分类： LINUX
2012-03-27 16:16:17
#cd linux-2.6.14
#vi Makefile
修改内容如下
ARCH =arm
CROSS_COMPILE = arm-linux-
下载交叉编译器arm-linux-gcc3.4.1
解压后,把交叉编译工具链的路径添加到环境变量中(修改/etc/bashrc文件)
2. 设置Nand Flash分区
修改linux-2.6.14\arch\arm\mach-s3c2410\devs.c
(1)添加下列头文件
#include
#include
#include
(2)添加NandFlash分区表
static struct mtd_partition partition_info[]={
{
name:"bootloader",
size:0x00020000,
offset:0x0,
},
{
name:"param",
size:0x00010000,
offset:0x00020000,
},
{
name:"kernel",
size:0x001c0000,
offset:0x00030000,
},
{
name:"rootfs",
size:0x03e00000,
offset:0x00200000,
//mask_flags:MTD_WRITEABLE,
}//,
};
其中
name: 代表分区名字
size: 代表flash分区大小(单位：字节)
offset: 代表flash分区的起始地址(相对于0x0的偏移)
(3)加入Nand Flash分区
struct s3c2410_nand_set nandset={
nr_partitions:4,//5,
partitions:partition_info,
};
nr_partitions: 指明partition_info中定义的分区数目
partitions: 分区信息表
(4)建立Nand Flash芯片支持
struct s3c2410_platform_nand superlpplatform={
tacls:0,
twrph0:30,
twrph1:0,
sets:&nandset,
nr_sets:1,
};
tacls, twrph0, twrph1的意思见S3C2410手册的63,
这3个值最后会被设置到NFCONF中,见S3C2410手册66.
sets: 支持的分区集
nr_sets:分区集的个数
(5)加入Nand Flash芯片支持到Nand Flash驱动
另外，还要修改此文件中的s3c_device_nand结构体变量,添加对dev成员的赋值
struct platform_device s3c_device_nand = {
.name = "s3c2410-nand",
.id = -1,
.num_resources = ARRAY_SIZE(s3c_nand_resource),
.resource = s3c_nand_resource,
.dev = {
.platform_data =&superlpplatform
}
};
name: 设备名称
id: 有效设备编号,如果只有唯一的一个设备为1,
有多个设备从0开始计数.
num_resource: 有几个寄存器区
resource: 寄存器区数组首地址
dev: 支持的Nand Flash设备
修改linux-2.6.14\arch\arm\mach-s3c2410\mach-smdk2410.c
static struct platform_device *smdk2410_devices[] __initdata = {
&s3c_device_usb,
&s3c_device_lcd,
&s3c_device_wdt,
&s3c_device_i2c,
&s3c_device_iis,
&s3c_device_nand,
};
(6)禁止Flash ECC校验
内核都是通过UBOOT写到Nand Flash的, UBOOT通过的软件ECC算法产生ECC校验码, 这与内核校验的ECC码不一样, 内核中的ECC码是由S3C2410中Nand Flash控制器产生的. 所以, 我们在这里选择禁止内核ECC校验.
修改linux-2.6.14\drivers\mtd\nand\s3c2410.c
找到函数s3c2410_nand_init_chip
最后一条语句 chip->eccmode = NAND_ECC_SOFT;
改为 chip->eccmode = NAND_ECC_NONE;
3. 配置内核选项
参考smdk2410开发板的配置文件,将其默认的配置文件复制到内核代码的根目录下,然后开始配置内核
#cd linux-2.6.14
#cp arch/arm/configs/smdk2410_defconfig .config
#make menuconfig
注意：在每个选项前有个方括号,其中[*]/<*>表示该选项加入内核编译;[ ]/<>表示不选择该选项;
Loadable module support ->
[*]Enable loadable module support #该选项的目的是使内核支持可加载模块,需要使用modprobe lsmod modinfo insmod rmmod等工具,所以必须选择;
[*]Module unloading #卸载模块选项
[*]Force module unloading#强制性卸载模块选项,如用rmmod –f 命令强制卸载;
[ ]Module versioning support
[ ]Source checksum for all modules
[*]Automatic kernel module loading#内核在任务中要使用一些被编译为模块的驱动或特性时,先使用modprobe命令来加载它,然后该选项自动调用modprobe加载需要的模块,所以该选项一定要选择.
加入内核对S3C2410 DMA(Direct Memory Access)的支持,配置如下：
System Type ->
[*]S3C2410 DMA support
修改命令行参数
Boot options ->
Default kernel command string
修改后内容如下:
noinitrd root=/dev/mtdblock3 init=/linuxrc console=ttySAC0,115200 mem=64M
或者
noinitrd root=/dev/mtdblock/3 init=/linuxrc console=ttySAC0,115200 mem=64M
mtdblock3表示使用Flash的第4个分区(就是rootfs分区), console=ttySAC0表示kernel启动期间的信息全部输出到串口0上,115200表示波特率为115200,
mem=64M表示内存大小为64MB
添加对浮点算法的支持
Floating point emulation ->
[*]NWFPE math emulation#支持NWFPE 浮点数,在许多情况下要使用,所以最好选上
对MTD(Memory Technology Devices)设备(如 Flash RAM等芯片)进行配置,选择配置如下
Device Drivers->
Memory Technology Devices(MTD)->
[*]MTD partitioning support
Device Drivers->
Memory Technology Devices(MTD)->
RAM/ROM/Flash chip drivers->
<*>Detect flash chips by Common Flash Interface(CFI)probe
<*>Detect non-CFI AMD/JEDEC-compatible flash chips
<*>Support for Intel/Sharp flash chips
<*>Support for AMD/Fujitsu flash chips
<*>Support for ROM chips in bus mapping
Device Drivers->
Memory Technology Devices(MTD)->
NAND Flash Device Drivers->
<*>NAND Device Support
<*>NAND Flash support for S3C2410/S3C2440 SoC
为了要内核支持devfs(Device Filesystem,设备文件系统),以及在启动时能自动加载/dev为devfs,需要对文件系统进行设置。
File systems ->
< >Second extended fs support#去除对ext2的支持
Linux2.6.14 fs/Kconfig没有支持devfs的选项(从2.6.13开始),所以make menuconfig时找不到相关菜单项
从2.6.10的fs/Kconfig中copy几项过来
config DEVFS_FS
bool "/dev file system support (OBSOLETE)"
depends on EXPERIMENTAL
help
This is support for devfs, a virtual file system (like /proc) which
provides the file system interface to device drivers, normally found
in /dev. Devfs does not depend on major and minor number
allocations. Device drivers register entries in /dev which then
appear automatically, which means that the system administrator does
not have to create character and block special device files in the
/dev directory using the mknod command (or MAKEDEV script) anymore.
