devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support
@ 2025-10-20 10:12 Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 01/15] mtd: rawnand: sunxi: Remove superfluous register readings Richard Genoud
                   ` (14 more replies)
  0 siblings, 15 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:12 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

Hi everyone,

This patch series introduce H6/H616 NAND controller support (but not yet
the DMA/MDMA part).

All the work was done on a H616 board with a Kioxia TC58NVG1S3HTA00 NAND
chip.
ECC is supported, as well as scrambling.

H6 SoC has not been tested, but it shares all registers with H616, plus
some registers dedicated to Embedded Crypto Engine that H616 lacks.

This IP has quite some register fields modified from the A10/A23, but in
the end, it works more or less the same.
Main differences with A10/A23 are:
- The need for 2 more clocks (for ECC and MBUS)
- The use of a new USER_DATA_LEN register
- More ECC strengths (44, 52, 68, 72, 76 and 80 bits / 1KB)
- MDMA based on chained buffers
- some registers layouts have changed (mainly due do the stronger ECC)

Patch 1 removes a superfluous call, and is independent from the others.

Patches 2-12 are paving the way to the introduction of H616 NAND
controller support.
They are small, without functional change and easier to review than a
big patch.
They move a fixed value (register offset, field mask, SRAM size) into
the struct sunxi_nfc_caps when this value is different on H6/H616.

Patch 13 introduces the support for H6/H616 NAND controller.

Patch 14 introduces allwinner,sun50i-h616-nand-controller compatible
in order to differentiate from the A10 and A23.

Patch 15 adds the NAND controller node to sun50i-h616.dtsi

Changes from v2:
- add missing 'status = "disabled";' in nand controller node.
- factorize back clock/clock-names in binding.
- pick-up reviewed-by

Changes from v1:
- reorder patches to have dtsi/bindings patches at the end.
- reorder nand-controller and pins nodes to respect the order.
- add /omit-if-no-ref/ on pins that may be unused.
- remove pinctrl from nand controller node (this should be added to device
 DT).
- rework dt binding.
- fix H616 comment on chained descriptors support.
- add missing mbus_clk description.
- make ECC clock mandatory for H6 (because it's indeed mandatory).
- harmonize new clock retrieving error messages with older ones.
- harmonize commits subjects (mtd: rawnand: sunxi).
- reword commit messages to use imperative mood.
- pick-up reviewed-by

Thanks Krzysztof, Jernej and Chen-Yu for the reviews!

Regards,
Richard

Richard Genoud (15):
  mtd: rawnand: sunxi: Remove superfluous register readings
  mtd: rawnand: sunxi: move ECC strenghts in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce reg_ecc_err_cnt in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce reg_user_data in sunxi_nfc_caps
  mtd: rawnand: sunxi: rework pattern found registers
  mtd: rawnand: sunxi: add has_ecc_block_512 capability
  mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce random en/dir in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce reg_pat_id in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce reg_spare_area in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce ecc_err_mask in sunxi_nfc_caps
  mtd: rawnand: sunxi: introduce sram_size in sunxi_nfc_caps
  mtd: rawnand: sunxi: Add support for H616 nand controller
  dt-bindings: mtd: sunxi: Add H616 compatible
  arm64: dts: allwinner: h616: add NAND controller

 .../mtd/allwinner,sun4i-a10-nand.yaml         |  41 +-
 .../arm64/boot/dts/allwinner/sun50i-h616.dtsi |  52 +++
 drivers/mtd/nand/raw/sunxi_nand.c             | 350 +++++++++++++++---
 3 files changed, 395 insertions(+), 48 deletions(-)


base-commit: 3a8660878839faadb4f1a6dd72c3179c1df56787

^ permalink raw reply	[flat|nested] 28+ messages in thread

* [PATCH v3 01/15] mtd: rawnand: sunxi: Remove superfluous register readings
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
@ 2025-10-20 10:12 ` Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 02/15] mtd: rawnand: sunxi: move ECC strenghts in sunxi_nfc_caps Richard Genoud
                   ` (13 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:12 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The register NFC_REG_ECC_CTL was read twice and the result was not used,
then a third time with a mask applied.
Removing those calls didn't change the behavior.

Tested on H616 SoC, scrambling enabled.

Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index f6a8e8ae819d..cb12179b63a5 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -623,13 +623,12 @@ static void sunxi_nfc_randomizer_config(struct nand_chip *nand, int page,
 					bool ecc)
 {
 	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
-	u32 ecc_ctl = readl(nfc->regs + NFC_REG_ECC_CTL);
+	u32 ecc_ctl;
 	u16 state;
 
 	if (!(nand->options & NAND_NEED_SCRAMBLING))
 		return;
 
-	ecc_ctl = readl(nfc->regs + NFC_REG_ECC_CTL);
 	state = sunxi_nfc_randomizer_state(nand, page, ecc);
 	ecc_ctl = readl(nfc->regs + NFC_REG_ECC_CTL) & ~NFC_RANDOM_SEED_MSK;
 	writel(ecc_ctl | NFC_RANDOM_SEED(state), nfc->regs + NFC_REG_ECC_CTL);

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 02/15] mtd: rawnand: sunxi: move ECC strenghts in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 01/15] mtd: rawnand: sunxi: Remove superfluous register readings Richard Genoud
@ 2025-10-20 10:12 ` Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 03/15] mtd: rawnand: sunxi: introduce reg_ecc_err_cnt " Richard Genoud
                   ` (12 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:12 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

H6/H616 has more ECC strenghts than A10/A23.

Move the ECC strenghts array to sunxi_nfc_caps to make it ready for
H6/H616 support.

No functional change.

Reviewed-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index cb12179b63a5..d3b17d54f04c 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -213,11 +213,15 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  *			through MBUS on A23/A33 needs extra configuration.
  * @reg_io_data:	I/O data register
  * @dma_maxburst:	DMA maxburst
+ * @ecc_strengths:	Available ECC strengths array
+ * @nstrengths:		Size of @ecc_strengths
  */
 struct sunxi_nfc_caps {
 	bool has_mdma;
 	unsigned int reg_io_data;
 	unsigned int dma_maxburst;
+	const u8 *ecc_strengths;
+	unsigned int nstrengths;
 };
 
 /**
@@ -1619,9 +1623,9 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 				       struct nand_ecc_ctrl *ecc,
 				       struct device_node *np)
 {
-	static const u8 strengths[] = { 16, 24, 28, 32, 40, 48, 56, 60, 64 };
 	struct sunxi_nand_chip *sunxi_nand = to_sunxi_nand(nand);
 	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
+	const u8 *strengths = nfc->caps->ecc_strengths;
 	struct mtd_info *mtd = nand_to_mtd(nand);
 	struct nand_device *nanddev = mtd_to_nanddev(mtd);
 	int nsectors;
@@ -1645,7 +1649,7 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 
 		ecc->strength = bytes * 8 / fls(8 * ecc->size);
 
-		for (i = 0; i < ARRAY_SIZE(strengths); i++) {
+		for (i = 0; i < nfc->caps->nstrengths; i++) {
 			if (strengths[i] > ecc->strength)
 				break;
 		}
@@ -1666,7 +1670,7 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 	}
 
 	/* Add ECC info retrieval from DT */
-	for (i = 0; i < ARRAY_SIZE(strengths); i++) {
+	for (i = 0; i < nfc->caps->nstrengths; i++) {
 		if (ecc->strength <= strengths[i]) {
 			/*
 			 * Update ecc->strength value with the actual strength
@@ -1677,7 +1681,7 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 		}
 	}
 
-	if (i >= ARRAY_SIZE(strengths)) {
+	if (i >= nfc->caps->nstrengths) {
 		dev_err(nfc->dev, "unsupported strength\n");
 		return -ENOTSUPP;
 	}
@@ -2167,15 +2171,23 @@ static void sunxi_nfc_remove(struct platform_device *pdev)
 		dma_release_channel(nfc->dmac);
 }
 
+static const u8 sunxi_ecc_strengths_a10[] = {
+	16, 24, 28, 32, 40, 48, 56, 60, 64
+};
+
 static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.dma_maxburst = 4,
+	.ecc_strengths = sunxi_ecc_strengths_a10,
+	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
 };
 
 static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.has_mdma = true,
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.dma_maxburst = 8,
+	.ecc_strengths = sunxi_ecc_strengths_a10,
+	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
 };
 
 static const struct of_device_id sunxi_nfc_ids[] = {

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 03/15] mtd: rawnand: sunxi: introduce reg_ecc_err_cnt in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 01/15] mtd: rawnand: sunxi: Remove superfluous register readings Richard Genoud
  2025-10-20 10:12 ` [PATCH v3 02/15] mtd: rawnand: sunxi: move ECC strenghts in sunxi_nfc_caps Richard Genoud
@ 2025-10-20 10:12 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data " Richard Genoud
                   ` (11 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:12 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 ECC_ERR_CNT register is not at the same offset as the
A10/A23 one, so move its offset into sunxi_nfc_caps

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index d3b17d54f04c..0285e4d0ca7f 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -46,7 +46,8 @@
 #define NFC_REG_ECC_CTL		0x0034
 #define NFC_REG_ECC_ST		0x0038
 #define NFC_REG_DEBUG		0x003C
-#define NFC_REG_ECC_ERR_CNT(x)	((0x0040 + (x)) & ~0x3)
+#define NFC_REG_A10_ECC_ERR_CNT	0x0040
+#define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
 #define NFC_REG_USER_DATA(x)	(0x0050 + ((x) * 4))
 #define NFC_REG_SPARE_AREA	0x00A0
 #define NFC_REG_PAT_ID		0x00A4
@@ -212,6 +213,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @has_mdma:		Use mbus dma mode, otherwise general dma
  *			through MBUS on A23/A33 needs extra configuration.
  * @reg_io_data:	I/O data register
+ * @reg_ecc_err_cnt:	ECC error counter register
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
  * @nstrengths:		Size of @ecc_strengths
@@ -219,6 +221,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
 struct sunxi_nfc_caps {
 	bool has_mdma;
 	unsigned int reg_io_data;
+	unsigned int reg_ecc_err_cnt;
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
 	unsigned int nstrengths;
@@ -790,7 +793,7 @@ static int sunxi_nfc_hw_ecc_correct(struct nand_chip *nand, u8 *data, u8 *oob,
 		return 0;
 	}
 
-	tmp = readl(nfc->regs + NFC_REG_ECC_ERR_CNT(step));
+	tmp = readl(nfc->regs + NFC_REG_ECC_ERR_CNT(nfc, step));
 
 	return NFC_ECC_ERR_CNT(step, tmp);
 }
@@ -2177,6 +2180,7 @@ static const u8 sunxi_ecc_strengths_a10[] = {
 
 static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
+	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
@@ -2185,6 +2189,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.has_mdma = true,
 	.reg_io_data = NFC_REG_A23_IO_DATA,
+	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (2 preceding siblings ...)
  2025-10-20 10:12 ` [PATCH v3 03/15] mtd: rawnand: sunxi: introduce reg_ecc_err_cnt " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-22  8:54   ` Miquel Raynal
  2025-10-20 10:13 ` [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers Richard Genoud
                   ` (10 subsequent siblings)
  14 siblings, 1 reply; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 USER_DATA register is not at the same offset as the
A10/A23 one, so move its offset into sunxi_nfc_caps

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 0285e4d0ca7f..8f5d8df19e33 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -48,7 +48,8 @@
 #define NFC_REG_DEBUG		0x003C
 #define NFC_REG_A10_ECC_ERR_CNT	0x0040
 #define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
-#define NFC_REG_USER_DATA(x)	(0x0050 + ((x) * 4))
+#define NFC_REG_A10_USER_DATA	0x0050
+#define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
 #define NFC_REG_SPARE_AREA	0x00A0
 #define NFC_REG_PAT_ID		0x00A4
 #define NFC_REG_MDMA_ADDR	0x00C0
@@ -214,6 +215,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  *			through MBUS on A23/A33 needs extra configuration.
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
+ * @reg_user_data:	User data register
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
  * @nstrengths:		Size of @ecc_strengths
@@ -222,6 +224,7 @@ struct sunxi_nfc_caps {
 	bool has_mdma;
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
+	unsigned int reg_user_data;
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
 	unsigned int nstrengths;
@@ -723,8 +726,8 @@ static void sunxi_nfc_hw_ecc_get_prot_oob_bytes(struct nand_chip *nand, u8 *oob,
 {
 	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
 
-	sunxi_nfc_user_data_to_buf(readl(nfc->regs + NFC_REG_USER_DATA(step)),
-				   oob);
+	sunxi_nfc_user_data_to_buf(readl(nfc->regs +
+					 NFC_REG_USER_DATA(nfc, step)), oob);
 
 	/* De-randomize the Bad Block Marker. */
 	if (bbm && (nand->options & NAND_NEED_SCRAMBLING))
@@ -746,7 +749,7 @@ static void sunxi_nfc_hw_ecc_set_prot_oob_bytes(struct nand_chip *nand,
 	}
 
 	writel(sunxi_nfc_buf_to_user_data(oob),
-	       nfc->regs + NFC_REG_USER_DATA(step));
+	       nfc->regs + NFC_REG_USER_DATA(nfc, step));
 }
 
 static void sunxi_nfc_hw_ecc_update_stats(struct nand_chip *nand,
@@ -2181,6 +2184,7 @@ static const u8 sunxi_ecc_strengths_a10[] = {
 static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
+	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
@@ -2190,6 +2194,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.has_mdma = true,
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
+	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (3 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-22  9:02   ` Miquel Raynal
  2025-10-22 10:05   ` Geert Uytterhoeven
  2025-10-20 10:13 ` [PATCH v3 06/15] mtd: rawnand: sunxi: add has_ecc_block_512 capability Richard Genoud
                   ` (9 subsequent siblings)
  14 siblings, 2 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

On H6/H616, the register ECC_PAT_FOUND is at its own address, and not
part of ECC status register.
So, introduce the pattern found register offset in sunxi_nfc_caps, along
with its mask.

