[RESEND PATCH v5 10/16] mtd: nand: qcom: add command elements in BAM transaction

[RESEND PATCH v5 10/16] mtd: nand: qcom: add command elements in BAM transaction

[Abhishek Sahu]

All the QPIC register read/write through BAM DMA requires
command descriptor which contains the array of command elements.

Reviewed-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Abhishek Sahu <absahu@codeaurora.org>
---

* Changes from v4: None

The BAM DMA patches [1] got merged in 4.14-rc1 so now all the build
dependencies are solved and this patch can be merged.
This patch can be cleanly applied over v4.14-rc1.

[1] http://www.spinics.net/lists/dmaengine/msg13665.html

 drivers/mtd/nand/qcom_nandc.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/qcom_nandc.c b/drivers/mtd/nand/qcom_nandc.c
index 7977a70..b0a4734 100644
--- a/drivers/mtd/nand/qcom_nandc.c
+++ b/drivers/mtd/nand/qcom_nandc.c
@@ -22,6 +22,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/delay.h>
+#include <linux/dma/qcom_bam_dma.h>
 
 /* NANDc reg offsets */
 #define	NAND_FLASH_CMD			0x00
@@ -199,6 +200,7 @@
  */
 #define dev_cmd_reg_addr(nandc, reg) ((nandc)->props->dev_cmd_reg_start + (reg))
 
+#define QPIC_PER_CW_CMD_ELEMENTS	32
 #define QPIC_PER_CW_CMD_SGL		32
 #define QPIC_PER_CW_DATA_SGL		8
 
@@ -221,8 +223,13 @@
 /*
  * This data type corresponds to the BAM transaction which will be used for all
  * NAND transfers.
+ * @bam_ce - the array of BAM command elements
  * @cmd_sgl - sgl for NAND BAM command pipe
  * @data_sgl - sgl for NAND BAM consumer/producer pipe
+ * @bam_ce_pos - the index in bam_ce which is available for next sgl
+ * @bam_ce_start - the index in bam_ce which marks the start position ce
+ *		   for current sgl. It will be used for size calculation
+ *		   for current sgl
  * @cmd_sgl_pos - current index in command sgl.
  * @cmd_sgl_start - start index in command sgl.
  * @tx_sgl_pos - current index in data sgl for tx.
@@ -231,8 +238,11 @@
  * @rx_sgl_start - start index in data sgl for rx.
  */
 struct bam_transaction {
+	struct bam_cmd_element *bam_ce;
 	struct scatterlist *cmd_sgl;
 	struct scatterlist *data_sgl;
+	u32 bam_ce_pos;
+	u32 bam_ce_start;
 	u32 cmd_sgl_pos;
 	u32 cmd_sgl_start;
 	u32 tx_sgl_pos;
@@ -462,7 +472,8 @@ static void free_bam_transaction(struct qcom_nand_controller *nandc)
 
 	bam_txn_size =
 		sizeof(*bam_txn) + num_cw *
-		((sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL) +
+		((sizeof(*bam_txn->bam_ce) * QPIC_PER_CW_CMD_ELEMENTS) +
+		(sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL) +
 		(sizeof(*bam_txn->data_sgl) * QPIC_PER_CW_DATA_SGL));
 
 	bam_txn_buf = devm_kzalloc(nandc->dev, bam_txn_size, GFP_KERNEL);
@@ -472,6 +483,10 @@ static void free_bam_transaction(struct qcom_nand_controller *nandc)
 	bam_txn = bam_txn_buf;
 	bam_txn_buf += sizeof(*bam_txn);
 
+	bam_txn->bam_ce = bam_txn_buf;
+	bam_txn_buf +=
+		sizeof(*bam_txn->bam_ce) * QPIC_PER_CW_CMD_ELEMENTS * num_cw;
+
 	bam_txn->cmd_sgl = bam_txn_buf;
 	bam_txn_buf +=
 		sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL * num_cw;
@@ -489,6 +504,8 @@ static void clear_bam_transaction(struct qcom_nand_controller *nandc)
 	if (!nandc->props->is_bam)
 		return;
 
+	bam_txn->bam_ce_pos = 0;
+	bam_txn->bam_ce_start = 0;
 	bam_txn->cmd_sgl_pos = 0;
 	bam_txn->cmd_sgl_start = 0;
 	bam_txn->tx_sgl_pos = 0;
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

[RESEND PATCH v5 11/16] mtd: nand: qcom: support for command descriptor formation

[Abhishek Sahu]

1. Add the function for command descriptor preparation which will
   be used only by BAM DMA and it will form the DMA descriptors
   containing command elements
2. DMA_PREP_CMD flag should be used for forming command DMA
   descriptors

Reviewed-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Abhishek Sahu <absahu@codeaurora.org>
---

* Changes from v4: None

The BAM DMA patches [1] got merged in 4.14-rc1 so now all the build
dependencies are solved and this patch can be merged.
This patch can be cleanly applied over v4.14-rc1.

