kernel BUG at page_alloc.c:213

billyhhzh

kernel BUG问题
运行一个程序，运行到memfree少于7400KB时，就出错，试了很多次都这样；
提示如下：

kernel BUG at page_alloc.c:213!
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c3610000
*pgd = 00000000, *pmd = 00000000
Internal error: Oops: ffffffff
CPU: 0
pc : [<c0021ad4>] lr : [<c002955c>] Not tainted
sp : c367fe64 ip : c367fe1c fp : c367fe74
r10: 00000aff r9 : 00000000 r8 : 00000001
r7 : 00000000 r6 : c01f4634 r5 : c01f461c r4 : 00000000
r3 : 00000000 r2 : 00000001 r1 : 00000001 r0 : 00000001
Flags: nZCv IRQs on FIQs on Mode SVC_32 Segment user
Control: C000317F Table: 33610000 DAC: 00000015
Process ac483mainloop (pid: 63, stackpage=c367f000)
Stack: (0xc367fe54 to 0xc3680000)
fe40: c002955c c0021ad4 60000013
fe60: ffffffff c026afc0 c367fea4 c367fe78 c0042b70 c0021aa0 20000013 00000101
fe80: c01f4838 00000000 00000000 000001d2 c01f4834 c01f461c c367fed4 c367fea8
fea0: c0042ed8 c004292c 00936000 00000000 00936000 00000000 c367ff20 c0945100
fec0: 00000000 00000340 c367fee4 c367fed8 c0042bdc c0042e94 c367ff50 c367fee8
fee0: c003daec c0042bcc ffffffff c367e000 00001000 c026b158 c03678d8 00000936
ff00: c0945168 00000340 fffffff4 00000300 c09451b0 c37d4940 40199300 c37d4920
ff20: 00000000 d567ff58 c0945100 c37d4920 ffffffea 00000000 40199000 c367e000
ff40: 4012726c c367ff68 c367ff54 c008e2c0 c003d6c8 00000640 c37d4920 c367ff78
ff60: c367ff6c c008e2a0 c008e2b4 c367ffa4 c367ff7c c0049968 c008e28c 4012726c
ff80: c367ffac 00000640 40199000 00000008 00000004 c001c9a4 00000000 c367ffa8
ffa0: c001c820 c004989c 00000640 c0022b28 00000008 40199000 00000640 00000000
ffc0: 00000640 40199000 00000008 00000640 00000001 02000cdc 4012726c bffffcdc
ffe0: 4003cd2c bffffcac 4002cab4 400ddf34 60000010 00000008 00000000 00000000
Backtrace:
Function entered at [<c0021a90>] from [<c0042b70>]
r4 = C026AFC0
Function entered at [<c004291c>] from [<c0042ed8>]
Function entered at [<c0042e84>] from [<c0042bdc>]
Function entered at [<c0042bbc>] from [<c003daec>]
Function entered at [<c003d6b8>] from [<c008e2c0>]
Function entered at [<c008e2a4>] from [<c008e2a0>]
r5 = C37D4920 r4 = 00000640
Function entered at [<c008e27c>] from [<c0049968>]
Function entered at [<c004988c>] from [<c001c820>]
r8 = C001C9A4 r7 = 00000004 r6 = 00000008 r5 = 40199000
r4 = 00000640
Code: eb001df9 e59f0014 eb001df7 e3a03000 (e5833000)
release ok

kernel BUG at page_alloc.c:213!
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c3610000
*pgd = 00000000, *pmd = 00000000
Internal error: Oops: ffffffff
CPU: 0

cat /proc/meminfo如下：

/proc #
/proc # cat meminfo

total: used: free: shared: buffers: cached:
Mem: 66924544 59400192 7524352 0 270336 48812032
Swap: 0 0 0
MemTotal: 65356 kB
MemFree: 7348 kB
MemShared: 0 kB
Buffers: 264 kB
Cached: 47668 kB
SwapCached: 0 kB
Active: 4864 kB
Inactive: 43400 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 65356 kB
LowFree: 7348 kB
SwapTotal: 0 kB
SwapFree: 0 kB
/proc #
/proc #
/proc #

主要是大量的使用fopen/fwrite/fclose的原因,消耗mem很快,半个小时就消耗完了;
kernel 2.4对内存回收是否有bug,要怎么解决?用kernel2.6可以解决吗?
急,请大侠们出手帮忙!谢谢!