This is work in progress. If you want to use this, you *must* read
the material in
the file README there.
Note that devfs no longer manages /dev/pts! If you are using UNIX98
ptys, you will also need to mount the /dev/pts filesystem (devpts).
Note that devfs has been obsoleted by udev,
It has been stripped down to a bare minimum and is only provided for
legacy installations that use its naming scheme which is
unfortunately different from the names normal Linux installations
use.
If unsure, say N.
config DEVFS_MOUNT
bool "Automatically mount at boot"
depends on DEVFS_FS
help
This option appears if you have CONFIG_DEVFS_FS enabled. Setting
this to 'Y' will make the kernel automatically mount devfs onto /dev
when the system is booted, before the init thread is started.
You can override this with the "devfs=nomount" boot option.
If unsure, say N.
config DEVFS_DEBUG
bool "Debug devfs"
depends on DEVFS_FS
help
If you say Y here, then the /dev file system code will generate
debugging messages. See the file
details.
If unsure, say N.
File systems ->
Pseudo filesystems ->
[*]/proc file system support
[*]/dev file system support(OBSOLETE)
[*]Automatically mount at boot
[*]Virtual memory file system support(former shm fs)
Miscellaneous filesystems ->
<*>Compressed ROM file system support(cramfs)
Network File Systems ->
<*>NFS file system support
除此之外,还需要配置一下选项支持S3C2410 RTC,USB,MMC/SD卡驱动,具体选项如下:
Device Drivers->
Character devices->
[*]Nonstandard serial port support
[*]S3C2410 RTC Driver
Device Drivers->
USB Support->
<*>Support for Host-side USB
Device Drivers->
MMC/SD Card Support
<*>MMC Support
<*>MMC block device driver
调试过程中出现的问题及解决办法
1. Starting kernel ...就打住了
应该是从u-boot跳转至内核的地址不正确,我下载的加了0x40字节头信息的内核镜像,下载到0x30008000,正确的跳转地址应该是0x30008040,在
u-boot-1.1.3\lib_arm\armlinux.c的do_bootm_linux函数代码如下：
void do_bootm_linux (cmd_tbl_t *cmdtp, int flag, int argc, char *argv[], ulong addr, ulong *len_ptr, int verify)
{
DECLARE_GLOBAL_DATA_PTR;
ulong len = 0, checksum;
ulong initrd_start, initrd_end;
ulong data;
void (*theKernel)(int zero, int arch, uint params);
image_header_t *hdr = &header;
bd_t *bd = gd->bd;
. . .
theKernel = (void (*)(int, int, uint))ntohl(hdr->ih_ep);
. . .
/* we assume that the kernel is in place */
printf ("\nStarting kernel ...\n\n");
#ifdef CONFIG_USB_DEVICE
{
extern void udc_disconnect (void);
udc_disconnect ();
}
#endif
cleanup_before_linux ();
theKernel (0, bd->bi_arch_number, bd->bi_boot_params);
}
theKernel = (void (*)(int, int, uint))ntohl(hdr->ih_ep);
typedef struct image_header {
uint32_t ih_magic; /* Image Header Magic Number */
uint32_t ih_hcrc; /* Image Header CRC Checksum */
uint32_t ih_time; /* Image Creation Timestamp */
uint32_t ih_size; /* Image Data Size */
uint32_t ih_load; /* Data Load Address */
uint32_t ih_ep; /* Entry Point Address */
uint32_t ih_dcrc; /* Image Data CRC Checksum */
uint8_t ih_os; /* Operating System */
uint8_t ih_arch; /* CPU architecture */
uint8_t ih_type; /* Image Type */
uint8_t ih_comp; /* Compression Type */
uint8_t ih_name[IH_NMLEN]; /* Image Name */
} image_header_t;
可以看出跳转地址是由image_header结构体的ih_ep(Entry Point Address)成员来确定的,而ih_ep是由mkimage时自己填写的,如下是我填充的mkimage参数,其中的
-e 0x30008000就是来填充ih_ep成员的.
mkimage -A arm -O Linux -T kernel -C none -a 0x30008000 -e 0x30008000 -n Linux-2.6.14 -d zImage uImage
显然这个值应该是0x30008040
mkimage -A arm -O Linux -T kernel -C none -a 0x30008000 -e 0x30008040 -n Linux-2.6.14 -d zImage uImage
修改后启动信息如下
## Booting image at 30008000 ...
Image Name: Linux-2.6.14
Created: 2008-07-14 8:59:37 UTC
Image Type: ARM Linux Kernel Image (uncompressed)
Data Size: 1028136 Bytes = 1004 kB
Load Address: 30008000
Entry Point: 30008040
Verifying Checksum ... OK
XIP Kernel Image ... OK
commandline:noinitrd root=/dev/mtdblock3 init=/linuxrc console= ttySAC0,115200 mem=64M
Starting kernel ...
