Subj : Re: What is the real costs of LOCK on x86 multiprocesor machine?
To   : comp.programming.threads
From : David Schwartz
Date : Sun Jul 31 2005 02:48 pm


"chris noonan" <usenet@leapheap.co.uk> wrote in message 
news:1122812800.698627.9550@g43g2000cwa.googlegroups.com...

>>     By the way, the way I do this in applications is by using an illegal
>> instruction. I catch the fault and replace the illegal instruction with
>> either a LOCK or  NOP at run time.
>>
>>[code snipped]

> Ingenious. I wonder if there might be a way of doing
> this for Microsoft Windows?

    I never tried. It comes down to whether you can catch the illegal 
instruction fault with enough information to fix it. My understanding is 
that you *can* catch faults with something like SetUnhandledExceptionFilter 
(or some function like that).

    DS

.