Also, introduce a non compile-time field_get() because FIELD_GET() and
u32_get_bits() don't work with non compile-time constant.
https://lore.kernel.org/lkml/cover.1739540679.git.geert+renesas@glider.be/

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 36 ++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 8f5d8df19e33..4cfb5d3e9c06 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -29,6 +29,9 @@
 #include <linux/iopoll.h>
 #include <linux/reset.h>
 
+/* non compile-time field get */
+#define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1))
+
 #define NFC_REG_CTL		0x0000
 #define NFC_REG_ST		0x0004
 #define NFC_REG_INT		0x0008
@@ -150,7 +153,13 @@
 /* define bit use in NFC_ECC_ST */
 #define NFC_ECC_ERR(x)		BIT(x)
 #define NFC_ECC_ERR_MSK		GENMASK(15, 0)
-#define NFC_ECC_PAT_FOUND(x)	BIT(x + 16)
+
+/*
+ * define bit use in NFC_REG_PAT_FOUND
+ * For A10/A23, NFC_REG_PAT_FOUND == NFC_ECC_ST register
+ */
+#define NFC_ECC_PAT_FOUND_MSK(nfc) (nfc->caps->pat_found_mask)
+
 #define NFC_ECC_ERR_CNT(b, x)	(((x) >> (((b) % 4) * 8)) & 0xff)
 
 #define NFC_DEFAULT_TIMEOUT_MS	1000
@@ -216,6 +225,8 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
+ * @reg_pat_found:	Data Pattern Status Register
+ * @pat_found_mask:	ECC_PAT_FOUND mask in NFC_REG_PAT_FOUND register
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
  * @nstrengths:		Size of @ecc_strengths
@@ -225,6 +236,8 @@ struct sunxi_nfc_caps {
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
+	unsigned int reg_pat_found;
+	unsigned int pat_found_mask;
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
 	unsigned int nstrengths;
@@ -766,7 +779,8 @@ static void sunxi_nfc_hw_ecc_update_stats(struct nand_chip *nand,
 }
 
 static int sunxi_nfc_hw_ecc_correct(struct nand_chip *nand, u8 *data, u8 *oob,
-				    int step, u32 status, bool *erased)
+				    int step, u32 status, u32 pattern_found,
+				    bool *erased)
 {
 	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
 	struct nand_ecc_ctrl *ecc = &nand->ecc;
@@ -777,7 +791,7 @@ static int sunxi_nfc_hw_ecc_correct(struct nand_chip *nand, u8 *data, u8 *oob,
 	if (status & NFC_ECC_ERR(step))
 		return -EBADMSG;
 
-	if (status & NFC_ECC_PAT_FOUND(step)) {
+	if (pattern_found & BIT(step)) {
 		u8 pattern;
 
 		if (unlikely(!(readl(nfc->regs + NFC_REG_PAT_ID) & 0x1))) {
@@ -811,6 +825,7 @@ static int sunxi_nfc_hw_ecc_read_chunk(struct nand_chip *nand,
 	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
 	struct nand_ecc_ctrl *ecc = &nand->ecc;
 	int raw_mode = 0;
+	u32 pattern_found;
 	bool erased;
 	int ret;
 
@@ -838,8 +853,12 @@ static int sunxi_nfc_hw_ecc_read_chunk(struct nand_chip *nand,
 
 	*cur_off = oob_off + ecc->bytes + 4;
 
+	pattern_found = readl(nfc->regs + nfc->caps->reg_pat_found);
+	pattern_found = field_get(NFC_ECC_PAT_FOUND_MSK(nfc), pattern_found);
+
 	ret = sunxi_nfc_hw_ecc_correct(nand, data, oob_required ? oob : NULL, 0,
 				       readl(nfc->regs + NFC_REG_ECC_ST),
+				       pattern_found,
 				       &erased);
 	if (erased)
 		return 1;
@@ -920,7 +939,7 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
 	unsigned int max_bitflips = 0;
 	int ret, i, raw_mode = 0;
 	struct scatterlist sg;
-	u32 status, wait;
+	u32 status, pattern_found, wait;
 
 	ret = sunxi_nfc_wait_cmd_fifo_empty(nfc);
 	if (ret)
@@ -961,6 +980,8 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
 		return ret;
 
 	status = readl(nfc->regs + NFC_REG_ECC_ST);
+	pattern_found = readl(nfc->regs + nfc->caps->reg_pat_found);
+	pattern_found = field_get(NFC_ECC_PAT_FOUND_MSK(nfc), pattern_found);
 
 	for (i = 0; i < nchunks; i++) {
 		int data_off = i * ecc->size;
@@ -971,7 +992,8 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
 
 		ret = sunxi_nfc_hw_ecc_correct(nand, randomized ? data : NULL,
 					       oob_required ? oob : NULL,
-					       i, status, &erased);
+					       i, status, pattern_found,
+					       &erased);
 
 		/* ECC errors are handled in the second loop. */
 		if (ret < 0)
@@ -2185,6 +2207,8 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_pat_found = NFC_REG_ECC_ST,
+	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
@@ -2195,6 +2219,8 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_pat_found = NFC_REG_ECC_ST,
+	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 06/15] mtd: rawnand: sunxi: add has_ecc_block_512 capability
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (4 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps Richard Genoud
                   ` (8 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H616 controller can't handle 512 bytes ECC block size. The
NFC_ECC_BLOCK_512 bit disappeared in H6, and NDFC_RANDOM_EN took its
place.

So, add has_ecc_block_512 capability to only set this bit on SoC having
it.

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 4cfb5d3e9c06..70c940840dda 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -222,6 +222,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  *
  * @has_mdma:		Use mbus dma mode, otherwise general dma
  *			through MBUS on A23/A33 needs extra configuration.
+ * @has_ecc_block_512:	If the ECC can handle 512B or only 1024B chuncks
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
@@ -233,6 +234,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  */
 struct sunxi_nfc_caps {
 	bool has_mdma;
+	bool has_ecc_block_512;
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
@@ -1748,8 +1750,14 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 	sunxi_nand->ecc.ecc_ctl = NFC_ECC_MODE(i) | NFC_ECC_EXCEPTION |
 				  NFC_ECC_PIPELINE | NFC_ECC_EN;
 
-	if (ecc->size == 512)
-		sunxi_nand->ecc.ecc_ctl |= NFC_ECC_BLOCK_512;
+	if (ecc->size == 512) {
+		if (nfc->caps->has_ecc_block_512) {
+			sunxi_nand->ecc.ecc_ctl |= NFC_ECC_BLOCK_512;
+		} else {
+			dev_err(nfc->dev, "512B ECC block not supported\n");
+			return -EOPNOTSUPP;
+		}
+	}
 
 	return 0;
 }
@@ -2204,6 +2212,7 @@ static const u8 sunxi_ecc_strengths_a10[] = {
 };
 
 static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
+	.has_ecc_block_512 = true,
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
@@ -2216,6 +2225,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 
 static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.has_mdma = true,
+	.has_ecc_block_512 = true,
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (5 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 06/15] mtd: rawnand: sunxi: add has_ecc_block_512 capability Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-22  9:05   ` Miquel Raynal
  2025-10-20 10:13 ` [PATCH v3 08/15] mtd: rawnand: sunxi: introduce random en/dir " Richard Genoud
                   ` (7 subsequent siblings)
  14 siblings, 1 reply; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 ECC_MODE field is not at the same offset, and has not the
same size.
So move the mask into sunxi_nfc_caps.

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 70c940840dda..4118307cac15 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -29,8 +29,9 @@
 #include <linux/iopoll.h>
 #include <linux/reset.h>
 
-/* non compile-time field get */
+/* non compile-time field get/prep */
 #define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1))
+#define field_prep(_mask, _val) (((_val) << (ffs(_mask) - 1)) & (_mask))
 
 #define NFC_REG_CTL		0x0000
 #define NFC_REG_ST		0x0004
@@ -145,8 +146,8 @@
 #define NFC_ECC_BLOCK_512	BIT(5)
 #define NFC_RANDOM_EN		BIT(9)
 #define NFC_RANDOM_DIRECTION	BIT(10)
-#define NFC_ECC_MODE_MSK	GENMASK(15, 12)
-#define NFC_ECC_MODE(x)		((x) << 12)
+#define NFC_ECC_MODE_MSK(nfc)	(nfc->caps->ecc_mode_mask)
+#define NFC_ECC_MODE(nfc, x)	field_prep(NFC_ECC_MODE_MSK(nfc), (x))
 #define NFC_RANDOM_SEED_MSK	GENMASK(30, 16)
 #define NFC_RANDOM_SEED(x)	((x) << 16)
 
@@ -227,6 +228,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
  * @reg_pat_found:	Data Pattern Status Register
+ * @ecc_mode_mask:	ECC_MODE mask in NFC_ECC_CTL register
  * @pat_found_mask:	ECC_PAT_FOUND mask in NFC_REG_PAT_FOUND register
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
@@ -239,6 +241,7 @@ struct sunxi_nfc_caps {
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
 	unsigned int reg_pat_found;
+	unsigned int ecc_mode_mask;
 	unsigned int pat_found_mask;
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
@@ -1747,7 +1750,7 @@ static int sunxi_nand_hw_ecc_ctrl_init(struct nand_chip *nand,
 	ecc->read_oob_raw = nand_read_oob_std;
 	ecc->write_oob_raw = nand_write_oob_std;
 
-	sunxi_nand->ecc.ecc_ctl = NFC_ECC_MODE(i) | NFC_ECC_EXCEPTION |
+	sunxi_nand->ecc.ecc_ctl = NFC_ECC_MODE(nfc, i) | NFC_ECC_EXCEPTION |
 				  NFC_ECC_PIPELINE | NFC_ECC_EN;
 
 	if (ecc->size == 512) {
@@ -2217,6 +2220,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.reg_pat_found = NFC_REG_ECC_ST,
+	.ecc_mode_mask = GENMASK(15, 12),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
@@ -2230,6 +2234,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.reg_pat_found = NFC_REG_ECC_ST,
+	.ecc_mode_mask = GENMASK(15, 12),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 08/15] mtd: rawnand: sunxi: introduce random en/dir in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (6 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 09/15] mtd: rawnand: sunxi: introduce reg_pat_id " Richard Genoud
                   ` (6 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 RANDOM EN/DIRECTION masks are different from A10/A23.
So move the masks into sunxi_nfc_caps.

