Path: news.mitre.org!blanket.mitre.org!nntprelay.mathworks.com!news-peer-east.sprintlink.net!news-peer.sprintlink.net!news-backup-west.sprintlink.net!news-in-west.sprintlink.net!news.sprintlink.net!Sprint!209.90.0.8!alpha.sky.net!newshub.cts.com!newsfeed.cts.com!cmkrnl!jeh
From: jeh@cmkrnl.com (Jamie Hanrahan)
Newsgroups: comp.os.ms-windows.programmer.nt.kernel-mode
Subject: Re: multiprocessor issue
Message-ID: <1998Jan25.113218.8445@cmkrnl>
Date: 25 Jan 98 11:32:18 PST
References: <34CB6BE2.25BF@tekram.com.tw>
Organization: Kernel Mode Systems, San Diego, CA
Lines: 27

In article <34CB6BE2.25BF@tekram.com.tw>, pkc <peng@tekram.com.tw> writes:
> Hi there,
> 
> I have written a (booting)device driver for NT 4.0. It works well
> under single processor kernel. But in dual processor environment, this
> driver traps when calling KeWaitForSingleObject(...) from within a
> thread created in my driver DriverEntry() routine. The thread priority
> has been changed to LOW_REALTIME_PRIORITY at the entrance, and I have
> tried to call KeAcquireSpinLock(...) before calling
> KeWaitForSingleObject(...), thought I don't know if this is necessary.
> 
> So how do i get around this?

Simple - DON'T call KeAcquireSpinLock (or KeRaiseIrql) before calling 
KeWaitForEtc.  

KeAcquireSpinLock raises IRQL to DISPATCH_LEVEL.  You can't enter a
wait from >= DISPATCH_LEVEL (though you can call KeWaitForEtc. from ==
DISPATCH_LEVEL if you know that it won't actually wait, e.g. if you
specify a timeout period of 0.) 

	--- Jamie Hanrahan, Kernel Mode Systems, San Diego CA
        Internet: jeh@cmkrnl.com (JH645)  CompuServe: 74140,2055  
drivers, internals, networks, applications, and training for VMS and Windows NT
NT driver FAQ, links, and other information:  http://www.cmkrnl.com/

If you post a reply in news, please don't e-mail it too.