linux/drivers/mtd/nand/raw/qcom_nandc.c

3053 lines
82 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* Copyright (c) 2016, The Linux Foundation. All rights reserved.
*/
#include <linux/clk.h>
#include <linux/slab.h>
#include <linux/bitops.h>
#include <linux/dma-mapping.h>
#include <linux/dmaengine.h>
#include <linux/module.h>
#include <linux/mtd/rawnand.h>
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
#include <linux/mtd/partitions.h>
#include <linux/of.h>
#include <linux/of_device.h>
#include <linux/delay.h>
#include <linux/dma/qcom_bam_dma.h>
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* NANDc reg offsets */
#define NAND_FLASH_CMD 0x00
#define NAND_ADDR0 0x04
#define NAND_ADDR1 0x08
#define NAND_FLASH_CHIP_SELECT 0x0c
#define NAND_EXEC_CMD 0x10
#define NAND_FLASH_STATUS 0x14
#define NAND_BUFFER_STATUS 0x18
#define NAND_DEV0_CFG0 0x20
#define NAND_DEV0_CFG1 0x24
#define NAND_DEV0_ECC_CFG 0x28
#define NAND_DEV1_ECC_CFG 0x2c
#define NAND_DEV1_CFG0 0x30
#define NAND_DEV1_CFG1 0x34
#define NAND_READ_ID 0x40
#define NAND_READ_STATUS 0x44
#define NAND_DEV_CMD0 0xa0
#define NAND_DEV_CMD1 0xa4
#define NAND_DEV_CMD2 0xa8
#define NAND_DEV_CMD_VLD 0xac
#define SFLASHC_BURST_CFG 0xe0
#define NAND_ERASED_CW_DETECT_CFG 0xe8
#define NAND_ERASED_CW_DETECT_STATUS 0xec
#define NAND_EBI2_ECC_BUF_CFG 0xf0
#define FLASH_BUF_ACC 0x100
#define NAND_CTRL 0xf00
#define NAND_VERSION 0xf08
#define NAND_READ_LOCATION_0 0xf20
#define NAND_READ_LOCATION_1 0xf24
#define NAND_READ_LOCATION_2 0xf28
#define NAND_READ_LOCATION_3 0xf2c
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* dummy register offsets, used by write_reg_dma */
#define NAND_DEV_CMD1_RESTORE 0xdead
#define NAND_DEV_CMD_VLD_RESTORE 0xbeef
/* NAND_FLASH_CMD bits */
#define PAGE_ACC BIT(4)
#define LAST_PAGE BIT(5)
/* NAND_FLASH_CHIP_SELECT bits */
#define NAND_DEV_SEL 0
#define DM_EN BIT(2)
/* NAND_FLASH_STATUS bits */
#define FS_OP_ERR BIT(4)
#define FS_READY_BSY_N BIT(5)
#define FS_MPU_ERR BIT(8)
#define FS_DEVICE_STS_ERR BIT(16)
#define FS_DEVICE_WP BIT(23)
/* NAND_BUFFER_STATUS bits */
#define BS_UNCORRECTABLE_BIT BIT(8)
#define BS_CORRECTABLE_ERR_MSK 0x1f
/* NAND_DEVn_CFG0 bits */
#define DISABLE_STATUS_AFTER_WRITE 4
#define CW_PER_PAGE 6
#define UD_SIZE_BYTES 9
#define ECC_PARITY_SIZE_BYTES_RS 19
#define SPARE_SIZE_BYTES 23
#define NUM_ADDR_CYCLES 27
#define STATUS_BFR_READ 30
#define SET_RD_MODE_AFTER_STATUS 31
/* NAND_DEVn_CFG0 bits */
#define DEV0_CFG1_ECC_DISABLE 0
#define WIDE_FLASH 1
#define NAND_RECOVERY_CYCLES 2
#define CS_ACTIVE_BSY 5
#define BAD_BLOCK_BYTE_NUM 6
#define BAD_BLOCK_IN_SPARE_AREA 16
#define WR_RD_BSY_GAP 17
#define ENABLE_BCH_ECC 27
/* NAND_DEV0_ECC_CFG bits */
#define ECC_CFG_ECC_DISABLE 0
#define ECC_SW_RESET 1
#define ECC_MODE 4
#define ECC_PARITY_SIZE_BYTES_BCH 8
#define ECC_NUM_DATA_BYTES 16
#define ECC_FORCE_CLK_OPEN 30
/* NAND_DEV_CMD1 bits */
#define READ_ADDR 0
/* NAND_DEV_CMD_VLD bits */
#define READ_START_VLD BIT(0)
#define READ_STOP_VLD BIT(1)
#define WRITE_START_VLD BIT(2)
#define ERASE_START_VLD BIT(3)
#define SEQ_READ_START_VLD BIT(4)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* NAND_EBI2_ECC_BUF_CFG bits */
#define NUM_STEPS 0
/* NAND_ERASED_CW_DETECT_CFG bits */
#define ERASED_CW_ECC_MASK 1
#define AUTO_DETECT_RES 0
#define MASK_ECC (1 << ERASED_CW_ECC_MASK)
#define RESET_ERASED_DET (1 << AUTO_DETECT_RES)
#define ACTIVE_ERASED_DET (0 << AUTO_DETECT_RES)
#define CLR_ERASED_PAGE_DET (RESET_ERASED_DET | MASK_ECC)
#define SET_ERASED_PAGE_DET (ACTIVE_ERASED_DET | MASK_ECC)
/* NAND_ERASED_CW_DETECT_STATUS bits */
#define PAGE_ALL_ERASED BIT(7)
#define CODEWORD_ALL_ERASED BIT(6)
#define PAGE_ERASED BIT(5)
#define CODEWORD_ERASED BIT(4)
#define ERASED_PAGE (PAGE_ALL_ERASED | PAGE_ERASED)
#define ERASED_CW (CODEWORD_ALL_ERASED | CODEWORD_ERASED)
/* NAND_READ_LOCATION_n bits */
#define READ_LOCATION_OFFSET 0
#define READ_LOCATION_SIZE 16
#define READ_LOCATION_LAST 31
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* Version Mask */
#define NAND_VERSION_MAJOR_MASK 0xf0000000
#define NAND_VERSION_MAJOR_SHIFT 28
#define NAND_VERSION_MINOR_MASK 0x0fff0000
#define NAND_VERSION_MINOR_SHIFT 16
/* NAND OP_CMDs */
#define OP_PAGE_READ 0x2
#define OP_PAGE_READ_WITH_ECC 0x3
#define OP_PAGE_READ_WITH_ECC_SPARE 0x4
#define OP_PROGRAM_PAGE 0x6
#define OP_PAGE_PROGRAM_WITH_ECC 0x7
#define OP_PROGRAM_PAGE_SPARE 0x9
#define OP_BLOCK_ERASE 0xa
#define OP_FETCH_ID 0xb
#define OP_RESET_DEVICE 0xd
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* Default Value for NAND_DEV_CMD_VLD */
#define NAND_DEV_CMD_VLD_VAL (READ_START_VLD | WRITE_START_VLD | \
ERASE_START_VLD | SEQ_READ_START_VLD)
/* NAND_CTRL bits */
#define BAM_MODE_EN BIT(0)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* the NAND controller performs reads/writes with ECC in 516 byte chunks.
* the driver calls the chunks 'step' or 'codeword' interchangeably
*/
#define NANDC_STEP_SIZE 512
/*
* the largest page size we support is 8K, this will have 16 steps/codewords
* of 512 bytes each
*/
#define MAX_NUM_STEPS (SZ_8K / NANDC_STEP_SIZE)
/* we read at most 3 registers per codeword scan */
#define MAX_REG_RD (3 * MAX_NUM_STEPS)
/* ECC modes supported by the controller */
#define ECC_NONE BIT(0)
#define ECC_RS_4BIT BIT(1)
#define ECC_BCH_4BIT BIT(2)
#define ECC_BCH_8BIT BIT(3)
#define nandc_set_read_loc(nandc, reg, offset, size, is_last) \
nandc_set_reg(nandc, NAND_READ_LOCATION_##reg, \
((offset) << READ_LOCATION_OFFSET) | \
((size) << READ_LOCATION_SIZE) | \
((is_last) << READ_LOCATION_LAST))
/*
* Returns the actual register address for all NAND_DEV_ registers
* (i.e. NAND_DEV_CMD0, NAND_DEV_CMD1, NAND_DEV_CMD2 and NAND_DEV_CMD_VLD)
*/
#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
#define QPIC_NAND_COMPLETION_TIMEOUT msecs_to_jiffies(2000)
/*
* Flags used in DMA descriptor preparation helper functions
* (i.e. read_reg_dma/write_reg_dma/read_data_dma/write_data_dma)
*/
/* Don't set the EOT in current tx BAM sgl */
#define NAND_BAM_NO_EOT BIT(0)
/* Set the NWD flag in current BAM sgl */
#define NAND_BAM_NWD BIT(1)
/* Finish writing in the current BAM sgl and start writing in another BAM sgl */
#define NAND_BAM_NEXT_SGL BIT(2)
/*
* Erased codeword status is being used two times in single transfer so this
* flag will determine the current value of erased codeword status register
*/
#define NAND_ERASED_CW_SET BIT(4)
/*
* 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.
* @tx_sgl_start - start index in data sgl for tx.
* @rx_sgl_pos - current index in data sgl for rx.
* @rx_sgl_start - start index in data sgl for rx.
* @wait_second_completion - wait for second DMA desc completion before making
* the NAND transfer completion.
* @txn_done - completion for NAND transfer.
* @last_data_desc - last DMA desc in data channel (tx/rx).
* @last_cmd_desc - last DMA desc in command channel.
*/
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;
u32 tx_sgl_start;
u32 rx_sgl_pos;
u32 rx_sgl_start;
bool wait_second_completion;
struct completion txn_done;
struct dma_async_tx_descriptor *last_data_desc;
struct dma_async_tx_descriptor *last_cmd_desc;
};
/*
* This data type corresponds to the nand dma descriptor
* @list - list for desc_info
* @dir - DMA transfer direction
* @adm_sgl - sgl which will be used for single sgl dma descriptor. Only used by
* ADM
* @bam_sgl - sgl which will be used for dma descriptor. Only used by BAM
* @sgl_cnt - number of SGL in bam_sgl. Only used by BAM
* @dma_desc - low level DMA engine descriptor
*/
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct desc_info {
struct list_head node;
enum dma_data_direction dir;
union {
struct scatterlist adm_sgl;
struct {
struct scatterlist *bam_sgl;
int sgl_cnt;
};
};
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct dma_async_tx_descriptor *dma_desc;
};
/*
* holds the current register values that we want to write. acts as a contiguous
* chunk of memory which we use to write the controller registers through DMA.
*/
struct nandc_regs {
__le32 cmd;
__le32 addr0;
__le32 addr1;
__le32 chip_sel;
__le32 exec;
__le32 cfg0;
__le32 cfg1;
__le32 ecc_bch_cfg;
__le32 clrflashstatus;
__le32 clrreadstatus;
__le32 cmd1;
__le32 vld;
__le32 orig_cmd1;
__le32 orig_vld;
__le32 ecc_buf_cfg;
__le32 read_location0;
__le32 read_location1;
__le32 read_location2;
__le32 read_location3;
__le32 erased_cw_detect_cfg_clr;
__le32 erased_cw_detect_cfg_set;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
};
/*
* NAND controller data struct
*
* @controller: base controller structure
* @host_list: list containing all the chips attached to the
* controller
* @dev: parent device
* @base: MMIO base
* @base_phys: physical base address of controller registers
* @base_dma: dma base address of controller registers
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
* @core_clk: controller clock
* @aon_clk: another controller clock
*
* @chan: dma channel
* @cmd_crci: ADM DMA CRCI for command flow control
* @data_crci: ADM DMA CRCI for data flow control
* @desc_list: DMA descriptor list (list of desc_infos)
*
* @data_buffer: our local DMA buffer for page read/writes,
* used when we can't use the buffer provided
* by upper layers directly
* @buf_size/count/start: markers for chip->legacy.read_buf/write_buf
* functions
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
* @reg_read_buf: local buffer for reading back registers via DMA
* @reg_read_dma: contains dma address for register read buffer
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
* @reg_read_pos: marker for data read in reg_read_buf
*
* @regs: a contiguous chunk of memory for DMA register
* writes. contains the register values to be
* written to controller
* @cmd1/vld: some fixed controller register values
* @props: properties of current NAND controller,
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
* initialized via DT match data
* @max_cwperpage: maximum QPIC codewords required. calculated
* from all connected NAND devices pagesize
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
struct qcom_nand_controller {
struct nand_controller controller;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct list_head host_list;
struct device *dev;
void __iomem *base;
phys_addr_t base_phys;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
dma_addr_t base_dma;
struct clk *core_clk;
struct clk *aon_clk;
union {
/* will be used only by QPIC for BAM DMA */
struct {
struct dma_chan *tx_chan;
struct dma_chan *rx_chan;
struct dma_chan *cmd_chan;
};
/* will be used only by EBI2 for ADM DMA */
struct {
struct dma_chan *chan;
unsigned int cmd_crci;
unsigned int data_crci;
};
};
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct list_head desc_list;
struct bam_transaction *bam_txn;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
u8 *data_buffer;
int buf_size;
int buf_count;
int buf_start;
unsigned int max_cwperpage;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
__le32 *reg_read_buf;
dma_addr_t reg_read_dma;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int reg_read_pos;
struct nandc_regs *regs;
u32 cmd1, vld;
const struct qcom_nandc_props *props;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
};
/*
* NAND chip structure
*
* @chip: base NAND chip structure
* @node: list node to add itself to host_list in
* qcom_nand_controller
*
* @cs: chip select value for this chip
* @cw_size: the number of bytes in a single step/codeword
* of a page, consisting of all data, ecc, spare
* and reserved bytes
* @cw_data: the number of bytes within a codeword protected
* by ECC
* @use_ecc: request the controller to use ECC for the
* upcoming read/write
* @bch_enabled: flag to tell whether BCH ECC mode is used
* @ecc_bytes_hw: ECC bytes used by controller hardware for this
* chip
* @status: value to be returned if NAND_CMD_STATUS command
* is executed
* @last_command: keeps track of last command on this chip. used
* for reading correct status
*
* @cfg0, cfg1, cfg0_raw..: NANDc register configurations needed for
* ecc/non-ecc mode for the current nand flash
* device
*/
struct qcom_nand_host {
struct nand_chip chip;
struct list_head node;
int cs;
int cw_size;
int cw_data;
bool use_ecc;
bool bch_enabled;
int ecc_bytes_hw;
int spare_bytes;
int bbm_size;
u8 status;
int last_command;
u32 cfg0, cfg1;
u32 cfg0_raw, cfg1_raw;
u32 ecc_buf_cfg;
u32 ecc_bch_cfg;
u32 clrflashstatus;
u32 clrreadstatus;
};
/*
* This data type corresponds to the NAND controller properties which varies
* among different NAND controllers.