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 4118307cac15..3ccb7ea7cf17 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -144,8 +144,8 @@
 #define NFC_ECC_EXCEPTION	BIT(4)
 #define NFC_ECC_BLOCK_SIZE_MSK	BIT(5)
 #define NFC_ECC_BLOCK_512	BIT(5)
-#define NFC_RANDOM_EN		BIT(9)
-#define NFC_RANDOM_DIRECTION	BIT(10)
+#define NFC_RANDOM_EN(nfc)	(nfc->caps->random_en_mask)
+#define NFC_RANDOM_DIRECTION(nfc) (nfc->caps->random_dir_mask)
 #define NFC_ECC_MODE_MSK(nfc)	(nfc->caps->ecc_mode_mask)
 #define NFC_ECC_MODE(nfc, x)	field_prep(NFC_ECC_MODE_MSK(nfc), (x))
 #define NFC_RANDOM_SEED_MSK	GENMASK(30, 16)
@@ -228,6 +228,8 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
  * @reg_pat_found:	Data Pattern Status Register
+ * @random_en_mask:	RANDOM_EN mask in NFC_ECC_CTL register
+ * @random_dir_mask:	RANDOM_DIRECTION mask in NFC_ECC_CTL register
  * @ecc_mode_mask:	ECC_MODE mask in NFC_ECC_CTL register
  * @pat_found_mask:	ECC_PAT_FOUND mask in NFC_REG_PAT_FOUND register
  * @dma_maxburst:	DMA maxburst
@@ -241,6 +243,8 @@ struct sunxi_nfc_caps {
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
 	unsigned int reg_pat_found;
+	unsigned int random_en_mask;
+	unsigned int random_dir_mask;
 	unsigned int ecc_mode_mask;
 	unsigned int pat_found_mask;
 	unsigned int dma_maxburst;
@@ -669,7 +673,7 @@ static void sunxi_nfc_randomizer_enable(struct nand_chip *nand)
 	if (!(nand->options & NAND_NEED_SCRAMBLING))
 		return;
 
-	writel(readl(nfc->regs + NFC_REG_ECC_CTL) | NFC_RANDOM_EN,
+	writel(readl(nfc->regs + NFC_REG_ECC_CTL) | NFC_RANDOM_EN(nfc),
 	       nfc->regs + NFC_REG_ECC_CTL);
 }
 
@@ -680,7 +684,7 @@ static void sunxi_nfc_randomizer_disable(struct nand_chip *nand)
 	if (!(nand->options & NAND_NEED_SCRAMBLING))
 		return;
 
-	writel(readl(nfc->regs + NFC_REG_ECC_CTL) & ~NFC_RANDOM_EN,
+	writel(readl(nfc->regs + NFC_REG_ECC_CTL) & ~NFC_RANDOM_EN(nfc),
 	       nfc->regs + NFC_REG_ECC_CTL);
 }
 
@@ -2220,6 +2224,8 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.reg_pat_found = NFC_REG_ECC_ST,
+	.random_en_mask = BIT(9),
+	.random_dir_mask = BIT(10),
 	.ecc_mode_mask = GENMASK(15, 12),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 4,
@@ -2234,6 +2240,8 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
 	.reg_pat_found = NFC_REG_ECC_ST,
+	.random_en_mask = BIT(9),
+	.random_dir_mask = BIT(10),
 	.ecc_mode_mask = GENMASK(15, 12),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 8,

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 09/15] mtd: rawnand: sunxi: introduce reg_pat_id in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (7 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 08/15] mtd: rawnand: sunxi: introduce random en/dir " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 10/15] mtd: rawnand: sunxi: introduce reg_spare_area " Richard Genoud
                   ` (5 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 pattern ID register is not at the same offset as the
A10/A23 one, so move its offset into sunxi_nfc_caps.

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 3ccb7ea7cf17..e8c08f0139d2 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -55,7 +55,8 @@
 #define NFC_REG_A10_USER_DATA	0x0050
 #define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
 #define NFC_REG_SPARE_AREA	0x00A0
-#define NFC_REG_PAT_ID		0x00A4
+#define NFC_REG_PAT_ID(nfc) (nfc->caps->reg_pat_id)
+#define NFC_REG_A10_PAT_ID	0x00A4
 #define NFC_REG_MDMA_ADDR	0x00C0
 #define NFC_REG_MDMA_CNT	0x00C4
 #define NFC_RAM0_BASE		0x0400
@@ -227,6 +228,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
+ * @reg_pat_id:		Pattern ID Register
  * @reg_pat_found:	Data Pattern Status Register
  * @random_en_mask:	RANDOM_EN mask in NFC_ECC_CTL register
  * @random_dir_mask:	RANDOM_DIRECTION mask in NFC_ECC_CTL register
@@ -242,6 +244,7 @@ struct sunxi_nfc_caps {
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
+	unsigned int reg_pat_id;
 	unsigned int reg_pat_found;
 	unsigned int random_en_mask;
 	unsigned int random_dir_mask;
@@ -803,7 +806,7 @@ static int sunxi_nfc_hw_ecc_correct(struct nand_chip *nand, u8 *data, u8 *oob,
 	if (pattern_found & BIT(step)) {
 		u8 pattern;
 
-		if (unlikely(!(readl(nfc->regs + NFC_REG_PAT_ID) & 0x1))) {
+		if (unlikely(!(readl(nfc->regs + NFC_REG_PAT_ID(nfc)) & 0x1))) {
 			pattern = 0x0;
 		} else {
 			pattern = 0xff;
@@ -2223,6 +2226,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_pat_id = NFC_REG_A10_PAT_ID,
 	.reg_pat_found = NFC_REG_ECC_ST,
 	.random_en_mask = BIT(9),
 	.random_dir_mask = BIT(10),
@@ -2239,6 +2243,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_pat_id = NFC_REG_A10_PAT_ID,
 	.reg_pat_found = NFC_REG_ECC_ST,
 	.random_en_mask = BIT(9),
 	.random_dir_mask = BIT(10),

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 10/15] mtd: rawnand: sunxi: introduce reg_spare_area in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (8 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 09/15] mtd: rawnand: sunxi: introduce reg_pat_id " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 11/15] mtd: rawnand: sunxi: introduce ecc_err_mask " Richard Genoud
                   ` (4 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 spare area register is not at the same offset as the
A10/A23 one, so move its offset into sunxi_nfc_caps.

No functional change.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index e8c08f0139d2..da7b8b81d39e 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -54,7 +54,8 @@
 #define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
 #define NFC_REG_A10_USER_DATA	0x0050
 #define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
-#define NFC_REG_SPARE_AREA	0x00A0
+#define NFC_REG_SPARE_AREA(nfc) (nfc->caps->reg_spare_area)
+#define NFC_REG_A10_SPARE_AREA	0x00A0
 #define NFC_REG_PAT_ID(nfc) (nfc->caps->reg_pat_id)
 #define NFC_REG_A10_PAT_ID	0x00A4
 #define NFC_REG_MDMA_ADDR	0x00C0
@@ -228,6 +229,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
+ * @reg_spare_area:	Spare Area Register
  * @reg_pat_id:		Pattern ID Register
  * @reg_pat_found:	Data Pattern Status Register
  * @random_en_mask:	RANDOM_EN mask in NFC_ECC_CTL register
@@ -244,6 +246,7 @@ struct sunxi_nfc_caps {
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
+	unsigned int reg_spare_area;
 	unsigned int reg_pat_id;
 	unsigned int reg_pat_found;
 	unsigned int random_en_mask;
@@ -466,7 +469,7 @@ static void sunxi_nfc_select_chip(struct nand_chip *nand, unsigned int cs)
 	if (sel->rb >= 0)
 		ctl |= NFC_RB_SEL(sel->rb);
 
-	writel(mtd->writesize, nfc->regs + NFC_REG_SPARE_AREA);
+	writel(mtd->writesize, nfc->regs + NFC_REG_SPARE_AREA(nfc));
 
 	if (nfc->clk_rate != sunxi_nand->clk_rate) {
 		clk_set_rate(nfc->mod_clk, sunxi_nand->clk_rate);
@@ -2226,6 +2229,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_spare_area = NFC_REG_A10_SPARE_AREA,
 	.reg_pat_id = NFC_REG_A10_PAT_ID,
 	.reg_pat_found = NFC_REG_ECC_ST,
 	.random_en_mask = BIT(9),
@@ -2243,6 +2247,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
+	.reg_spare_area = NFC_REG_A10_SPARE_AREA,
 	.reg_pat_id = NFC_REG_A10_PAT_ID,
 	.reg_pat_found = NFC_REG_ECC_ST,
 	.random_en_mask = BIT(9),

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 11/15] mtd: rawnand: sunxi: introduce ecc_err_mask in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (9 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 10/15] mtd: rawnand: sunxi: introduce reg_spare_area " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 12/15] mtd: rawnand: sunxi: introduce sram_size " Richard Genoud
                   ` (3 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 error mask register is bigger than the A10/A23 one, so move
its mask into sunxi_nfc_caps.

No functional change

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index da7b8b81d39e..3de5642c05a8 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -155,7 +155,7 @@
 
 /* define bit use in NFC_ECC_ST */
 #define NFC_ECC_ERR(x)		BIT(x)
-#define NFC_ECC_ERR_MSK		GENMASK(15, 0)
+#define NFC_ECC_ERR_MSK(nfc)	(nfc->caps->ecc_err_mask)
 
 /*
  * define bit use in NFC_REG_PAT_FOUND
@@ -235,6 +235,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @random_en_mask:	RANDOM_EN mask in NFC_ECC_CTL register
  * @random_dir_mask:	RANDOM_DIRECTION mask in NFC_ECC_CTL register
  * @ecc_mode_mask:	ECC_MODE mask in NFC_ECC_CTL register
+ * @ecc_err_mask:	NFC_ECC_ERR mask in NFC_ECC_ST register
  * @pat_found_mask:	ECC_PAT_FOUND mask in NFC_REG_PAT_FOUND register
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
@@ -252,6 +253,7 @@ struct sunxi_nfc_caps {
 	unsigned int random_en_mask;
 	unsigned int random_dir_mask;
 	unsigned int ecc_mode_mask;
+	unsigned int ecc_err_mask;
 	unsigned int pat_found_mask;
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
@@ -1030,7 +1032,7 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
 		sunxi_nfc_hw_ecc_update_stats(nand, &max_bitflips, ret);
 	}
 
-	if (status & NFC_ECC_ERR_MSK) {
+	if (status & NFC_ECC_ERR_MSK(nfc)) {
 		for (i = 0; i < nchunks; i++) {
 			int data_off = i * ecc->size;
 			int oob_off = i * (ecc->bytes + 4);
@@ -2235,6 +2237,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.random_en_mask = BIT(9),
 	.random_dir_mask = BIT(10),
 	.ecc_mode_mask = GENMASK(15, 12),
+	.ecc_err_mask = GENMASK(15, 0),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
@@ -2253,6 +2256,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.random_en_mask = BIT(9),
 	.random_dir_mask = BIT(10),
 	.ecc_mode_mask = GENMASK(15, 12),
+	.ecc_err_mask = GENMASK(15, 0),
 	.pat_found_mask = GENMASK(31, 16),
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 12/15] mtd: rawnand: sunxi: introduce sram_size in sunxi_nfc_caps
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (10 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 11/15] mtd: rawnand: sunxi: introduce ecc_err_mask " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller Richard Genoud
                   ` (2 subsequent siblings)
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H6/H616 the SRAM is bigger than the A10/A23 one, so move its size
into sunxi_nfc_caps.

