On Tuesday 22 February 2011 09:59:21 Marti Raudsepp wrote:
> On Tue, Feb 22, 2011 at 07:38, Fujii Masao <masao.fujii@gmail.com> wrote:
> > + SpinLockAcquire(&WalSndCtl->ctlmutex);
> > + result = WalSndCtl->sync_rep_service_available;
> > + SpinLockRelease(&WalSndCtl->ctlmutex);
> >
> > volatile pointer needs to be used to prevent code rearrangement.
>
> I don't think that's necessary. Spinlock functions already prevent
> reordering using __asm__ __volatile__
>
> Otherwise, surely they would be utterly broken?
Its not the spinlock thats the problem but that "result" may be already loaded
into a register. Thats not prohibited by __asm__ __volatile__.
Andres