Uncompressing Linux.................................................................... done, booting the kernel.
2.Starting kernel ...
Uncompressing Linux............................................................
......... done, booting the kernel.
Error: unrecognized/unsupported machine ID (r1 = 0x000000c2).
Available machine support:
ID (hex) NAME
000000c1 SMDK2410
Please check your kernel config and/or bootloader.
修改linux-2.6.14\arch\arm\kernel\head.S,如下,蓝色字体为添加内容
ENTRY(stext)
msr cpsr_c, #PSR_F_BIT | PSR_I_BIT | MODE_SVC @ ensure svc mode
@ and irqs disabled
bl __lookup_processor_type @ r5=procinfo r9=cpuid
movs r10, r5 @ invalid processor (r5=0)?
beq __error_p @ yes, error 'p'
mov r1, #0xc1 @ 参考linux-2.6.14\include\asm-arm\ mach-types.h
bl __lookup_machine_type @ r5=machinfo
movs r8, r5 @ invalid machine (r5=0)?
beq __error_a @ yes, error 'a'
3. Starting kernel ...
Uncompressing Linux.................................................
......... done, booting the kernel.
然后就停止了
查看commandline完全正确,从上面的打印信息可以看出,内核已经解压成功了,这样只能进入内核跟踪调试了
在内核中arch/arm/kernel/debug.S中存在一个debug函数叫做printascii,使用方法如下：
adr r0, str_p1
bl printascii
str_p1:.asciz"\nError: unrecognized/unsupported process type\n"
用printascii在arch/arm/kernel/head.S里跟踪调试,直到跳转到start_kernel都没有发现问题,查到printascii在进入start_kernel之后也可以使用,说明如下:
mmu已经开启后,系统中的物理地址都变成虚拟地址了,因此原来基于物理地址的调试方案将都会失败,但可以使用printascii继续调试,该调试功能同时支持物理地址与虚拟地址,并且提供了一个解决方案就是将printascii加入到printk的vsprintf()之里。
修改linux-2.6.14\kernel\printk.c,蓝色字体为添加内容
extern void printascii(const char*);
asmlinkage int vprintk(const char *fmt, va_list args)
{
unsigned long flags;
int printed_len;
char *p;
static char printk_buf[1024];
static int log_level_unknown = 1;
preempt_disable();
if (unlikely(oops_in_progress) && printk_cpu == smp_processor_id())
zap_locks();
spin_lock_irqsave(&logbuf_lock, flags);
printk_cpu = smp_processor_id();
/* Emit the output into the temporary buffer */
printed_len = vscnprintf(printk_buf, sizeof(printk_buf), fmt, args);
printascii(printk_buf);
...
...
}
修改后内核启动成功,超级终端出来一大堆打印信息
## Booting image at 30008000 ...
Image Name: Linux-2.6.14
Created: 2008-07-14 8:59:37 UTC
Image Type: ARM Linux Kernel Image (uncompressed)
Data Size: 1028136 Bytes = 1004 kB
Load Address: 30008000
Entry Point: 30008040
Verifying Checksum ... OK
XIP Kernel Image ... OK
Starting kernel ...
Uncompressing Linux.................................................................... done, booting the kernel.
<5>Linux version 2.6.14 (root@localhost.localdomain) (gcc version 3.4.1) #3 Mon Jul 14 04:59:17 EDT 2008
CPU: ARM920Tid(wb) [41129200] revision 0 (ARMv4T)
Machine: SMDK2410
Memory policy: ECC disabled, Data cache writeback
CPU S3C2410A (id 0x32410002)
S3C2410: core 202.800 MHz, memory 101.400 MHz, peripheral 50.700 MHz
<6>S3C2410 Clocks, (c) 2004 Simtec Electronics
CLOCK: Slow mode (1.500 MHz), fast, MPLL on, UPLL on
<7>On node 0 totalpages: 16384
<7> DMA zone: 16384 pages, LIFO batch:7
<7> Normal zone: 0 pages, LIFO batch:1
<7> HighMem zone: 0 pages, LIFO batch:1
CPU0: D VIVT write-back cache
CPU0: I cache: 16384 bytes, associativity 64, 32 byte lines, 8 sets
CPU0: D cache: 16384 bytes, associativity 64, 32 byte lines, 8 sets
Built 1 zonelists
<5>Kernel command line: noinitrd root=/dev/mtdblock3 init=/linuxrc console= ttySAC0,115200 mem=64M
irq: clearing subpending status 00000003
PID hash table entries: 512 (order: 9, 8192 bytes)
timer tcon=00500000, tcnt a509, tcfg 00000200,00000000, usec 00001e4c
Console: colour dummy device 80x30
Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
<6>Memory: 64MB = 64MB total
<5>Memory: 62592KB available (1708K code, 381K data, 92K init)
<7>Calibrating delay loop... 101.17 BogoMIPS (lpj=252928)
Mount-cache hash table entries: 512
<6>CPU: Testing write buffer coherency: ok
softlockup thread 0 started up.