No functional change

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index 3de5642c05a8..e69fb0ac22b8 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -167,8 +167,6 @@
 
 #define NFC_DEFAULT_TIMEOUT_MS	1000
 
-#define NFC_SRAM_SIZE		1024
-
 #define NFC_MAX_CS		7
 
 /**
@@ -240,6 +238,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
  * @nstrengths:		Size of @ecc_strengths
+ * @sram_size:		Size of the NAND controller SRAM
  */
 struct sunxi_nfc_caps {
 	bool has_mdma;
@@ -258,6 +257,7 @@ struct sunxi_nfc_caps {
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
 	unsigned int nstrengths;
+	int sram_size;
 };
 
 /**
@@ -495,7 +495,7 @@ static void sunxi_nfc_read_buf(struct nand_chip *nand, uint8_t *buf, int len)
 	while (len > offs) {
 		bool poll = false;
 
-		cnt = min(len - offs, NFC_SRAM_SIZE);
+		cnt = min(len - offs, nfc->caps->sram_size);
 
 		ret = sunxi_nfc_wait_cmd_fifo_empty(nfc);
 		if (ret)
@@ -533,7 +533,7 @@ static void sunxi_nfc_write_buf(struct nand_chip *nand, const uint8_t *buf,
 	while (len > offs) {
 		bool poll = false;
 
-		cnt = min(len - offs, NFC_SRAM_SIZE);
+		cnt = min(len - offs, nfc->caps->sram_size);
 
 		ret = sunxi_nfc_wait_cmd_fifo_empty(nfc);
 		if (ret)
@@ -1861,7 +1861,7 @@ static int sunxi_nfc_exec_subop(struct nand_chip *nand,
 		case NAND_OP_DATA_OUT_INSTR:
 			start = nand_subop_get_data_start_off(subop, i);
 			remaining = nand_subop_get_data_len(subop, i);
-			cnt = min_t(u32, remaining, NFC_SRAM_SIZE);
+			cnt = min_t(u32, remaining, nfc->caps->sram_size);
 			cmd |= NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD;
 
 			if (instr->type == NAND_OP_DATA_OUT_INSTR) {
@@ -2242,6 +2242,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
+	.sram_size = 1024,
 };
 
 static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
@@ -2261,6 +2262,7 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
+	.sram_size = 1024,
 };
 
 static const struct of_device_id sunxi_nfc_ids[] = {

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (11 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 12/15] mtd: rawnand: sunxi: introduce sram_size " Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-22  9:18   ` Miquel Raynal
  2025-10-20 10:13 ` [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible Richard Genoud
  2025-10-20 10:13 ` [PATCH v3 15/15] arm64: dts: allwinner: h616: add NAND controller Richard Genoud
  14 siblings, 1 reply; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H616 nand controller has the same base as A10/A23, with some
differences:
- mdma is based on chained buffers
- its ECC supports up to 80bit per 1024bytes
- some registers layouts are a bit different, mainly due do the stronger
  ECC.
- it uses USER_DATA_LEN registers along USER_DATA registers.
- it needs a specific clock for ECC and MBUS.

Introduce the basic support, with ECC and scrambling, but without
DMA/MDMA.

Tested on Whatsminer H616 board (with and without scrambling, ECC)

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 drivers/mtd/nand/raw/sunxi_nand.c | 190 ++++++++++++++++++++++++++++--
 1 file changed, 182 insertions(+), 8 deletions(-)

diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
index e69fb0ac22b8..29d17d4a0913 100644
--- a/drivers/mtd/nand/raw/sunxi_nand.c
+++ b/drivers/mtd/nand/raw/sunxi_nand.c
@@ -49,17 +49,40 @@
 #define NFC_REG_A23_IO_DATA	0x0300
 #define NFC_REG_ECC_CTL		0x0034
 #define NFC_REG_ECC_ST		0x0038
-#define NFC_REG_DEBUG		0x003C
+#define NFC_REG_H6_PAT_FOUND	0x003C
 #define NFC_REG_A10_ECC_ERR_CNT	0x0040
+#define NFC_REG_H6_ECC_ERR_CNT	0x0050
 #define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
+#define NFC_REG_H6_RDATA_CTL	0x0044
+#define NFC_REG_H6_RDATA_0	0x0048
+#define NFC_REG_H6_RDATA_1	0x004C
 #define NFC_REG_A10_USER_DATA	0x0050
+#define NFC_REG_H6_USER_DATA	0x0080
 #define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
+#define NFC_REG_H6_USER_DATA_LEN 0x0070
+/* A USER_DATA_LEN register can hold the length of 8 USER_DATA registers */
+#define NFC_REG_USER_DATA_LEN_CAPACITY 8
+#define NFC_REG_USER_DATA_LEN(nfc, step) \
+	 (nfc->caps->reg_user_data_len + \
+	 ((step) / NFC_REG_USER_DATA_LEN_CAPACITY) * 4)
 #define NFC_REG_SPARE_AREA(nfc) (nfc->caps->reg_spare_area)
 #define NFC_REG_A10_SPARE_AREA	0x00A0
 #define NFC_REG_PAT_ID(nfc) (nfc->caps->reg_pat_id)
 #define NFC_REG_A10_PAT_ID	0x00A4
 #define NFC_REG_MDMA_ADDR	0x00C0
 #define NFC_REG_MDMA_CNT	0x00C4
+#define NFC_REG_H6_EFNAND_STATUS 0x0110
+#define NFC_REG_H6_SPARE_AREA	0x0114
+#define NFC_REG_H6_PAT_ID	0x0118
+#define NFC_REG_H6_DDR2_SPEC_CTL 0x011C
+#define NFC_REG_H6_NDMA_MODE_CTL 0x0120
+#define NFC_REG_H6_MDMA_DLBA_REG 0x0200
+#define NFC_REG_H6_MDMA_STA	0x0204
+#define NFC_REG_H6_MDMA_INT_MAS	0x0208
+#define NFC_REG_H6_MDMA_DESC_ADDR 0x020C
+#define NFC_REG_H6_MDMA_BUF_ADDR 0x0210
+#define NFC_REG_H6_MDMA_CNT	0x0214
+
 #define NFC_RAM0_BASE		0x0400
 #define NFC_RAM1_BASE		0x0800
 
@@ -71,6 +94,7 @@
 #define NFC_BUS_WIDTH_16	(1 << 2)
 #define NFC_RB_SEL_MSK		BIT(3)
 #define NFC_RB_SEL(x)		((x) << 3)
+/* CE_SEL BIT 27 is meant to be used for GPIO chipselect */
 #define NFC_CE_SEL_MSK		GENMASK(26, 24)
 #define NFC_CE_SEL(x)		((x) << 24)
 #define NFC_CE_CTL		BIT(6)
@@ -89,6 +113,9 @@
 #define NFC_STA			BIT(4)
 #define NFC_NATCH_INT_FLAG	BIT(5)
 #define NFC_RB_STATE(x)		BIT(x + 8)
+#define NFC_RB_STATE_MSK	GENMASK(11, 8)
+#define NDFC_RDATA_STA_1	BIT(12)
+#define NDFC_RDATA_STA_0	BIT(13)
 
 /* define bit use in NFC_INT */
 #define NFC_B2R_INT_ENABLE	BIT(0)
@@ -100,6 +127,7 @@
 
 /* define bit use in NFC_TIMING_CTL */
 #define NFC_TIMING_CTL_EDO	BIT(8)
+#define NFC_TIMING_CTL_E_EDO	BIT(9)
 
 /* define NFC_TIMING_CFG register layout */
 #define NFC_TIMING_CFG(tWB, tADL, tWHR, tRHW, tCAD)		\
@@ -107,9 +135,15 @@
 	(((tWHR) & 0x3) << 4) | (((tRHW) & 0x3) << 6) |		\
 	(((tCAD) & 0x7) << 8))
 
+#define NFC_TIMING_CFG2(tCDQSS, tSC, tCLHZ, tCSS, tWC)		\
+	((((tCDQSS) & 0x1) << 11) | (((tSC) & 0x3) << 12) |	\
+	 (((tCLHZ) & 0x3) << 14) | (((tCSS) & 0x3) << 16) |	\
+	 (((tWC) & 0x3) << 18))
+
 /* define bit use in NFC_CMD */
 #define NFC_CMD_LOW_BYTE_MSK	GENMASK(7, 0)
-#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)
+#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)  // 15-10 reserved on H6
+#define NFC_CMD_ADR_NUM_MSK	GENMASK(9, 8)
 #define NFC_CMD(x)		(x)
 #define NFC_ADR_NUM_MSK		GENMASK(18, 16)
 #define NFC_ADR_NUM(x)		(((x) - 1) << 16)
@@ -122,6 +156,7 @@
 #define NFC_SEQ			BIT(25)
 #define NFC_DATA_SWAP_METHOD	BIT(26)
 #define NFC_ROW_AUTO_INC	BIT(27)
+#define NFC_H6_SEND_RND_CMD2	BIT(27)
 #define NFC_SEND_CMD3		BIT(28)
 #define NFC_SEND_CMD4		BIT(29)
 #define NFC_CMD_TYPE_MSK	GENMASK(31, 30)
@@ -133,6 +168,7 @@
 #define NFC_READ_CMD_MSK	GENMASK(7, 0)
 #define NFC_RND_READ_CMD0_MSK	GENMASK(15, 8)
 #define NFC_RND_READ_CMD1_MSK	GENMASK(23, 16)
+#define NFC_RND_READ_CMD2_MSK	GENMASK(31, 24)
 
 /* define bit use in NFC_WCMD_SET */
 #define NFC_PROGRAM_CMD_MSK	GENMASK(7, 0)
@@ -150,6 +186,9 @@
 #define NFC_RANDOM_DIRECTION(nfc) (nfc->caps->random_dir_mask)
 #define NFC_ECC_MODE_MSK(nfc)	(nfc->caps->ecc_mode_mask)
 #define NFC_ECC_MODE(nfc, x)	field_prep(NFC_ECC_MODE_MSK(nfc), (x))
+/* RANDOM_PAGE_SIZE: 0: ECC block size  1: page size */
+#define NFC_A23_RANDOM_PAGE_SIZE	BIT(11)
+#define NFC_H6_RANDOM_PAGE_SIZE	BIT(7)
 #define NFC_RANDOM_SEED_MSK	GENMASK(30, 16)
 #define NFC_RANDOM_SEED(x)	((x) << 16)
 
@@ -165,6 +204,9 @@
 
 #define NFC_ECC_ERR_CNT(b, x)	(((x) >> (((b) % 4) * 8)) & 0xff)
 
+#define NFC_USER_DATA_LEN_MSK(step) \
+	(0xf << (((step) % NFC_REG_USER_DATA_LEN_CAPACITY) * 4))
+
 #define NFC_DEFAULT_TIMEOUT_MS	1000
 
 #define NFC_MAX_CS		7
@@ -224,9 +266,12 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @has_mdma:		Use mbus dma mode, otherwise general dma
  *			through MBUS on A23/A33 needs extra configuration.
  * @has_ecc_block_512:	If the ECC can handle 512B or only 1024B chuncks
+ * @has_ecc_clk:	If the controller needs an ECC clock.
+ * @has_mbus_clk:	If the controller needs a mbus clock.
  * @reg_io_data:	I/O data register
  * @reg_ecc_err_cnt:	ECC error counter register
  * @reg_user_data:	User data register
+ * @reg_user_data_len:	User data length register
  * @reg_spare_area:	Spare Area Register
  * @reg_pat_id:		Pattern ID Register
  * @reg_pat_found:	Data Pattern Status Register
@@ -238,14 +283,24 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
  * @dma_maxburst:	DMA maxburst
  * @ecc_strengths:	Available ECC strengths array
  * @nstrengths:		Size of @ecc_strengths
+ * @max_ecc_steps:	Maximum supported steps for ECC, this is also the
+ *			number of user data registers
+ * @user_data_len_tab:  Table of lenghts supported by USER_DATA_LEN register
+ *			The table index is the value to set in NFC_USER_DATA_LEN
+ *			registers, and the corresponding value is the number of
+ *			bytes to write
+ * @nuser_data_tab:	Size of @user_data_len_tab
  * @sram_size:		Size of the NAND controller SRAM
  */
 struct sunxi_nfc_caps {
 	bool has_mdma;
 	bool has_ecc_block_512;
+	bool has_ecc_clk;
+	bool has_mbus_clk;
 	unsigned int reg_io_data;
 	unsigned int reg_ecc_err_cnt;
 	unsigned int reg_user_data;
+	unsigned int reg_user_data_len;
 	unsigned int reg_spare_area;
 	unsigned int reg_pat_id;
 	unsigned int reg_pat_found;
@@ -257,6 +312,9 @@ struct sunxi_nfc_caps {
 	unsigned int dma_maxburst;
 	const u8 *ecc_strengths;
 	unsigned int nstrengths;
+	const u8 *user_data_len_tab;
+	unsigned int nuser_data_tab;
+	unsigned int max_ecc_steps;
 	int sram_size;
 };
 
@@ -268,6 +326,8 @@ struct sunxi_nfc_caps {
  * @regs: NAND controller registers
  * @ahb_clk: NAND controller AHB clock
  * @mod_clk: NAND controller mod clock
+ * @ecc_clk: NAND controller ECC clock
+ * @mbus_clk: NAND controller MBUS clock
  * @reset: NAND controller reset line
  * @assigned_cs: bitmask describing already assigned CS lines
  * @clk_rate: NAND controller current clock rate
@@ -283,6 +343,8 @@ struct sunxi_nfc {
 	void __iomem *regs;
 	struct clk *ahb_clk;
 	struct clk *mod_clk;
+	struct clk *ecc_clk;
+	struct clk *mbus_clk;
 	struct reset_control *reset;
 	unsigned long assigned_cs;
 	unsigned long clk_rate;
@@ -764,6 +826,53 @@ static void sunxi_nfc_hw_ecc_get_prot_oob_bytes(struct nand_chip *nand, u8 *oob,
 		sunxi_nfc_randomize_bbm(nand, page, oob);
 }
 
