Hello,
I am the developer of a driver for a network adapter device that does scatter-gather DMA. Driver is not newly developed, but has had some recent changes.
My client has a puzzling BSOD on one x64 machine after about 18 hours of successful operation. My x64 Win 7 machine runs fine forever. Client's box crashes in seemingly random places that indicate mem corruption. Mem test on machine was clean. Minidump is attached. WinDbg stack doesn't trace back to my code - it just shows the calls to handle the exception. Any clues y'all can see?
System info from WinDbg:
Windows 7 Kernel Version 7601 (Service Pack 1) MP (4 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 7601.17640.amd64fre.win7sp1_gdr.110622-1506
One difference between client's x64 and mine: Theirs has a valid DMA MapRegisterBase pointer from a AllocateAdapterChannel call, where mine (and my x86 systems) always have a null DMA MapRegisterBase pointer.
Thanks.
I am the developer of a driver for a network adapter device that does scatter-gather DMA. Driver is not newly developed, but has had some recent changes.
My client has a puzzling BSOD on one x64 machine after about 18 hours of successful operation. My x64 Win 7 machine runs fine forever. Client's box crashes in seemingly random places that indicate mem corruption. Mem test on machine was clean. Minidump is attached. WinDbg stack doesn't trace back to my code - it just shows the calls to handle the exception. Any clues y'all can see?
System info from WinDbg:
Windows 7 Kernel Version 7601 (Service Pack 1) MP (4 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 7601.17640.amd64fre.win7sp1_gdr.110622-1506
One difference between client's x64 and mine: Theirs has a valid DMA MapRegisterBase pointer from a AllocateAdapterChannel call, where mine (and my x86 systems) always have a null DMA MapRegisterBase pointer.
Thanks.