[1] http://www.spinics.net/lists/dmaengine/msg13665.html

 drivers/mtd/nand/qcom_nandc.c | 108 +++++++++++++++++++++++++++++++++++-------
 1 file changed, 92 insertions(+), 16 deletions(-)

diff --git a/drivers/mtd/nand/qcom_nandc.c b/drivers/mtd/nand/qcom_nandc.c
index b0a4734..52d9fae 100644
--- a/drivers/mtd/nand/qcom_nandc.c
+++ b/drivers/mtd/nand/qcom_nandc.c
@@ -200,6 +200,14 @@
  */
 #define dev_cmd_reg_addr(nandc, reg) ((nandc)->props->dev_cmd_reg_start + (reg))
 
+/* Returns the NAND register physical address */
+#define nandc_reg_phys(chip, offset) ((chip)->base_phys + (offset))
+
+/* Returns the dma address for reg read buffer */
+#define reg_buf_dma_addr(chip, vaddr) \
+	((chip)->reg_read_dma + \
+	((uint8_t *)(vaddr) - (uint8_t *)(chip)->reg_read_buf))
+
 #define QPIC_PER_CW_CMD_ELEMENTS	32
 #define QPIC_PER_CW_CMD_SGL		32
 #define QPIC_PER_CW_DATA_SGL		8
@@ -317,7 +325,8 @@ struct nandc_regs {
  *				controller
  * @dev:			parent device
  * @base:			MMIO base
- * @base_dma:			physical base address of controller registers
+ * @base_phys:			physical base address of controller registers
+ * @base_dma:			dma base address of controller registers
  * @core_clk:			controller clock
  * @aon_clk:			another controller clock
  *
@@ -350,6 +359,7 @@ struct qcom_nand_controller {
 	struct device *dev;
 
 	void __iomem *base;
+	phys_addr_t base_phys;
 	dma_addr_t base_dma;
 
 	struct clk *core_clk;
@@ -751,6 +761,66 @@ static int prepare_bam_async_desc(struct qcom_nand_controller *nandc,
 }
 
 /*
+ * Prepares the command descriptor for BAM DMA which will be used for NAND
+ * register reads and writes. The command descriptor requires the command
+ * to be formed in command element type so this function uses the command
+ * element from bam transaction ce array and fills the same with required
+ * data. A single SGL can contain multiple command elements so
+ * NAND_BAM_NEXT_SGL will be used for starting the separate SGL
+ * after the current command element.
+ */
+static int prep_bam_dma_desc_cmd(struct qcom_nand_controller *nandc, bool read,
+				 int reg_off, const void *vaddr,
+				 int size, unsigned int flags)
+{
+	int bam_ce_size;
+	int i, ret;
+	struct bam_cmd_element *bam_ce_buffer;
+	struct bam_transaction *bam_txn = nandc->bam_txn;
+
+	bam_ce_buffer = &bam_txn->bam_ce[bam_txn->bam_ce_pos];
+
+	/* fill the command desc */
+	for (i = 0; i < size; i++) {
+		if (read)
+			bam_prep_ce(&bam_ce_buffer[i],
+				    nandc_reg_phys(nandc, reg_off + 4 * i),
+				    BAM_READ_COMMAND,
+				    reg_buf_dma_addr(nandc,
+						     (__le32 *)vaddr + i));
+		else
+			bam_prep_ce_le32(&bam_ce_buffer[i],
+					 nandc_reg_phys(nandc, reg_off + 4 * i),
+					 BAM_WRITE_COMMAND,
+					 *((__le32 *)vaddr + i));
+	}
+
+	bam_txn->bam_ce_pos += size;
+
+	/* use the separate sgl after this command */
+	if (flags & NAND_BAM_NEXT_SGL) {
+		bam_ce_buffer = &bam_txn->bam_ce[bam_txn->bam_ce_start];
+		bam_ce_size = (bam_txn->bam_ce_pos -
+				bam_txn->bam_ce_start) *
+				sizeof(struct bam_cmd_element);
+		sg_set_buf(&bam_txn->cmd_sgl[bam_txn->cmd_sgl_pos],
+			   bam_ce_buffer, bam_ce_size);
+		bam_txn->cmd_sgl_pos++;
+		bam_txn->bam_ce_start = bam_txn->bam_ce_pos;
+
+		if (flags & NAND_BAM_NWD) {
+			ret = prepare_bam_async_desc(nandc, nandc->cmd_chan,
+						     DMA_PREP_FENCE |
+						     DMA_PREP_CMD);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+/*
  * Prepares the data descriptor for BAM DMA which will be used for NAND
  * data reads and writes.
  */
@@ -868,19 +938,22 @@ static int read_reg_dma(struct qcom_nand_controller *nandc, int first,
 {
 	bool flow_control = false;
 	void *vaddr;
-	int size;
 