<6>NET: Registered protocol family 16
S3C2410: Initialising architecture
<6>usbcore: registered new driver usbfs
<6>usbcore: registered new driver hub
S3C2410 DMA Driver, (c) 2003-2004 Simtec Electronics
DMA channel 0 at c4800000, irq 33
DMA channel 1 at c4800040, irq 34
DMA channel 2 at c4800080, irq 35
DMA channel 3 at c48000c0, irq 36
<4>NetWinder Floating Point Emulator V0.97 (double precision)
Console: switching to colour frame buffer device 80x25
<6>fb0: Virtual frame buffer device, using 1024K of video memory
S3C2410 RTC, (c) 2004 Simtec Electronics
<6>s3c2410_serial0 at MMIO 0x50000000 (irq = 70) is a S3C2410
<6>s3c2410_serial1 at MMIO 0x50004000 (irq = 73) is a S3C2410
<6>s3c2410_serial2 at MMIO 0x50008000 (irq = 76) is a S3C2410
<6>io scheduler noop registered
<6>io scheduler anticipatory registered
<6>io scheduler deadline registered
<6>io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
S3C24XX NAND Driver, (c) 2004 Simtec Electronics
<6>s3c2410-nand: mapped registers at c4980000
<6>s3c2410-nand: timing: Tacls 10ns, Twrph0 10ns, Twrph1 10ns
<4>No NAND device found!!!
<5>usbmon: debugfs is not available
<6>mice: PS/2 mouse device common for all mice
<6>NET: Registered protocol family 2
IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
TCP established hash table entries: 4096 (order: 2, 16384 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
<6>TCP: Hash tables configured (established 4096 bind 4096)
<6>TCP reno registered
<6>TCP bic registered
<6>NET: Registered protocol family 1
<3>Root-NFS: No NFS server available, giving up.
<3>VFS: Unable to mount root fs via NFS, trying floppy.
VFS: Cannot open root device "mtdblock3" or unknown-block(2,0)
Please append a correct "root=" boot option
<0>Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(2,0)
那么就说明内核没有问题,而是commandline参数有问题,仔细看,才发现
noinitrd root=/dev/mtdblock3 init=/linuxrc console= ttySAC0,115200 mem=
64M
console=之后多了一个空格,但是不确定是否是这个原因,重新设置bootargs参数,去掉空格
setenv bootargs noinitrd root=/dev/mtdblock3 init=/linuxrc console=ttyS
AC0,115200 mem=64M
重新下载,启动成功.
在网上查到有说u-boot和内核频率不一致可能导致串口没有打印信息,最好将u-boot的频率设为200M,不过我试过了, 202.800 MHz也没有问题
Uncompressing Linux............................................................ done, booting the kernel.
Linux version 2.6.14 (root@localhost.localdomain) (gcc version 3.4.1) #3 Mon Jul 14 04:59:17 EDT 2008
CPU: ARM920Tid(wb) [41129200] revision 0 (ARMv4T)
Machine: SMDK2410
Memory policy: ECC disabled, Data cache writeback
CPU S3C2410A (id 0x32410002)
S3C2410: core 202.800 MHz, memory 101.400 MHz, peripheral 50.700 MHz
Starting kernel ...
Uncompressing Linux................................................................... done, booting the kernel.
Linux version 2.6.14 (root@localhost.localdomain) (gcc version 3.4.1) #20 Sat Jul 5 10:01:30 EDT 2008
CPU: ARM920Tid(wb) [41129200] revision 0 (ARMv4T)
Machine: SMDK2410
Memory policy: ECC disabled, Data cache writeback
CPU S3C2410A (id 0x32410002)
S3C2410: core 200.000 MHz, memory 100.000 MHz, peripheral 50.000 MHz
1. S3C24XX NAND Driver, (c) 2004 Simtec Electronics
s3c2410-nand: mapped registers at c4980000
s3c2410-nand: timing: Tacls 10ns, Twrph0 10ns, Twrph1 10ns
No NAND device found!!!
usbmon: debugfs is not available
mice: PS/2 mouse device common for all mice
NET: Registered protocol family 2
IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
TCP established hash table entries: 4096 (order: 2, 16384 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
TCP: Hash tables configured (established 4096 bind 4096)
TCP reno registered
TCP bic registered
NET: Registered protocol family 1
Root-NFS: No NFS server available, giving up.
VFS: Unable to mount root fs via NFS, trying floppy.
VFS: Cannot open root device "mtdblock3" or unknown-block(2,0)
Please append a correct "root=" boot option
Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(2,0)
原因可能是一下几点：
(1)commandline中的root=/dev/mtdblock3表示rootfs在Flash的第4分区,检查确认rootfs是否是Flash的第4分区(第一个分区号为0)
(2)Flash的rootfs分区中没有下载好的根文件系统,确认是否下载好根文件系统
(3)Flash的驱动有问题,无法对Flash进行读写
我的NAND Flash分区如下:
mtdpart info. (5 partitions)
name offset size flag
------------------------------------------------
vivi : 0x00000000 0x00020000 0 128k
param : 0x00020000 0x00010000 0 64k
kernel : 0x00030000 0x001c0000 0 1M+768k
rootfs : 0x00200000 0x02000000 0 32M
user : 0x02200000 0x01e00000 0 30M
所以commandline中的mtdblock3没有错误
我已下载好根文件系统至rootfs分区,再看看串口的打印信息,注意红色的部分
No NAND device found!!!
