Skip to contentSkip to conversation

tango NFC support + bindings doc

PATCH In review Marc Gonzalez 8 participants 18 messages Oct 25, 2016 – Dec 19, 2016 7 messages not in the archive

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v7 cover tango NFC support + bindings doc

Hello,

This patch series adds support for the NAND Flash controller on recent Tango chips, along with the bindings documentation.

Marc Gonzalez (2):
mtd: nand: add tango NFC dt bindings doc
mtd: nand: add tango NAND flash controller support

 Documentation/devicetree/bindings/mtd/tango-nand.txt |  38 ++
 drivers/mtd/nand/Kconfig                             |   6 +
 drivers/mtd/nand/Makefile                            |   1 +
 drivers/mtd/nand/tango_nand.c                        | 654 +++++++++++++++++++++++++++++
 4 files changed, 699 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/mtd/tango-nand.txt
 create mode 100644 drivers/mtd/nand/tango_nand.c

Regards.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

#1 lore raw

PATCH v7 1/2 mtd: nand: add tango NFC dt bindings doc

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +38 -0
new file mode 100644
index 000000000000..3cbf95d6595a
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;
+		dmas = <&dma0 3>;
+		dma-names = "nfc_sbox";
+		clocks = <&clkgen SYS_CLK>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		nand@0 {
+			reg = <0>; /* CS0 */
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+
+		nand@1 {
+			reg = <1>; /* CS1 */
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+	};
Signature · 5 lines
2.9.0

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

A message that is not in the archive (sent off-list or not yet received): [email protected]

#3 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 21 lines · On Tue, 25 Oct 2016 15:15:50 +0200 Marc Gonzalez…

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
new file mode 100644
index 000000000000..3cbf95d6595a
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"

Documentation/devicetree/bindings/mtd/tango-nand.txt line 18

Quoted 10 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +10 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
	nandc: nand-controller@2c000 {

No need to resend a new version for that (I can fix it when applying), unless I find other things in patch 2.

Quoted 10 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +10 -0
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;
+		dmas = <&dma0 3>;
+		dma-names = "nfc_sbox";
+		clocks = <&clkgen SYS_CLK>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		nand@0 {
+			reg = <0>; /* CS0 */

Documentation/devicetree/bindings/mtd/tango-nand.txt line 38

Quoted 10 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +10 -0
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+
+		nand@1 {
+			reg = <1>; /* CS1 */
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+	};

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Acked-by
#4 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

In reply to a message not in the archive
Quoted 6 lines · Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#5 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

In reply to a message not in the archive
Quoted 3 lines · On Tue, 25 Oct 2016 15:15:50 +0200 Marc Gonzalez…

Add the tango NAND Flash Controller dt bindings documentation.

Applied after renaming the nand controller node.

Thanks,

Boris

Quoted 38 lines
---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +28 -0
new file mode 100644
index 000000000000..3cbf95d6595a
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;
+		dmas = <&dma0 3>;
+		dma-names = "nfc_sbox";
+		clocks = <&clkgen SYS_CLK>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		nand@0 {
+			reg = <0>; /* CS0 */

Documentation/devicetree/bindings/mtd/tango-nand.txt line 38

Quoted 10 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +10 -0
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+
+		nand@1 {
+			reg = <1>; /* CS1 */
+			nand-ecc-strength = <14>;
+			nand-ecc-step-size = <1024>;
+		};
+	};

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#6 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

In reply to a message not in the archive
Quoted 13 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"

Drop the "nfc_" prefix here, it seems redundant.

Documentation/devicetree/bindings/mtd/tango-nand.txt line 20

Quoted 12 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +12 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;

It would be nicer to write this as

	reg = <0x2c000 0x30>, <0x2d000 0x800>, <0x20000 0x1000>;

which is identical in binary format.

Arnd
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#7 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Quoted 16 lines
Quoted 6 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"

Drop the "nfc_" prefix here, it seems redundant.

I don't think it's redundant; there are switch boxes for several different HW blocks; nfc_sbox is the first one to be exposed.

Quoted 15 lines

Documentation/devicetree/bindings/mtd/tango-nand.txt line 20

Quoted 12 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +12 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;

It would be nicer to write this as

Quoted 3 lines
	reg = <0x2c000 0x30>, <0x2d000 0x800>, <0x20000 0x1000>;

which is identical in binary format.

I didn't know that, thanks for pointing it out.

Unfortunately, Boris already accepted the patch yesterday :-(

Boris, do you fixup patches in your tree?

Regards.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

A message that is not in the archive (sent off-list or not yet received): [email protected]

#9 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 26 lines · On Mon, 7 Nov 2016 10:26:48 +0100 Marc Gonzalez…
Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Quoted 16 lines
Quoted 6 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"  

Drop the "nfc_" prefix here, it seems redundant.

Quoted 2 lines

I don't think it's redundant; there are switch boxes for several different HW blocks; nfc_sbox is the first one to be exposed.

The dma-names are local to the device, so I think the nfc_ prefix is indeed not needed.

Quoted 18 lines
Quoted 18 lines

Documentation/devicetree/bindings/mtd/tango-nand.txt line 20

Quoted 13 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +12 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;  

It would be nicer to write this as

	reg = <0x2c000 0x30>, <0x2d000 0x800>, <0x20000 0x1000>;
Quoted 7 lines
Quoted 1 line

which is identical in binary format.

I didn't know that, thanks for pointing it out.

Unfortunately, Boris already accepted the patch yesterday :-(

Boris, do you fixup patches in your tree?

I usually try to avoid that, unless one of the patches breaks bisectibility (which is not the case here). Please send new patches to fix that.

Thanks,

Boris

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#10 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

In reply to Arnd Bergmann
Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Quoted 16 lines
Quoted 6 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"

Drop the "nfc_" prefix here, it seems redundant.

I took a closer look at Documentation/devicetree/bindings and arch/arm/boot/dts and it seems the overwhelming nomenclature is "rx", "tx" (with variants around these two).