* @ecc_modes - ecc mode for NAND
* @is_bam - whether NAND controller is using BAM
* @dev_cmd_reg_start - NAND_DEV_CMD_* registers starting offset
*/
struct qcom_nandc_props {
u32 ecc_modes;
bool is_bam;
u32 dev_cmd_reg_start;
};
/* Frees the BAM transaction memory */
static void free_bam_transaction(struct qcom_nand_controller *nandc)
{
struct bam_transaction *bam_txn = nandc->bam_txn;
devm_kfree(nandc->dev, bam_txn);
}
/* Allocates and Initializes the BAM transaction */
static struct bam_transaction *
alloc_bam_transaction(struct qcom_nand_controller *nandc)
{
struct bam_transaction *bam_txn;
size_t bam_txn_size;
unsigned int num_cw = nandc->max_cwperpage;
void *bam_txn_buf;
bam_txn_size =
sizeof(*bam_txn) + num_cw *
((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);
if (!bam_txn_buf)
return NULL;
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;
bam_txn->data_sgl = bam_txn_buf;
init_completion(&bam_txn->txn_done);
return bam_txn;
}
/* Clears the BAM transaction indexes */
static void clear_bam_transaction(struct qcom_nand_controller *nandc)
{
struct bam_transaction *bam_txn = nandc->bam_txn;
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;
bam_txn->tx_sgl_start = 0;
bam_txn->rx_sgl_pos = 0;
bam_txn->rx_sgl_start = 0;
bam_txn->last_data_desc = NULL;
bam_txn->wait_second_completion = false;
sg_init_table(bam_txn->cmd_sgl, nandc->max_cwperpage *
QPIC_PER_CW_CMD_SGL);
sg_init_table(bam_txn->data_sgl, nandc->max_cwperpage *
QPIC_PER_CW_DATA_SGL);
reinit_completion(&bam_txn->txn_done);
}
/* Callback for DMA descriptor completion */
static void qpic_bam_dma_done(void *data)
{
struct bam_transaction *bam_txn = data;
/*
* In case of data transfer with NAND, 2 callbacks will be generated.
* One for command channel and another one for data channel.
* If current transaction has data descriptors
* (i.e. wait_second_completion is true), then set this to false
* and wait for second DMA descriptor completion.
*/
if (bam_txn->wait_second_completion)
bam_txn->wait_second_completion = false;
else
complete(&bam_txn->txn_done);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
static inline struct qcom_nand_host *to_qcom_nand_host(struct nand_chip *chip)
{
return container_of(chip, struct qcom_nand_host, chip);
}
static inline struct qcom_nand_controller *
get_qcom_nand_controller(struct nand_chip *chip)
{
return container_of(chip->controller, struct qcom_nand_controller,
controller);
}
static inline u32 nandc_read(struct qcom_nand_controller *nandc, int offset)
{
return ioread32(nandc->base + offset);
}
static inline void nandc_write(struct qcom_nand_controller *nandc, int offset,
u32 val)
{
iowrite32(val, nandc->base + offset);
}
static inline void nandc_read_buffer_sync(struct qcom_nand_controller *nandc,
bool is_cpu)
{
if (!nandc->props->is_bam)
return;
if (is_cpu)
dma_sync_single_for_cpu(nandc->dev, nandc->reg_read_dma,
MAX_REG_RD *
sizeof(*nandc->reg_read_buf),
DMA_FROM_DEVICE);
else
dma_sync_single_for_device(nandc->dev, nandc->reg_read_dma,
MAX_REG_RD *
sizeof(*nandc->reg_read_buf),
DMA_FROM_DEVICE);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
static __le32 *offset_to_nandc_reg(struct nandc_regs *regs, int offset)
{
switch (offset) {
case NAND_FLASH_CMD:
return &regs->cmd;
case NAND_ADDR0:
return &regs->addr0;
case NAND_ADDR1:
return &regs->addr1;
case NAND_FLASH_CHIP_SELECT:
return &regs->chip_sel;
case NAND_EXEC_CMD:
return &regs->exec;
case NAND_FLASH_STATUS:
return &regs->clrflashstatus;
case NAND_DEV0_CFG0:
return &regs->cfg0;
case NAND_DEV0_CFG1:
return &regs->cfg1;
case NAND_DEV0_ECC_CFG:
return &regs->ecc_bch_cfg;
case NAND_READ_STATUS:
return &regs->clrreadstatus;
case NAND_DEV_CMD1:
return &regs->cmd1;
case NAND_DEV_CMD1_RESTORE:
return &regs->orig_cmd1;
case NAND_DEV_CMD_VLD:
return &regs->vld;
case NAND_DEV_CMD_VLD_RESTORE:
return &regs->orig_vld;
case NAND_EBI2_ECC_BUF_CFG:
return &regs->ecc_buf_cfg;
case NAND_READ_LOCATION_0:
return &regs->read_location0;
case NAND_READ_LOCATION_1:
return &regs->read_location1;
case NAND_READ_LOCATION_2:
return &regs->read_location2;
case NAND_READ_LOCATION_3:
return &regs->read_location3;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
default:
return NULL;
}
}
static void nandc_set_reg(struct qcom_nand_controller *nandc, int offset,
u32 val)
{
struct nandc_regs *regs = nandc->regs;
__le32 *reg;
reg = offset_to_nandc_reg(regs, offset);
if (reg)
*reg = cpu_to_le32(val);
}
/* helper to configure address register values */
static void set_address(struct qcom_nand_host *host, u16 column, int page)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
if (chip->options & NAND_BUSWIDTH_16)
column >>= 1;
nandc_set_reg(nandc, NAND_ADDR0, page << 16 | column);
nandc_set_reg(nandc, NAND_ADDR1, page >> 16 & 0xff);
}
/*
* update_rw_regs: set up read/write register values, these will be
* written to the NAND controller registers via DMA
*
* @num_cw: number of steps for the read/write operation
* @read: read or write operation
*/
static void update_rw_regs(struct qcom_nand_host *host, int num_cw, bool read)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
u32 cmd, cfg0, cfg1, ecc_bch_cfg;
if (read) {
if (host->use_ecc)
cmd = OP_PAGE_READ_WITH_ECC | PAGE_ACC | LAST_PAGE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
else
cmd = OP_PAGE_READ | PAGE_ACC | LAST_PAGE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
} else {
cmd = OP_PROGRAM_PAGE | PAGE_ACC | LAST_PAGE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
if (host->use_ecc) {
cfg0 = (host->cfg0 & ~(7U << CW_PER_PAGE)) |
(num_cw - 1) << CW_PER_PAGE;
cfg1 = host->cfg1;
ecc_bch_cfg = host->ecc_bch_cfg;
} else {
cfg0 = (host->cfg0_raw & ~(7U << CW_PER_PAGE)) |
(num_cw - 1) << CW_PER_PAGE;
cfg1 = host->cfg1_raw;
ecc_bch_cfg = 1 << ECC_CFG_ECC_DISABLE;
}
nandc_set_reg(nandc, NAND_FLASH_CMD, cmd);
nandc_set_reg(nandc, NAND_DEV0_CFG0, cfg0);
nandc_set_reg(nandc, NAND_DEV0_CFG1, cfg1);
nandc_set_reg(nandc, NAND_DEV0_ECC_CFG, ecc_bch_cfg);
nandc_set_reg(nandc, NAND_EBI2_ECC_BUF_CFG, host->ecc_buf_cfg);
nandc_set_reg(nandc, NAND_FLASH_STATUS, host->clrflashstatus);
nandc_set_reg(nandc, NAND_READ_STATUS, host->clrreadstatus);
nandc_set_reg(nandc, NAND_EXEC_CMD, 1);
if (read)
nandc_set_read_loc(nandc, 0, 0, host->use_ecc ?
host->cw_data : host->cw_size, 1);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* Maps the scatter gather list for DMA transfer and forms the DMA descriptor
* for BAM. This descriptor will be added in the NAND DMA descriptor queue
* which will be submitted to DMA engine.
*/
static int prepare_bam_async_desc(struct qcom_nand_controller *nandc,
struct dma_chan *chan,
unsigned long flags)
{
struct desc_info *desc;
struct scatterlist *sgl;
unsigned int sgl_cnt;
int ret;
struct bam_transaction *bam_txn = nandc->bam_txn;
enum dma_transfer_direction dir_eng;
struct dma_async_tx_descriptor *dma_desc;
desc = kzalloc(sizeof(*desc), GFP_KERNEL);
if (!desc)
return -ENOMEM;
if (chan == nandc->cmd_chan) {
sgl = &bam_txn->cmd_sgl[bam_txn->cmd_sgl_start];
sgl_cnt = bam_txn->cmd_sgl_pos - bam_txn->cmd_sgl_start;
bam_txn->cmd_sgl_start = bam_txn->cmd_sgl_pos;
dir_eng = DMA_MEM_TO_DEV;
desc->dir = DMA_TO_DEVICE;
} else if (chan == nandc->tx_chan) {
sgl = &bam_txn->data_sgl[bam_txn->tx_sgl_start];
sgl_cnt = bam_txn->tx_sgl_pos - bam_txn->tx_sgl_start;
bam_txn->tx_sgl_start = bam_txn->tx_sgl_pos;
dir_eng = DMA_MEM_TO_DEV;
desc->dir = DMA_TO_DEVICE;
} else {
sgl = &bam_txn->data_sgl[bam_txn->rx_sgl_start];
sgl_cnt = bam_txn->rx_sgl_pos - bam_txn->rx_sgl_start;
bam_txn->rx_sgl_start = bam_txn->rx_sgl_pos;
dir_eng = DMA_DEV_TO_MEM;
desc->dir = DMA_FROM_DEVICE;
}
sg_mark_end(sgl + sgl_cnt - 1);
ret = dma_map_sg(nandc->dev, sgl, sgl_cnt, desc->dir);
if (ret == 0) {
dev_err(nandc->dev, "failure in mapping desc\n");
kfree(desc);
return -ENOMEM;
}
desc->sgl_cnt = sgl_cnt;
desc->bam_sgl = sgl;
dma_desc = dmaengine_prep_slave_sg(chan, sgl, sgl_cnt, dir_eng,
flags);
if (!dma_desc) {
dev_err(nandc->dev, "failure in prep desc\n");
dma_unmap_sg(nandc->dev, sgl, sgl_cnt, desc->dir);
kfree(desc);
return -EINVAL;
}
desc->dma_desc = dma_desc;
/* update last data/command descriptor */
if (chan == nandc->cmd_chan)
bam_txn->last_cmd_desc = dma_desc;
else
bam_txn->last_data_desc = dma_desc;
list_add_tail(&desc->node, &nandc->desc_list);
return 0;
}
/*
* 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.