有点奇怪,其实我不太相信linux的源代码会有问题,就在根文件系统的加载这一步拖了好多天,一开始由于自己的疏忽,make menuconfig时把有些选项选成模块了,也没有搞清楚<*>和
linux-2.6.14\drivers\mtd\nand\ s3c2410.c中s3c2410_nand_init函数对NAND Flash初始化
static int __init s3c2410_nand_init(void)
{
printk("S3C24XX NAND Driver, (c) 2004 Simtec Electronics\n");
driver_register(&s3c2440_nand_driver);
return driver_register(&s3c2410_nand_driver);
}
static struct device_driver s3c2410_nand_driver = {
.name = "s3c2410-nand",
.bus = &platform_bus_type,
.probe = s3c2410_nand_probe,
.remove = s3c2410_nand_remove,
};
static int s3c2410_nand_probe(struct device *dev)
{
return s3c24xx_nand_probe(dev, 0);
}
static int s3c24xx_nand_probe(struct device *dev, int is_s3c2440)
{
struct platform_device *pdev = to_platform_device(dev);
struct s3c2410_platform_nand *plat = to_nand_plat(dev);
struct s3c2410_nand_info *info;
struct s3c2410_nand_mtd *nmtd;
struct s3c2410_nand_set *sets;
struct resource *res;
. . .
for (setno = 0; setno < nr_sets; setno++, nmtd++) {
pr_debug("initialising set %d (%p, info %p)\n",
setno, nmtd, info);
s3c2410_nand_init_chip(info, nmtd, sets);
nmtd->scan_res = nand_scan(&nmtd->mtd,
(sets) ? sets->nr_chips : 1);
if (nmtd->scan_res == 0) {
s3c2410_nand_add_partition(info, nmtd, sets);
}
if (sets != NULL)
sets++;
}
. . .
}
经过串口打印出nmtd->scan_res的值为1,根据
if (nmtd->scan_res == 0) {
s3c2410_nand_add_partition(info, nmtd, sets);
}
来判断,应该返回0.
进入nand_scan
int nand_scan (struct mtd_info *mtd, int maxchips)
{
int i, nand_maf_id, nand_dev_id, busw, maf_id;
struct nand_chip *this = mtd->priv;
. . .
/* Select the device */
this->select_chip(mtd, 0);
/* Send the command for reading device ID */
this->cmdfunc (mtd, NAND_CMD_READID, 0x00, -1);
/* Read manufacturer and device IDs */
nand_maf_id = this->read_byte(mtd);
nand_dev_id = this->read_byte(mtd);
. . .
/* Print and store flash device information */
for (i = 0; nand_flash_ids[i].name != NULL; i++) {
if (nand_dev_id != nand_flash_ids[i].id)
continue;
if (!mtd->name) mtd->name = nand_flash_ids[i].name;
this->chipsize = nand_flash_ids[i].chipsize << 20;
. . .
/* Try to identify manufacturer */
for (maf_id = 0; nand_manuf_ids[maf_id].id != 0x0; maf_id++) {
if (nand_manuf_ids[maf_id].id == nand_maf_id)
break;
}
. . .
printk (KERN_INFO "NAND device: Manufacturer ID:"
" 0x%02x, Chip ID: 0x%02x (%s %s)\n", nand_maf_id, nand_dev_id,
nand_manuf_ids[maf_id].name , nand_flash_ids[i].name);
break;
}
if (!nand_flash_ids[i].name) {
printk (KERN_WARNING "No NAND device found!!!\n");
this->select_chip(mtd, -1);
return 1;
}
. . .
}
根据上面的代码可以看出没有在nand_flash_ids数组中找到nand_dev_id匹配值,记得在u-boot中也有nand_probe函数检测NAND Flash,打印出2个ID值如下
NAND:Flash chip found: Manufacturer ID: 0xEC, Chip ID: 0x76
Linux中nand_flash_ids和nand_manuf_ids数组内容如下:
/*
* Chip ID list*
* Name. ID code, pagesize, chipsize in MegaByte, eraseblock size,
*/
struct nand_flash_dev nand_flash_ids[] = {
{"NAND 1MiB 5V 8-bit", 0x6e, 256, 1, 0x1000, 0},
{"NAND 2MiB 5V 8-bit", 0x64, 256, 2, 0x1000, 0},
{"NAND 4MiB 5V 8-bit", 0x6b, 512, 4, 0x2000, 0},
{"NAND 1MiB 3,3V 8-bit", 0xe8, 256, 1, 0x1000, 0},
{"NAND 1MiB 3,3V 8-bit", 0xec, 256, 1, 0x1000, 0},
{"NAND 2MiB 3,3V 8-bit", 0xea, 256, 2, 0x1000, 0},
{"NAND 4MiB 3,3V 8-bit", 0xd5, 512, 4, 0x2000, 0},
{"NAND 4MiB 3,3V 8-bit", 0xe3, 512, 4, 0x2000, 0},
{"NAND 4MiB 3,3V 8-bit", 0xe5, 512, 4, 0x2000, 0},
. . .
{"NAND 64MiB 3,3V 8-bit", 0x76, 512, 64, 0x4000, 0},
. . .
