IMX6SX u-boot gpmi NAND not found

[peterbärtsch]

Hello

On our custom board we use a Spansion S34MS02G1 NAND. But u-boot (fslc-2015-10) can not find it.

CPU:   Freescale i.MX6SX rev1.2 at 792MHz

CPU:   Industrial temperature grade (-40C to 105C) at 42C

Reset cause: POR

Board: TEM IMX6SX Modul

I2C:   ready

DRAM:  256 MiB

PMIC: PFUZE3000 DEV_ID=0x30 REV_ID=0x11

NAND:  0 MiB

Our config file:

/* NAND */

#define CONFIG_CMD_NAND

#define CONFIG_NAND_MXS

#define CONFIG_SYS_MAX_NAND_DEVICE     1

#define CONFIG_SYS_NAND_BASE           0x40000000

#define CONFIG_SYS_NAND_5_ADDR_CYCLE

#define CONFIG_SYS_NAND_ONFI_DETECTION

/* DMA stuff, needed for GPMI/MXS NAND support */

#define CONFIG_APBH_DMA

#define CONFIG_APBH_DMA_BURST

#define CONFIG_APBH_DMA_BURST8

/* MTD/UBI/UBIFS config */

#define CONFIG_LZO

#define CONFIG_MTD_DEVICE

#define CONFIG_MTD_PARTITIONS

#define CONFIG_RBTREE

On linux there is no problem:

nand: device found, Manufacturer ID: 0x01, Chip ID: 0xaa

nand: AMD/Spansion S34MS02G1

nand: 256MiB, SLC, page size: 2048, OOB size: 64

Bad block table found at page 131008, version 0x01

Bad block table found at page 130944, version 0x01

2 ofpart partitions found on MTD device gpmi-nand

Creating 2 MTD partitions on "gpmi-nand":

0x0000005a0000-0x0000069a0000 : "root"

0x0000069a0000-0x000007da0000 : "data"

Has somebody any idea?

Peter

[fabio_estevam]

Hi Peter,

Care to submit this fix to u-boot@lists.denx.de ? Please generate it against master branch.

Thanks,

Fabio Estevam

[fabio_estevam]

This fix has been applied into u-boot-imx branch and should be part of 2016.01 version, thanks:

git.denx.de Git - u-boot/u-boot-imx.git/commitdiff

[igorpadykov]

for uboot NAND integration one can look at FSL uboot-imx for sabre ai board,

please look at attached Yocto Guide p.4, sect.5.4 U-Boot configuration

Best regards

igor

[peterbärtsch · Accepted solution]

Hi

I found the problem. First the pin mux for ALE signal was wrong and I add follow patch. Now it works :smileyhappy:

Thanks for help

Index: git/arch/arm/cpu/armv7/mx6/clock.c

===================================================================

--- git.orig/arch/arm/cpu/armv7/mx6/clock.c    2015-11-24 10:06:02.814884669 +0100

+++ git/arch/arm/cpu/armv7/mx6/clock.c    2015-11-24 10:06:03.442884668 +0100

@@ -47,6 +47,17 @@

              MXC_CCM_CCGR4_RAWNAND_U_GPMI_INPUT_APB_MASK |

              MXC_CCM_CCGR4_PL301_MX6QPER1_BCH_MASK);

+#ifdef CONFIG_MX6SX

+    clrbits_le32(&imx_ccm->CCGR4, MXC_CCM_CCGR4_QSPI2_ENFC_MASK);

+

+    clrsetbits_le32(&imx_ccm->cs2cdr,

+            MXC_CCM_CS2CDR_QSPI2_CLK_PODF_MASK |

+            MXC_CCM_CS2CDR_QSPI2_CLK_PRED_MASK |

+            MXC_CCM_CS2CDR_QSPI2_CLK_SEL_MASK,

+            cfg);

+

+    setbits_le32(&imx_ccm->CCGR4, MXC_CCM_CCGR4_QSPI2_ENFC_MASK);

+#else

     clrbits_le32(&imx_ccm->CCGR2, MXC_CCM_CCGR2_IOMUX_IPT_CLK_IO_MASK);

     clrsetbits_le32(&imx_ccm->cs2cdr,

@@ -56,6 +67,7 @@

             cfg);

     setbits_le32(&imx_ccm->CCGR2, MXC_CCM_CCGR2_IOMUX_IPT_CLK_IO_MASK);

+#endif

     setbits_le32(&imx_ccm->CCGR4,

              MXC_CCM_CCGR4_RAWNAND_U_BCH_INPUT_APB_MASK |

              MXC_CCM_CCGR4_RAWNAND_U_GPMI_BCH_INPUT_BCH_MASK |

[KursadOney]

NAND size is detected via the nand_flash_detect_onfi() function in both u-boot and linux. You can add some debug statements in the function and see why the size is being reported as zero. This could be due to a Spansion/AMD weirdness or one of the necessary clocks not being enabled.