*/
static int prep_bam_dma_desc_data(struct qcom_nand_controller *nandc, bool read,
const void *vaddr,
int size, unsigned int flags)
{
int ret;
struct bam_transaction *bam_txn = nandc->bam_txn;
if (read) {
sg_set_buf(&bam_txn->data_sgl[bam_txn->rx_sgl_pos],
vaddr, size);
bam_txn->rx_sgl_pos++;
} else {
sg_set_buf(&bam_txn->data_sgl[bam_txn->tx_sgl_pos],
vaddr, size);
bam_txn->tx_sgl_pos++;
/*
* BAM will only set EOT for DMA_PREP_INTERRUPT so if this flag
* is not set, form the DMA descriptor
*/
if (!(flags & NAND_BAM_NO_EOT)) {
ret = prepare_bam_async_desc(nandc, nandc->tx_chan,
DMA_PREP_INTERRUPT);
if (ret)
return ret;
}
}
return 0;
}
static int prep_adm_dma_desc(struct qcom_nand_controller *nandc, bool read,
int reg_off, const void *vaddr, int size,
bool flow_control)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct desc_info *desc;
struct dma_async_tx_descriptor *dma_desc;
struct scatterlist *sgl;
struct dma_slave_config slave_conf;
enum dma_transfer_direction dir_eng;
int ret;
desc = kzalloc(sizeof(*desc), GFP_KERNEL);
if (!desc)
return -ENOMEM;
sgl = &desc->adm_sgl;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
sg_init_one(sgl, vaddr, size);
if (read) {
dir_eng = DMA_DEV_TO_MEM;
desc->dir = DMA_FROM_DEVICE;
} else {
dir_eng = DMA_MEM_TO_DEV;
desc->dir = DMA_TO_DEVICE;
}
ret = dma_map_sg(nandc->dev, sgl, 1, desc->dir);
if (ret == 0) {
ret = -ENOMEM;
goto err;
}
memset(&slave_conf, 0x00, sizeof(slave_conf));
slave_conf.device_fc = flow_control;
if (read) {
slave_conf.src_maxburst = 16;
slave_conf.src_addr = nandc->base_dma + reg_off;
slave_conf.slave_id = nandc->data_crci;
} else {
slave_conf.dst_maxburst = 16;
slave_conf.dst_addr = nandc->base_dma + reg_off;
slave_conf.slave_id = nandc->cmd_crci;
}
ret = dmaengine_slave_config(nandc->chan, &slave_conf);
if (ret) {
dev_err(nandc->dev, "failed to configure dma channel\n");
goto err;
}
dma_desc = dmaengine_prep_slave_sg(nandc->chan, sgl, 1, dir_eng, 0);
if (!dma_desc) {
dev_err(nandc->dev, "failed to prepare desc\n");
ret = -EINVAL;
goto err;
}
desc->dma_desc = dma_desc;
list_add_tail(&desc->node, &nandc->desc_list);
return 0;
err:
kfree(desc);
return ret;
}
/*
* read_reg_dma: prepares a descriptor to read a given number of
* contiguous registers to the reg_read_buf pointer
*
* @first: offset of the first register in the contiguous block
* @num_regs: number of registers to read
* @flags: flags to control DMA descriptor preparation
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int read_reg_dma(struct qcom_nand_controller *nandc, int first,
int num_regs, unsigned int flags)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
bool flow_control = false;
void *vaddr;
vaddr = nandc->reg_read_buf + nandc->reg_read_pos;
nandc->reg_read_pos += num_regs;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (first == NAND_DEV_CMD_VLD || first == NAND_DEV_CMD1)
first = dev_cmd_reg_addr(nandc, first);
if (nandc->props->is_bam)
return prep_bam_dma_desc_cmd(nandc, true, first, vaddr,
num_regs, flags);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
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);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* write_reg_dma: prepares a descriptor to write a given number of
* contiguous registers
*
* @first: offset of the first register in the contiguous block
* @num_regs: number of registers to write
* @flags: flags to control DMA descriptor preparation
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int write_reg_dma(struct qcom_nand_controller *nandc, int first,
int num_regs, unsigned int flags)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
bool flow_control = false;
struct nandc_regs *regs = nandc->regs;
void *vaddr;
vaddr = offset_to_nandc_reg(regs, first);
if (first == NAND_ERASED_CW_DETECT_CFG) {
if (flags & NAND_ERASED_CW_SET)
vaddr = &regs->erased_cw_detect_cfg_set;
else
vaddr = &regs->erased_cw_detect_cfg_clr;
}
if (first == NAND_EXEC_CMD)
flags |= NAND_BAM_NWD;
if (first == NAND_DEV_CMD1_RESTORE || first == NAND_DEV_CMD1)
first = dev_cmd_reg_addr(nandc, NAND_DEV_CMD1);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (first == NAND_DEV_CMD_VLD_RESTORE || first == NAND_DEV_CMD_VLD)
first = dev_cmd_reg_addr(nandc, NAND_DEV_CMD_VLD);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
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;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return prep_adm_dma_desc(nandc, false, first, vaddr,
num_regs * sizeof(u32), flow_control);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* read_data_dma: prepares a DMA descriptor to transfer data from the
* controller's internal buffer to the buffer 'vaddr'
*
* @reg_off: offset within the controller's data buffer
* @vaddr: virtual address of the buffer we want to write to
* @size: DMA transaction size in bytes
* @flags: flags to control DMA descriptor preparation
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int read_data_dma(struct qcom_nand_controller *nandc, int reg_off,
const u8 *vaddr, int size, unsigned int flags)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
if (nandc->props->is_bam)
return prep_bam_dma_desc_data(nandc, true, vaddr, size, flags);
return prep_adm_dma_desc(nandc, true, reg_off, vaddr, size, false);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* write_data_dma: prepares a DMA descriptor to transfer data from
* 'vaddr' to the controller's internal buffer
*
* @reg_off: offset within the controller's data buffer
* @vaddr: virtual address of the buffer we want to read from
* @size: DMA transaction size in bytes
* @flags: flags to control DMA descriptor preparation
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int write_data_dma(struct qcom_nand_controller *nandc, int reg_off,
const u8 *vaddr, int size, unsigned int flags)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
if (nandc->props->is_bam)
return prep_bam_dma_desc_data(nandc, false, vaddr, size, flags);
return prep_adm_dma_desc(nandc, false, reg_off, vaddr, size, false);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* Helper to prepare DMA descriptors for configuring registers
* before reading a NAND page.
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static void config_nand_page_read(struct qcom_nand_controller *nandc)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
write_reg_dma(nandc, NAND_ADDR0, 2, 0);
write_reg_dma(nandc, NAND_DEV0_CFG0, 3, 0);
write_reg_dma(nandc, NAND_EBI2_ECC_BUF_CFG, 1, 0);
write_reg_dma(nandc, NAND_ERASED_CW_DETECT_CFG, 1, 0);
write_reg_dma(nandc, NAND_ERASED_CW_DETECT_CFG, 1,
NAND_ERASED_CW_SET | NAND_BAM_NEXT_SGL);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* Helper to prepare DMA descriptors for configuring registers
* before reading each codeword in NAND page.
*/
static void
config_nand_cw_read(struct qcom_nand_controller *nandc, bool use_ecc)
{
if (nandc->props->is_bam)
write_reg_dma(nandc, NAND_READ_LOCATION_0, 4,
NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_FLASH_CMD, 1, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (use_ecc) {
read_reg_dma(nandc, NAND_FLASH_STATUS, 2, 0);
read_reg_dma(nandc, NAND_ERASED_CW_DETECT_STATUS, 1,
NAND_BAM_NEXT_SGL);
} else {
read_reg_dma(nandc, NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* Helper to prepare dma descriptors to configure registers needed for reading a
* single codeword in page
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static void
config_nand_single_cw_page_read(struct qcom_nand_controller *nandc,
bool use_ecc)
{
config_nand_page_read(nandc);
config_nand_cw_read(nandc, use_ecc);
}
/*
* Helper to prepare DMA descriptors used to configure registers needed for
* before writing a NAND page.
*/
static void config_nand_page_write(struct qcom_nand_controller *nandc)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
write_reg_dma(nandc, NAND_ADDR0, 2, 0);
write_reg_dma(nandc, NAND_DEV0_CFG0, 3, 0);
write_reg_dma(nandc, NAND_EBI2_ECC_BUF_CFG, 1,
NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* Helper to prepare DMA descriptors for configuring registers
* before writing each codeword in NAND page.
*/
static void config_nand_cw_write(struct qcom_nand_controller *nandc)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
write_reg_dma(nandc, NAND_FLASH_CMD, 1, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_reg_dma(nandc, NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
write_reg_dma(nandc, NAND_FLASH_STATUS, 1, 0);
write_reg_dma(nandc, NAND_READ_STATUS, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* the following functions are used within chip->legacy.cmdfunc() to
* perform different NAND_CMD_* commands
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
/* sets up descriptors for NAND_CMD_PARAM */
static int nandc_param(struct qcom_nand_host *host)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
/*
* NAND_CMD_PARAM is called before we know much about the FLASH chip
* in use. we configure the controller to perform a raw read of 512
* bytes to read onfi params
*/
nandc_set_reg(nandc, NAND_FLASH_CMD, OP_PAGE_READ | PAGE_ACC | LAST_PAGE);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_ADDR0, 0);
nandc_set_reg(nandc, NAND_ADDR1, 0);
nandc_set_reg(nandc, NAND_DEV0_CFG0, 0 << CW_PER_PAGE
| 512 << UD_SIZE_BYTES
| 5 << NUM_ADDR_CYCLES
| 0 << SPARE_SIZE_BYTES);
nandc_set_reg(nandc, NAND_DEV0_CFG1, 7 << NAND_RECOVERY_CYCLES
| 0 << CS_ACTIVE_BSY
| 17 << BAD_BLOCK_BYTE_NUM
| 1 << BAD_BLOCK_IN_SPARE_AREA
| 2 << WR_RD_BSY_GAP
| 0 << WIDE_FLASH
| 1 << DEV0_CFG1_ECC_DISABLE);
nandc_set_reg(nandc, NAND_EBI2_ECC_BUF_CFG, 1 << ECC_CFG_ECC_DISABLE);
/* configure CMD1 and VLD for ONFI param probing */
nandc_set_reg(nandc, NAND_DEV_CMD_VLD,
(nandc->vld & ~READ_START_VLD));
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_DEV_CMD1,
(nandc->cmd1 & ~(0xFF << READ_ADDR))
| NAND_CMD_PARAM << READ_ADDR);
nandc_set_reg(nandc, NAND_EXEC_CMD, 1);
nandc_set_reg(nandc, NAND_DEV_CMD1_RESTORE, nandc->cmd1);
nandc_set_reg(nandc, NAND_DEV_CMD_VLD_RESTORE, nandc->vld);
nandc_set_read_loc(nandc, 0, 0, 512, 1);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
write_reg_dma(nandc, NAND_DEV_CMD_VLD, 1, 0);
write_reg_dma(nandc, NAND_DEV_CMD1, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc->buf_count = 512;
memset(nandc->data_buffer, 0xff, nandc->buf_count);
config_nand_single_cw_page_read(nandc, false);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_data_dma(nandc, FLASH_BUF_ACC, nandc->data_buffer,
nandc->buf_count, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* restore CMD1 and VLD regs */
write_reg_dma(nandc, NAND_DEV_CMD1_RESTORE, 1, 0);
write_reg_dma(nandc, NAND_DEV_CMD_VLD_RESTORE, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
/* sets up descriptors for NAND_CMD_ERASE1 */
static int erase_block(struct qcom_nand_host *host, int page_addr)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
nandc_set_reg(nandc, NAND_FLASH_CMD,
OP_BLOCK_ERASE | PAGE_ACC | LAST_PAGE);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_ADDR0, page_addr);
nandc_set_reg(nandc, NAND_ADDR1, 0);
nandc_set_reg(nandc, NAND_DEV0_CFG0,
host->cfg0_raw & ~(7 << CW_PER_PAGE));
nandc_set_reg(nandc, NAND_DEV0_CFG1, host->cfg1_raw);
nandc_set_reg(nandc, NAND_EXEC_CMD, 1);
nandc_set_reg(nandc, NAND_FLASH_STATUS, host->clrflashstatus);
nandc_set_reg(nandc, NAND_READ_STATUS, host->clrreadstatus);
write_reg_dma(nandc, NAND_FLASH_CMD, 3, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_DEV0_CFG0, 2, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_reg_dma(nandc, NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
write_reg_dma(nandc, NAND_FLASH_STATUS, 1, 0);
write_reg_dma(nandc, NAND_READ_STATUS, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
/* sets up descriptors for NAND_CMD_READID */
static int read_id(struct qcom_nand_host *host, int column)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
if (column == -1)
return 0;
nandc_set_reg(nandc, NAND_FLASH_CMD, OP_FETCH_ID);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_ADDR0, column);
nandc_set_reg(nandc, NAND_ADDR1, 0);
nandc_set_reg(nandc, NAND_FLASH_CHIP_SELECT,
nandc->props->is_bam ? 0 : DM_EN);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_EXEC_CMD, 1);
write_reg_dma(nandc, NAND_FLASH_CMD, 4, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_reg_dma(nandc, NAND_READ_ID, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
/* sets up descriptors for NAND_CMD_RESET */
static int reset(struct qcom_nand_host *host)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
nandc_set_reg(nandc, NAND_FLASH_CMD, OP_RESET_DEVICE);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc_set_reg(nandc, NAND_EXEC_CMD, 1);
write_reg_dma(nandc, NAND_FLASH_CMD, 1, NAND_BAM_NEXT_SGL);
write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_reg_dma(nandc, NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
/* helpers to submit/free our list of dma descriptors */
static int submit_descs(struct qcom_nand_controller *nandc)
{
struct desc_info *desc;
dma_cookie_t cookie = 0;
struct bam_transaction *bam_txn = nandc->bam_txn;
int r;
if (nandc->props->is_bam) {
if (bam_txn->rx_sgl_pos > bam_txn->rx_sgl_start) {
r = prepare_bam_async_desc(nandc, nandc->rx_chan, 0);
if (r)
return r;
}
if (bam_txn->tx_sgl_pos > bam_txn->tx_sgl_start) {
r = prepare_bam_async_desc(nandc, nandc->tx_chan,
DMA_PREP_INTERRUPT);
if (r)
return r;
}
if (bam_txn->cmd_sgl_pos > bam_txn->cmd_sgl_start) {
r = prepare_bam_async_desc(nandc, nandc->cmd_chan,
DMA_PREP_CMD);
if (r)
return r;
}
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
list_for_each_entry(desc, &nandc->desc_list, node)
cookie = dmaengine_submit(desc->dma_desc);
if (nandc->props->is_bam) {
bam_txn->last_cmd_desc->callback = qpic_bam_dma_done;
bam_txn->last_cmd_desc->callback_param = bam_txn;
if (bam_txn->last_data_desc) {
bam_txn->last_data_desc->callback = qpic_bam_dma_done;
bam_txn->last_data_desc->callback_param = bam_txn;
bam_txn->wait_second_completion = true;
}
dma_async_issue_pending(nandc->tx_chan);
dma_async_issue_pending(nandc->rx_chan);
dma_async_issue_pending(nandc->cmd_chan);
if (!wait_for_completion_timeout(&bam_txn->txn_done,
QPIC_NAND_COMPLETION_TIMEOUT))
return -ETIMEDOUT;
} else {
if (dma_sync_wait(nandc->chan, cookie) != DMA_COMPLETE)
return -ETIMEDOUT;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
static void free_descs(struct qcom_nand_controller *nandc)
{
struct desc_info *desc, *n;
list_for_each_entry_safe(desc, n, &nandc->desc_list, node) {
list_del(&desc->node);
if (nandc->props->is_bam)
dma_unmap_sg(nandc->dev, desc->bam_sgl,
desc->sgl_cnt, desc->dir);
else
dma_unmap_sg(nandc->dev, &desc->adm_sgl, 1,
desc->dir);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
kfree(desc);
}
}
/* reset the register read buffer for next NAND operation */
static void clear_read_regs(struct qcom_nand_controller *nandc)
{
nandc->reg_read_pos = 0;
nandc_read_buffer_sync(nandc, false);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
static void pre_command(struct qcom_nand_host *host, int command)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
nandc->buf_count = 0;
nandc->buf_start = 0;
host->use_ecc = false;
host->last_command = command;
clear_read_regs(nandc);
if (command == NAND_CMD_RESET || command == NAND_CMD_READID ||
command == NAND_CMD_PARAM || command == NAND_CMD_ERASE1)
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* this is called after NAND_CMD_PAGEPROG and NAND_CMD_ERASE1 to set our
* privately maintained status byte, this status byte can be read after
* NAND_CMD_STATUS is called
*/
static void parse_erase_write_errors(struct qcom_nand_host *host, int command)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int num_cw;
int i;
num_cw = command == NAND_CMD_PAGEPROG ? ecc->steps : 1;
nandc_read_buffer_sync(nandc, true);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
for (i = 0; i < num_cw; i++) {
u32 flash_status = le32_to_cpu(nandc->reg_read_buf[i]);
if (flash_status & FS_MPU_ERR)
host->status &= ~NAND_STATUS_WP;
if (flash_status & FS_OP_ERR || (i == (num_cw - 1) &&
(flash_status &
FS_DEVICE_STS_ERR)))
host->status |= NAND_STATUS_FAIL;
}
}
static void post_command(struct qcom_nand_host *host, int command)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
switch (command) {
case NAND_CMD_READID:
nandc_read_buffer_sync(nandc, true);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
memcpy(nandc->data_buffer, nandc->reg_read_buf,
nandc->buf_count);
break;
case NAND_CMD_PAGEPROG:
case NAND_CMD_ERASE1:
parse_erase_write_errors(host, command);
break;
default:
break;
}
}
/*
* Implements chip->legacy.cmdfunc. It's only used for a limited set of
* commands. The rest of the commands wouldn't be called by upper layers.