{NULL,}
};
/*
* Manufacturer ID list
*/
struct nand_manufacturers nand_manuf_ids[] = {
{NAND_MFR_TOSHIBA, "Toshiba"},
{NAND_MFR_SAMSUNG, "Samsung"},
{NAND_MFR_FUJITSU, "Fujitsu"},
{NAND_MFR_NATIONAL, "National"},
{NAND_MFR_RENESAS, "Renesas"},
{NAND_MFR_STMICRO, "ST Micro"},
{NAND_MFR_HYNIX, "Hynix"},
{0x0, "Unknown"}
};
/*
* NAND Flash Manufacturer ID Codes
*/
#define NAND_MFR_TOSHIBA 0x98
#define NAND_MFR_SAMSUNG 0xec
#define NAND_MFR_FUJITSU 0x04
#define NAND_MFR_NATIONAL 0x8f
#define NAND_MFR_RENESAS 0x07
#define NAND_MFR_STMICRO 0x20
#define NAND_MFR_HYNIX 0xad
这样就说明nand_scan中读取NAND Flash的id值是错误的,不是0xec和0x76,从串口打印出来的id值确实不正确,首先我的NAND Flash肯定正常工作,在u-boot中一直是好的,那么可能是Linux中对Flash读写的出问题了,怀疑是发出命令后的延时有问题,向NAND Flash发出读取id命令后的延时太短,导致读出来的值错误, nand_wait_ready这个就是延时等待命令执行完成的函数,经过测试没有问题,到这个时候,真有点崩溃了.然后估计是NAND Flash的NFCONF寄存器设置有问题, NFCONF寄存器在函数s3c2410_nand_inithw中初始化,s3c2410_nand_inithw中将NFCONF寄存器的值打印出来,如下:
s3c2410-nand: mapped registers at c4980000
s3c2410-nand: timing: Tacls 10ns, Twrph0 10ns, Twrph1 10ns
s3c2410-nand: NF_CONF is 0x8000
与u-boot中设置的NFCONF寄存器值比较发现,不同的是TACLS TWRPH0 TWRPH1以及是否初始化ECC,nFCE的值
U-boot中TACLS TWRPH0 TWRPH1值分别为0 4 2,初始化ECC, nFCE=1(inactive)
Linux中TACLS TWRPH0 TWRPH1值分别为0 0 0,未初始化ECC, nFCE=0(active)
应该是TACLS TWRPH0 TWRPH1的问题,查看S3C2410 user manual和NAND Flash用户手册Samsung K9F1208U0M user manual,找到三张图,如下:
图1 s3c2410 user manual, nand flash memory timing
显然s3c2410 user manual 推荐值是TACLS=0 TWRPH0=1 TWRPH1=0
TACLS:1个HCLK TWRPH0:2个HCLK TWRPH1:1个HCLK
内核的频率为200MHZ,即FCLK=200MHZ,则HCLK=100MHZ(周期为10ns),所以
TACLS:10ns TWRPH0:20ns TWRPH1:10ns
图2 NAND Flash(K9F1208U0M) user manual,Read and Write timing
tCLS等价于图1中的TACLS;tWP等价于图1中的TWRPH0;tCLH等价于图1中的TWRPH1
图3 NAND Flash(K9F1208U0M) user manual
结合前3张图,可以看出对NAND Flash(K9F1208U0M)进行读写的顺序要求为:
TACLS <----------->tCLS,最小值为0ns
TWRPH0<----------->tWP,最小值为25ns
TWRPH1<----------->tCLH, 最小值为10ns
内核的频率为200MHZ,即FCLK=200MHZ,则HCLK=100MHZ(周期为10ns)
TACLS=0, TWRPH0=2,TWRPH1=0即可满足时序要求,此时
TACLS:10ns TWRPH0:30ns TWRPH1:10ns
现在再看看先前串口输出的信息:
s3c2410-nand: mapped registers at c4980000
s3c2410-nand: timing: Tacls 10ns, Twrph0 10ns, Twrph1 10ns
No NAND device found!!!
很显然Twrph0 10ns <25ns,原来如此!!!!
不过我试了一下TACLS=0, TWRPH0=1,TWRPH1=0, TACLS:10ns TWRPH0:20ns TWRPH1:10ns,NAND Flash照样正常工作.不知道为什么.
现在要说明一下我是如何修改的, TACLS TWRPH0 TWRPH1 的值在linux-2.6.14\arch\arm\mach-s3c2410\ devs.c中已经设置,代码如下:
struct s3c2410_platform_nand superlpplatform={
tacls:0,
twrph0:3,
twrph1:0,
sets:&nandset,
nr_sets:1,
};
这里设置的值和上面的分析是一致的,应该没有问题,但是为什么打印出来的却是
s3c2410-nand: timing: Tacls 10ns, Twrph0 10ns, Twrph1 10ns
s3c2410_nand_inithw函数对NFCONF寄存器初始化,代码如下:
static int s3c2410_nand_inithw(struct s3c2410_nand_info *info,
struct device *dev)
{
struct s3c2410_platform_nand *plat = to_nand_plat(dev);
unsigned int tacls, twrph0, twrph1;
unsigned long clkrate = clk_get_rate(info->clk);
unsigned long cfg;
/* calculate the timing information for the controller */
if (plat != NULL) {
tacls = s3c2410_nand_calc_rate(plat->tacls, clkrate, 4);
twrph0 = s3c2410_nand_calc_rate(plat->twrph0, clkrate, 8);
twrph1 = s3c2410_nand_calc_rate(plat->twrph1, clkrate, 8);
printk("plat->tacls:%d\n",plat->tacls);
printk("plat->twrph0:%d\n",plat->twrph0);
printk("plat->twrph1:%d\n",plat->twrph1);
printk("tacls:%d\n",tacls);
printk("twrph0:%d\n",twrph0);
printk("twrph1:%d\n",twrph1);
printk("clkrate:%d\n",clkrate);
tacls =1;
twrph0 =2;
twrph1 =1;
} else {
/* default timings */
tacls = 4;
twrph0 = 8;
twrph1 = 8;
}
if (tacls < 0 || twrph0 < 0 || twrph1 < 0) {
printk(KERN_ERR PFX "cannot get timings suitable for board\n");
return -EINVAL;
}
printk(KERN_INFO PFX "timing: Tacls %ldns, Twrph0 %ldns, Twrph1 %ldns\n",
to_ns(tacls, clkrate),
to_ns(twrph0, clkrate),
to_ns(twrph1, clkrate));
if (!info->is_s3c2440) {
cfg = S3C2410_NFCONF_EN;
cfg |= S3C2410_NFCONF_TACLS(tacls-1);