Should I just use "rxtx" for my driver?

Quoted 15 lines

Documentation/devicetree/bindings/mtd/tango-nand.txt line 20

Quoted 12 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +12 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;

It would be nicer to write this as

Quoted 3 lines
	reg = <0x2c000 0x30>, <0x2d000 0x800>, <0x20000 0x1000>;

which is identical in binary format.

This is indeed a marked improvement. Boris, if I fix these two nits, can I submit a single patch, or do you want them split?

Regards.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

#11 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 26 lines · Marc Gonzalez
Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Quoted 16 lines
Quoted 6 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"

Drop the "nfc_" prefix here, it seems redundant.

Quoted 5 lines

I took a closer look at Documentation/devicetree/bindings and arch/arm/boot/dts and it seems the overwhelming nomenclature is "rx", "tx" (with variants around these two).

Should I just use "rxtx" for my driver?

On IRC, Arnd wrote:

"The string for dma-names is not important, it just needs to be documented in the binding. If you have hardware specifications, use the name that is next to the wire for the dma-request line. "dma-names" is only required so you can connect a single dmarq to multiple dma engines (most chips only connect each rq to one dmaengine though)."

In my system, peripheral devices are connected to DMA channels via a multiplexer called the "host switch box". The documentation calls the ports FOO_SBOX and SBOX_FOO.

For MLC NAND Flash controller 0, the ports are mlc_flash0_sbox and sbox_mlc_flash0.

Additionally, the DMA driver handles both directions, so the DT only defines a single duplex channel.

I originally used "nfc_sbox" (nfc = NAND Flash controller) to stick to the HW naming, but I am willing to change it to "rxtx" if that's what Boris prefers, (or just "sbox", but Mans pointed out that this was too specific, and future HW might do away with the switch box).

I'm open to comments.

Regards.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

A message that is not in the archive (sent off-list or not yet received): ee9efd0a-59f5-8c21-0cad-40ca30e3a3e7-y1yR0Z3OICC7zZZRDBGcUA@public.gmane.org

#13 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 18 lines · Marc Gonzalez
Quoted 1 line · Marc Gonzalez

Should I just use "rxtx" for my driver?

On IRC, Arnd wrote:

"The string for dma-names is not important, it just needs to be documented in the binding. If you have hardware specifications, use the name that is next to the wire for the dma-request line. "dma-names" is only required so you can connect a single dmarq to multiple dma engines (most chips only connect each rq to one dmaengine though)."

In my system, peripheral devices are connected to DMA channels via a multiplexer called the "host switch box". The documentation calls the ports FOO_SBOX and SBOX_FOO.

For MLC NAND Flash controller 0, the ports are mlc_flash0_sbox and sbox_mlc_flash0.

Quoted 7 lines

Additionally, the DMA driver handles both directions, so the DT only defines a single duplex channel.

I originally used "nfc_sbox" (nfc = NAND Flash controller) to stick to the HW naming, but I am willing to change it to "rxtx" if that's what Boris prefers, (or just "sbox", but Mans pointed out that this was too specific, and future HW might do away with the switch box).

"rxtx" is the best option, if you need a name at all.

The name is relative to the device instance, so "nfc_" is unnecessary; we know the node is an NFC controller. The sbox is part of the SoC wiring rather than being part of the NFC controller, so the "sbox" part also shouldn't be part of the name.

Likewise for the NAND flash controller. There, the "0" instance number also shouldn't have been there -- the name is relative to the instance, and two instances should use the same names. Too late now, I guess. :(

Thanks,
Mark.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): d3cbfda6-23bf-4ad1-b1ce-c179b765272f-y1yR0Z3OICC7zZZRDBGcUA@public.gmane.org

#15 lore raw

Re: [PATCH v7 1/2] mtd: nand: add tango NFC dt bindings doc

Quoted 47 lines · On Fri, 16 Dec 2016 11:56:38 +0100 Marc Gonzalez…
Quoted 8 lines · Arnd Bergmann
Quoted 7 lines · CET Marc Gonzalez

Add the tango NAND Flash Controller dt bindings documentation.

---
 Documentation/devicetree/bindings/mtd/tango-nand.txt | 38 ++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
Quoted 16 lines
Quoted 6 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +0 -0
new file mode 100644
index 000000000000..3cbf95d6595a

Documentation/devicetree/bindings/mtd/tango-nand.txt line 8

Quoted 8 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +8 -0
@@ -0,0 +1,38 @@
+Sigma Designs Tango4 NAND Flash Controller (NFC)
+
+Required properties:
+
+- compatible: "sigma,smp8758-nand"
+- reg: address/size of nfc_reg, nfc_mem, and pbus_reg
+- dmas: reference to the DMA channel used by the controller
+- dma-names: "nfc_sbox"  

Drop the "nfc_" prefix here, it seems redundant.

I took a closer look at Documentation/devicetree/bindings and arch/arm/boot/dts and it seems the overwhelming nomenclature is "rx", "tx" (with variants around these two).

Should I just use "rxtx" for my driver?

Quoted 15 lines

Documentation/devicetree/bindings/mtd/tango-nand.txt line 20

Quoted 12 lines
Documentation/devicetree/bindings/mtd/tango-nand.txt new file +12 -0
+- clocks: reference to the system clock
+- #address-cells: <1>
+- #size-cells: <0>
+
+Children nodes represent the available NAND chips.
+See Documentation/devicetree/bindings/mtd/nand.txt for generic bindings.
+
+Example:
+
+	nand: nand@2c000 {
+		compatible = "sigma,smp8758-nand";
+		reg = <0x2c000 0x30 0x2d000 0x800 0x20000 0x1000>;  

It would be nicer to write this as

Quoted 6 lines
Quoted 3 lines
	reg = <0x2c000 0x30>, <0x2d000 0x800>, <0x20000 0x1000>;

which is identical in binary format.

This is indeed a marked improvement. Boris, if I fix these two nits, can I submit a single patch, or do you want them split?