* For example, NAND_CMD_READOOB would never be called because we have our own
* versions of read_oob ops for nand_ecc_ctrl.
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static void qcom_nandc_command(struct nand_chip *chip, unsigned int command,
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int column, int page_addr)
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
bool wait = false;
int ret = 0;
pre_command(host, command);
switch (command) {
case NAND_CMD_RESET:
ret = reset(host);
wait = true;
break;
case NAND_CMD_READID:
nandc->buf_count = 4;
ret = read_id(host, column);
wait = true;
break;
case NAND_CMD_PARAM:
ret = nandc_param(host);
wait = true;
break;
case NAND_CMD_ERASE1:
ret = erase_block(host, page_addr);
wait = true;
break;
case NAND_CMD_READ0:
/* we read the entire page for now */
WARN_ON(column != 0);
host->use_ecc = true;
set_address(host, 0, page_addr);
update_rw_regs(host, ecc->steps, true);
break;
case NAND_CMD_SEQIN:
WARN_ON(column != 0);
set_address(host, 0, page_addr);
break;
case NAND_CMD_PAGEPROG:
case NAND_CMD_STATUS:
case NAND_CMD_NONE:
default:
break;
}
if (ret) {
dev_err(nandc->dev, "failure executing command %d\n",
command);
free_descs(nandc);
return;
}
if (wait) {
ret = submit_descs(nandc);
if (ret)
dev_err(nandc->dev,
"failure submitting descs for command %d\n",
command);
}
free_descs(nandc);
post_command(host, command);
}
/*
* when using BCH ECC, the HW flags an error in NAND_FLASH_STATUS if it read
* an erased CW, and reports an erased CW in NAND_ERASED_CW_DETECT_STATUS.
*
* when using RS ECC, the HW reports the same erros when reading an erased CW,
* but it notifies that it is an erased CW by placing special characters at
* certain offsets in the buffer.
*
* verify if the page is erased or not, and fix up the page for RS ECC by
* replacing the special characters with 0xff.
*/
static bool erased_chunk_check_and_fixup(u8 *data_buf, int data_len)
{
u8 empty1, empty2;
/*
* an erased page flags an error in NAND_FLASH_STATUS, check if the page
* is erased by looking for 0x54s at offsets 3 and 175 from the
* beginning of each codeword
*/
empty1 = data_buf[3];
empty2 = data_buf[175];
/*
* if the erased codework markers, if they exist override them with
* 0xffs
*/
if ((empty1 == 0x54 && empty2 == 0xff) ||
(empty1 == 0xff && empty2 == 0x54)) {
data_buf[3] = 0xff;
data_buf[175] = 0xff;
}
/*
* check if the entire chunk contains 0xffs or not. if it doesn't, then
* restore the original values at the special offsets
*/
if (memchr_inv(data_buf, 0xff, data_len)) {
data_buf[3] = empty1;
data_buf[175] = empty2;
return false;
}
return true;
}
struct read_stats {
__le32 flash;
__le32 buffer;
__le32 erased_cw;
};
/* reads back FLASH_STATUS register set by the controller */
static int check_flash_errors(struct qcom_nand_host *host, int cw_cnt)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
int i;
for (i = 0; i < cw_cnt; i++) {
u32 flash = le32_to_cpu(nandc->reg_read_buf[i]);
if (flash & (FS_OP_ERR | FS_MPU_ERR))
return -EIO;
}
return 0;
}
/* performs raw read for one codeword */
static int
qcom_nandc_read_cw_raw(struct mtd_info *mtd, struct nand_chip *chip,
u8 *data_buf, u8 *oob_buf, int page, int cw)
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int data_size1, data_size2, oob_size1, oob_size2;
int ret, reg_off = FLASH_BUF_ACC, read_loc = 0;
nand_read_page_op(chip, page, 0, NULL, 0);
host->use_ecc = false;
clear_bam_transaction(nandc);
set_address(host, host->cw_size * cw, page);
update_rw_regs(host, 1, true);
config_nand_page_read(nandc);
data_size1 = mtd->writesize - host->cw_size * (ecc->steps - 1);
oob_size1 = host->bbm_size;
if (cw == (ecc->steps - 1)) {
data_size2 = ecc->size - data_size1 -
((ecc->steps - 1) * 4);
oob_size2 = (ecc->steps * 4) + host->ecc_bytes_hw +
host->spare_bytes;
} else {
data_size2 = host->cw_data - data_size1;
oob_size2 = host->ecc_bytes_hw + host->spare_bytes;
}
if (nandc->props->is_bam) {
nandc_set_read_loc(nandc, 0, read_loc, data_size1, 0);
read_loc += data_size1;
nandc_set_read_loc(nandc, 1, read_loc, oob_size1, 0);
read_loc += oob_size1;
nandc_set_read_loc(nandc, 2, read_loc, data_size2, 0);
read_loc += data_size2;
nandc_set_read_loc(nandc, 3, read_loc, oob_size2, 1);
}
config_nand_cw_read(nandc, false);
read_data_dma(nandc, reg_off, data_buf, data_size1, 0);
reg_off += data_size1;
read_data_dma(nandc, reg_off, oob_buf, oob_size1, 0);
reg_off += oob_size1;
read_data_dma(nandc, reg_off, data_buf + data_size1, data_size2, 0);
reg_off += data_size2;
read_data_dma(nandc, reg_off, oob_buf + oob_size1, oob_size2, 0);
ret = submit_descs(nandc);
free_descs(nandc);
if (ret) {
dev_err(nandc->dev, "failure to read raw cw %d\n", cw);
return ret;
}
return check_flash_errors(host, 1);
}
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
/*
* Bitflips can happen in erased codewords also so this function counts the
* number of 0 in each CW for which ECC engine returns the uncorrectable
* error. The page will be assumed as erased if this count is less than or
* equal to the ecc->strength for each CW.
*
* 1. Both DATA and OOB need to be checked for number of 0. The
* top-level API can be called with only data buf or OOB buf so use
* chip->data_buf if data buf is null and chip->oob_poi if oob buf
* is null for copying the raw bytes.
* 2. Perform raw read for all the CW which has uncorrectable errors.
* 3. For each CW, check the number of 0 in cw_data and usable OOB bytes.
* The BBM and spare bytes bit flip wont affect the ECC so dont check
* the number of bitflips in this area.
*/
static int
check_for_erased_page(struct qcom_nand_host *host, u8 *data_buf,
u8 *oob_buf, unsigned long uncorrectable_cws,
int page, unsigned int max_bitflips)
{
struct nand_chip *chip = &host->chip;
struct mtd_info *mtd = nand_to_mtd(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
u8 *cw_data_buf, *cw_oob_buf;
int cw, data_size, oob_size, ret = 0;
if (!data_buf)
data_buf = nand_get_data_buf(chip);
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
if (!oob_buf) {
nand_get_data_buf(chip);
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
oob_buf = chip->oob_poi;
}
for_each_set_bit(cw, &uncorrectable_cws, ecc->steps) {
if (cw == (ecc->steps - 1)) {
data_size = ecc->size - ((ecc->steps - 1) * 4);
oob_size = (ecc->steps * 4) + host->ecc_bytes_hw;
} else {
data_size = host->cw_data;
oob_size = host->ecc_bytes_hw;
}
/* determine starting buffer address for current CW */
cw_data_buf = data_buf + (cw * host->cw_data);
cw_oob_buf = oob_buf + (cw * ecc->bytes);
ret = qcom_nandc_read_cw_raw(mtd, chip, cw_data_buf,
cw_oob_buf, page, cw);
if (ret)
return ret;
/*
* make sure it isn't an erased page reported
* as not-erased by HW because of a few bitflips
*/
ret = nand_check_erased_ecc_chunk(cw_data_buf, data_size,
cw_oob_buf + host->bbm_size,
oob_size, NULL,
0, ecc->strength);
if (ret < 0) {
mtd->ecc_stats.failed++;
} else {
mtd->ecc_stats.corrected += ret;
max_bitflips = max_t(unsigned int, max_bitflips, ret);
}
}
return max_bitflips;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* reads back status registers set by the controller to notify page read
* errors. this is equivalent to what 'ecc->correct()' would do.
*/
static int parse_read_errors(struct qcom_nand_host *host, u8 *data_buf,
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
u8 *oob_buf, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct mtd_info *mtd = nand_to_mtd(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
unsigned int max_bitflips = 0, uncorrectable_cws = 0;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct read_stats *buf;
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
bool flash_op_err = false, erased;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int i;
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
u8 *data_buf_start = data_buf, *oob_buf_start = oob_buf;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
buf = (struct read_stats *)nandc->reg_read_buf;
nandc_read_buffer_sync(nandc, true);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
for (i = 0; i < ecc->steps; i++, buf++) {
u32 flash, buffer, erased_cw;
int data_len, oob_len;
if (i == (ecc->steps - 1)) {
data_len = ecc->size - ((ecc->steps - 1) << 2);
oob_len = ecc->steps << 2;
} else {
data_len = host->cw_data;
oob_len = 0;
}
flash = le32_to_cpu(buf->flash);
buffer = le32_to_cpu(buf->buffer);
erased_cw = le32_to_cpu(buf->erased_cw);
/*
* Check ECC failure for each codeword. ECC failure can
* happen in either of the following conditions
* 1. If number of bitflips are greater than ECC engine
* capability.
* 2. If this codeword contains all 0xff for which erased
* codeword detection check will be done.
*/
if ((flash & FS_OP_ERR) && (buffer & BS_UNCORRECTABLE_BIT)) {
/*
* For BCH ECC, ignore erased codeword errors, if
* ERASED_CW bits are set.
*/
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (host->bch_enabled) {
erased = (erased_cw & ERASED_CW) == ERASED_CW ?
true : false;
/*
* For RS ECC, HW reports the erased CW by placing
* special characters at certain offsets in the buffer.
* These special characters will be valid only if
* complete page is read i.e. data_buf is not NULL.