-	if (first == NAND_READ_ID || first == NAND_FLASH_STATUS)
-		flow_control = true;
+	vaddr = nandc->reg_read_buf + nandc->reg_read_pos;
+	nandc->reg_read_pos += num_regs;
 
 	if (first == NAND_DEV_CMD_VLD || first == NAND_DEV_CMD1)
 		first = dev_cmd_reg_addr(nandc, first);
 
-	size = num_regs * sizeof(u32);
-	vaddr = nandc->reg_read_buf + nandc->reg_read_pos;
-	nandc->reg_read_pos += num_regs;
+	if (nandc->props->is_bam)
+		return prep_bam_dma_desc_cmd(nandc, true, first, vaddr,
+					     num_regs, flags);
 
-	return prep_adm_dma_desc(nandc, true, first, vaddr, size, flow_control);
+	if (first == NAND_READ_ID || first == NAND_FLASH_STATUS)
+		flow_control = true;
+
+	return prep_adm_dma_desc(nandc, true, first, vaddr,
+				 num_regs * sizeof(u32), flow_control);
 }
 
 /*
@@ -897,13 +970,9 @@ static int write_reg_dma(struct qcom_nand_controller *nandc, int first,
 	bool flow_control = false;
 	struct nandc_regs *regs = nandc->regs;
 	void *vaddr;
-	int size;
 
 	vaddr = offset_to_nandc_reg(regs, first);
 
-	if (first == NAND_FLASH_CMD)
-		flow_control = true;
-
 	if (first == NAND_ERASED_CW_DETECT_CFG) {
 		if (flags & NAND_ERASED_CW_SET)
 			vaddr = &regs->erased_cw_detect_cfg_set;
@@ -920,10 +989,15 @@ static int write_reg_dma(struct qcom_nand_controller *nandc, int first,
 	if (first == NAND_DEV_CMD_VLD_RESTORE || first == NAND_DEV_CMD_VLD)
 		first = dev_cmd_reg_addr(nandc, NAND_DEV_CMD_VLD);
 
-	size = num_regs * sizeof(u32);
+	if (nandc->props->is_bam)
+		return prep_bam_dma_desc_cmd(nandc, false, first, vaddr,
+					     num_regs, flags);
+
+	if (first == NAND_FLASH_CMD)
+		flow_control = true;
 
-	return prep_adm_dma_desc(nandc, false, first, vaddr, size,
-				 flow_control);
+	return prep_adm_dma_desc(nandc, false, first, vaddr,
+				 num_regs * sizeof(u32), flow_control);
 }
 
 /*
@@ -1187,7 +1261,8 @@ static int submit_descs(struct qcom_nand_controller *nandc)
 		}
 
 		if (bam_txn->cmd_sgl_pos > bam_txn->cmd_sgl_start) {
-			r = prepare_bam_async_desc(nandc, nandc->cmd_chan, 0);
+			r = prepare_bam_async_desc(nandc, nandc->cmd_chan,
+						   DMA_PREP_CMD);
 			if (r)
 				return r;
 		}
@@ -2722,6 +2797,7 @@ static int qcom_nandc_probe(struct platform_device *pdev)
 	if (IS_ERR(nandc->base))
 		return PTR_ERR(nandc->base);
 
+	nandc->base_phys = res->start;
 	nandc->base_dma = phys_to_dma(dev, (phys_addr_t)res->start);
 
 	nandc->core_clk = devm_clk_get(dev, "core");
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

Re: [RESEND PATCH v5 10/16] mtd: nand: qcom: add command elements in BAM transaction

[Boris Brezillon]

On Mon, 25 Sep 2017 13:21:25 +0530
Abhishek Sahu <absahu@codeaurora.org> wrote:

> All the QPIC register read/write through BAM DMA requires
> command descriptor which contains the array of command elements.
> 
> Reviewed-by: Archit Taneja <architt@codeaurora.org>
> Signed-off-by: Abhishek Sahu <absahu@codeaurora.org>

Applied both.

