Re: frlock and barrier discussion

From: Andrea Arcangeli (andrea@suse.de)
Date: Thu Jan 30 2003 - 19:57:47 EST


On Thu, Jan 30, 2003 at 04:41:20PM -0800, Richard Henderson wrote:
> On Thu, Jan 30, 2003 at 02:52:19AM +0100, Andrea Arcangeli wrote:
> > you're missing xtimensec is a write.
>
> Eh? xtimensec is a register.

then it's fine, the register will be flushed to ram eventually and it
will be within the two wmb(), just write in the example also the line
where the xtimensec ""register"" is flushed to ram and it will be in the
right place

if the register isn't flushed to ram eventually, it will be discared and
the whole critical section is a noop from the point of view of the other
cpus and no wmb() or rmb() or mb() would be needed in the first place

Andrea
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Jan 31 2003 - 22:00:24 EST