*/
} else if (data_buf) {
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
erased = erased_chunk_check_and_fixup(data_buf,
data_len);
} else {
erased = false;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
if (!erased)
uncorrectable_cws |= BIT(i);
/*
* Check if MPU or any other operational error (timeout,
* device failure, etc.) happened for this codeword and
* make flash_op_err true. If flash_op_err is set, then
* EIO will be returned for page read.
*/
} else if (flash & (FS_OP_ERR | FS_MPU_ERR)) {
flash_op_err = true;
/*
* No ECC or operational errors happened. Check the number of
* bits corrected and update the ecc_stats.corrected.
*/
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
} else {
unsigned int stat;
stat = buffer & BS_CORRECTABLE_ERR_MSK;
mtd->ecc_stats.corrected += stat;
max_bitflips = max(max_bitflips, stat);
}
if (data_buf)
data_buf += data_len;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (oob_buf)
oob_buf += oob_len + ecc->bytes;
}
if (flash_op_err)
return -EIO;
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
if (!uncorrectable_cws)
return max_bitflips;
return check_for_erased_page(host, data_buf_start, oob_buf_start,
uncorrectable_cws, page,
max_bitflips);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* helper to perform the actual page read operation, used by ecc->read_page(),
* ecc->read_oob()
*/
static int read_page_ecc(struct qcom_nand_host *host, u8 *data_buf,
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
u8 *oob_buf, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
u8 *data_buf_start = data_buf, *oob_buf_start = oob_buf;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int i, ret;
config_nand_page_read(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* queue cmd descs for each codeword */
for (i = 0; i < ecc->steps; i++) {
int data_size, oob_size;
if (i == (ecc->steps - 1)) {
data_size = ecc->size - ((ecc->steps - 1) << 2);
oob_size = (ecc->steps << 2) + host->ecc_bytes_hw +
host->spare_bytes;
} else {
data_size = host->cw_data;
oob_size = host->ecc_bytes_hw + host->spare_bytes;
}
if (nandc->props->is_bam) {
if (data_buf && oob_buf) {
nandc_set_read_loc(nandc, 0, 0, data_size, 0);
nandc_set_read_loc(nandc, 1, data_size,
oob_size, 1);
} else if (data_buf) {
nandc_set_read_loc(nandc, 0, 0, data_size, 1);
} else {
nandc_set_read_loc(nandc, 0, data_size,
oob_size, 1);
}
}
config_nand_cw_read(nandc, true);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (data_buf)
read_data_dma(nandc, FLASH_BUF_ACC, data_buf,
data_size, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* when ecc is enabled, the controller doesn't read the real
* or dummy bad block markers in each chunk. To maintain a
* consistent layout across RAW and ECC reads, we just
* leave the real/dummy BBM offsets empty (i.e, filled with
* 0xffs)
*/
if (oob_buf) {
int j;
for (j = 0; j < host->bbm_size; j++)
*oob_buf++ = 0xff;
read_data_dma(nandc, FLASH_BUF_ACC + data_size,
oob_buf, oob_size, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
if (data_buf)
data_buf += data_size;
if (oob_buf)
oob_buf += oob_size;
}
ret = submit_descs(nandc);
free_descs(nandc);
if (ret) {
dev_err(nandc->dev, "failure to read page/oob\n");
return ret;
}
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
return parse_read_errors(host, data_buf_start, oob_buf_start, page);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* a helper that copies the last step/codeword of a page (containing free oob)
* into our local buffer
*/
static int copy_last_cw(struct qcom_nand_host *host, int page)
{
struct nand_chip *chip = &host->chip;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int size;
int ret;
clear_read_regs(nandc);
size = host->use_ecc ? host->cw_data : host->cw_size;
/* prepare a clean read buffer */
memset(nandc->data_buffer, 0xff, size);
set_address(host, host->cw_size * (ecc->steps - 1), page);
update_rw_regs(host, 1, true);
config_nand_single_cw_page_read(nandc, host->use_ecc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
read_data_dma(nandc, FLASH_BUF_ACC, nandc->data_buffer, size, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = submit_descs(nandc);
if (ret)
dev_err(nandc->dev, "failed to copy last codeword\n");
free_descs(nandc);
return ret;
}
/* implements ecc->read_page() */
static int qcom_nandc_read_page(struct nand_chip *chip, uint8_t *buf,
int oob_required, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
u8 *data_buf, *oob_buf = NULL;
mtd: nand: force drivers to explicitly send READ/PROG commands The core currently send the READ0 and SEQIN+PAGEPROG commands in nand_do_read/write_ops(). This is inconsistent with ->read/write_oob[_raw]() hooks behavior which are expected to send these commands. There's already a flag (NAND_ECC_CUSTOM_PAGE_ACCESS) to inform the core that a specific controller wants to send the READ/SEQIN+PAGEPROG commands on its own, but it's an opt-in flag, and existing drivers are unlikely to be updated to pass it. Moreover, some controllers cannot dissociate the READ/PAGEPROG commands from the associated data transfer and ECC engine activation, and developers have to hack things in their ->cmdfunc() implementation to handle such complex cases, or have to accept the perf penalty of sending twice the same command. To address this problem we are planning on adding a new interface which is passed all information about a NAND operation (including the amount of data to transfer) and replacing all calls to ->cmdfunc() to calls to this new ->exec_op() hook. But, in order to do that, we need to have all ->cmdfunc() calls placed near their associated ->read/write_buf/byte() calls. Modify the core and relevant drivers to make NAND_ECC_CUSTOM_PAGE_ACCESS the default case, and remove this flag. Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> [miquel.raynal@free-electrons.com: tested, fixed and rebased on nand/next] Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2017-12-01 01:01:30 +08:00
nand_read_page_op(chip, page, 0, NULL, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
data_buf = buf;
oob_buf = oob_required ? chip->oob_poi : NULL;
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
return read_page_ecc(host, data_buf, oob_buf, page);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/* implements ecc->read_page_raw() */
static int qcom_nandc_read_page_raw(struct nand_chip *chip, uint8_t *buf,
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int oob_required, int page)
{
struct mtd_info *mtd = nand_to_mtd(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int cw, ret;
u8 *data_buf = buf, *oob_buf = chip->oob_poi;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
for (cw = 0; cw < ecc->steps; cw++) {
ret = qcom_nandc_read_cw_raw(mtd, chip, data_buf, oob_buf,
page, cw);
if (ret)
return ret;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
data_buf += host->cw_data;
oob_buf += ecc->bytes;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/* implements ecc->read_oob() */
static int qcom_nandc_read_oob(struct nand_chip *chip, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
clear_read_regs(nandc);
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
host->use_ecc = true;
set_address(host, 0, page);
update_rw_regs(host, ecc->steps, true);
mtd: rawnand: qcom: erased page bitflips detection NAND parts can have bitflips in an erased page due to the process technology used. In this case, QCOM NAND controller is not able to identify that page as an erased page. Currently the driver calls nand_check_erased_ecc_chunk() for identifying the erased pages but this won’t work always since the checking is being with ECC engine returned data. In case of bitflips, the ECC engine tries to correct the data and then it generates the uncorrectable error. Now, this data is not equal to original raw data. For erased CW identification, the raw data should be read again from NAND device and this nand_check_erased_ecc_chunk function() should be called for raw data only. Now following logic is being added to identify the erased codeword bitflips. 1. In most of the cases, not all the codewords will have bitflips and only single CW will have bitflips. So, there is no need to read the complete raw page data. The NAND raw read can be scheduled for any CW in page. The NAND controller works on CW basis and it will update the status register after each CW read. Maintain the bitmask for the CW which generated the uncorrectable error. 2. Do raw read for all the CW's which generated the uncorrectable error. 3. Both DATA and OOB need to be checked for number of 0. The top-level API can be called with only data buf or OOB buf so use chip->databuf if data buf is null and chip->oob_poi if OOB buf is null for copying the raw bytes temporarily. 4. For each CW, check the number of 0 in cw_data and usable oob bytes, The bbm and spare (unused) bytes bit flip won’t affect the ECC so don’t check the number of bitflips in this area. Signed-off-by: Abhishek Sahu <absahu@codeaurora.org> Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
2018-07-03 20:06:03 +08:00
return read_page_ecc(host, NULL, chip->oob_poi, page);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/* implements ecc->write_page() */
static int qcom_nandc_write_page(struct nand_chip *chip, const uint8_t *buf,
int oob_required, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
u8 *data_buf, *oob_buf;
int i, ret;
mtd: nand: force drivers to explicitly send READ/PROG commands The core currently send the READ0 and SEQIN+PAGEPROG commands in nand_do_read/write_ops(). This is inconsistent with ->read/write_oob[_raw]() hooks behavior which are expected to send these commands. There's already a flag (NAND_ECC_CUSTOM_PAGE_ACCESS) to inform the core that a specific controller wants to send the READ/SEQIN+PAGEPROG commands on its own, but it's an opt-in flag, and existing drivers are unlikely to be updated to pass it. Moreover, some controllers cannot dissociate the READ/PAGEPROG commands from the associated data transfer and ECC engine activation, and developers have to hack things in their ->cmdfunc() implementation to handle such complex cases, or have to accept the perf penalty of sending twice the same command. To address this problem we are planning on adding a new interface which is passed all information about a NAND operation (including the amount of data to transfer) and replacing all calls to ->cmdfunc() to calls to this new ->exec_op() hook. But, in order to do that, we need to have all ->cmdfunc() calls placed near their associated ->read/write_buf/byte() calls. Modify the core and relevant drivers to make NAND_ECC_CUSTOM_PAGE_ACCESS the default case, and remove this flag. Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> [miquel.raynal@free-electrons.com: tested, fixed and rebased on nand/next] Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2017-12-01 01:01:30 +08:00
nand_prog_page_begin_op(chip, page, 0, NULL, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
clear_read_regs(nandc);
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
data_buf = (u8 *)buf;
oob_buf = chip->oob_poi;
host->use_ecc = true;
update_rw_regs(host, ecc->steps, false);
config_nand_page_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
for (i = 0; i < ecc->steps; i++) {
int data_size, oob_size;
if (i == (ecc->steps - 1)) {
data_size = ecc->size - ((ecc->steps - 1) << 2);
oob_size = (ecc->steps << 2) + host->ecc_bytes_hw +
host->spare_bytes;
} else {
data_size = host->cw_data;
oob_size = ecc->bytes;
}
write_data_dma(nandc, FLASH_BUF_ACC, data_buf, data_size,
i == (ecc->steps - 1) ? NAND_BAM_NO_EOT : 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* when ECC is enabled, we don't really need to write anything
* to oob for the first n - 1 codewords since these oob regions
* just contain ECC bytes that's written by the controller
* itself. For the last codeword, we skip the bbm positions and
* write to the free oob area.