Sure, just describe the different binding changes in your commit message.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#16 lore raw

PATCH v7 2/2 mtd: nand: add tango NAND flash controller support

In reply to Marc Gonzalez

This driver supports the NAND Flash controller embedded in recent Tango chips, such as SMP8758 and SMP8759.

---
 drivers/mtd/nand/Kconfig      |   6 +
 drivers/mtd/nand/Makefile     |   1 +
 drivers/mtd/nand/tango_nand.c | 654 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 661 insertions(+)
drivers/mtd/nand/Kconfig +6 -0
index 7b7a887b4709..844ab20a23a4 100644
@@ -205,6 +205,12 @@ config MTD_NAND_S3C2410_CLKSTOP
 	  when the is NAND chip selected or released, but will save
 	  approximately 5mA of power when there is nothing happening.
 
+config MTD_NAND_TANGO
+	tristate "NAND Flash support for Tango chips"
+	depends on ARCH_TANGO
+	help
+	  Enables the NAND Flash controller on Tango chips.
+
 config MTD_NAND_DISKONCHIP
 	tristate "DiskOnChip 2000, Millennium and Millennium Plus (NAND reimplementation)"
 	depends on HAS_IOMEM
drivers/mtd/nand/Makefile +1 -0
index cafde6f3d957..4904ad3614fb 100644
@@ -16,6 +16,7 @@ obj-$(CONFIG_MTD_NAND_DENALI_DT)	+= denali_dt.o
 obj-$(CONFIG_MTD_NAND_AU1550)		+= au1550nd.o
 obj-$(CONFIG_MTD_NAND_BF5XX)		+= bf5xx_nand.o
 obj-$(CONFIG_MTD_NAND_S3C2410)		+= s3c2410.o
+obj-$(CONFIG_MTD_NAND_TANGO)		+= tango_nand.o
 obj-$(CONFIG_MTD_NAND_DAVINCI)		+= davinci_nand.o
 obj-$(CONFIG_MTD_NAND_DISKONCHIP)	+= diskonchip.o
 obj-$(CONFIG_MTD_NAND_DOCG4)		+= docg4.o
drivers/mtd/nand/tango_nand.c new file +654 -0

654 lines not on this page · raw message

Signature · 5 lines
2.9.0

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

A message that is not in the archive (sent off-list or not yet received): [email protected]

#18 lore raw

PATCH v8 1/1 mtd: nand: add tango NAND flash controller support

This driver supports the NAND Flash controller embedded in recent Tango chips, such as SMP8758 and SMP8759.

---
 drivers/mtd/nand/Kconfig      |   7 +
 drivers/mtd/nand/Makefile     |   1 +
 drivers/mtd/nand/tango_nand.c | 654 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 662 insertions(+)
 create mode 100644 drivers/mtd/nand/tango_nand.c
drivers/mtd/nand/Kconfig +7 -0
index 7b7a887b4709..84ce3fe6e5e2 100644
@@ -205,6 +205,13 @@ config MTD_NAND_S3C2410_CLKSTOP
 	  when the is NAND chip selected or released, but will save
 	  approximately 5mA of power when there is nothing happening.
 
+config MTD_NAND_TANGO
+	tristate "NAND Flash support for Tango chips"
+	depends on ARCH_TANGO || COMPILE_TEST
+	depends on HAS_DMA
+	help
+	  Enables the NAND Flash controller on Tango chips.
+
 config MTD_NAND_DISKONCHIP
 	tristate "DiskOnChip 2000, Millennium and Millennium Plus (NAND reimplementation)"
 	depends on HAS_IOMEM
drivers/mtd/nand/Makefile +1 -0
index cafde6f3d957..4904ad3614fb 100644
@@ -16,6 +16,7 @@ obj-$(CONFIG_MTD_NAND_DENALI_DT)	+= denali_dt.o
 obj-$(CONFIG_MTD_NAND_AU1550)		+= au1550nd.o
 obj-$(CONFIG_MTD_NAND_BF5XX)		+= bf5xx_nand.o
 obj-$(CONFIG_MTD_NAND_S3C2410)		+= s3c2410.o
+obj-$(CONFIG_MTD_NAND_TANGO)		+= tango_nand.o
 obj-$(CONFIG_MTD_NAND_DAVINCI)		+= davinci_nand.o
 obj-$(CONFIG_MTD_NAND_DISKONCHIP)	+= diskonchip.o
 obj-$(CONFIG_MTD_NAND_DOCG4)		+= docg4.o
drivers/mtd/nand/tango_nand.c new file +654 -0

654 lines not on this page · raw message

Signature · 5 lines
2.9.0
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): [email protected]

#20 lore raw

Re: [PATCH v8] mtd: nand: add tango NAND flash controller support

Quoted 4 lines · On Tue, 25 Oct 2016 18:10:47 +0200 Marc Gonzalez…

This driver supports the NAND Flash controller embedded in recent Tango chips, such as SMP8758 and SMP8759.

Applied after fixing a few coding style issues to make checkpatch happy.

Thanks,

Boris

Quoted 687 lines
---
 drivers/mtd/nand/Kconfig      |   7 +
 drivers/mtd/nand/Makefile     |   1 +
 drivers/mtd/nand/tango_nand.c | 654 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 662 insertions(+)
 create mode 100644 drivers/mtd/nand/tango_nand.c
drivers/mtd/nand/Kconfig +7 -0
index 7b7a887b4709..84ce3fe6e5e2 100644
@@ -205,6 +205,13 @@ config MTD_NAND_S3C2410_CLKSTOP
 	  when the is NAND chip selected or released, but will save
 	  approximately 5mA of power when there is nothing happening.
 