+/*
+ * On H6/H6 the user_data length has to be set in specific registers
+ * before writing.
+ */
+static void sunxi_nfc_reset_user_data_len(struct sunxi_nfc *nfc)
+{
+	int loop_step = NFC_REG_USER_DATA_LEN_CAPACITY;
+
+	/* not all SoCs have this register */
+	if (!nfc->caps->reg_user_data_len)
+		return;
+
+	for (int i = 0; i < nfc->caps->max_ecc_steps; i += loop_step)
+		writel(0, nfc->regs + NFC_REG_USER_DATA_LEN(nfc, i));
+}
+
+static void sunxi_nfc_set_user_data_len(struct sunxi_nfc *nfc,
+					int len, int step)
+{
+	bool found = false;
+	u32 val;
+	int i;
+
+	/* not all SoCs have this register */
+	if (!nfc->caps->reg_user_data_len)
+		return;
+
+	for (i = 0; i < nfc->caps->nuser_data_tab; i++) {
+		if (len == nfc->caps->user_data_len_tab[i]) {
+			found = true;
+			break;
+		}
+	}
+
+	if (!found) {
+		dev_warn(nfc->dev,
+			 "Unsupported length for user data reg: %d\n", len);
+		return;
+	}
+
+	val = readl(nfc->regs + NFC_REG_USER_DATA_LEN(nfc, step));
+
+	val &= ~NFC_USER_DATA_LEN_MSK(step);
+	val |= field_prep(NFC_USER_DATA_LEN_MSK(step), i);
+	writel(val, nfc->regs + NFC_REG_USER_DATA_LEN(nfc, step));
+}
+
 static void sunxi_nfc_hw_ecc_set_prot_oob_bytes(struct nand_chip *nand,
 						const u8 *oob, int step,
 						bool bbm, int page)
@@ -858,6 +967,8 @@ static int sunxi_nfc_hw_ecc_read_chunk(struct nand_chip *nand,
 	if (ret)
 		return ret;
 
+	sunxi_nfc_reset_user_data_len(nfc);
+	sunxi_nfc_set_user_data_len(nfc, 4, 0);
 	sunxi_nfc_randomizer_config(nand, page, false);
 	sunxi_nfc_randomizer_enable(nand);
 	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD | NFC_ECC_OP,
@@ -968,6 +1079,8 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
 		return ret;
 
 	sunxi_nfc_hw_ecc_enable(nand);
+	sunxi_nfc_reset_user_data_len(nfc);
+	sunxi_nfc_set_user_data_len(nfc, 4, 0);
 	sunxi_nfc_randomizer_config(nand, page, false);
 	sunxi_nfc_randomizer_enable(nand);
 
@@ -1100,6 +1213,8 @@ static int sunxi_nfc_hw_ecc_write_chunk(struct nand_chip *nand,
 
 	sunxi_nfc_randomizer_config(nand, page, false);
 	sunxi_nfc_randomizer_enable(nand);
+	sunxi_nfc_reset_user_data_len(nfc);
+	sunxi_nfc_set_user_data_len(nfc, 4, 0);
 	sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, 0, bbm, page);
 
 	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD |
@@ -1344,10 +1459,12 @@ static int sunxi_nfc_hw_ecc_write_page_dma(struct nand_chip *nand,
 	if (ret)
 		goto pio_fallback;
 
+	sunxi_nfc_reset_user_data_len(nfc);
 	for (i = 0; i < ecc->steps; i++) {
 		const u8 *oob = nand->oob_poi + (i * (ecc->bytes + 4));
 
 		sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, i, !i, page);
+		sunxi_nfc_set_user_data_len(nfc, 4, i);
 	}
 
 	nand_prog_page_begin_op(nand, page, 0, NULL, 0);
@@ -2148,6 +2265,10 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return irq;
 
+	nfc->caps = of_device_get_match_data(dev);
+	if (!nfc->caps)
+		return -EINVAL;
+
 	nfc->ahb_clk = devm_clk_get_enabled(dev, "ahb");
 	if (IS_ERR(nfc->ahb_clk)) {
 		dev_err(dev, "failed to retrieve ahb clk\n");
@@ -2160,6 +2281,22 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
 		return PTR_ERR(nfc->mod_clk);
 	}
 
+	if (nfc->caps->has_ecc_clk) {
+		nfc->ecc_clk = devm_clk_get_enabled(dev, "ecc");
+		if (IS_ERR(nfc->ecc_clk)) {
+			dev_err(dev, "failed to retrieve ecc clk\n");
+			return PTR_ERR(nfc->ecc_clk);
+		}
+	}
+
+	if (nfc->caps->has_mbus_clk) {
+		nfc->mbus_clk = devm_clk_get_enabled(dev, "mbus");
+		if (IS_ERR(nfc->mbus_clk)) {
+			dev_err(dev, "failed to retrieve mbus clk\n");
+			return PTR_ERR(nfc->mbus_clk);
+		}
+	}
+
 	nfc->reset = devm_reset_control_get_optional_exclusive(dev, "ahb");
 	if (IS_ERR(nfc->reset))
 		return PTR_ERR(nfc->reset);
@@ -2170,12 +2307,6 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	nfc->caps = of_device_get_match_data(&pdev->dev);
-	if (!nfc->caps) {
-		ret = -EINVAL;
-		goto out_ahb_reset_reassert;
-	}
-
 	ret = sunxi_nfc_rst(nfc);
 	if (ret)
 		goto out_ahb_reset_reassert;
@@ -2226,8 +2357,18 @@ static const u8 sunxi_ecc_strengths_a10[] = {
 	16, 24, 28, 32, 40, 48, 56, 60, 64
 };
 
+static const u8 sunxi_ecc_strengths_h6[] = {
+	16, 24, 28, 32, 40, 44, 48, 52, 56, 60, 64, 68, 72, 76, 80
+};
+
+static const u8 sunxi_user_data_len_h6[] = {
+	0, 4, 8, 12, 16, 20, 24, 28, 32
+};
+
 static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.has_ecc_block_512 = true,
+	.has_ecc_clk = false,
+	.has_mbus_clk = false,
 	.reg_io_data = NFC_REG_A10_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
 	.reg_user_data = NFC_REG_A10_USER_DATA,
@@ -2242,11 +2383,14 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
 	.dma_maxburst = 4,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
+	.max_ecc_steps = 16,
 	.sram_size = 1024,
 };
 
 static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.has_mdma = true,
+	.has_ecc_clk = false,
+	.has_mbus_clk = false,
 	.has_ecc_block_512 = true,
 	.reg_io_data = NFC_REG_A23_IO_DATA,
 	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
@@ -2262,9 +2406,35 @@ static const struct sunxi_nfc_caps sunxi_nfc_a23_caps = {
 	.dma_maxburst = 8,
 	.ecc_strengths = sunxi_ecc_strengths_a10,
 	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
+	.max_ecc_steps = 16,
 	.sram_size = 1024,
 };
 
+static const struct sunxi_nfc_caps sunxi_nfc_h616_caps = {
+	.has_mdma = false, // H616 supports only chained descriptors
+	.has_ecc_clk = true,
+	.has_mbus_clk = true,
+	.reg_io_data = NFC_REG_A23_IO_DATA,
+	.reg_ecc_err_cnt = NFC_REG_H6_ECC_ERR_CNT,
+	.reg_user_data = NFC_REG_H6_USER_DATA,
+	.reg_user_data_len = NFC_REG_H6_USER_DATA_LEN,
+	.reg_spare_area = NFC_REG_H6_SPARE_AREA,
+	.reg_pat_id = NFC_REG_H6_PAT_ID,
+	.reg_pat_found = NFC_REG_H6_PAT_FOUND,
+	.random_en_mask = BIT(5),
+	.random_dir_mask = BIT(6),
+	.ecc_mode_mask = GENMASK(15, 8),
+	.ecc_err_mask = GENMASK(31, 0),
+	.pat_found_mask = GENMASK(31, 0),
+	.dma_maxburst = 8,
+	.ecc_strengths = sunxi_ecc_strengths_h6,
+	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_h6),
+	.user_data_len_tab = sunxi_user_data_len_h6,
+	.nuser_data_tab = ARRAY_SIZE(sunxi_user_data_len_h6),
+	.max_ecc_steps = 32,
+	.sram_size = 8192,
+};
+
 static const struct of_device_id sunxi_nfc_ids[] = {
 	{
 		.compatible = "allwinner,sun4i-a10-nand",
@@ -2274,6 +2444,10 @@ static const struct of_device_id sunxi_nfc_ids[] = {
 		.compatible = "allwinner,sun8i-a23-nand-controller",
 		.data = &sunxi_nfc_a23_caps,
 	},
+	{
+		.compatible = "allwinner,sun50i-h616-nand-controller",
+		.data = &sunxi_nfc_h616_caps,
+	},
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, sunxi_nfc_ids);

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (12 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  2025-10-20 17:18   ` Conor Dooley
  2025-10-22  9:20   ` Miquel Raynal
  2025-10-20 10:13 ` [PATCH v3 15/15] arm64: dts: allwinner: h616: add NAND controller Richard Genoud
  14 siblings, 2 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H616 NAND controller is quite different from the A10 and A23 ones,
some registers offset changed, and some new one are introduced.
Also, the DMA handling is different (it uses chained descriptors)

So, introduce a new compatible to represent this version of the IP.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 .../mtd/allwinner,sun4i-a10-nand.yaml         | 41 +++++++++++++++++--
 1 file changed, 38 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/mtd/allwinner,sun4i-a10-nand.yaml b/Documentation/devicetree/bindings/mtd/allwinner,sun4i-a10-nand.yaml
index 054b6b8bf9b9..9d061e2216cb 100644
--- a/Documentation/devicetree/bindings/mtd/allwinner,sun4i-a10-nand.yaml
+++ b/Documentation/devicetree/bindings/mtd/allwinner,sun4i-a10-nand.yaml
@@ -6,9 +6,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: Allwinner A10 NAND Controller
 
-allOf:
-  - $ref: nand-controller.yaml
-
 maintainers:
   - Chen-Yu Tsai <wens@csie.org>
   - Maxime Ripard <mripard@kernel.org>
@@ -18,6 +15,8 @@ properties:
     enum:
       - allwinner,sun4i-a10-nand
       - allwinner,sun8i-a23-nand-controller
+      - allwinner,sun50i-h616-nand-controller
+
   reg:
     maxItems: 1
 
@@ -25,14 +24,20 @@ properties:
     maxItems: 1
 
   clocks:
+    minItems: 2
     items:
       - description: Bus Clock
       - description: Module Clock
+      - description: ECC Clock
+      - description: MBus Clock
 
   clock-names:
+    minItems: 2
     items:
       - const: ahb
       - const: mod
+      - const: ecc
+      - const: mbus
 
   resets:
     maxItems: 1
@@ -85,6 +90,36 @@ required:
 
 unevaluatedProperties: false
 
+allOf:
+  - $ref: nand-controller.yaml
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - allwinner,sun4i-a10-nand
+              - allwinner,sun8i-a23-nand-controller
+    then:
+      properties:
+        clocks:
+          maxItems: 2
+        clock-names:
+          maxItems: 2
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - allwinner,sun50i-h616-nand-controller
+    then:
+      properties:
+        clocks:
+          minItems: 4
+        clock-names:
+          minItems: 4
+
 examples:
   - |
     #include <dt-bindings/interrupt-controller/arm-gic.h>

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* [PATCH v3 15/15] arm64: dts: allwinner: h616: add NAND controller
  2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
                   ` (13 preceding siblings ...)
  2025-10-20 10:13 ` [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible Richard Genoud
@ 2025-10-20 10:13 ` Richard Genoud
  14 siblings, 0 replies; 28+ messages in thread
From: Richard Genoud @ 2025-10-20 10:13 UTC (permalink / raw)
  To: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland
  Cc: Uwe Kleine-König, Wentao Liang, Johan Hovold, Maxime Ripard,
	Thomas Petazzoni, linux-mtd, devicetree, linux-arm-kernel,
	linux-sunxi, linux-kernel, Richard Genoud

The H616 has a NAND controller quite similar to the A10/A23 ones, but
with some register differences, more clocks (for ECC and MBUS), more ECC
strengths, so this requires a new compatible string.