*/
if (i == (ecc->steps - 1)) {
oob_buf += host->bbm_size;
write_data_dma(nandc, FLASH_BUF_ACC + data_size,
oob_buf, oob_size, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
config_nand_cw_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
data_buf += data_size;
oob_buf += oob_size;
}
ret = submit_descs(nandc);
if (ret)
dev_err(nandc->dev, "failure to write page\n");
free_descs(nandc);
mtd: nand: force drivers to explicitly send READ/PROG commands The core currently send the READ0 and SEQIN+PAGEPROG commands in nand_do_read/write_ops(). This is inconsistent with ->read/write_oob[_raw]() hooks behavior which are expected to send these commands. There's already a flag (NAND_ECC_CUSTOM_PAGE_ACCESS) to inform the core that a specific controller wants to send the READ/SEQIN+PAGEPROG commands on its own, but it's an opt-in flag, and existing drivers are unlikely to be updated to pass it. Moreover, some controllers cannot dissociate the READ/PAGEPROG commands from the associated data transfer and ECC engine activation, and developers have to hack things in their ->cmdfunc() implementation to handle such complex cases, or have to accept the perf penalty of sending twice the same command. To address this problem we are planning on adding a new interface which is passed all information about a NAND operation (including the amount of data to transfer) and replacing all calls to ->cmdfunc() to calls to this new ->exec_op() hook. But, in order to do that, we need to have all ->cmdfunc() calls placed near their associated ->read/write_buf/byte() calls. Modify the core and relevant drivers to make NAND_ECC_CUSTOM_PAGE_ACCESS the default case, and remove this flag. Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> [miquel.raynal@free-electrons.com: tested, fixed and rebased on nand/next] Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2017-12-01 01:01:30 +08:00
if (!ret)
ret = nand_prog_page_end_op(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return ret;
}
/* implements ecc->write_page_raw() */
static int qcom_nandc_write_page_raw(struct nand_chip *chip,
const uint8_t *buf, int oob_required,
int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct mtd_info *mtd = nand_to_mtd(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
u8 *data_buf, *oob_buf;
int i, ret;
mtd: nand: force drivers to explicitly send READ/PROG commands The core currently send the READ0 and SEQIN+PAGEPROG commands in nand_do_read/write_ops(). This is inconsistent with ->read/write_oob[_raw]() hooks behavior which are expected to send these commands. There's already a flag (NAND_ECC_CUSTOM_PAGE_ACCESS) to inform the core that a specific controller wants to send the READ/SEQIN+PAGEPROG commands on its own, but it's an opt-in flag, and existing drivers are unlikely to be updated to pass it. Moreover, some controllers cannot dissociate the READ/PAGEPROG commands from the associated data transfer and ECC engine activation, and developers have to hack things in their ->cmdfunc() implementation to handle such complex cases, or have to accept the perf penalty of sending twice the same command. To address this problem we are planning on adding a new interface which is passed all information about a NAND operation (including the amount of data to transfer) and replacing all calls to ->cmdfunc() to calls to this new ->exec_op() hook. But, in order to do that, we need to have all ->cmdfunc() calls placed near their associated ->read/write_buf/byte() calls. Modify the core and relevant drivers to make NAND_ECC_CUSTOM_PAGE_ACCESS the default case, and remove this flag. Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> [miquel.raynal@free-electrons.com: tested, fixed and rebased on nand/next] Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2017-12-01 01:01:30 +08:00
nand_prog_page_begin_op(chip, page, 0, NULL, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
clear_read_regs(nandc);
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
data_buf = (u8 *)buf;
oob_buf = chip->oob_poi;
host->use_ecc = false;
update_rw_regs(host, ecc->steps, false);
config_nand_page_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
for (i = 0; i < ecc->steps; i++) {
int data_size1, data_size2, oob_size1, oob_size2;
int reg_off = FLASH_BUF_ACC;
data_size1 = mtd->writesize - host->cw_size * (ecc->steps - 1);
oob_size1 = host->bbm_size;
if (i == (ecc->steps - 1)) {
data_size2 = ecc->size - data_size1 -
((ecc->steps - 1) << 2);
oob_size2 = (ecc->steps << 2) + host->ecc_bytes_hw +
host->spare_bytes;
} else {
data_size2 = host->cw_data - data_size1;
oob_size2 = host->ecc_bytes_hw + host->spare_bytes;
}
write_data_dma(nandc, reg_off, data_buf, data_size1,
NAND_BAM_NO_EOT);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
reg_off += data_size1;
data_buf += data_size1;
write_data_dma(nandc, reg_off, oob_buf, oob_size1,
NAND_BAM_NO_EOT);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
reg_off += oob_size1;
oob_buf += oob_size1;
write_data_dma(nandc, reg_off, data_buf, data_size2,
NAND_BAM_NO_EOT);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
reg_off += data_size2;
data_buf += data_size2;
write_data_dma(nandc, reg_off, oob_buf, oob_size2, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
oob_buf += oob_size2;
config_nand_cw_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
ret = submit_descs(nandc);
if (ret)
dev_err(nandc->dev, "failure to write raw page\n");
free_descs(nandc);
mtd: nand: force drivers to explicitly send READ/PROG commands The core currently send the READ0 and SEQIN+PAGEPROG commands in nand_do_read/write_ops(). This is inconsistent with ->read/write_oob[_raw]() hooks behavior which are expected to send these commands. There's already a flag (NAND_ECC_CUSTOM_PAGE_ACCESS) to inform the core that a specific controller wants to send the READ/SEQIN+PAGEPROG commands on its own, but it's an opt-in flag, and existing drivers are unlikely to be updated to pass it. Moreover, some controllers cannot dissociate the READ/PAGEPROG commands from the associated data transfer and ECC engine activation, and developers have to hack things in their ->cmdfunc() implementation to handle such complex cases, or have to accept the perf penalty of sending twice the same command. To address this problem we are planning on adding a new interface which is passed all information about a NAND operation (including the amount of data to transfer) and replacing all calls to ->cmdfunc() to calls to this new ->exec_op() hook. But, in order to do that, we need to have all ->cmdfunc() calls placed near their associated ->read/write_buf/byte() calls. Modify the core and relevant drivers to make NAND_ECC_CUSTOM_PAGE_ACCESS the default case, and remove this flag. Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com> [miquel.raynal@free-electrons.com: tested, fixed and rebased on nand/next] Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2017-12-01 01:01:30 +08:00
if (!ret)
ret = nand_prog_page_end_op(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return ret;
}
/*
* implements ecc->write_oob()
*
* the NAND controller cannot write only data or only OOB within a codeword
* since ECC is calculated for the combined codeword. So update the OOB from
* chip->oob_poi, and pad the data area with OxFF before writing.
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int qcom_nandc_write_oob(struct nand_chip *chip, int page)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct mtd_info *mtd = nand_to_mtd(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
u8 *oob = chip->oob_poi;
int data_size, oob_size;
int ret;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
host->use_ecc = true;
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* calculate the data and oob size for the last codeword/step */
data_size = ecc->size - ((ecc->steps - 1) << 2);
oob_size = mtd->oobavail;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
memset(nandc->data_buffer, 0xff, host->cw_data);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* override new oob content to last codeword */
mtd_ooblayout_get_databytes(mtd, nandc->data_buffer + data_size, oob,
0, mtd->oobavail);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
set_address(host, host->cw_size * (ecc->steps - 1), page);
update_rw_regs(host, 1, false);
config_nand_page_write(nandc);
write_data_dma(nandc, FLASH_BUF_ACC,
nandc->data_buffer, data_size + oob_size, 0);
config_nand_cw_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = submit_descs(nandc);
free_descs(nandc);
if (ret) {
dev_err(nandc->dev, "failure to write oob\n");
return -EIO;
}
return nand_prog_page_end_op(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
static int qcom_nandc_block_bad(struct nand_chip *chip, loff_t ofs)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct mtd_info *mtd = nand_to_mtd(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int page, ret, bbpos, bad = 0;
page = (int)(ofs >> chip->page_shift) & chip->pagemask;
/*
* configure registers for a raw sub page read, the address is set to
* the beginning of the last codeword, we don't care about reading ecc
* portion of oob. we just want the first few bytes from this codeword
* that contains the BBM
*/
host->use_ecc = false;
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = copy_last_cw(host, page);
if (ret)
goto err;
if (check_flash_errors(host, 1)) {
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
dev_warn(nandc->dev, "error when trying to read BBM\n");
goto err;
}
bbpos = mtd->writesize - host->cw_size * (ecc->steps - 1);
bad = nandc->data_buffer[bbpos] != 0xff;
if (chip->options & NAND_BUSWIDTH_16)
bad = bad || (nandc->data_buffer[bbpos + 1] != 0xff);
err:
return bad;
}
static int qcom_nandc_block_markbad(struct nand_chip *chip, loff_t ofs)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
int page, ret;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
clear_read_regs(nandc);
clear_bam_transaction(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* to mark the BBM as bad, we flash the entire last codeword with 0s.
* we don't care about the rest of the content in the codeword since
* we aren't going to use this block again
*/
memset(nandc->data_buffer, 0x00, host->cw_size);
page = (int)(ofs >> chip->page_shift) & chip->pagemask;
/* prepare write */
host->use_ecc = false;
set_address(host, host->cw_size * (ecc->steps - 1), page);
update_rw_regs(host, 1, false);
config_nand_page_write(nandc);
write_data_dma(nandc, FLASH_BUF_ACC,
nandc->data_buffer, host->cw_size, 0);
config_nand_cw_write(nandc);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = submit_descs(nandc);
free_descs(nandc);
if (ret) {
dev_err(nandc->dev, "failure to update BBM\n");
return -EIO;
}
return nand_prog_page_end_op(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/*
* the three functions below implement chip->legacy.read_byte(),
* chip->legacy.read_buf() and chip->legacy.write_buf() respectively. these
* aren't used for reading/writing page data, they are used for smaller data
* like reading id, status etc
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static uint8_t qcom_nandc_read_byte(struct nand_chip *chip)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
u8 *buf = nandc->data_buffer;
u8 ret = 0x0;
if (host->last_command == NAND_CMD_STATUS) {
ret = host->status;
host->status = NAND_STATUS_READY | NAND_STATUS_WP;
return ret;
}
if (nandc->buf_start < nandc->buf_count)
ret = buf[nandc->buf_start++];
return ret;
}
static void qcom_nandc_read_buf(struct nand_chip *chip, uint8_t *buf, int len)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
int real_len = min_t(size_t, len, nandc->buf_count - nandc->buf_start);
memcpy(buf, nandc->data_buffer + nandc->buf_start, real_len);
nandc->buf_start += real_len;
}
static void qcom_nandc_write_buf(struct nand_chip *chip, const uint8_t *buf,
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
int len)
{
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
int real_len = min_t(size_t, len, nandc->buf_count - nandc->buf_start);
memcpy(nandc->data_buffer + nandc->buf_start, buf, real_len);
nandc->buf_start += real_len;
}
/* we support only one external chip for now */
static void qcom_nandc_select_chip(struct nand_chip *chip, int chipnr)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
if (chipnr <= 0)
return;
dev_warn(nandc->dev, "invalid chip select\n");
}
/*
* NAND controller page layout info
*
* Layout with ECC enabled:
*
* |----------------------| |---------------------------------|
* | xx.......yy| | *********xx.......yy|
* | DATA xx..ECC..yy| | DATA **SPARE**xx..ECC..yy|
* | (516) xx.......yy| | (516-n*4) **(n*4)**xx.......yy|
* | xx.......yy| | *********xx.......yy|
* |----------------------| |---------------------------------|
* codeword 1,2..n-1 codeword n
* <---(528/532 Bytes)--> <-------(528/532 Bytes)--------->
*
* n = Number of codewords in the page
* . = ECC bytes
* * = Spare/free bytes
* x = Unused byte(s)
* y = Reserved byte(s)
*
* 2K page: n = 4, spare = 16 bytes
* 4K page: n = 8, spare = 32 bytes
* 8K page: n = 16, spare = 64 bytes
*
* the qcom nand controller operates at a sub page/codeword level. each
* codeword is 528 and 532 bytes for 4 bit and 8 bit ECC modes respectively.
* the number of ECC bytes vary based on the ECC strength and the bus width.
*
* the first n - 1 codewords contains 516 bytes of user data, the remaining
* 12/16 bytes consist of ECC and reserved data. The nth codeword contains
* both user data and spare(oobavail) bytes that sum up to 516 bytes.
*
* When we access a page with ECC enabled, the reserved bytes(s) are not
* accessible at all. When reading, we fill up these unreadable positions
* with 0xffs. When writing, the controller skips writing the inaccessible
* bytes.
*
* Layout with ECC disabled:
*
* |------------------------------| |---------------------------------------|
* | yy xx.......| | bb *********xx.......|
* | DATA1 yy DATA2 xx..ECC..| | DATA1 bb DATA2 **SPARE**xx..ECC..|
* | (size1) yy (size2) xx.......| | (size1) bb (size2) **(n*4)**xx.......|
* | yy xx.......| | bb *********xx.......|
* |------------------------------| |---------------------------------------|
* codeword 1,2..n-1 codeword n
* <-------(528/532 Bytes)------> <-----------(528/532 Bytes)----------->
*
* n = Number of codewords in the page
* . = ECC bytes
* * = Spare/free bytes
* x = Unused byte(s)
* y = Dummy Bad Bock byte(s)
* b = Real Bad Block byte(s)
* size1/size2 = function of codeword size and 'n'
*
* when the ECC block is disabled, one reserved byte (or two for 16 bit bus
* width) is now accessible. For the first n - 1 codewords, these are dummy Bad
* Block Markers. In the last codeword, this position contains the real BBM
*
* In order to have a consistent layout between RAW and ECC modes, we assume
* the following OOB layout arrangement:
*
* |-----------| |--------------------|
* |yyxx.......| |bb*********xx.......|
* |yyxx..ECC..| |bb*FREEOOB*xx..ECC..|
* |yyxx.......| |bb*********xx.......|
* |yyxx.......| |bb*********xx.......|
* |-----------| |--------------------|
* first n - 1 nth OOB region
* OOB regions
*
* n = Number of codewords in the page
* . = ECC bytes
* * = FREE OOB bytes
* y = Dummy bad block byte(s) (inaccessible when ECC enabled)
* x = Unused byte(s)
* b = Real bad block byte(s) (inaccessible when ECC enabled)
*
* This layout is read as is when ECC is disabled. When ECC is enabled, the
* inaccessible Bad Block byte(s) are ignored when we write to a page/oob,
* and assumed as 0xffs when we read a page/oob. The ECC, unused and
* dummy/real bad block bytes are grouped as ecc bytes (i.e, ecc->bytes is
* the sum of the three).
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
*/
static int qcom_nand_ooblayout_ecc(struct mtd_info *mtd, int section,
struct mtd_oob_region *oobregion)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct nand_chip *chip = mtd_to_nand(mtd);
struct qcom_nand_host *host = to_qcom_nand_host(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct nand_ecc_ctrl *ecc = &chip->ecc;
if (section > 1)
return -ERANGE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (!section) {
oobregion->length = (ecc->bytes * (ecc->steps - 1)) +
host->bbm_size;
oobregion->offset = 0;
} else {
oobregion->length = host->ecc_bytes_hw + host->spare_bytes;
oobregion->offset = mtd->oobsize - oobregion->length;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
return 0;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
static int qcom_nand_ooblayout_free(struct mtd_info *mtd, int section,
struct mtd_oob_region *oobregion)
{
struct nand_chip *chip = mtd_to_nand(mtd);
struct qcom_nand_host *host = to_qcom_nand_host(chip);
struct nand_ecc_ctrl *ecc = &chip->ecc;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (section)
return -ERANGE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
oobregion->length = ecc->steps * 4;
oobregion->offset = ((ecc->steps - 1) * ecc->bytes) + host->bbm_size;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
static const struct mtd_ooblayout_ops qcom_nand_ooblayout_ops = {
.ecc = qcom_nand_ooblayout_ecc,
.free = qcom_nand_ooblayout_free,
};
static int
qcom_nandc_calc_ecc_bytes(int step_size, int strength)
{
return strength == 4 ? 12 : 16;
}
NAND_ECC_CAPS_SINGLE(qcom_nandc_ecc_caps, qcom_nandc_calc_ecc_bytes,
NANDC_STEP_SIZE, 4, 8);
static int qcom_nand_attach_chip(struct nand_chip *chip)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct mtd_info *mtd = nand_to_mtd(chip);
struct qcom_nand_host *host = to_qcom_nand_host(chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct nand_ecc_ctrl *ecc = &chip->ecc;
struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
int cwperpage, bad_block_byte, ret;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
bool wide_bus;
int ecc_mode = 1;
/* controller only supports 512 bytes data steps */
ecc->size = NANDC_STEP_SIZE;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
wide_bus = chip->options & NAND_BUSWIDTH_16 ? true : false;
cwperpage = mtd->writesize / NANDC_STEP_SIZE;
/*
* Each CW has 4 available OOB bytes which will be protected with ECC
* so remaining bytes can be used for ECC.