+config MTD_NAND_TANGO
+	tristate "NAND Flash support for Tango chips"
+	depends on ARCH_TANGO || COMPILE_TEST
+	depends on HAS_DMA
+	help
+	  Enables the NAND Flash controller on Tango chips.
+
 config MTD_NAND_DISKONCHIP
 	tristate "DiskOnChip 2000, Millennium and Millennium Plus (NAND reimplementation)"
 	depends on HAS_IOMEM
drivers/mtd/nand/Makefile +1 -0
index cafde6f3d957..4904ad3614fb 100644
@@ -16,6 +16,7 @@ obj-$(CONFIG_MTD_NAND_DENALI_DT)	+= denali_dt.o
 obj-$(CONFIG_MTD_NAND_AU1550)		+= au1550nd.o
 obj-$(CONFIG_MTD_NAND_BF5XX)		+= bf5xx_nand.o
 obj-$(CONFIG_MTD_NAND_S3C2410)		+= s3c2410.o
+obj-$(CONFIG_MTD_NAND_TANGO)		+= tango_nand.o
 obj-$(CONFIG_MTD_NAND_DAVINCI)		+= davinci_nand.o
 obj-$(CONFIG_MTD_NAND_DISKONCHIP)	+= diskonchip.o
 obj-$(CONFIG_MTD_NAND_DOCG4)		+= docg4.o
drivers/mtd/nand/tango_nand.c new file +644 -0
new file mode 100644
index 000000000000..74e39a92771c
@@ -0,0 +1,654 @@
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/clk.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/mtd/nand.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
+
+/* Offsets relative to chip->base */
+#define PBUS_CMD	0
+#define PBUS_ADDR	4
+#define PBUS_DATA	8
+
+/* Offsets relative to reg_base */
+#define NFC_STATUS	0x00
+#define NFC_FLASH_CMD	0x04
+#define NFC_DEVICE_CFG	0x08
+#define NFC_TIMING1	0x0c
+#define NFC_TIMING2	0x10
+#define NFC_XFER_CFG	0x14
+#define NFC_PKT_0_CFG	0x18
+#define NFC_PKT_N_CFG	0x1c
+#define NFC_BB_CFG	0x20
+#define NFC_ADDR_PAGE	0x24
+#define NFC_ADDR_OFFSET	0x28
+#define NFC_XFER_STATUS	0x2c
+
+/* NFC_STATUS values */
+#define CMD_READY	BIT(31)
+
+/* NFC_FLASH_CMD values */
+#define NFC_READ	1
+#define NFC_WRITE	2
+
+/* NFC_XFER_STATUS values */
+#define PAGE_IS_EMPTY	BIT(16)
+
+/* Offsets relative to mem_base */
+#define METADATA	0x000
+#define ERROR_REPORT	0x1c0
+
+/*
+ * Error reports are split in two bytes:
+ * byte 0 for the first packet in the page (PKT_0)
+ * byte 1 for other packets in the page (PKT_N, for N > 0)
+ * ERR_COUNT_PKT_N is the max error count over all but the first packet.
+ */
+#define DECODE_OK_PKT_0(v)	(v & BIT(7))
+#define DECODE_OK_PKT_N(v)	(v & BIT(15))
+#define ERR_COUNT_PKT_0(v)	((v >> 0) & 0x3f)
+#define ERR_COUNT_PKT_N(v)	((v >> 8) & 0x3f)
+
+/* Offsets relative to pbus_base */
+#define PBUS_CS_CTRL	0x83c
+#define PBUS_PAD_MODE	0x8f0
+
+/* PBUS_CS_CTRL values */
+#define PBUS_IORDY	BIT(31)
+
+/*
+ * PBUS_PAD_MODE values
+ * In raw mode, the driver communicates directly with the NAND chips.
+ * In NFC mode, the NAND Flash controller manages the communication.
+ * We use NFC mode for read and write; raw mode for everything else.
+ */
+#define MODE_RAW	0
+#define MODE_NFC	BIT(31)
+
+#define METADATA_SIZE	4
+#define BBM_SIZE	6
+#define FIELD_ORDER	15
+
+#define MAX_CS		4
+
+struct tango_nfc {
+	struct nand_hw_control hw;
+	void __iomem *reg_base;
+	void __iomem *mem_base;
+	void __iomem *pbus_base;
+	struct tango_chip *chips[MAX_CS];
+	struct dma_chan *chan;
+	int freq_kHz;
+};
+
+#define to_tango_nfc(ptr) container_of(ptr, struct tango_nfc, hw)
+
+struct tango_chip {
+	struct nand_chip nand_chip;
+	void __iomem *base;
+	u32 timing1;
+	u32 timing2;
+	u32 xfer_cfg;
+	u32 pkt_0_cfg;
+	u32 pkt_n_cfg;
+	u32 bb_cfg;
+};
+
+#define to_tango_chip(ptr) container_of(ptr, struct tango_chip, nand_chip)
+
+#define XFER_CFG(cs, page_count, steps, metadata_size)	\
+	((cs) << 24 | (page_count) << 16 | (steps) << 8 | (metadata_size))
+
+#define PKT_CFG(size, strength) ((size) << 16 | (strength))
+
+#define BB_CFG(bb_offset, bb_size) ((bb_offset) << 16 | (bb_size))
+
+#define TIMING(t0, t1, t2, t3) ((t0) << 24 | (t1) << 16 | (t2) << 8 | (t3))
+
+static void tango_cmd_ctrl(struct mtd_info *mtd, int dat, unsigned int ctrl)
+{
+	struct tango_chip *tchip = to_tango_chip(mtd_to_nand(mtd));
+
+	if (ctrl & NAND_CLE)
+		writeb_relaxed(dat, tchip->base + PBUS_CMD);
+
+	if (ctrl & NAND_ALE)
+		writeb_relaxed(dat, tchip->base + PBUS_ADDR);
+}
+
+static int tango_dev_ready(struct mtd_info *mtd)
+{
+	struct nand_chip *chip = mtd_to_nand(mtd);
+	struct tango_nfc *nfc = to_tango_nfc(chip->controller);
+
+	return readl_relaxed(nfc->pbus_base + PBUS_CS_CTRL) & PBUS_IORDY;
+}
+
+static uint8_t tango_read_byte(struct mtd_info *mtd)
+{
+	struct tango_chip *tchip = to_tango_chip(mtd_to_nand(mtd));
+
+	return readb_relaxed(tchip->base + PBUS_DATA);
+}
+
+static void tango_read_buf(struct mtd_info *mtd, uint8_t *buf, int len)
+{
+	struct tango_chip *tchip = to_tango_chip(mtd_to_nand(mtd));