Add the NAND controller node and pins in the device tree.

Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
---
 .../arm64/boot/dts/allwinner/sun50i-h616.dtsi | 52 +++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/arch/arm64/boot/dts/allwinner/sun50i-h616.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-h616.dtsi
index ceedae9e399b..2a4d70298655 100644
--- a/arch/arm64/boot/dts/allwinner/sun50i-h616.dtsi
+++ b/arch/arm64/boot/dts/allwinner/sun50i-h616.dtsi
@@ -304,6 +304,42 @@ mmc2_pins: mmc2-pins {
 				bias-pull-up;
 			};
 
+			/omit-if-no-ref/
+			nand_pins: nand-pins {
+				pins = "PC0", "PC1", "PC2", "PC5", "PC8", "PC9",
+				       "PC10", "PC11", "PC12", "PC13", "PC14",
+				       "PC15", "PC16";
+				function = "nand0";
+			};
+
+			/omit-if-no-ref/
+			nand_cs0_pin: nand-cs0-pin {
+				pins = "PC4";
+				function = "nand0";
+				bias-pull-up;
+			};
+
+			/omit-if-no-ref/
+			nand_cs1_pin: nand-cs1-pin {
+				pins = "PC3";
+				function = "nand0";
+				bias-pull-up;
+			};
+
+			/omit-if-no-ref/
+			nand_rb0_pin: nand-rb0-pin {
+				pins = "PC6";
+				function = "nand0";
+				bias-pull-up;
+			};
+
+			/omit-if-no-ref/
+			nand_rb1_pin: nand-rb1-pin {
+				pins = "PC7";
+				function = "nand0";
+				bias-pull-up;
+			};
+
 			/omit-if-no-ref/
 			spi0_pins: spi0-pins {
 				pins = "PC0", "PC2", "PC4";
@@ -377,6 +413,22 @@ iommu: iommu@30f0000 {
 			#iommu-cells = <1>;
 		};
 
+		nfc: nand-controller@4011000 {
+			compatible = "allwinner,sun50i-h616-nand-controller";
+			reg = <0x04011000 0x1000>;
+			interrupts = <GIC_SPI 34 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&ccu CLK_BUS_NAND>, <&ccu CLK_NAND0>,
+				<&ccu CLK_NAND1>, <&ccu CLK_MBUS_NAND>;
+			clock-names = "ahb", "mod", "ecc", "mbus";
+			resets = <&ccu RST_BUS_NAND>;
+			reset-names = "ahb";
+			dmas = <&dma 10>;
+			dma-names = "rxtx";
+			#address-cells = <1>;
+			#size-cells = <0>;
+			status = "disabled";
+		};
+
 		mmc0: mmc@4020000 {
 			compatible = "allwinner,sun50i-h616-mmc",
 				     "allwinner,sun50i-a100-mmc";

^ permalink raw reply related	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible
  2025-10-20 10:13 ` [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible Richard Genoud
@ 2025-10-20 17:18   ` Conor Dooley
  2025-10-22  9:20   ` Miquel Raynal
  1 sibling, 0 replies; 28+ messages in thread
From: Conor Dooley @ 2025-10-20 17:18 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland, Uwe Kleine-König,
	Wentao Liang, Johan Hovold, Maxime Ripard, Thomas Petazzoni,
	linux-mtd, devicetree, linux-arm-kernel, linux-sunxi,
	linux-kernel

[-- Attachment #1: Type: text/plain, Size: 75 bytes --]

Acked-by: Conor Dooley <conor.dooley@microchip.com>
pw-bot: not-applicable

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data in sunxi_nfc_caps
  2025-10-20 10:13 ` [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data " Richard Genoud
@ 2025-10-22  8:54   ` Miquel Raynal
  2025-10-22  9:07     ` Johan Hovold
  2025-10-24  7:20     ` Richard GENOUD
  0 siblings, 2 replies; 28+ messages in thread
From: Miquel Raynal @ 2025-10-22  8:54 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi Richard,

On 20/10/2025 at 12:13:00 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> The H6/H616 USER_DATA register is not at the same offset as the
> A10/A23 one, so move its offset into sunxi_nfc_caps
>
> No functional change.
>
> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
> ---
>  drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
>  1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
> index 0285e4d0ca7f..8f5d8df19e33 100644
> --- a/drivers/mtd/nand/raw/sunxi_nand.c
> +++ b/drivers/mtd/nand/raw/sunxi_nand.c
> @@ -48,7 +48,8 @@
>  #define NFC_REG_DEBUG		0x003C
>  #define NFC_REG_A10_ECC_ERR_CNT	0x0040
>  #define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
> -#define NFC_REG_USER_DATA(x)	(0x0050 + ((x) * 4))
> +#define NFC_REG_A10_USER_DATA	0x0050
> +#define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
>  #define NFC_REG_SPARE_AREA	0x00A0
>  #define NFC_REG_PAT_ID		0x00A4
>  #define NFC_REG_MDMA_ADDR	0x00C0
> @@ -214,6 +215,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
>   *			through MBUS on A23/A33 needs extra configuration.
>   * @reg_io_data:	I/O data register
>   * @reg_ecc_err_cnt:	ECC error counter register
> + * @reg_user_data:	User data register
>   * @dma_maxburst:	DMA maxburst
>   * @ecc_strengths:	Available ECC strengths array
>   * @nstrengths:		Size of @ecc_strengths
> @@ -222,6 +224,7 @@ struct sunxi_nfc_caps {
>  	bool has_mdma;
>  	unsigned int reg_io_data;
>  	unsigned int reg_ecc_err_cnt;
> +	unsigned int reg_user_data;
>  	unsigned int dma_maxburst;
>  	const u8 *ecc_strengths;
>  	unsigned int nstrengths;
> @@ -723,8 +726,8 @@ static void sunxi_nfc_hw_ecc_get_prot_oob_bytes(struct nand_chip *nand, u8 *oob,
>  {
>  	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
>  
> -	sunxi_nfc_user_data_to_buf(readl(nfc->regs + NFC_REG_USER_DATA(step)),
> -				   oob);
> +	sunxi_nfc_user_data_to_buf(readl(nfc->regs +
> +					 NFC_REG_USER_DATA(nfc, step)),
> oob);

Minor nit, column limit is 100 now, so typically for this kind of
situation everything would fit on a single line.

Don't respin just for that if there is nothing else later, but if a v4
is needed you can change it.

Looks neat otherwise so far.

Thanks,
Miquèl

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers
  2025-10-20 10:13 ` [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers Richard Genoud
@ 2025-10-22  9:02   ` Miquel Raynal
  2025-10-22 10:05   ` Geert Uytterhoeven
  1 sibling, 0 replies; 28+ messages in thread
From: Miquel Raynal @ 2025-10-22  9:02 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi Richard,

On 20/10/2025 at 12:13:01 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> On H6/H616, the register ECC_PAT_FOUND is at its own address, and not
> part of ECC status register.
> So, introduce the pattern found register offset in sunxi_nfc_caps, along
> with its mask.
>
> Also, introduce a non compile-time field_get() because FIELD_GET() and
> u32_get_bits() don't work with non compile-time constant.
> https://lore.kernel.org/lkml/cover.1739540679.git.geert+renesas@glider.be/
>
> No functional change.
>
> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
> ---
>  drivers/mtd/nand/raw/sunxi_nand.c | 36 ++++++++++++++++++++++++++-----
>  1 file changed, 31 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
> index 8f5d8df19e33..4cfb5d3e9c06 100644
> --- a/drivers/mtd/nand/raw/sunxi_nand.c
> +++ b/drivers/mtd/nand/raw/sunxi_nand.c
> @@ -29,6 +29,9 @@
>  #include <linux/iopoll.h>
>  #include <linux/reset.h>
>  
> +/* non compile-time field get */
> +#define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) -
> 1))

I see there are already 26 users for this, let's hope it gets merged
soon and this one dropped.

Not a blocker for me.

Miquèl

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps
  2025-10-20 10:13 ` [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps Richard Genoud
@ 2025-10-22  9:05   ` Miquel Raynal
  2025-10-24  7:52     ` Richard GENOUD
  0 siblings, 1 reply; 28+ messages in thread
From: Miquel Raynal @ 2025-10-22  9:05 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi Richard,

On 20/10/2025 at 12:13:03 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> The H6/H616 ECC_MODE field is not at the same offset, and has not the
> same size.
> So move the mask into sunxi_nfc_caps.
>
> No functional change.
>
> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
> ---
>  drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
>  1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
> index 70c940840dda..4118307cac15 100644
> --- a/drivers/mtd/nand/raw/sunxi_nand.c
> +++ b/drivers/mtd/nand/raw/sunxi_nand.c
> @@ -29,8 +29,9 @@
>  #include <linux/iopoll.h>
>  #include <linux/reset.h>
>  
> -/* non compile-time field get */
> +/* non compile-time field get/prep */
>  #define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1))
> +#define field_prep(_mask, _val) (((_val) << (ffs(_mask) - 1)) &
> (_mask))

This could probably require a mention in the commit message, as for the
introduction of the field_get() helper.

No need to resend just for this either.

Thanks,
Miquèl

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data in sunxi_nfc_caps
  2025-10-22  8:54   ` Miquel Raynal
@ 2025-10-22  9:07     ` Johan Hovold
  2025-10-24  7:20     ` Richard GENOUD
  1 sibling, 0 replies; 28+ messages in thread
From: Johan Hovold @ 2025-10-22  9:07 UTC (permalink / raw)
  To: Miquel Raynal
  Cc: Richard Genoud, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland, Uwe Kleine-König,
	Wentao Liang, Maxime Ripard, Thomas Petazzoni, linux-mtd,
	devicetree, linux-arm-kernel, linux-sunxi, linux-kernel

On Wed, Oct 22, 2025 at 10:54:02AM +0200, Miquel Raynal wrote:
> On 20/10/2025 at 12:13:00 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> > @@ -723,8 +726,8 @@ static void sunxi_nfc_hw_ecc_get_prot_oob_bytes(struct nand_chip *nand, u8 *oob,
> >  {
> >  	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
> >  
> > -	sunxi_nfc_user_data_to_buf(readl(nfc->regs + NFC_REG_USER_DATA(step)),
> > -				   oob);
> > +	sunxi_nfc_user_data_to_buf(readl(nfc->regs +
> > +					 NFC_REG_USER_DATA(nfc, step)),
> > oob);
> 
> Minor nit, column limit is 100 now, so typically for this kind of
> situation everything would fit on a single line.

Just a drive-by comment:

The preferred limit is still 80 chars, unless exceeding it significantly
increases readability.

I guess here such an exception may be warranted.

Johan

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller
  2025-10-20 10:13 ` [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller Richard Genoud
@ 2025-10-22  9:18   ` Miquel Raynal
  2025-10-28  7:30     ` Richard GENOUD
  0 siblings, 1 reply; 28+ messages in thread
From: Miquel Raynal @ 2025-10-22  9:18 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hello Richard,

On 20/10/2025 at 12:13:09 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> The H616 nand controller has the same base as A10/A23, with some
> differences:
> - mdma is based on chained buffers
> - its ECC supports up to 80bit per 1024bytes
> - some registers layouts are a bit different, mainly due do the stronger
>   ECC.
> - it uses USER_DATA_LEN registers along USER_DATA registers.
> - it needs a specific clock for ECC and MBUS.
>
> Introduce the basic support, with ECC and scrambling, but without
> DMA/MDMA.
>
> Tested on Whatsminer H616 board (with and without scrambling, ECC)
>
> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>

...

