Subj : Re: What does Memory Barriers mean ??
To   : comp.programming.threads
From : SenderX
Date : Sat Jan 08 2005 09:38 am

>> I have had some compiler reordering issues a while ago when I was testing
>> and developing the RCU algorithm that my library uses. I disassembled 
>> some
>> key stuff and found that a hazards reference was being performed before 
>> the
>> RCU quiescent state changed to active. MSVC++ did not reorder the
>> instructions, but GCC on linux did.
>
> I bet the error was on your side.

Yes this was a problem with trying to use pure C with threads....


My RCU algorithm requires that the quiescent state change to "active" 
happens before "any" hazard reference can be attempted. GCC on Linux 
reordered the instructions so a hazard was executed before the GC was able 
to protect it. Kind of like a hazard pointer being set after the hazard 
reference was executed/accessed. You go boom after that....


;)

.