+
+	ioread8_rep(tchip->base + PBUS_DATA, buf, len);
+}
+
+static void tango_write_buf(struct mtd_info *mtd, const uint8_t *buf, int len)
+{
+	struct tango_chip *tchip = to_tango_chip(mtd_to_nand(mtd));
+
+	iowrite8_rep(tchip->base + PBUS_DATA, buf, len);
+}
+
+static void tango_select_chip(struct mtd_info *mtd, int idx)
+{
+	struct nand_chip *chip = mtd_to_nand(mtd);
+	struct tango_nfc *nfc = to_tango_nfc(chip->controller);
+	struct tango_chip *tchip = to_tango_chip(chip);
+
+	if (idx < 0)
+		return; /* No "chip unselect" function */
+
+	writel_relaxed(tchip->timing1, nfc->reg_base + NFC_TIMING1);
+	writel_relaxed(tchip->timing2, nfc->reg_base + NFC_TIMING2);
+	writel_relaxed(tchip->xfer_cfg, nfc->reg_base + NFC_XFER_CFG);
+	writel_relaxed(tchip->pkt_0_cfg, nfc->reg_base + NFC_PKT_0_CFG);
+	writel_relaxed(tchip->pkt_n_cfg, nfc->reg_base + NFC_PKT_N_CFG);
+	writel_relaxed(tchip->bb_cfg, nfc->reg_base + NFC_BB_CFG);
+}
+
+/*
+ * The controller does not check for bitflips in erased pages,
+ * therefore software must check instead.
+ */
+static int check_erased_page(struct nand_chip *chip, u8 *buf)
+{
+	u8 *meta = chip->oob_poi + BBM_SIZE;
+	u8 *ecc = chip->oob_poi + BBM_SIZE + METADATA_SIZE;
+	const int ecc_size = chip->ecc.bytes;
+	const int pkt_size = chip->ecc.size;
+	int i, res, meta_len, bitflips = 0;
+
+	for (i = 0; i < chip->ecc.steps; ++i) {
+		meta_len = i ? 0 : METADATA_SIZE;
+		res = nand_check_erased_ecc_chunk(buf, pkt_size, ecc, ecc_size,
+				meta, meta_len, chip->ecc.strength);
+		if (res < 0)
+			chip->mtd.ecc_stats.failed++;
+
+		bitflips = max(res, bitflips);
+		buf += pkt_size;
+		ecc += ecc_size;
+	}
+
+	return bitflips;
+}
+
+static int decode_error_report(struct tango_nfc *nfc)
+{
+	u32 status, res;
+
+	status = readl_relaxed(nfc->reg_base + NFC_XFER_STATUS);
+	if (status & PAGE_IS_EMPTY)
+		return 0;
+
+	res = readl_relaxed(nfc->mem_base + ERROR_REPORT);
+
+	if (DECODE_OK_PKT_0(res) && DECODE_OK_PKT_N(res))
+		return max(ERR_COUNT_PKT_0(res), ERR_COUNT_PKT_N(res));
+
+	return -EBADMSG;
+}
+
+static void tango_dma_callback(void *arg)
+{
+	complete(arg);
+}
+
+static int do_dma(struct tango_nfc *nfc, int dir, int cmd,
+		const void *buf, int len, int page)
+{
+	void __iomem *addr = nfc->reg_base + NFC_STATUS;
+	struct dma_chan *chan = nfc->chan;
+	struct dma_async_tx_descriptor *desc;
+	struct scatterlist sg;
+	struct completion tx_done;
+	int err = -EIO;
+	u32 res, val;
+
+	sg_init_one(&sg, buf, len);
+	if (dma_map_sg(chan->device->dev, &sg, 1, dir) != 1)
+		return -EIO;
+
+	desc = dmaengine_prep_slave_sg(chan, &sg, 1, dir, DMA_PREP_INTERRUPT);
+	if (!desc)
+		goto dma_unmap;
+
+	desc->callback = tango_dma_callback;
+	desc->callback_param = &tx_done;
+	init_completion(&tx_done);
+
+	writel_relaxed(MODE_NFC, nfc->pbus_base + PBUS_PAD_MODE);
+
+	writel_relaxed(page, nfc->reg_base + NFC_ADDR_PAGE);
+	writel_relaxed(0, nfc->reg_base + NFC_ADDR_OFFSET);
+	writel_relaxed(cmd, nfc->reg_base + NFC_FLASH_CMD);
+
+	dmaengine_submit(desc);
+	dma_async_issue_pending(chan);
+
+	res = wait_for_completion_timeout(&tx_done, HZ);
+	if (res > 0)
+		err = readl_poll_timeout(addr, val, val & CMD_READY, 0, 1000);
+
+	writel_relaxed(MODE_RAW, nfc->pbus_base + PBUS_PAD_MODE);
+
+dma_unmap:
+	dma_unmap_sg(chan->device->dev, &sg, 1, dir);
+
+	return err;
+}
+
+static int tango_read_page(struct mtd_info *mtd, struct nand_chip *chip,
+		uint8_t *buf, int oob_required, int page)
+{
+	struct tango_nfc *nfc = to_tango_nfc(chip->controller);
+	int err, res, len = mtd->writesize;
+
+	if (oob_required)
+		chip->ecc.read_oob(mtd, chip, page);
+
+	err = do_dma(nfc, DMA_FROM_DEVICE, NFC_READ, buf, len, page);
+	if (err)
+		return err;
+
+	res = decode_error_report(nfc);
+	if (res < 0) {
+		chip->ecc.read_oob_raw(mtd, chip, page);
+		res = check_erased_page(chip, buf);
+	}
+
+	return res;
+}
+
+static int tango_write_page(struct mtd_info *mtd, struct nand_chip *chip,
+		const uint8_t *buf, int oob_required, int page)
+{
+	struct tango_nfc *nfc = to_tango_nfc(chip->controller);
+	int err, len = mtd->writesize;
+
+	/* Calling tango_write_oob() would send PAGEPROG twice */
+	if (oob_required)
+		return -ENOTSUPP;
+
+	writel_relaxed(0xffffffff, nfc->mem_base + METADATA);
+	err = do_dma(nfc, DMA_TO_DEVICE, NFC_WRITE, buf, len, page);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void aux_read(struct nand_chip *chip, u8 **buf, int len, int *pos)
+{
+	*pos += len;
+
+	if (*buf == NULL) {