>  
> +#define NFC_TIMING_CFG2(tCDQSS, tSC, tCLHZ, tCSS, tWC)		\
> +	((((tCDQSS) & 0x1) << 11) | (((tSC) & 0x3) << 12) |	\
> +	 (((tCLHZ) & 0x3) << 14) | (((tCSS) & 0x3) << 16) |	\
> +	 (((tWC) & 0x3) << 18))
> +
>  /* define bit use in NFC_CMD */
>  #define NFC_CMD_LOW_BYTE_MSK	GENMASK(7, 0)
> -#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)
> +#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)  // 15-10 reserved on H6

Wrong comment type :-)

> +#define NFC_CMD_ADR_NUM_MSK	GENMASK(9, 8)
>  #define NFC_CMD(x)		(x)
>  #define NFC_ADR_NUM_MSK		GENMASK(18, 16)
>  #define NFC_ADR_NUM(x)		(((x) - 1) << 16)
> @@ -122,6 +156,7 @@
>  #define NFC_SEQ			BIT(25)
>  #define NFC_DATA_SWAP_METHOD	BIT(26)
>  #define NFC_ROW_AUTO_INC	BIT(27)
> +#define NFC_H6_SEND_RND_CMD2	BIT(27)
>  #define NFC_SEND_CMD3		BIT(28)
>  #define NFC_SEND_CMD4		BIT(29)
>  #define NFC_CMD_TYPE_MSK	GENMASK(31, 30)
> @@ -133,6 +168,7 @@
>  #define NFC_READ_CMD_MSK	GENMASK(7, 0)
>  #define NFC_RND_READ_CMD0_MSK	GENMASK(15, 8)
>  #define NFC_RND_READ_CMD1_MSK	GENMASK(23, 16)
> +#define NFC_RND_READ_CMD2_MSK	GENMASK(31, 24)

...

> @@ -858,6 +967,8 @@ static int sunxi_nfc_hw_ecc_read_chunk(struct nand_chip *nand,
>  	if (ret)
>  		return ret;
>  
> +	sunxi_nfc_reset_user_data_len(nfc);
> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);

I'm not sure I understand this properly. Why isn't this a fixed setting?
Also, what is 4? It is not obvious to me and my require either a comment
or a define (or maybe a sizeof()).

>  	sunxi_nfc_randomizer_config(nand, page, false);
>  	sunxi_nfc_randomizer_enable(nand);
>  	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD | NFC_ECC_OP,
> @@ -968,6 +1079,8 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
>  		return ret;
>  
>  	sunxi_nfc_hw_ecc_enable(nand);
> +	sunxi_nfc_reset_user_data_len(nfc);
> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);
>  	sunxi_nfc_randomizer_config(nand, page, false);
>  	sunxi_nfc_randomizer_enable(nand);
>  
> @@ -1100,6 +1213,8 @@ static int sunxi_nfc_hw_ecc_write_chunk(struct nand_chip *nand,
>  
>  	sunxi_nfc_randomizer_config(nand, page, false);
>  	sunxi_nfc_randomizer_enable(nand);
> +	sunxi_nfc_reset_user_data_len(nfc);
> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);
>  	sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, 0, bbm, page);
>  
>  	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD |
> @@ -1344,10 +1459,12 @@ static int sunxi_nfc_hw_ecc_write_page_dma(struct nand_chip *nand,
>  	if (ret)
>  		goto pio_fallback;
>  
> +	sunxi_nfc_reset_user_data_len(nfc);
>  	for (i = 0; i < ecc->steps; i++) {
>  		const u8 *oob = nand->oob_poi + (i * (ecc->bytes + 4));
>  
>  		sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, i, !i, page);
> +		sunxi_nfc_set_user_data_len(nfc, 4, i);

Here you use it differently, maybe a bit of explanation in a comment
could help.

>  	}
>  
>  	nand_prog_page_begin_op(nand, page, 0, NULL, 0);
> @@ -2148,6 +2265,10 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
>  	if (irq < 0)
>  		return irq;
>  
> +	nfc->caps = of_device_get_match_data(dev);
> +	if (!nfc->caps)
> +		return -EINVAL;
> +
>  	nfc->ahb_clk = devm_clk_get_enabled(dev, "ahb");
>  	if (IS_ERR(nfc->ahb_clk)) {
>  		dev_err(dev, "failed to retrieve ahb clk\n");
> @@ -2160,6 +2281,22 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
>  		return PTR_ERR(nfc->mod_clk);
>  	}
>  

...

>  static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
>  	.has_ecc_block_512 = true,
> +	.has_ecc_clk = false,
> +	.has_mbus_clk = false,

As you want, but setting these fields (and below) to false is not
strictly required as they will be set to 0 (which means false,
automatically).

>  	.reg_io_data = NFC_REG_A10_IO_DATA,
>  	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
>  	.reg_user_data = NFC_REG_A10_USER_DATA,
> @@ -2242,11 +2383,14 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
>  	.dma_maxburst = 4,
>  	.ecc_strengths = sunxi_ecc_strengths_a10,
>  	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
> +	.max_ecc_steps = 16,
>  	.sram_size = 1024,
>  };
>  

> +static const struct sunxi_nfc_caps sunxi_nfc_h616_caps = {
> +	.has_mdma = false, // H616 supports only chained descriptors

Wrong comment type :-)

LGTM otherwise.

Thanks,
Miquèl

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible
  2025-10-20 10:13 ` [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible Richard Genoud
  2025-10-20 17:18   ` Conor Dooley
@ 2025-10-22  9:20   ` Miquel Raynal
  2025-10-24  9:08     ` Richard GENOUD
  1 sibling, 1 reply; 28+ messages in thread
From: Miquel Raynal @ 2025-10-22  9:20 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi Richard,

On 20/10/2025 at 12:13:10 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:

> The H616 NAND controller is quite different from the A10 and A23 ones,
> some registers offset changed, and some new one are introduced.
> Also, the DMA handling is different (it uses chained descriptors)
>
> So, introduce a new compatible to represent this version of the IP.
>
> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
> ---
>  .../mtd/allwinner,sun4i-a10-nand.yaml         | 41 +++++++++++++++++--
>  1 file changed, 38 insertions(+), 3 deletions(-)

Please move the binding patch first in your series (binding then driver
changes aligned with the introduced binding and then DT changes using
it).

Thanks,
Miquèl

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers
  2025-10-20 10:13 ` [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers Richard Genoud
  2025-10-22  9:02   ` Miquel Raynal
@ 2025-10-22 10:05   ` Geert Uytterhoeven
  1 sibling, 0 replies; 28+ messages in thread
From: Geert Uytterhoeven @ 2025-10-22 10:05 UTC (permalink / raw)
  To: Richard Genoud
  Cc: Miquel Raynal, Richard Weinberger, Vignesh Raghavendra,
	Rob Herring, Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai,
	Jernej Skrabec, Samuel Holland, Uwe Kleine-König,
	Wentao Liang, Johan Hovold, Maxime Ripard, Thomas Petazzoni,
	linux-mtd, devicetree, linux-arm-kernel, linux-sunxi,
	linux-kernel

On Tue, 21 Oct 2025 at 22:51, Richard Genoud <richard.genoud@bootlin.com> wrote:
> On H6/H616, the register ECC_PAT_FOUND is at its own address, and not
> part of ECC status register.
> So, introduce the pattern found register offset in sunxi_nfc_caps, along
> with its mask.
>
> Also, introduce a non compile-time field_get() because FIELD_GET() and
> u32_get_bits() don't work with non compile-time constant.
> https://lore.kernel.org/lkml/cover.1739540679.git.geert+renesas@glider.be/

FTR, v4 at
https://lore.kernel.org/all/cover.1760696560.git.geert+renesas@glider.be

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data in sunxi_nfc_caps
  2025-10-22  8:54   ` Miquel Raynal
  2025-10-22  9:07     ` Johan Hovold
@ 2025-10-24  7:20     ` Richard GENOUD
  1 sibling, 0 replies; 28+ messages in thread
From: Richard GENOUD @ 2025-10-24  7:20 UTC (permalink / raw)
  To: Miquel Raynal
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi,
Le 22/10/2025 à 10:54, Miquel Raynal a écrit :
> Hi Richard,
> 
> On 20/10/2025 at 12:13:00 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:
> 
>> The H6/H616 USER_DATA register is not at the same offset as the
>> A10/A23 one, so move its offset into sunxi_nfc_caps
>>
>> No functional change.
>>
>> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
>> ---
>>   drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
>>   1 file changed, 9 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
>> index 0285e4d0ca7f..8f5d8df19e33 100644
>> --- a/drivers/mtd/nand/raw/sunxi_nand.c
>> +++ b/drivers/mtd/nand/raw/sunxi_nand.c
>> @@ -48,7 +48,8 @@
>>   #define NFC_REG_DEBUG		0x003C
>>   #define NFC_REG_A10_ECC_ERR_CNT	0x0040
>>   #define NFC_REG_ECC_ERR_CNT(nfc, x)	((nfc->caps->reg_ecc_err_cnt + (x)) & ~0x3)
>> -#define NFC_REG_USER_DATA(x)	(0x0050 + ((x) * 4))
>> +#define NFC_REG_A10_USER_DATA	0x0050
>> +#define NFC_REG_USER_DATA(nfc, x)	(nfc->caps->reg_user_data + ((x) * 4))
>>   #define NFC_REG_SPARE_AREA	0x00A0
>>   #define NFC_REG_PAT_ID		0x00A4
>>   #define NFC_REG_MDMA_ADDR	0x00C0
>> @@ -214,6 +215,7 @@ static inline struct sunxi_nand_chip *to_sunxi_nand(struct nand_chip *nand)
>>    *			through MBUS on A23/A33 needs extra configuration.
>>    * @reg_io_data:	I/O data register
>>    * @reg_ecc_err_cnt:	ECC error counter register
>> + * @reg_user_data:	User data register
>>    * @dma_maxburst:	DMA maxburst
>>    * @ecc_strengths:	Available ECC strengths array
>>    * @nstrengths:		Size of @ecc_strengths
>> @@ -222,6 +224,7 @@ struct sunxi_nfc_caps {
>>   	bool has_mdma;
>>   	unsigned int reg_io_data;
>>   	unsigned int reg_ecc_err_cnt;
>> +	unsigned int reg_user_data;
>>   	unsigned int dma_maxburst;
>>   	const u8 *ecc_strengths;
>>   	unsigned int nstrengths;
>> @@ -723,8 +726,8 @@ static void sunxi_nfc_hw_ecc_get_prot_oob_bytes(struct nand_chip *nand, u8 *oob,
>>   {
>>   	struct sunxi_nfc *nfc = to_sunxi_nfc(nand->controller);
>>   
>> -	sunxi_nfc_user_data_to_buf(readl(nfc->regs + NFC_REG_USER_DATA(step)),
>> -				   oob);
>> +	sunxi_nfc_user_data_to_buf(readl(nfc->regs +
>> +					 NFC_REG_USER_DATA(nfc, step)),
>> oob);
> 
> Minor nit, column limit is 100 now, so typically for this kind of
> situation everything would fit on a single line.
Indeed, the 80 column limit has been loosened (but braille displays are 
still 80 cells max AFAIK).
Anyway, you're right, the 80-rule could be bent here for readability.

> 
> Don't respin just for that if there is nothing else later, but if a v4
> is needed you can change it.
> 
> Looks neat otherwise so far.
> 
> Thanks,
> Miquèl

Thanks!


-- 
Richard Genoud, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps
  2025-10-22  9:05   ` Miquel Raynal
@ 2025-10-24  7:52     ` Richard GENOUD
  0 siblings, 0 replies; 28+ messages in thread
From: Richard GENOUD @ 2025-10-24  7:52 UTC (permalink / raw)
  To: Miquel Raynal
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Hi,
Le 22/10/2025 à 11:05, Miquel Raynal a écrit :
> Hi Richard,
> 
> On 20/10/2025 at 12:13:03 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:
> 
>> The H6/H616 ECC_MODE field is not at the same offset, and has not the
>> same size.
>> So move the mask into sunxi_nfc_caps.
>>
>> No functional change.
>>
>> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
>> ---
>>   drivers/mtd/nand/raw/sunxi_nand.c | 13 +++++++++----
>>   1 file changed, 9 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/mtd/nand/raw/sunxi_nand.c b/drivers/mtd/nand/raw/sunxi_nand.c
>> index 70c940840dda..4118307cac15 100644
>> --- a/drivers/mtd/nand/raw/sunxi_nand.c
>> +++ b/drivers/mtd/nand/raw/sunxi_nand.c
>> @@ -29,8 +29,9 @@
>>   #include <linux/iopoll.h>
>>   #include <linux/reset.h>
>>   
>> -/* non compile-time field get */
>> +/* non compile-time field get/prep */
>>   #define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1))
>> +#define field_prep(_mask, _val) (((_val) << (ffs(_mask) - 1)) &
>> (_mask))
> 
> This could probably require a mention in the commit message, as for the
> introduction of the field_get() helper.
> 
> No need to resend just for this either.
Ok, will do.

