STM32F103 ucLinux开发之三（内核启动后不正常）（完结）

STM32F103 ucLinux内核没有完全启动

从BOOT跳转到内核后，执行一长段的汇编语言，然后来到startkernel函数，开启C语言之旅。

但是内核输出不正常，如下所示：

Linux version 2.6.26-uc0 ([email protected]) (gcc version 4.3.3 (Sourcery G++ Lite 2009q1-163) ) #2 Wed Sep 2 10:09:23 UTC 2009
CPU: ARMv7-M Processor [411fc231] revision 1 (ARMv?(11)M)
Machine: STM3210E-EVAL
Warning: bad configuration page, trying to continue
SRAM Config: bank[0] @ 0x68000000 (size: 1024KB) - bank[1] @ 0x20000000 (size: 64KB).
Ignoring unrecognised tag 0x00000000
Built 1 zonelists in Zone order, mobility grouping off. Total pages: 254
Kernel command line:
PID hash table entries: 16 (order: 4, 64 bytes)
console [ttyS0] enabled
Dentry cache hash table entries: 1024 (order: 0, 4096 bytes)
Inode-cache hash table entries: 1024 (order: 0, 4096 bytes)
Bad page state in process ‘swapper‘
page:68901180 flags:0x2f55030e mapping:00000000 mapcount:0 count:0
Trying to fix it up, but a reboot is needed
Backtrace:
Bad page state in process ‘swapper‘
page:689011c0 flags:0x08a8420a mapping:00000000 mapcount:0 count:134218033
Trying to fix it up, but a reboot is needed
Backtrace:
Bad page state in process ‘swapper‘
page:689011c0 flags:0x08a8420a mapping:00000000 mapcount:0 count:134218033
Trying to fix it up, but a reboot is needed
Backtrace:
Bad page state in process ‘swapper‘
page:689011e0 flags:0x2eb8430e mapping:00000000 mapcount:0 count:134218033
Trying to fix it up, but a reboot is needed
Backtrace:
Bad page state in process ‘swapper‘
page:689011e0 flags:0x2eb8430e mapping:00000000 mapcount:0 count:134218033
Trying to fix it up, but a reboot is needed
Backtrace:
Bad page state in process ‘swapper‘
page:68901200 flags:0x0185000c mapping:00000000 mapcount:0 count:0
Trying to fix it up, but a reboot is needed
Backtrace:
CPU: 0 Tainted: G B (2.6.26-uc0 #2)
pc : [<08039bc2>] lr : [<08039cb1>] psr: 0100000b
sp : 20001f50 ip : 00000000 fp : 68901000
r10: 2000c28c r9 : 00000006 r8 : 00900000
r7 : 00000000 r6 : 00020000 r5 : 0c955a82 r4 : 68901220
r3 : 2984a715 r2 : ffffffff r1 : 9e375c8e r0 : 689
01220

跟踪调试内核，发现在mem_init→ free_all_bootmem_node→  free_all_bootmem_core→  __free_pages_bootmem→ __free_page→ __free_pages

→free_hot_page→ free_hot_cold_page→ free_pages_check→ bad_page

在bad_pages函数最终输出了下面的信息

Bad page state in process ‘swapper‘
page:68901180 flags:0x2f55030e mapping:00000000 mapcount:0 count:0
Trying to fix it up, but a reboot is needed
Backtrace:

而导致上面的原因是在free_all_bootmem_core函数中

pfn_to_page→ __pfn_to_page（根据推断是它，但看编译又不是）

#define __pfn_to_page(pfn) (mem_map + ((pfn) - ARCH_PFN_OFFSET))
#define ARCH_PFN_OFFSET PHYS_PFN_OFFSET
#define PHYS_PFN_OFFSET (PHYS_OFFSET >> PAGE_SHIFT)
#define PHYS_OFFSET (CONFIG_DRAM_BASE)
#define CONFIG_DRAM_BASE 0x20000000

struct page *mem_map，page结构体占用32个字节，mem_map的地址是0x6800 0000，ARCH_PFN_OFFSET根据上面的推算为2000 0

若pfn是6800 0，则内存0x6800 0000对应的页管理地址是

0x6800 0000 +(6800 0-2000 0)*32 = 6890 0000

而6890 0000这个地址实际是不存在的。

综上所述，很奇怪ST官网上的开发板可以正常运行。

我做了如下尝试：

1、将官网上的BOOT、内核下载到自己的设备上，串口输出乱码。原因是官网上是无源的8M晶振，而我们使用的是有源的11.0592M，

     这样根据BOOT配置，系统主频变成了99M，但是串口的波特率应该不会改变啊，输出却是乱码，试了几个波特率都不对。

2、内核中，在调用控制台初始化时，我对串口1重新配置，修改它的波特率，仍然不起作用，输出乱码。

3、更换晶振，但是手边没有8M有源的，无源又需要电容搭接电路等，现在暂时没有试验出来。

STM32F103 ucLinux开发之三（内核启动后不正常）（完结）,古老的榕树,5-wow.com