首页
社区
课程
招聘
[求助]用了自旋锁,数据还是丢失,蓝屏!
发表于: 2009-3-28 17:15 6227

[求助]用了自旋锁,数据还是丢失,蓝屏!

2009-3-28 17:15
6227
还是那个代码的问题,上一贴在:http://bbs.pediy.com/showthread.php?t=83735

后来我在插入链表时使用了自旋锁,用的是 ExInterlockedInsertTailList 函数,
在WINXP和WIN2003下,已经不会再出现链表中数据丢失的情况了,
但在WIN2000下,还是会出现,出现的环境是卡巴和DebugView都运行的情况下,
当数据丢失后,试图通过DebugView打印时蓝屏了。
但如果关闭卡巴和DebugView任意一个,则不会蓝屏,只是传递到用户层的数据是空的。

蓝屏如下:

 
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 50, {ff974cb0, 0, 80461e23, 0}
*** ERROR: Module load completed but symbols could not be loaded for Dbgv.sys
*** ERROR: Module load completed but symbols could not be loaded for klif.sys
*** WARNING: Unable to verify checksum for irp4exe.exe
*** ERROR: Module load completed but symbols could not be loaded for irp4exe.exe
*************************************************************************
***                                                                   ***
***                                                                   ***
***    Your debugger is not using the correct symbols                 ***
***                                                                   ***
***    In order for this command to work properly, your symbol path   ***
***    must point to .pdb files that have full type information.      ***
***                                                                   ***
***    Certain .pdb files (such as the public OS symbols) do not      ***
***    contain the required information.  Contact the group that      ***
***    provided you with these symbols if you need this command to    ***
***    work.                                                          ***
***                                                                   ***
***    Type referenced: kernel32!pNlsUserInfo                         ***
***                                                                   ***
*************************************************************************
Probably caused by : Dbgv.sys ( Dbgv+992 )
Followup: MachineOwner
---------
nt!RtlpBreakWithStatusInstruction:
80455558 cc              int     3
kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
PAGE_FAULT_IN_NONPAGED_AREA (50)
Invalid system memory was referenced.  This cannot be protected by try-except,
it must be protected by a Probe.  Typically the address is just plain bad or it
is pointing at freed memory.
Arguments:
Arg1: ff974cb0, memory referenced.
Arg2: 00000000, value 0 = read operation, 1 = write operation.
Arg3: 80461e23, If non-zero, the instruction address which referenced the bad memory
 address.
Arg4: 00000000, (reserved)
Debugging Details:
------------------
READ_ADDRESS:  ff974cb0 Nonpaged pool
FAULTING_IP: 
nt!_output+434
80461e23 66833800        cmp     word ptr [eax],0
MM_INTERNAL_CODE:  0
DEFAULT_BUCKET_ID:  DRIVER_FAULT
BUGCHECK_STR:  0x50
PROCESS_NAME:  irp4exe.exe
TRAP_FRAME:  be1204d4 -- (.trap 0xffffffffbe1204d4)
ErrCode = 00000000
eax=ff974cb0 ebx=80064b53 ecx=ff974cb0 edx=7fffffff esi=7ffffffe edi=00000001
eip=80461e23 esp=be120548 ebp=be12079c iopl=0         nv up ei pl nz na pe nc
cs=0008  ss=0010  ds=0023  es=0023  fs=0030  gs=0000             efl=00010206
nt!_output+0x434:
[COLOR=red]80461e23 66833800        cmp     word ptr [eax],0         ds:0023:ff974cb0=????[/COLOR]
Resetting default scope
LAST_CONTROL_TRANSFER:  from 8042a9eb to 80455558
STACK_TEXT:  
be1200b8 8042a9eb 00000003 be120100 ff974cb0 nt!RtlpBreakWithStatusInstruction
be1200e8 8042afde 00000003 c03fe5d0 80064b7c nt!KiBugCheckDebugBreak+0x31
be120474 80449d19 00000001 ff974cb0 00000000 nt!KeBugCheckEx+0x390
be1204bc 80467cbf 00000000 ff974cb0 00000000 nt!MmAccessFault+0x757
be1204bc 80461e23 00000000 ff974cb0 00000000 nt!KiTrap0E+0xc3
be12079c 80461626 be1207b4 f41a69de be120b00 nt!_output+0x434
be1207d4 be0e9992 be1207ec f41a69d0 be120afc nt!vsprintf+0x2d
WARNING: Stack unwind information not available. Following frames may be wrong.
be120af0 f41a6acd f41a69d0 ff974cb0 ff974cb0 Dbgv+0x992
be120bb8 8041ddf1 814ef030 815bef68 815bef68 irp4!MyDeviceIoControl+0xdd [e:\mydriver\irp\4\sys\irp4.c @ 172]
be120bcc 804ae9bc 815befd8 00000000 815bef68 nt!IopfCallDriver+0x35
be120be0 804af826 814ef030 815bef68 812e76e8 nt!IopSynchronousServiceTail+0x60
be120cb8 804a73ae 000003d4 00000000 00000000 nt!IopXxxControlFile+0x5e4
be120cec be6580ce 000003d4 00000000 00000000 nt!NtDeviceIoControlFile+0x28
be120d34 80465024 000003d4 00000000 00000000 klif+0x100ce
be120d34 77f88403 000003d4 00000000 00000000 nt!KiSystemService+0xc4
0012feb8 77e6951b 000003d4 00000000 00000000 ntdll!ZwDeviceIoControlFile+0xb
0012ff1c 00401072 000003d4 00222400 00000000 KERNEL32!DeviceIoControl+0xf8
0012ffc0 77e889d5 00000006 00000008 7ffdf000 irp4exe+0x1072
0012fff0 00000000 00401181 00000000 000000c8 KERNEL32!BaseProcessStart+0x3d

