[Xen-devel] [PATCH] cpu barriers moved and x86-64 barriers add

[Jerone Young <jyoung5@xxxxxxxxxx>]

We actually should have wmb() be defined as "asm volatile("":::"memory")
since we are not using out of order io.

The patch attached removes all the instances and places the definitions
in libxc/xc.h as per the discussion last week between Anthony & Keir.


Signed-off-by: Jerone Young <jyoung5@xxxxxxxxxx>


On Tue, 2005-03-01 at 15:23 -0600, Anthony Liguori wrote:
> Keir Fraser wrote:
> 
> >
> > On 1 Mar 2005, at 19:56, Jerone Young wrote:
> >
> >>     I would like to discuss is this correct, dead wrong, or even 
> >> needed at
> >> all?
> >
> >
> > x86/64 has proper barrier instructions -- see 
> > include/asm-x86_64/system.h in Linux. It is from there that we should 
> > pull our definitions. Barrier macros are defined in a few places in 
> > the tools -- we ought to pull them all into one single header 
> > incorporated by all tools that need it.
> 
> I was just about to send this out myself :-)  I think the consensus was 
> to put them all in xc.h.  I posted a patch recently that made everything 
> include asm/system.h.  All it should take is replacing asm/system.h with 
> xc.h in the patch and then making the necessary changes to xc.h.
> 
> For reference, the proper x86-64 barriers are:
> 
> #define mb()     asm volatile("mfence":::"memory")
> #define rmb()    asm volatile("lfence":::"memory")
> #define wmb()    asm volatile("sfence":::"memory")
> 
> Regards,
> Anthony Liguori
> 
-- 
Jerone Young
Open Virtualization
IBM Linux Technology Center
jyoung5@xxxxxxxxxx
512-838-1157 (T/L: 678-1157)

Attachment: cpu_barrier_patch.diff
Description: Text Data