cfg |= S3C2410_NFCONF_TWRPH0(twrph0-1);
cfg |= S3C2410_NFCONF_TWRPH1(twrph1-1);
} else {
cfg = S3C2440_NFCONF_TACLS(tacls-1);
cfg |= S3C2440_NFCONF_TWRPH0(twrph0-1);
cfg |= S3C2440_NFCONF_TWRPH1(twrph1-1);
}
pr_debug(PFX "NF_CONF is 0x%lx\n", cfg);
writel(cfg, info->regs + S3C2410_NFCONF);
return 0;
}
我加了一点代码将相关参数打印出来:
plat->tacls:0
plat->twrph0:3
plat->twrph1:0
tacls:1
twrph0:1
twrph1:1
clkrate:100000000
真是搞不懂plat->tacls, plat->twrph0, plat->twrph1的值和前面初始化的值是一样的,但是经过s3c2410_nand_calc_rate函数处理后就出问题了,正确的输出值应该是
tacls:1 twrph0:4 twrph1:1
算了,对tacls,twrph0,twrph1三个变量强制性赋值
tacls =1;
twrph0 =2;
twrph1 =1;
修改后NAND Flash可以正常工作了,启动信息如下:
Linux version 2.6.14 (root@localhost.localdomain) (gcc version 3.4.1) #7 Mon Jul 14 09:34:58 EDT 2008
CPU: ARM920Tid(wb) [41129200] revision 0 (ARMv4T)
Machine: SMDK2410
Memory policy: ECC disabled, Data cache writeback
CPU S3C2410A (id 0x32410002)
S3C2410: core 200.000 MHz, memory 100.000 MHz, peripheral 50.000 MHz
S3C2410 Clocks, (c) 2004 Simtec Electronics
CLOCK: Slow mode (1.500 MHz), fast, MPLL on, UPLL on
CPU0: D VIVT write-back cache
CPU0: I cache: 16384 bytes, associativity 64, 32 byte lines, 8 sets
CPU0: D cache: 16384 bytes, associativity 64, 32 byte lines, 8 sets
Built 1 zonelists
Kernel command line: noinitrd root=/dev/mtdblock3 init=/linuxrc console=ttySAC0,115200 mem=64M
irq: clearing subpending status 00000003
irq: clearing subpending status 00000002
PID hash table entries: 512 (order: 9, 8192 bytes)
timer tcon=00500000, tcnt a2c1, tcfg 00000200,00000000, usec 00001eb8
Console: colour dummy device 80x30
Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
Memory: 64MB = 64MB total
Memory: 62592KB available (1708K code, 381K data, 92K init)
Mount-cache hash table entries: 512
CPU: Testing write buffer coherency: ok
softlockup thread 0 started up.
NET: Registered protocol family 16
S3C2410: Initialising architecture
usbcore: registered new driver usbfs
usbcore: registered new driver hub
S3C2410 DMA Driver, (c) 2003-2004 Simtec Electronics
DMA channel 0 at c4800000, irq 33
DMA channel 1 at c4800040, irq 34
DMA channel 2 at c4800080, irq 35
DMA channel 3 at c48000c0, irq 36
NetWinder Floating Point Emulator V0.97 (double precision)
Console: switching to colour frame buffer device 80x25
fb0: Virtual frame buffer device, using 1024K of video memory
S3C2410 RTC, (c) 2004 Simtec Electronics
s3c2410_serial0 at MMIO 0x50000000 (irq = 70) is a S3C2410
s3c2410_serial1 at MMIO 0x50004000 (irq = 73) is a S3C2410
s3c2410_serial2 at MMIO 0x50008000 (irq = 76) is a S3C2410
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
S3C24XX NAND Driver, (c) 2004 Simtec Electronics
s3c2410-nand: mapped registers at c4980000
s3c2410-nand: timing: Tacls 10ns, Twrph0 20ns, Twrph1 10ns
s3c2410-nand: NF_CONF is 0x8010
NAND device: Manufacturer ID: 0xec, Chip ID: 0x76 (Samsung NAND 64MiB 3,3V 8-bit)
NAND_ECC_NONE selected by board driver. This is not recommended !!
Scanning device for bad blocks
Creating 5 MTD partitions on "NAND 64MiB 3,3V 8-bit":
0x00000000-0x00020000 : "bootloader"
0x00020000-0x00030000 : "param"
0x00030000-0x001f0000 : "kernel"
0x00200000-0x02200000 : "rootfs"
0x02200000-0x04000000 : "user"
usbmon: debugfs is not available
mice: PS/2 mouse device common for all mice
NET: Registered protocol family 2
IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
TCP established hash table entries: 4096 (order: 2, 16384 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
TCP: Hash tables configured (established 4096 bind 4096)
TCP reno registered
TCP bic registered
NET: Registered protocol family 1
Reading data from NAND FLASH without ECC is not recommended
VFS: Mounted root (cramfs filesystem) readonly.
Freeing init memory: 92K
Warning: unable to open an initial console.