*/
ret = nand_ecc_choose_conf(chip, &qcom_nandc_ecc_caps,
mtd->oobsize - (cwperpage * 4));
if (ret) {
dev_err(nandc->dev, "No valid ECC settings possible\n");
return ret;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (ecc->strength >= 8) {
/* 8 bit ECC defaults to BCH ECC on all platforms */
host->bch_enabled = true;
ecc_mode = 1;
if (wide_bus) {
host->ecc_bytes_hw = 14;
host->spare_bytes = 0;
host->bbm_size = 2;
} else {
host->ecc_bytes_hw = 13;
host->spare_bytes = 2;
host->bbm_size = 1;
}
} else {
/*
* if the controller supports BCH for 4 bit ECC, the controller
* uses lesser bytes for ECC. If RS is used, the ECC bytes is
* always 10 bytes
*/
if (nandc->props->ecc_modes & ECC_BCH_4BIT) {
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* BCH */
host->bch_enabled = true;
ecc_mode = 0;
if (wide_bus) {
host->ecc_bytes_hw = 8;
host->spare_bytes = 2;
host->bbm_size = 2;
} else {
host->ecc_bytes_hw = 7;
host->spare_bytes = 4;
host->bbm_size = 1;
}
} else {
/* RS */
host->ecc_bytes_hw = 10;
if (wide_bus) {
host->spare_bytes = 0;
host->bbm_size = 2;
} else {
host->spare_bytes = 1;
host->bbm_size = 1;
}
}
}
/*
* we consider ecc->bytes as the sum of all the non-data content in a
* step. It gives us a clean representation of the oob area (even if
* all the bytes aren't used for ECC).It is always 16 bytes for 8 bit
* ECC and 12 bytes for 4 bit ECC
*/
ecc->bytes = host->ecc_bytes_hw + host->spare_bytes + host->bbm_size;
ecc->read_page = qcom_nandc_read_page;
ecc->read_page_raw = qcom_nandc_read_page_raw;
ecc->read_oob = qcom_nandc_read_oob;
ecc->write_page = qcom_nandc_write_page;
ecc->write_page_raw = qcom_nandc_write_page_raw;
ecc->write_oob = qcom_nandc_write_oob;
ecc->mode = NAND_ECC_HW;
mtd_set_ooblayout(mtd, &qcom_nand_ooblayout_ops);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc->max_cwperpage = max_t(unsigned int, nandc->max_cwperpage,
cwperpage);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* DATA_UD_BYTES varies based on whether the read/write command protects
* spare data with ECC too. We protect spare data by default, so we set
* it to main + spare data, which are 512 and 4 bytes respectively.
*/
host->cw_data = 516;
/*
* total bytes in a step, either 528 bytes for 4 bit ECC, or 532 bytes
* for 8 bit ECC
*/
host->cw_size = host->cw_data + ecc->bytes;
bad_block_byte = mtd->writesize - host->cw_size * (cwperpage - 1) + 1;
host->cfg0 = (cwperpage - 1) << CW_PER_PAGE
| host->cw_data << UD_SIZE_BYTES
| 0 << DISABLE_STATUS_AFTER_WRITE
| 5 << NUM_ADDR_CYCLES
| host->ecc_bytes_hw << ECC_PARITY_SIZE_BYTES_RS
| 0 << STATUS_BFR_READ
| 1 << SET_RD_MODE_AFTER_STATUS
| host->spare_bytes << SPARE_SIZE_BYTES;
host->cfg1 = 7 << NAND_RECOVERY_CYCLES
| 0 << CS_ACTIVE_BSY
| bad_block_byte << BAD_BLOCK_BYTE_NUM
| 0 << BAD_BLOCK_IN_SPARE_AREA
| 2 << WR_RD_BSY_GAP
| wide_bus << WIDE_FLASH
| host->bch_enabled << ENABLE_BCH_ECC;
host->cfg0_raw = (cwperpage - 1) << CW_PER_PAGE
| host->cw_size << UD_SIZE_BYTES
| 5 << NUM_ADDR_CYCLES
| 0 << SPARE_SIZE_BYTES;
host->cfg1_raw = 7 << NAND_RECOVERY_CYCLES
| 0 << CS_ACTIVE_BSY
| 17 << BAD_BLOCK_BYTE_NUM
| 1 << BAD_BLOCK_IN_SPARE_AREA
| 2 << WR_RD_BSY_GAP
| wide_bus << WIDE_FLASH
| 1 << DEV0_CFG1_ECC_DISABLE;
host->ecc_bch_cfg = !host->bch_enabled << ECC_CFG_ECC_DISABLE
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
| 0 << ECC_SW_RESET
| host->cw_data << ECC_NUM_DATA_BYTES
| 1 << ECC_FORCE_CLK_OPEN
| ecc_mode << ECC_MODE
| host->ecc_bytes_hw << ECC_PARITY_SIZE_BYTES_BCH;
host->ecc_buf_cfg = 0x203 << NUM_STEPS;
host->clrflashstatus = FS_READY_BSY_N;
host->clrreadstatus = 0xc0;
nandc->regs->erased_cw_detect_cfg_clr =
cpu_to_le32(CLR_ERASED_PAGE_DET);
nandc->regs->erased_cw_detect_cfg_set =
cpu_to_le32(SET_ERASED_PAGE_DET);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
dev_dbg(nandc->dev,
"cfg0 %x cfg1 %x ecc_buf_cfg %x ecc_bch cfg %x cw_size %d cw_data %d strength %d parity_bytes %d steps %d\n",
host->cfg0, host->cfg1, host->ecc_buf_cfg, host->ecc_bch_cfg,
host->cw_size, host->cw_data, ecc->strength, ecc->bytes,
cwperpage);
return 0;
}
static const struct nand_controller_ops qcom_nandc_ops = {
.attach_chip = qcom_nand_attach_chip,
};
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
static int qcom_nandc_alloc(struct qcom_nand_controller *nandc)
{
int ret;
ret = dma_set_coherent_mask(nandc->dev, DMA_BIT_MASK(32));
if (ret) {
dev_err(nandc->dev, "failed to set DMA mask\n");
return ret;
}
/*
* we use the internal buffer for reading ONFI params, reading small
* data like ID and status, and preforming read-copy-write operations
* when writing to a codeword partially. 532 is the maximum possible
* size of a codeword for our nand controller
*/
nandc->buf_size = 532;
nandc->data_buffer = devm_kzalloc(nandc->dev, nandc->buf_size,
GFP_KERNEL);
if (!nandc->data_buffer)
return -ENOMEM;
nandc->regs = devm_kzalloc(nandc->dev, sizeof(*nandc->regs),
GFP_KERNEL);
if (!nandc->regs)
return -ENOMEM;
treewide: devm_kzalloc() -> devm_kcalloc() The devm_kzalloc() function has a 2-factor argument form, devm_kcalloc(). This patch replaces cases of: devm_kzalloc(handle, a * b, gfp) with: devm_kcalloc(handle, a * b, gfp) as well as handling cases of: devm_kzalloc(handle, a * b * c, gfp) with: devm_kzalloc(handle, array3_size(a, b, c), gfp) as it's slightly less ugly than: devm_kcalloc(handle, array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: devm_kzalloc(handle, 4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. Some manual whitespace fixes were needed in this patch, as Coccinelle really liked to write "=devm_kcalloc..." instead of "= devm_kcalloc...". The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ expression HANDLE; type TYPE; expression THING, E; @@ ( devm_kzalloc(HANDLE, - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | devm_kzalloc(HANDLE, - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression HANDLE; expression COUNT; typedef u8; typedef __u8; @@ ( devm_kzalloc(HANDLE, - sizeof(u8) * (COUNT) + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(__u8) * (COUNT) + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(char) * (COUNT) + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(unsigned char) * (COUNT) + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(u8) * COUNT + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(__u8) * COUNT + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(char) * COUNT + COUNT , ...) | devm_kzalloc(HANDLE, - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ expression HANDLE; type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ expression HANDLE; identifier SIZE, COUNT; @@ - devm_kzalloc + devm_kcalloc (HANDLE, - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression HANDLE; expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( devm_kzalloc(HANDLE, - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression HANDLE; expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( devm_kzalloc(HANDLE, - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | devm_kzalloc(HANDLE, - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | devm_kzalloc(HANDLE, - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ expression HANDLE; identifier STRIDE, SIZE, COUNT; @@ ( devm_kzalloc(HANDLE, - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | devm_kzalloc(HANDLE, - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression HANDLE; expression E1, E2, E3; constant C1, C2, C3; @@ ( devm_kzalloc(HANDLE, C1 * C2 * C3, ...) | devm_kzalloc(HANDLE, - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | devm_kzalloc(HANDLE, - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | devm_kzalloc(HANDLE, - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | devm_kzalloc(HANDLE, - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression HANDLE; expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( devm_kzalloc(HANDLE, sizeof(THING) * C2, ...) | devm_kzalloc(HANDLE, sizeof(TYPE) * C2, ...) | devm_kzalloc(HANDLE, C1 * C2 * C3, ...) | devm_kzalloc(HANDLE, C1 * C2, ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - (E1) * E2 + E1, E2 , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - (E1) * (E2) + E1, E2 , ...) | - devm_kzalloc + devm_kcalloc (HANDLE, - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 05:07:58 +08:00
nandc->reg_read_buf = devm_kcalloc(nandc->dev,
MAX_REG_RD, sizeof(*nandc->reg_read_buf),
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
GFP_KERNEL);
if (!nandc->reg_read_buf)
return -ENOMEM;
if (nandc->props->is_bam) {
nandc->reg_read_dma =
dma_map_single(nandc->dev, nandc->reg_read_buf,
MAX_REG_RD *
sizeof(*nandc->reg_read_buf),
DMA_FROM_DEVICE);
if (dma_mapping_error(nandc->dev, nandc->reg_read_dma)) {
dev_err(nandc->dev, "failed to DMA MAP reg buffer\n");
return -EIO;
}
nandc->tx_chan = dma_request_slave_channel(nandc->dev, "tx");
if (!nandc->tx_chan) {
dev_err(nandc->dev, "failed to request tx channel\n");
return -ENODEV;
}
nandc->rx_chan = dma_request_slave_channel(nandc->dev, "rx");
if (!nandc->rx_chan) {
dev_err(nandc->dev, "failed to request rx channel\n");
return -ENODEV;
}
nandc->cmd_chan = dma_request_slave_channel(nandc->dev, "cmd");
if (!nandc->cmd_chan) {
dev_err(nandc->dev, "failed to request cmd channel\n");
return -ENODEV;
}
/*
* Initially allocate BAM transaction to read ONFI param page.
