[cpp-threads] modes, pass 2

[Alexander Terekhov]

On 5/9/05, Doug Lea <dl at cs.oswego.edu> wrote:
> 
> >
> > No. I mean
> >On 5/9/05, Doug Lea <dl at cs.oswego.edu> wrote:
> 
> >
> > No. I mean
> >
> > ----
> > "However, an ExitEnter (StoreLoad) barrier is in general still
> >  needed between successive MonitorExits and MonitorEnters. "
> 
> The underlying need here is (of course) for at least one StoreLoad
> between a lock release and a lock acquire.

That's (only) because revised Java does RCsc, not RCpc. Right?

Can we agree once and for all that MT memory model in threaded 
C/C++ (and sorta "revised" pthreads/POSIX as well) will resemble 
RCpc, not RCsc?

Please.

> I should revise that passage because the example doesn't
> illustrate the intended issue -- that If you don't know for sure you
> have it on one side, you need it on the other side.
> So the "in general" disclaimer covers only those cases
> where you don't know if a lock has been used correctly
> (i.e., the monitorEnter/monitorExits aren't properly nested)
> which is (as of JVMS spec v2) not even legal.
> 
> >>
> >>Every time a question like this has arisen about PPC, I have asked
> >>people at IBM. And I generally revise cookbook with the answers I get,
> >>which have not always been consistent. I keep thinking that I should
> >>just kill all mention of PPC rather than try to take responsibility for
> >>it. If anyone would like to help collect the authoritative answers here,
> >>please feel free to help me out!
> >
> >
> > No comment. ;-) Well, but if you have access to IBM's w3forums, see my
> > "PMFJI" message in "Apparent memory consistency in SMP Power5
> > environment" thread at forums.hardware.powerpc...
> >
> 
> I don't think I do.
> If you think it would help clear things up for the (many!) people
> who read cookbook, could you please ask permission to send
> me whatever clarifications this entails?

Well, I've just rambled a bit in reply to John McCalpin's message 
about memory model on Power and lwsync + isync. I'll ask him if 
it's OK to <Forward Quoted> his message to this list.

regards,
alexander.