+		/* skip over "len" bytes */
+		chip->cmdfunc(&chip->mtd, NAND_CMD_RNDOUT, *pos, -1);
+	} else {
+		tango_read_buf(&chip->mtd, *buf, len);
+		*buf += len;
+	}
+}
+
+static void aux_write(struct nand_chip *chip, const u8 **buf, int len, int *pos)
+{
+	*pos += len;
+
+	if (*buf == NULL) {
+		/* skip over "len" bytes */
+		chip->cmdfunc(&chip->mtd, NAND_CMD_SEQIN, *pos, -1);
+	} else {
+		tango_write_buf(&chip->mtd, *buf, len);
+		*buf += len;
+	}
+}
+
+/*
+ * Physical page layout (not drawn to scale)
+ *
+ * NB: Bad Block Marker area splits PKT_N in two (N1, N2).
+ *
+ * +---+-----------------+-------+-----+-----------+-----+----+-------+
+ * | M |      PKT_0      | ECC_0 | ... |     N1    | BBM | N2 | ECC_N |
+ * +---+-----------------+-------+-----+-----------+-----+----+-------+
+ *
+ * Logical page layout:
+ *
+ *       +-----+---+-------+-----+-------+
+ * oob = | BBM | M | ECC_0 | ... | ECC_N |
+ *       +-----+---+-------+-----+-------+
+ *
+ *       +-----------------+-----+-----------------+
+ * buf = |      PKT_0      | ... |      PKT_N      |
+ *       +-----------------+-----+-----------------+
+ */
+static int raw_read(struct nand_chip *chip, u8 *buf, u8 *oob)
+{
+	u8 *oob_orig = oob;
+	const int page_size = chip->mtd.writesize;
+	const int ecc_size = chip->ecc.bytes;
+	const int pkt_size = chip->ecc.size;
+	int pos = 0; /* position within physical page */
+	int rem = page_size; /* bytes remaining until BBM area */
+
+	if (oob != NULL)
+		oob += BBM_SIZE;
+
+	aux_read(chip, &oob, METADATA_SIZE, &pos);
+
+	while (rem > pkt_size) {
+		aux_read(chip, &buf, pkt_size, &pos);
+		aux_read(chip, &oob, ecc_size, &pos);
+		rem = page_size - pos;
+	}
+
+	aux_read(chip, &buf, rem, &pos);
+	aux_read(chip, &oob_orig, BBM_SIZE, &pos);
+	aux_read(chip, &buf, pkt_size - rem, &pos);
+	aux_read(chip, &oob, ecc_size, &pos);
+
+	return 0;
+}
+
+static int raw_write(struct nand_chip *chip, const u8 *buf, const u8 *oob)
+{
+	const u8 *oob_orig = oob;
+	const int page_size = chip->mtd.writesize;
+	const int ecc_size = chip->ecc.bytes;
+	const int pkt_size = chip->ecc.size;
+	int pos = 0; /* position within physical page */
+	int rem = page_size; /* bytes remaining until BBM area */
+
+	if (oob != NULL)
+		oob += BBM_SIZE;
+
+	aux_write(chip, &oob, METADATA_SIZE, &pos);
+
+	while (rem > pkt_size) {
+		aux_write(chip, &buf, pkt_size, &pos);
+		aux_write(chip, &oob, ecc_size, &pos);
+		rem = page_size - pos;
+	}
+
+	aux_write(chip, &buf, rem, &pos);
+	aux_write(chip, &oob_orig, BBM_SIZE, &pos);
+	aux_write(chip, &buf, pkt_size - rem, &pos);
+	aux_write(chip, &oob, ecc_size, &pos);
+
+	return 0;
+}
+
+static int tango_read_page_raw(struct mtd_info *mtd, struct nand_chip *chip,
+		uint8_t *buf, int oob_required, int page)
+{
+	return raw_read(chip, buf, chip->oob_poi);
+}
+
+static int tango_write_page_raw(struct mtd_info *mtd, struct nand_chip *chip,
+		const uint8_t *buf, int oob_required, int page)
+{
+	return raw_write(chip, buf, chip->oob_poi);
+}
+
+static int tango_read_oob(struct mtd_info *mtd, struct nand_chip *chip, int page)
+{
+	chip->cmdfunc(mtd, NAND_CMD_READ0, 0, page);
+	return raw_read(chip, NULL, chip->oob_poi);
+}
+
+static int tango_write_oob(struct mtd_info *mtd, struct nand_chip *chip, int page)
+{
+	chip->cmdfunc(mtd, NAND_CMD_SEQIN, 0, page);
+	raw_write(chip, NULL, chip->oob_poi);
+	chip->cmdfunc(mtd, NAND_CMD_PAGEPROG, -1, -1);
+	chip->waitfunc(mtd, chip);
+	return 0;
+}
+
+static int oob_ecc(struct mtd_info *mtd, int idx, struct mtd_oob_region *res)
+{
+	struct nand_chip *chip = mtd_to_nand(mtd);
+	struct nand_ecc_ctrl *ecc = &chip->ecc;
+
+	if (idx >= ecc->steps)
+		return -ERANGE;
+
+	res->offset = BBM_SIZE + METADATA_SIZE + ecc->bytes * idx;
+	res->length = ecc->bytes;
+
+	return 0;
+}
+
+static int oob_free(struct mtd_info *mtd, int idx, struct mtd_oob_region *res)
+{
+	return -ERANGE; /* no free space in spare area */
+}
+
+static const struct mtd_ooblayout_ops tango_nand_ooblayout_ops = {
+	.ecc	= oob_ecc,
+	.free	= oob_free,
+};
+
+static u32 to_ticks(int kHz, int ps)
+{
+	return DIV_ROUND_UP_ULL((u64)kHz * ps, NSEC_PER_SEC);
+}
+
+static int tango_set_timings(struct mtd_info *mtd,
+		const struct nand_data_interface *conf, bool check_only)
+{
+	const struct nand_sdr_timings *sdr = nand_get_sdr_timings(conf);
+	struct nand_chip *chip = mtd_to_nand(mtd);
+	struct tango_nfc *nfc = to_tango_nfc(chip->controller);
+	struct tango_chip *tchip = to_tango_chip(chip);