Thanks,

Boris

> ---
> 
> * Changes from v4: None
> 
> The BAM DMA patches [1] got merged in 4.14-rc1 so now all the build
> dependencies are solved and this patch can be merged.
> This patch can be cleanly applied over v4.14-rc1.
> 
> [1] http://www.spinics.net/lists/dmaengine/msg13665.html
> 
>  drivers/mtd/nand/qcom_nandc.c | 19 ++++++++++++++++++-
>  1 file changed, 18 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mtd/nand/qcom_nandc.c b/drivers/mtd/nand/qcom_nandc.c
> index 7977a70..b0a4734 100644
> --- a/drivers/mtd/nand/qcom_nandc.c
> +++ b/drivers/mtd/nand/qcom_nandc.c
> @@ -22,6 +22,7 @@
>  #include <linux/of.h>
>  #include <linux/of_device.h>
>  #include <linux/delay.h>
> +#include <linux/dma/qcom_bam_dma.h>
>  
>  /* NANDc reg offsets */
>  #define	NAND_FLASH_CMD			0x00
> @@ -199,6 +200,7 @@
>   */
>  #define dev_cmd_reg_addr(nandc, reg) ((nandc)->props->dev_cmd_reg_start + (reg))
>  
> +#define QPIC_PER_CW_CMD_ELEMENTS	32
>  #define QPIC_PER_CW_CMD_SGL		32
>  #define QPIC_PER_CW_DATA_SGL		8
>  
> @@ -221,8 +223,13 @@
>  /*
>   * This data type corresponds to the BAM transaction which will be used for all
>   * NAND transfers.
> + * @bam_ce - the array of BAM command elements
>   * @cmd_sgl - sgl for NAND BAM command pipe
>   * @data_sgl - sgl for NAND BAM consumer/producer pipe
> + * @bam_ce_pos - the index in bam_ce which is available for next sgl
> + * @bam_ce_start - the index in bam_ce which marks the start position ce
> + *		   for current sgl. It will be used for size calculation
> + *		   for current sgl
>   * @cmd_sgl_pos - current index in command sgl.
>   * @cmd_sgl_start - start index in command sgl.
>   * @tx_sgl_pos - current index in data sgl for tx.
> @@ -231,8 +238,11 @@
>   * @rx_sgl_start - start index in data sgl for rx.
>   */
>  struct bam_transaction {
> +	struct bam_cmd_element *bam_ce;
>  	struct scatterlist *cmd_sgl;
>  	struct scatterlist *data_sgl;
> +	u32 bam_ce_pos;
> +	u32 bam_ce_start;
>  	u32 cmd_sgl_pos;
>  	u32 cmd_sgl_start;
>  	u32 tx_sgl_pos;
> @@ -462,7 +472,8 @@ static void free_bam_transaction(struct qcom_nand_controller *nandc)
>  
>  	bam_txn_size =
>  		sizeof(*bam_txn) + num_cw *
> -		((sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL) +
> +		((sizeof(*bam_txn->bam_ce) * QPIC_PER_CW_CMD_ELEMENTS) +
> +		(sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL) +
>  		(sizeof(*bam_txn->data_sgl) * QPIC_PER_CW_DATA_SGL));
>  
>  	bam_txn_buf = devm_kzalloc(nandc->dev, bam_txn_size, GFP_KERNEL);
> @@ -472,6 +483,10 @@ static void free_bam_transaction(struct qcom_nand_controller *nandc)
>  	bam_txn = bam_txn_buf;
>  	bam_txn_buf += sizeof(*bam_txn);
>  
> +	bam_txn->bam_ce = bam_txn_buf;
> +	bam_txn_buf +=
> +		sizeof(*bam_txn->bam_ce) * QPIC_PER_CW_CMD_ELEMENTS * num_cw;
> +
>  	bam_txn->cmd_sgl = bam_txn_buf;
>  	bam_txn_buf +=
>  		sizeof(*bam_txn->cmd_sgl) * QPIC_PER_CW_CMD_SGL * num_cw;
> @@ -489,6 +504,8 @@ static void clear_bam_transaction(struct qcom_nand_controller *nandc)
>  	if (!nandc->props->is_bam)
>  		return;
>  
> +	bam_txn->bam_ce_pos = 0;
> +	bam_txn->bam_ce_start = 0;
>  	bam_txn->cmd_sgl_pos = 0;
>  	bam_txn->cmd_sgl_start = 0;
>  	bam_txn->tx_sgl_pos = 0;