STACK_COMMAND:  kb
FOLLOWUP_IP: 
Dbgv+992
be0e9992 83c40c          add     esp,0Ch
SYMBOL_STACK_INDEX:  7
SYMBOL_NAME:  Dbgv+992
FOLLOWUP_NAME:  MachineOwner
MODULE_NAME: Dbgv
IMAGE_NAME:  Dbgv.sys
DEBUG_FLR_IMAGE_TIMESTAMP:  4436c4b6
FAILURE_BUCKET_ID:  0x50_Dbgv+992
BUCKET_ID:  0x50_Dbgv+992
Followup: MachineOwner
---------



请稍微指点一下!

[培训]内核驱动高级班,冲击BAT一流互联网大厂工作,每周日13:00-18:00直播授课

收藏
免费 0
支持
分享
最新回复 (8)
雪    币: 8835
活跃值: (2404)
能力值: ( LV12,RANK:760 )
在线值:
发帖
回帖
粉丝
2
DbgPrint不是一个和谐的函数~~
2009-3-28 17:43
0
雪    币: 8835
活跃值: (2404)
能力值: ( LV12,RANK:760 )
在线值:
发帖
回帖
粉丝
3
用KdPrint吧,windbg调试,比dbgview舒服~
而且不蓝屏~
2009-3-28 17:44
0
雪    币: 26
活跃值: (28)
能力值: ( LV2,RANK:10 )
在线值:
发帖
回帖
粉丝
4
既然有悖于当今时代主题?!可我的数据哪里去了?锁了已经。
上面的蓝屏分析也是在WinDbg双机调试模式下的。
可我还是想弄明白。
2009-3-28 17:46
0
雪    币: 26
活跃值: (28)
能力值: ( LV2,RANK:10 )
在线值:
发帖
回帖
粉丝
5
DDK编译环境是WinXP-Checked。
2009-3-29 00:20
0
雪    币: 354
活跃值: (10)
能力值: ( LV8,RANK:120 )
在线值:
发帖
回帖
粉丝
6
Windows XP and later versions of Windows provide optimized versions of the sequenced singly-linked list functions that are not available in Windows 2000. If your driver uses these functions and also must run with Windows 2000, the driver must define the _WIN2K_COMPAT_SLIST_USAGE flag, as follows:

#define _WIN2K_COMPAT_SLIST_USAGE

For x86-based processors, this flag causes the compiler to use versions of the sequenced singly-linked list functions that are compatible with Windows 2000.
2009-3-29 02:14
0
雪    币: 26
活跃值: (28)
能力值: ( LV2,RANK:10 )
在线值:
发帖
回帖
粉丝
7
朋友你说的是 Sequenced Singly-Linked Lists ,
但我代码中用的是 Doubly-Linked Lists 。
2009-3-29 10:00
0
雪    币: 26
活跃值: (28)
能力值: ( LV2,RANK:10 )
在线值:
发帖
回帖
粉丝
8
是什么原因导致ff974cb0这个地方数据消失呢?
即使数据消失了,为何访问它就蓝屏呢?
2009-3-29 23:53
0
雪    币: 26
活跃值: (28)
能力值: ( LV2,RANK:10 )
在线值:
发帖
回帖
粉丝
9
抓狂,还是分析不出来。
2009-3-30 22:11
0
游客
登录 | 注册 方可回帖
返回
//