+	u32 Trdy, Textw, Twc, Twpw, Tacc, Thold, Trpw, Textr;
+	int kHz = nfc->freq_kHz;
+
+	if (IS_ERR(sdr))
+		return PTR_ERR(sdr);
+
+	if (check_only)
+		return 0;
+
+	Trdy = to_ticks(kHz, sdr->tCEA_max - sdr->tREA_max);
+	Textw = to_ticks(kHz, sdr->tWB_max);
+	Twc = to_ticks(kHz, sdr->tWC_min);
+	Twpw = to_ticks(kHz, sdr->tWC_min - sdr->tWP_min);
+
+	Tacc = to_ticks(kHz, sdr->tREA_max);
+	Thold = to_ticks(kHz, sdr->tREH_min);
+	Trpw = to_ticks(kHz, sdr->tRC_min - sdr->tREH_min);
+	Textr = to_ticks(kHz, sdr->tRHZ_max);
+
+	tchip->timing1 = TIMING(Trdy, Textw, Twc, Twpw);
+	tchip->timing2 = TIMING(Tacc, Thold, Trpw, Textr);
+
+	return 0;
+}
+
+static int chip_init(struct device *dev, struct device_node *np)
+{
+	u32 cs;
+	int err, res;
+	struct mtd_info *mtd;
+	struct nand_chip *chip;
+	struct tango_chip *tchip;
+	struct nand_ecc_ctrl *ecc;
+	struct tango_nfc *nfc = dev_get_drvdata(dev);
+
+	tchip = devm_kzalloc(dev, sizeof(*tchip), GFP_KERNEL);
+	if (!tchip)
+		return -ENOMEM;
+
+	res = of_property_count_u32_elems(np, "reg");
+	if (res < 0)
+		return res;
+
+	if (res != 1)
+		return -ENOTSUPP; /* Multi-CS chips are not supported */
+
+	err = of_property_read_u32_index(np, "reg", 0, &cs);
+	if (err)
+		return err;
+
+	if (cs >= MAX_CS)
+		return -EINVAL;
+
+	chip = &tchip->nand_chip;
+	ecc = &chip->ecc;
+	mtd = &chip->mtd;
+
+	chip->read_byte = tango_read_byte;
+	chip->write_buf = tango_write_buf;
+	chip->read_buf = tango_read_buf;
+	chip->select_chip = tango_select_chip;
+	chip->cmd_ctrl = tango_cmd_ctrl;
+	chip->dev_ready = tango_dev_ready;
+	chip->setup_data_interface = tango_set_timings;
+	chip->options = NAND_USE_BOUNCE_BUFFER
+		| NAND_NO_SUBPAGE_WRITE
+		| NAND_WAIT_TCCS;
+	chip->controller = &nfc->hw;
+	tchip->base = nfc->pbus_base + (cs * 256);
+
+	nand_set_flash_node(chip, np);
+	mtd_set_ooblayout(mtd, &tango_nand_ooblayout_ops);
+	mtd->dev.parent = dev;
+
+	err = nand_scan_ident(mtd, 1, NULL);
+	if (err)
+		return err;
+
+	ecc->mode = NAND_ECC_HW;
+	ecc->algo = NAND_ECC_BCH;
+	ecc->bytes = DIV_ROUND_UP(ecc->strength * FIELD_ORDER, BITS_PER_BYTE);
+
+	ecc->read_page_raw = tango_read_page_raw;
+	ecc->write_page_raw = tango_write_page_raw;
+	ecc->read_page = tango_read_page;
+	ecc->write_page = tango_write_page;
+	ecc->read_oob = tango_read_oob;
+	ecc->write_oob = tango_write_oob;
+
+	err = nand_scan_tail(mtd);
+	if (err)
+		return err;
+
+	tchip->xfer_cfg = XFER_CFG(cs, 1, ecc->steps, METADATA_SIZE);
+	tchip->pkt_0_cfg = PKT_CFG(ecc->size + METADATA_SIZE, ecc->strength);
+	tchip->pkt_n_cfg = PKT_CFG(ecc->size, ecc->strength);
+	tchip->bb_cfg = BB_CFG(mtd->writesize, BBM_SIZE);
+
+	err = mtd_device_register(mtd, NULL, 0);
+	if (err)
+		return err;
+
+	nfc->chips[cs] = tchip;
+
+	return 0;
+}
+
+static int tango_nand_remove(struct platform_device *pdev)
+{
+	int cs;
+	struct tango_nfc *nfc = platform_get_drvdata(pdev);
+
+	dma_release_channel(nfc->chan);
+
+	for (cs = 0; cs < MAX_CS; ++cs) {
+		if (nfc->chips[cs] != NULL)
+			nand_release(&nfc->chips[cs]->nand_chip.mtd);
+	}
+
+	return 0;
+}
+
+static int tango_nand_probe(struct platform_device *pdev)
+{
+	int err;
+	struct clk *clk;
+	struct resource *res;
+	struct tango_nfc *nfc;
+	struct device_node *np;
+
+	nfc = devm_kzalloc(&pdev->dev, sizeof(*nfc), GFP_KERNEL);
+	if (!nfc)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	nfc->reg_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(nfc->reg_base))
+		return PTR_ERR(nfc->reg_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	nfc->mem_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(nfc->mem_base))
+		return PTR_ERR(nfc->mem_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 2);
+	nfc->pbus_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(nfc->pbus_base))
+		return PTR_ERR(nfc->pbus_base);
+
+	clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+
+	nfc->chan = dma_request_chan(&pdev->dev, "nfc_sbox");
+	if (IS_ERR(nfc->chan))
+		return PTR_ERR(nfc->chan);
+
+	platform_set_drvdata(pdev, nfc);
+	nand_hw_control_init(&nfc->hw);
+	nfc->freq_kHz = clk_get_rate(clk) / 1000;
+
+	for_each_child_of_node(pdev->dev.of_node, np) {
+		err = chip_init(&pdev->dev, np);
+		if (err) {
+			tango_nand_remove(pdev);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static const struct of_device_id tango_nand_ids[] = {
+	{ .compatible = "sigma,smp8758-nand" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver tango_nand_driver = {
+	.probe	= tango_nand_probe,
+	.remove	= tango_nand_remove,
+	.driver	= {