* After detecting all the devices, this BAM transaction will
* be freed and the next BAM tranasction will be allocated with
* maximum codeword size
*/
nandc->max_cwperpage = 1;
nandc->bam_txn = alloc_bam_transaction(nandc);
if (!nandc->bam_txn) {
dev_err(nandc->dev,
"failed to allocate bam transaction\n");
return -ENOMEM;
}
} else {
nandc->chan = dma_request_slave_channel(nandc->dev, "rxtx");
if (!nandc->chan) {
dev_err(nandc->dev,
"failed to request slave channel\n");
return -ENODEV;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
INIT_LIST_HEAD(&nandc->desc_list);
INIT_LIST_HEAD(&nandc->host_list);
nand_controller_init(&nandc->controller);
nandc->controller.ops = &qcom_nandc_ops;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
static void qcom_nandc_unalloc(struct qcom_nand_controller *nandc)
{
if (nandc->props->is_bam) {
if (!dma_mapping_error(nandc->dev, nandc->reg_read_dma))
dma_unmap_single(nandc->dev, nandc->reg_read_dma,
MAX_REG_RD *
sizeof(*nandc->reg_read_buf),
DMA_FROM_DEVICE);
if (nandc->tx_chan)
dma_release_channel(nandc->tx_chan);
if (nandc->rx_chan)
dma_release_channel(nandc->rx_chan);
if (nandc->cmd_chan)
dma_release_channel(nandc->cmd_chan);
} else {
if (nandc->chan)
dma_release_channel(nandc->chan);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/* one time setup of a few nand controller registers */
static int qcom_nandc_setup(struct qcom_nand_controller *nandc)
{
u32 nand_ctrl;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* kill onenand */
nandc_write(nandc, SFLASHC_BURST_CFG, 0);
nandc_write(nandc, dev_cmd_reg_addr(nandc, NAND_DEV_CMD_VLD),
NAND_DEV_CMD_VLD_VAL);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* enable ADM or BAM DMA */
if (nandc->props->is_bam) {
nand_ctrl = nandc_read(nandc, NAND_CTRL);
nandc_write(nandc, NAND_CTRL, nand_ctrl | BAM_MODE_EN);
} else {
nandc_write(nandc, NAND_FLASH_CHIP_SELECT, DM_EN);
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/* save the original values of these registers */
nandc->cmd1 = nandc_read(nandc, dev_cmd_reg_addr(nandc, NAND_DEV_CMD1));
nandc->vld = NAND_DEV_CMD_VLD_VAL;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
static int qcom_nand_host_init_and_register(struct qcom_nand_controller *nandc,
struct qcom_nand_host *host,
struct device_node *dn)
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{
struct nand_chip *chip = &host->chip;
struct mtd_info *mtd = nand_to_mtd(chip);
struct device *dev = nandc->dev;
int ret;
ret = of_property_read_u32(dn, "reg", &host->cs);
if (ret) {
dev_err(dev, "can't get chip-select\n");
return -ENXIO;
}
nand_set_flash_node(chip, dn);
mtd->name = devm_kasprintf(dev, GFP_KERNEL, "qcom_nand.%d", host->cs);
if (!mtd->name)
return -ENOMEM;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
mtd->owner = THIS_MODULE;
mtd->dev.parent = dev;
chip->legacy.cmdfunc = qcom_nandc_command;
chip->legacy.select_chip = qcom_nandc_select_chip;
chip->legacy.read_byte = qcom_nandc_read_byte;
chip->legacy.read_buf = qcom_nandc_read_buf;
chip->legacy.write_buf = qcom_nandc_write_buf;
chip->legacy.set_features = nand_get_set_features_notsupp;
chip->legacy.get_features = nand_get_set_features_notsupp;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* the bad block marker is readable only when we read the last codeword
* of a page with ECC disabled. currently, the nand_base and nand_bbt
* helpers don't allow us to read BB from a nand chip with ECC
* disabled (MTD_OPS_PLACE_OOB is set by default). use the block_bad
* and block_markbad helpers until we permanently switch to using
* MTD_OPS_RAW for all drivers (with the help of badblockbits)
*/
chip->legacy.block_bad = qcom_nandc_block_bad;
chip->legacy.block_markbad = qcom_nandc_block_markbad;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
chip->controller = &nandc->controller;
chip->options |= NAND_NO_SUBPAGE_WRITE | NAND_USE_BOUNCE_BUFFER |
NAND_SKIP_BBTSCAN;
/* set up initial status value */
host->status = NAND_STATUS_READY | NAND_STATUS_WP;
ret = nand_scan(chip, 1);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
if (ret)
return ret;
mtd: rawnand: qcom: fix memory corruption that causes panic This patch fixes a memory corruption that occurred in the qcom-nandc driver since it was converted to nand_scan(). On boot, an affected device will panic from a NPE at a weird place: | Unable to handle kernel NULL pointer dereference at virtual address 0 | pgd = (ptrval) | [00000000] *pgd=00000000 | Internal error: Oops: 80000005 [#1] SMP ARM | CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.19.9 #0 | Hardware name: Generic DT based system | PC is at (null) | LR is at nand_block_isbad+0x90/0xa4 | pc : [<00000000>] lr : [<c0592240>] psr: 80000013 | sp : cf839d40 ip : 00000000 fp : cfae9e20 | r10: cf815810 r9 : 00000000 r8 : 00000000 | r7 : 00000000 r6 : 00000000 r5 : 00000001 r4 : cf815810 | r3 : 00000000 r2 : cfae9810 r1 : ffffffff r0 : cf815810 | Flags: Nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none | Control: 10c5387d Table: 8020406a DAC: 00000051 | Process swapper/0 (pid: 1, stack limit = 0x(ptrval)) | [<c0592240>] (nand_block_isbad) from [<c0580a94>] | [<c0580a94>] (allocate_partition) from [<c05811e4>] | [<c05811e4>] (add_mtd_partitions) from [<c0581164>] | [<c0581164>] (parse_mtd_partitions) from [<c057def4>] | [<c057def4>] (mtd_device_parse_register) from [<c059d274>] | [<c059d274>] (qcom_nandc_probe) from [<c0567f00>] The problem is that the nand_scan()'s qcom_nand_attach_chip callback is updating the nandc->max_cwperpage from 1 to 4. This causes the sg_init_table of clear_bam_transaction() in the driver's qcom_nandc_block_bad() to memset much more than what was initially allocated by alloc_bam_transaction(). This patch restores the old behavior by reallocating the shared bam transaction alloc_bam_transaction() after the chip was identified, but before mtd_device_parse_register() (which is an alias for mtd_device_register() - see panic) gets called. This fixes the corruption and the driver is working again. Cc: stable@vger.kernel.org Fixes: 6a3cec64f18c ("mtd: rawnand: qcom: convert driver to nand_scan()") Signed-off-by: Christian Lamparter <chunkeey@gmail.com> Acked-by: Miquel Raynal <miquel.raynal@bootlin.com> Signed-off-by: Boris Brezillon <bbrezillon@kernel.org>
2018-12-23 08:31:26 +08:00
if (nandc->props->is_bam) {
free_bam_transaction(nandc);
nandc->bam_txn = alloc_bam_transaction(nandc);
if (!nandc->bam_txn) {
dev_err(nandc->dev,
"failed to allocate bam transaction\n");
return -ENOMEM;
}
}
ret = mtd_device_register(mtd, NULL, 0);
if (ret)
nand_cleanup(chip);
return ret;
}
static int qcom_probe_nand_devices(struct qcom_nand_controller *nandc)
{
struct device *dev = nandc->dev;
struct device_node *dn = dev->of_node, *child;
struct qcom_nand_host *host;
int ret;
for_each_available_child_of_node(dn, child) {
host = devm_kzalloc(dev, sizeof(*host), GFP_KERNEL);
if (!host) {
of_node_put(child);
return -ENOMEM;
}
ret = qcom_nand_host_init_and_register(nandc, host, child);
if (ret) {
devm_kfree(dev, host);
continue;
}
list_add_tail(&host->node, &nandc->host_list);
}
if (list_empty(&nandc->host_list))
return -ENODEV;
return 0;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
/* parse custom DT properties here */
static int qcom_nandc_parse_dt(struct platform_device *pdev)
{
struct qcom_nand_controller *nandc = platform_get_drvdata(pdev);
struct device_node *np = nandc->dev->of_node;
int ret;
if (!nandc->props->is_bam) {
ret = of_property_read_u32(np, "qcom,cmd-crci",
&nandc->cmd_crci);
if (ret) {
dev_err(nandc->dev, "command CRCI unspecified\n");
return ret;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = of_property_read_u32(np, "qcom,data-crci",
&nandc->data_crci);
if (ret) {
dev_err(nandc->dev, "data CRCI unspecified\n");
return ret;
}
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
}
return 0;
}
static int qcom_nandc_probe(struct platform_device *pdev)
{
struct qcom_nand_controller *nandc;
const void *dev_data;
struct device *dev = &pdev->dev;
struct resource *res;
int ret;
nandc = devm_kzalloc(&pdev->dev, sizeof(*nandc), GFP_KERNEL);
if (!nandc)
return -ENOMEM;
platform_set_drvdata(pdev, nandc);
nandc->dev = dev;
dev_data = of_device_get_match_data(dev);
if (!dev_data) {
dev_err(&pdev->dev, "failed to get device data\n");
return -ENODEV;
}
nandc->props = dev_data;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
nandc->core_clk = devm_clk_get(dev, "core");
if (IS_ERR(nandc->core_clk))
return PTR_ERR(nandc->core_clk);
nandc->aon_clk = devm_clk_get(dev, "aon");
if (IS_ERR(nandc->aon_clk))
return PTR_ERR(nandc->aon_clk);
ret = qcom_nandc_parse_dt(pdev);
if (ret)
return ret;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
nandc->base = devm_ioremap_resource(dev, res);
if (IS_ERR(nandc->base))
return PTR_ERR(nandc->base);
nandc->base_phys = res->start;
nandc->base_dma = dma_map_resource(dev, res->start,
resource_size(res),
DMA_BIDIRECTIONAL, 0);
if (!nandc->base_dma)
return -ENXIO;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = qcom_nandc_alloc(nandc);
if (ret)
goto err_nandc_alloc;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
ret = clk_prepare_enable(nandc->core_clk);
if (ret)
goto err_core_clk;
ret = clk_prepare_enable(nandc->aon_clk);
if (ret)
goto err_aon_clk;
ret = qcom_nandc_setup(nandc);
if (ret)
goto err_setup;
ret = qcom_probe_nand_devices(nandc);
if (ret)
goto err_setup;
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
err_setup:
clk_disable_unprepare(nandc->aon_clk);
err_aon_clk:
clk_disable_unprepare(nandc->core_clk);
err_core_clk:
qcom_nandc_unalloc(nandc);
err_nandc_alloc:
dma_unmap_resource(dev, res->start, resource_size(res),
DMA_BIDIRECTIONAL, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return ret;
}
static int qcom_nandc_remove(struct platform_device *pdev)
{
struct qcom_nand_controller *nandc = platform_get_drvdata(pdev);
struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
struct qcom_nand_host *host;
list_for_each_entry(host, &nandc->host_list, node)
nand_release(&host->chip);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
qcom_nandc_unalloc(nandc);
clk_disable_unprepare(nandc->aon_clk);
clk_disable_unprepare(nandc->core_clk);
dma_unmap_resource(&pdev->dev, nandc->base_dma, resource_size(res),
DMA_BIDIRECTIONAL, 0);
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
return 0;
}
static const struct qcom_nandc_props ipq806x_nandc_props = {
.ecc_modes = (ECC_RS_4BIT | ECC_BCH_8BIT),
.is_bam = false,
.dev_cmd_reg_start = 0x0,
};
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
static const struct qcom_nandc_props ipq4019_nandc_props = {
.ecc_modes = (ECC_BCH_4BIT | ECC_BCH_8BIT),
.is_bam = true,
.dev_cmd_reg_start = 0x0,
};
static const struct qcom_nandc_props ipq8074_nandc_props = {
.ecc_modes = (ECC_BCH_4BIT | ECC_BCH_8BIT),
.is_bam = true,
.dev_cmd_reg_start = 0x7000,
};
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
/*
* data will hold a struct pointer containing more differences once we support
* more controller variants
*/
static const struct of_device_id qcom_nandc_of_match[] = {
{
.compatible = "qcom,ipq806x-nand",
.data = &ipq806x_nandc_props,
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
},
{
.compatible = "qcom,ipq4019-nand",
.data = &ipq4019_nandc_props,
},
{
.compatible = "qcom,ipq8074-nand",
.data = &ipq8074_nandc_props,
},
mtd: nand: Qualcomm NAND controller driver The Qualcomm NAND controller is found in SoCs like IPQ806x, MSM7xx, MDM9x15 series. It exists as a sub block inside the IPs EBI2 (External Bus Interface 2) and QPIC (Qualcomm Parallel Interface Controller). These IPs provide a broader interface for external slow peripheral devices such as LCD and NAND/NOR flash memory or SRAM like interfaces. We add support for the NAND controller found within EBI2. For the SoCs of our interest, we only use the NAND controller within EBI2. Therefore, it's safe for us to assume that the NAND controller is a standalone block within the SoC. The controller supports 512B, 2kB, 4kB and 8kB page 8-bit and 16-bit NAND flash devices. It contains a HW ECC block that supports BCH ECC (4, 8 and 16 bit correction/step) and RS ECC(4 bit correction/step) that covers main and spare data. The controller contains an internal 512 byte page buffer to which we read/write via DMA. The EBI2 type NAND controller uses ADM DMA for register read/write and data transfers. The controller performs page reads and writes at a codeword/step level of 512 bytes. It can support up to 2 external chips of different configurations. The driver prepares register read and write configuration descriptors for each codeword, followed by data descriptors to read or write data from the controller's internal buffer. It uses a single ADM DMA channel that we get via dmaengine API. The controller requires 2 ADM CRCIs for command and data flow control. These are passed via DT. The ecc layout used by the controller is syndrome like, but we can't use the standard syndrome ecc ops because of several reasons. First, the amount of data bytes covered by ecc isn't same in each step. Second, writing to free oob space requires us writing to the entire step in which the oob lies. This forces us to create our own ecc ops. One more difference is how the controller accesses the bad block marker. The controller ignores reading the marker when ECC is enabled. ECC needs to be explicity disabled to read or write to the bad block marker. The nand_bbt helpers library hence can't access BBMs for the controller. For now, we skip the creation of BBT and populate chip->block_bad and chip->block_markbad helpers instead. Reviewed-by: Andy Gross <agross@codeaurora.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Archit Taneja <architt@codeaurora.org> Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2016-02-03 16:59:50 +08:00
{}
};
MODULE_DEVICE_TABLE(of, qcom_nandc_of_match);
static struct platform_driver qcom_nandc_driver = {
.driver = {
.name = "qcom-nandc",
.of_match_table = qcom_nandc_of_match,
},
.probe = qcom_nandc_probe,
.remove = qcom_nandc_remove,
};
module_platform_driver(qcom_nandc_driver);
MODULE_AUTHOR("Archit Taneja <architt@codeaurora.org>");
MODULE_DESCRIPTION("Qualcomm NAND Controller driver");
MODULE_LICENSE("GPL v2");