Reading data from NAND FLASH without ECC is not recommended
sd_mod: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
usb_storage: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
usbvideo: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
ov511: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
今天写总结再来分析一下s3c2410_nand_calc_rate函数,终于知道问题在哪了,之前的打印内容如下:
plat->tacls:0
plat->twrph0:3
plat->twrph1:0
tacls:1
twrph0:1
twrph1:1
clkrate:100000000/*注意是8个0,实际就是是HCLK=100MHz*/
#define NS_IN_KHZ 10000000 /*注意是7个0*/
static int s3c2410_nand_calc_rate(int wanted, unsigned long clk, int max)
{
int result;
result = (wanted * NS_IN_KHZ) / clk;
result++;
pr_debug("result %d from %ld, %d\n", result, clk, wanted);
if (result > max) {
printk("%d ns is too big for current clock rate %ld\n",
wanted, clk);
return -1;
}
if (result < 1)
result = 1;
return result;
}
s3c2410_nand_calc_rate(plat->twrph0, clkrate, 8)
即s3c2410_nand_calc_rate(3, 100000000, 8)
则wanted=3, clk=100000000, max=8
result = (wanted * NS_IN_KHZ) / clk
result = (3 * 10000000)/ 100000000 显然结果为0, result++;后返回值为1
按照这个函数的本意应该是要返回4的,如果要修改的话, 将它的值NS_IN_KHZ 改为与HCLK相同的值,就可以了,我本来想改为100000000,想想如果uboot的FCLK改成202.8MHz,那么HCLK值为101.4MHz,则 clkrate 为101400000
result = (wanted * NS_IN_KHZ) / clk=(3 * 100000000)/ 101400000=0
还是不行啊,真该死,终于找到原因了,原来linux源码是没有问题的,在
linux-2.6.14\linux-2.6.14\arch\arm\mach-s3c2410\ devs.c中
struct s3c2410_platform_nand superlpplatform={
tacls:0,
twrph0:30,/*原来写的是3*/
twrph1:0,
sets:&nandset,
nr_sets:1,
};
串口打印出来内容如下:
plat->tacls:0
plat->twrph0:30
plat->twrph1:0
tacls:1
twrph0:3
twrph1:1
clkrate:101400000
s3c2410-nand: timing: Tacls 10ns, Twrph0 30ns, Twrph1 10ns
5.Reading data from NAND FLASH without ECC is not recommended
VFS: Mounted root (cramfs filesystem) readonly.
Freeing init memory: 92K
Warning: unable to open an initial console.
Reading data from NAND FLASH without ECC is not recommended
sd_mod: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
usb_storage: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
usbvideo: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
ov511: version magic '2.6.8.1-ptx1 ARMv4 gcc-3.3' should be '2.6.14 ARMv4 gcc-3.4'
说明一下:之前并没有碰到这个问题,写总结时故意去掉对devfs的支持,出现这样的错误
Warning: unable to open an initial console.
原因是文件系统的dev目录下没有console节点,解决办法有2个,
(1)制作文件系统时在dev目录下建立console节点
(2)make menuconfig时加上对devfs的支持,就可以正常启动
加上对devfs的支持,系统会根据需要自动创建所需节点
用ramdisk作根文件系统:
(1)make menuconfig时要选择支持ext2文件系统，因为ramdisk大多用ext2;
(2)make menuconfig时要选择RAM disk support
(3)make menuconfig时还要选择初始化ramdisk(initrd)
Device Drivers ->
Block Device->
<*>RAM disk support
[*]Initial RAM disk(initrd)support
(4)make menuconfig时还要设置好RAM disk的大小
Device Drivers ->
Block Device->
<*>RAM disk support
(16)Default number of RAM disks
(4096)Default RAM disk size(kbytes)
加载ramdisk时出现如下问题：
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
RAMDISK: Compressed image found at block 0
RAMDISK: incomplete write (-28 != 32768) 4194304
fs_names=ext2
root_mount_data=
VFS: Mounted root (ext2 filesystem).
Mounted devfs on /dev
Freeing init memory: 92K
attempt to access beyond end of device
ram0: rw=0, want=16518, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block - inode=1944, block=8258
Failed to execute /linuxrc. Attempting defaults...
attempt to access beyond end of device
ram0: rw=0, want=16514, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block - inode=1922, block=8256
Kernel panic - not syncing: No init found. Try passing init= option to kernel.
解决办法：修改内核关于Ramdisk的设置
make menuconfig
Device Drivers ->
Block Device->
<*>RAM disk support
(16)Default number of RAM disks 16改为8
(4096)Default RAM disk size(kbytes) 4096改为8192
板子的SDRAM是64M,16*4096kB = 8*8192kB = 64M
问题依然：
RAMDISK driver initialized: 8 RAM disks of 8192K size 1024 blocksize
Mounted devfs on /dev
Freeing init memory: 92K
attempt to access beyond end of device
ram0: rw=0, want=16518, limit=16384
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block - inode=1944, block=8258
Failed to execute /linuxrc. Attempting defaults...
attempt to access beyond end of device
ram0: rw=0, want=16514, limit=16384
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block - inode=1922, block=8256
Kernel panic - not syncing: No init found. Try passing init= option to kernel.
修改
Default number of RAM disks 值改为4
Default RAM disk size(kbytes) 值改为16384
问题解决
为什么是16384呢?16384kb=16M,我制作的ramdisk镜像大小为15M,这里的15M是未用gzip压缩前的大小.
另外关于commandline的值:initrd=0x30800040,0x400000 root=/dev/ram init=/linuxrc console=ttySAC0
initrd=0x30800040是ramdisk在SDRAM中的地址,我用mkimage将ramdisk加了64(0x40)字节头信息,制作成uboot可以识别的uRamdisk,然后下载到0x30800000,所以ramdisk的正真地址为0x30800040
0x400000(4M)为ramdisk的大小,要大于压缩后的大小,我制作好的ramdisk大小为15M，压缩后1.2M
root=/dev/ram 也可以是root=/dev/ram0
我在nand flash的第四个分区下载了cramfs文件系统,可以通过mount命令挂载该分区到tmp目录下
# cd dev
# ls
console kmem mmc ptmx rd tty vcc
fb kmsg mtdblock pts root urandom zero
full mem null pty shm usb
input misc port random tts vc
#mount /dev/mtdblock/3 /tmp
另一种方法可以在系统启动时就挂载该分区
修改/etc/init.d/rcS文件
内容如下(蓝色字体为添加的内容):
#!/bin/sh
echo "mount cramfs from mtdblock/3"
/bin/mount /dev/mtdblock/3 /tmp
echo "mount all"
/bin/mount –a
说明:/etc/init.d/rcS是可执行的二进制文件,系统启动时被执行