drivers/mtd/nand/tango_nand.c line 654

Quoted 10 lines
drivers/mtd/nand/tango_nand.c new file +10 -0
+		.name		= "tango-nand",
+		.of_match_table	= tango_nand_ids,
+	},
+};
+
+module_platform_driver(tango_nand_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sigma Designs");
+MODULE_DESCRIPTION("Tango4 NAND Flash controller driver");

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#21 lore raw

Re: [PATCH v8] mtd: nand: add tango NAND flash controller support

Quoted 1 line · Boris Brezillon

Applied after fixing a few coding style issues to make checkpatch happy.

First of all, I want to thank you for all the help you provided along the way.

I have to admit that I'm a bit frustrated by some of the changes you made to the patch.

Specifically, changing

from: if (ptr_expr == NULL)
to: if (!ptr_expr)

I dislike the second form, because it "hides" what is being tested. I've even seen people use !count to mean count == 0, and I find that very unintuitive.

I also have a hard time discerning a '!' after a '('

The other change is the chip->options initialization. I thought you said on IRC I could keep my preferred formatting.

In the end, these are tiny issues in the grand scheme of things, but it was important for me to point them out.

Regards.

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): [email protected]

#23 lore raw

Re: [PATCH v8] mtd: nand: add tango NAND flash controller support

Hi Marc,

Quoted 10 lines · On Fri, 11 Nov 2016 17:01:52 +0100 Mason…
Quoted 1 line · Boris Brezillon

Applied after fixing a few coding style issues to make checkpatch happy.

First of all, I want to thank you for all the help you provided along the way.

I have to admit that I'm a bit frustrated by some of the changes you made to the patch.

Quoted 8 lines

Specifically, changing

from: if (ptr_expr == NULL)
to: if (!ptr_expr)

I dislike the second form, because it "hides" what is being tested. I've even seen people use !count to mean count == 0, and I find that very unintuitive.

I just fixed checkpatch --strict complaints.

Quoted 6 lines

I also have a hard time discerning a '!' after a '('

The other change is the chip->options initialization. I thought you said on IRC I could keep my preferred formatting.

Hm, I said that you could use one flag per line, and I kept that. I just moved the 'or' operator which is, IMO, better placed at the end of the line, and I aligned the following flags with the first one to make checkpatch happy.

Quoted 4 lines

In the end, these are tiny issues in the grand scheme of things, but it was important for me to point them out.

Some maintainers don't care much, but I think enforcing these rules is what brings consistency in the kernel code, that's why I try to make checkpatch happy when I submit new code and ask developers to do so when they submit code to the NAND subsystem.

I know you often disagree with the Linux coding style rules, but you're a contributor to this project, and as such, you should commit to the defined rules (and this includes the coding style rules). And it appears that, instead of complying to these rules when you're asked to, you keep arguing endlessly that your own coding style is better. You'd better spend time on things that really matters.

Note that most of the time, when the contribution is looking good, but a few coding style issue remain, I'm fixing those issues myself because I know how annoying it can be to ask someone to post a new version to fix such minor things.
But since you complain about that, next time I'll ask you to make "checkpatch --strict" happy before accepting your patch.

On a side note, as I already told you privately, you seem to be a smart guy (the evolution of the tango NAND driver shows that you're able to understand how things work, and take comments/reviews into account). But you have the bad habit of consistently arguing when you're asked to change something (and especially on minor things that are easy to fix). This is really the kind of behavior that can upset maintainers (me included). So please, try to work on that.

Regards,

Boris
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#24 lore raw

Re: [PATCH v8] mtd: nand: add tango NAND flash controller support

In reply to a message not in the archive

Hi,

Quoted 5 lines · Marc Gonzalez

This driver supports the NAND Flash controller embedded in recent Tango chips, such as SMP8758 and SMP8759.

---

...

drivers/mtd/nand/tango_nand.c

Quoted 6 lines
drivers/mtd/nand/tango_nand.c new file +0 -0
new file mode 100644
index 000000000000..74e39a92771c
@@ -0,0 +1,654 @@

This driver is being introduced with no copyright or license header. At least it has an appropriate MODULE_LICENSE().

Can the original author(s) please provide a follow-up patch with one? Thanks.

Quoted 9 lines
drivers/mtd/nand/tango_nand.c new file +9 -0
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/clk.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/mtd/nand.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>

...

drivers/mtd/nand/tango_nand.c line 654

Quoted 3 lines
drivers/mtd/nand/tango_nand.c new file +3 -0
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sigma Designs");
+MODULE_DESCRIPTION("Tango4 NAND Flash controller driver");

Brian
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html