&flash->lock);
+	if (!ret)
+		*retlen += len;
+	else
+		goto fail;
+
+	ret = mchp48l640_waitforbit(flash, MCHP48L640_STATUS_WEL, false);
+	if (ret)
+		goto fail;
+
+	kfree(cmd);
+	return 0;
+fail:
+	kfree(cmd);
+	dev_err(&flash->spi->dev, "write fail with: %d", ret);
+	return ret;
+};
+
+static int mchp48l640_write(struct mtd_info *mtd, loff_t to, size_t len,
+			    size_t *retlen, const unsigned char *buf)
+{
+	struct mchp48l640_flash *flash = to_mchp48l640_flash(mtd);
+	int ret;
+	size_t wlen = 0;
+	loff_t woff = to;
+	size_t ws;
+	int i;
+	int page_sz = flash->caps->page_size;
+
+	/*
+	 * we set PRO bit (page rollover), but writing length > page size
+	 * does result in total chaos, so write in 32 byte chunks.
+	 */
+	while (wlen < len) {
+		ws = min((len - wlen), page_sz);
+		ret = mchp48l640_write_page(mtd, woff, ws, retlen, &buf[wlen]);
+		if (ret)
+			return ret;
+		wlen += ws;
+		woff += ws;
+	}
+
+	return ret;
+}
+
+static int mchp48l640_read_page(struct mtd_info *mtd, loff_t from, size_t len,
+			   size_t *retlen, unsigned char *buf)
+{
+	struct mchp48l640_flash *flash = to_mchp48l640_flash(mtd);
+	unsigned char *cmd;
+	int ret;
+	int cmdlen;
+
+	cmd = kmalloc((3 + len), GFP_KERNEL | GFP_DMA);
+	if (!cmd)
+		return -ENOMEM;
+
+	ret = mchp48l640_wait_rdy(flash);
+	if (ret)
+		goto fail;
+
+	mutex_lock(&flash->lock);
+	cmdlen = mchp48l640_mkcmd(flash, MCHP48L640_CMD_READ, from, cmd);
+	ret = spi_write_then_read(flash->spi, cmd, cmdlen, buf, len);
+	mutex_unlock(&flash->lock);
+	if (!ret)
+		*retlen += len;
+
+	return ret;
+
+fail:
+	kfree(cmd);
+	dev_err(&flash->spi->dev, "read fail with: %d", ret);
+	return ret;
+}
+
+static int mchp48l640_read(struct mtd_info *mtd, loff_t from, size_t len,
+			   size_t *retlen, unsigned char *buf)
+{
+	struct mchp48l640_flash *flash = to_mchp48l640_flash(mtd);
+	int ret;
+	size_t wlen = 0;
+	loff_t woff = from;
+	size_t ws;
+	int i;
+	int page_sz = flash->caps->page_size;
+
+	/*
+	 * we set PRO bit (page rollover), but if read length > page size
+	 * does result in total chaos in result ...
+	 */
+	dev_dbg(&flash->spi->dev, "read len: %d from: %llx", len, from);
+	while (wlen < len) {
+		ws = min((len - wlen), page_sz);
+		ret = mchp48l640_read_page(mtd, woff, ws, retlen, &buf[wlen]);
+		if (ret)
+			return ret;
+		wlen += ws;
+		woff += ws;
+	}
+
+	return ret;
+};
+
+static const struct mchp48_caps mchp48l640_caps = {
+	.size = SZ_8K,
+	.page_size = 32,
+};
+
+static int mchp48l640_probe(struct spi_device *spi)
+{
+	struct mchp48l640_flash *flash;
+	struct flash_platform_data *data;
+	int err;
+	int status;
+
+	flash = devm_kzalloc(&spi->dev, sizeof(*flash), GFP_KERNEL);
+	if (!flash)
+		return -ENOMEM;
+
+	flash->spi = spi;
+	mutex_init(&flash->lock);
+	spi_set_drvdata(spi, flash);
+
+	err = mchp48l640_read_status(flash, &status);
+	if (err)
+		return err;
+
+	err = mchp48l640_set_mode(flash);
+	if (err)
+		return err;
+
+	data = dev_get_platdata(&spi->dev);
+
+	flash->caps = of_device_get_match_data(&spi->dev);
+	if (!flash->caps)
+		flash->caps = &mchp48l640_caps;
+
+	mtd_set_of_node(&flash->mtd, spi->dev.of_node);
+	flash->mtd.dev.parent	= &spi->dev;
+	flash->mtd.type		= MTD_RAM;
+	flash->mtd.flags	= MTD_CAP_RAM;
+	flash->mtd.writesize	= flash->caps->page_size;
+	flash->mtd.size		= flash->caps->size;
+	flash->mtd._read	= mchp48l640_read;
+	flash->mtd._write	= mchp48l640_write;
+
+	err = mtd_device_register(&flash->mtd, data ? data->parts : NULL,
+				  data ? data->nr_parts : 0);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int mchp48l640_remove(struct spi_device *spi)
+{
+	struct mchp48l640_flash *flash = spi_get_drvdata(spi);
+
+	return mtd_device_unregister(&flash->mtd);
+}
+
+static const struct of_device_id mchp48l640_of_table[] = {
+	{
+		.compatible = "microchip,48l640",
+		.data = &mchp48l640_caps,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(of, mchp48l640_of_table);
+
+static struct spi_driver mchp48l640_driver = {
+	.driver = {
+		.name	= "mchp48l640",
+		.of_match_table = of_match_ptr(mchp48l640_of_table),
+	},
+	.probe		= mchp48l640_probe,
+	.remove		= mchp48l640_remove,
+};
+
+module_spi_driver(mchp48l640_driver);
+
+MODULE_DESCRIPTION("MTD SPI driver for Microchip 48l640 EERAM chips");
+MODULE_AUTHOR("Heiko Schocher <hs@denx.de>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("spi:mchp48l640");
-- 
2.30.2

Re: [PATCH v1 2/2] mtd: devices: add support for microchip 48l640 EERAM

[Fabio Estevam]

Hi Heiko,

On Fri, May 28, 2021 at 4:26 AM Heiko Schocher <hs@denx.de> wrote:
>
> The Microchip 48l640 is a 8KByte EERAM connected via SPI.
>
> Signed-off-by: Heiko Schocher <hs@denx.de>
> ---
> I tried to use drivers/mtd/devices/mchp23k256.c but this
> driver does not use any status register and there seems
> slight differences in registers (no write enable register
> for example), so I decided to make a new driver.
>
> This driver sets the continuous mode bit in Status register,
> which states you can write continuous ... but after writting
> 32 bytes the chip goes into an undefined state, so driver now
> writes data in 32 byte chunks.
>
> I also tried to use regmap, but it leads in a lot of more code,
> and as this chip has only spi interface it makes no sense, or?
>
> Tested  this driver on board imx8mp-phyboard-pollux-rdk
> board, which is already in mainline.
>
> Made some tbot tests, which write at random offset random
> length bytes with dd and and random content. Reread the data
> after a reboot and compare with the written data. Works fine.

Works for me too on an imx7d based board, thanks:

Tested-by: Fabio Estevam <festevam@denx.de>