> 
> Thanks,
> Miquèl
> 
Thanks!

-- 
Richard Genoud, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible
  2025-10-22  9:20   ` Miquel Raynal
@ 2025-10-24  9:08     ` Richard GENOUD
  0 siblings, 0 replies; 28+ messages in thread
From: Richard GENOUD @ 2025-10-24  9:08 UTC (permalink / raw)
  To: Miquel Raynal
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Le 22/10/2025 à 11:20, Miquel Raynal a écrit :
> Hi Richard,
> 
> On 20/10/2025 at 12:13:10 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:
> 
>> The H616 NAND controller is quite different from the A10 and A23 ones,
>> some registers offset changed, and some new one are introduced.
>> Also, the DMA handling is different (it uses chained descriptors)
>>
>> So, introduce a new compatible to represent this version of the IP.
>>
>> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
>> ---
>>   .../mtd/allwinner,sun4i-a10-nand.yaml         | 41 +++++++++++++++++--
>>   1 file changed, 38 insertions(+), 3 deletions(-)
> 
> Please move the binding patch first in your series (binding then driver
> changes aligned with the introduced binding and then DT changes using
> it).
Yes, sorry, it took time but now I got it :)

> 
> Thanks,
> Miquèl
> 

Thanks!

-- 
Richard Genoud, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

^ permalink raw reply	[flat|nested] 28+ messages in thread

* Re: [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller
  2025-10-22  9:18   ` Miquel Raynal
@ 2025-10-28  7:30     ` Richard GENOUD
  0 siblings, 0 replies; 28+ messages in thread
From: Richard GENOUD @ 2025-10-28  7:30 UTC (permalink / raw)
  To: Miquel Raynal
  Cc: Richard Weinberger, Vignesh Raghavendra, Rob Herring,
	Krzysztof Kozlowski, Conor Dooley, Chen-Yu Tsai, Jernej Skrabec,
	Samuel Holland, Uwe Kleine-König, Wentao Liang, Johan Hovold,
	Maxime Ripard, Thomas Petazzoni, linux-mtd, devicetree,
	linux-arm-kernel, linux-sunxi, linux-kernel

Le 22/10/2025 à 11:18, Miquel Raynal a écrit :
> Hello Richard,
> 
> On 20/10/2025 at 12:13:09 +02, Richard Genoud <richard.genoud@bootlin.com> wrote:
> 
>> The H616 nand controller has the same base as A10/A23, with some
>> differences:
>> - mdma is based on chained buffers
>> - its ECC supports up to 80bit per 1024bytes
>> - some registers layouts are a bit different, mainly due do the stronger
>>    ECC.
>> - it uses USER_DATA_LEN registers along USER_DATA registers.
>> - it needs a specific clock for ECC and MBUS.
>>
>> Introduce the basic support, with ECC and scrambling, but without
>> DMA/MDMA.
>>
>> Tested on Whatsminer H616 board (with and without scrambling, ECC)
>>
>> Signed-off-by: Richard Genoud <richard.genoud@bootlin.com>
> 
> ...
> 
>>   
>> +#define NFC_TIMING_CFG2(tCDQSS, tSC, tCLHZ, tCSS, tWC)		\
>> +	((((tCDQSS) & 0x1) << 11) | (((tSC) & 0x3) << 12) |	\
>> +	 (((tCLHZ) & 0x3) << 14) | (((tCSS) & 0x3) << 16) |	\
>> +	 (((tWC) & 0x3) << 18))
>> +
>>   /* define bit use in NFC_CMD */
>>   #define NFC_CMD_LOW_BYTE_MSK	GENMASK(7, 0)
>> -#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)
>> +#define NFC_CMD_HIGH_BYTE_MSK	GENMASK(15, 8)  // 15-10 reserved on H6
> 
> Wrong comment type :-)
ok

> 
>> +#define NFC_CMD_ADR_NUM_MSK	GENMASK(9, 8)
>>   #define NFC_CMD(x)		(x)
>>   #define NFC_ADR_NUM_MSK		GENMASK(18, 16)
>>   #define NFC_ADR_NUM(x)		(((x) - 1) << 16)
>> @@ -122,6 +156,7 @@
>>   #define NFC_SEQ			BIT(25)
>>   #define NFC_DATA_SWAP_METHOD	BIT(26)
>>   #define NFC_ROW_AUTO_INC	BIT(27)
>> +#define NFC_H6_SEND_RND_CMD2	BIT(27)
>>   #define NFC_SEND_CMD3		BIT(28)
>>   #define NFC_SEND_CMD4		BIT(29)
>>   #define NFC_CMD_TYPE_MSK	GENMASK(31, 30)
>> @@ -133,6 +168,7 @@
>>   #define NFC_READ_CMD_MSK	GENMASK(7, 0)
>>   #define NFC_RND_READ_CMD0_MSK	GENMASK(15, 8)
>>   #define NFC_RND_READ_CMD1_MSK	GENMASK(23, 16)
>> +#define NFC_RND_READ_CMD2_MSK	GENMASK(31, 24)
> 
> ...
> 
>> @@ -858,6 +967,8 @@ static int sunxi_nfc_hw_ecc_read_chunk(struct nand_chip *nand,
>>   	if (ret)
>>   		return ret;
>>   
>> +	sunxi_nfc_reset_user_data_len(nfc);
>> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);
> 
> I'm not sure I understand this properly. Why isn't this a fixed setting?
> Also, what is 4? It is not obvious to me and my require either a comment
> or a define (or maybe a sizeof()).

Yes, indeed. There are '4' all over the place, maybe a bit of tidy up is 
needed in a separate patch.
(and some explanations)

I'll resend a v4 with that.

> 
>>   	sunxi_nfc_randomizer_config(nand, page, false);
>>   	sunxi_nfc_randomizer_enable(nand);
>>   	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD | NFC_ECC_OP,
>> @@ -968,6 +1079,8 @@ static int sunxi_nfc_hw_ecc_read_chunks_dma(struct nand_chip *nand, uint8_t *buf
>>   		return ret;
>>   
>>   	sunxi_nfc_hw_ecc_enable(nand);
>> +	sunxi_nfc_reset_user_data_len(nfc);
>> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);
>>   	sunxi_nfc_randomizer_config(nand, page, false);
>>   	sunxi_nfc_randomizer_enable(nand);
>>   
>> @@ -1100,6 +1213,8 @@ static int sunxi_nfc_hw_ecc_write_chunk(struct nand_chip *nand,
>>   
>>   	sunxi_nfc_randomizer_config(nand, page, false);
>>   	sunxi_nfc_randomizer_enable(nand);
>> +	sunxi_nfc_reset_user_data_len(nfc);
>> +	sunxi_nfc_set_user_data_len(nfc, 4, 0);
>>   	sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, 0, bbm, page);
>>   
>>   	writel(NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD |
>> @@ -1344,10 +1459,12 @@ static int sunxi_nfc_hw_ecc_write_page_dma(struct nand_chip *nand,
>>   	if (ret)
>>   		goto pio_fallback;
>>   
>> +	sunxi_nfc_reset_user_data_len(nfc);
>>   	for (i = 0; i < ecc->steps; i++) {
>>   		const u8 *oob = nand->oob_poi + (i * (ecc->bytes + 4));
>>   
>>   		sunxi_nfc_hw_ecc_set_prot_oob_bytes(nand, oob, i, !i, page);
>> +		sunxi_nfc_set_user_data_len(nfc, 4, i);
> 
> Here you use it differently, maybe a bit of explanation in a comment
> could help.
> 
yes, definitely

>>   	}
>>   
>>   	nand_prog_page_begin_op(nand, page, 0, NULL, 0);
>> @@ -2148,6 +2265,10 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
>>   	if (irq < 0)
>>   		return irq;
>>   
>> +	nfc->caps = of_device_get_match_data(dev);
>> +	if (!nfc->caps)
>> +		return -EINVAL;
>> +
>>   	nfc->ahb_clk = devm_clk_get_enabled(dev, "ahb");
>>   	if (IS_ERR(nfc->ahb_clk)) {
>>   		dev_err(dev, "failed to retrieve ahb clk\n");
>> @@ -2160,6 +2281,22 @@ static int sunxi_nfc_probe(struct platform_device *pdev)
>>   		return PTR_ERR(nfc->mod_clk);
>>   	}
>>   
> 
> ...
> 
>>   static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
>>   	.has_ecc_block_512 = true,
>> +	.has_ecc_clk = false,
>> +	.has_mbus_clk = false,
> 
> As you want, but setting these fields (and below) to false is not
> strictly required as they will be set to 0 (which means false,
> automatically).
> 
yes, I'll remove them.

>>   	.reg_io_data = NFC_REG_A10_IO_DATA,
>>   	.reg_ecc_err_cnt = NFC_REG_A10_ECC_ERR_CNT,
>>   	.reg_user_data = NFC_REG_A10_USER_DATA,
>> @@ -2242,11 +2383,14 @@ static const struct sunxi_nfc_caps sunxi_nfc_a10_caps = {
>>   	.dma_maxburst = 4,
>>   	.ecc_strengths = sunxi_ecc_strengths_a10,
>>   	.nstrengths = ARRAY_SIZE(sunxi_ecc_strengths_a10),
>> +	.max_ecc_steps = 16,
>>   	.sram_size = 1024,
>>   };
>>   
> 
>> +static const struct sunxi_nfc_caps sunxi_nfc_h616_caps = {
>> +	.has_mdma = false, // H616 supports only chained descriptors
> 
> Wrong comment type :-)
I'll drop this line also.

> 
> LGTM otherwise.
> 
> Thanks,
> Miquèl
Thanks!

-- 
Richard Genoud, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

^ permalink raw reply	[flat|nested] 28+ messages in thread

end of thread, other threads:[~2025-10-28  7:30 UTC | newest]

Thread overview: 28+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-10-20 10:12 [PATCH v3 00/15] Introduce Allwinner H6/H616 NAND controller support Richard Genoud
2025-10-20 10:12 ` [PATCH v3 01/15] mtd: rawnand: sunxi: Remove superfluous register readings Richard Genoud
2025-10-20 10:12 ` [PATCH v3 02/15] mtd: rawnand: sunxi: move ECC strenghts in sunxi_nfc_caps Richard Genoud
2025-10-20 10:12 ` [PATCH v3 03/15] mtd: rawnand: sunxi: introduce reg_ecc_err_cnt " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 04/15] mtd: rawnand: sunxi: introduce reg_user_data " Richard Genoud
2025-10-22  8:54   ` Miquel Raynal
2025-10-22  9:07     ` Johan Hovold
2025-10-24  7:20     ` Richard GENOUD
2025-10-20 10:13 ` [PATCH v3 05/15] mtd: rawnand: sunxi: rework pattern found registers Richard Genoud
2025-10-22  9:02   ` Miquel Raynal
2025-10-22 10:05   ` Geert Uytterhoeven
2025-10-20 10:13 ` [PATCH v3 06/15] mtd: rawnand: sunxi: add has_ecc_block_512 capability Richard Genoud
2025-10-20 10:13 ` [PATCH v3 07/15] mtd: rawnand: sunxi: introduce ecc_mode_mask in sunxi_nfc_caps Richard Genoud
2025-10-22  9:05   ` Miquel Raynal
2025-10-24  7:52     ` Richard GENOUD
2025-10-20 10:13 ` [PATCH v3 08/15] mtd: rawnand: sunxi: introduce random en/dir " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 09/15] mtd: rawnand: sunxi: introduce reg_pat_id " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 10/15] mtd: rawnand: sunxi: introduce reg_spare_area " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 11/15] mtd: rawnand: sunxi: introduce ecc_err_mask " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 12/15] mtd: rawnand: sunxi: introduce sram_size " Richard Genoud
2025-10-20 10:13 ` [PATCH v3 13/15] mtd: rawnand: sunxi: Add support for H616 nand controller Richard Genoud
2025-10-22  9:18   ` Miquel Raynal
2025-10-28  7:30     ` Richard GENOUD
2025-10-20 10:13 ` [PATCH v3 14/15] dt-bindings: mtd: sunxi: Add H616 compatible Richard Genoud
2025-10-20 17:18   ` Conor Dooley
2025-10-22  9:20   ` Miquel Raynal
2025-10-24  9:08     ` Richard GENOUD
2025-10-20 10:13 ` [PATCH v3 15/15] arm64: dts: allwinner: h616: add NAND controller Richard Genoud

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).