On Mon, Sep 30, 2024 at 02:06:12PM +0200, Jonas Oberhauser wrote:
>
>
> Am 9/30/2024 um 1:04 PM schrieb Paul E. McKenney:
> > On Mon, Sep 30, 2024 at 11:42:11AM +0200, Jonas Oberhauser wrote:
> > >
> > >
> > > I also currently don't see any major difference between the constant and
> > > register case. The point is that the address is known before loading into
> > > b,
> > > and hence the compiler + hardware can speculatively load *b before loading
> > > into b.
> >
> > In theory, true. In practice, in the register case, you need a little
> > more bad luck for the compiler to be able to exploit your mistake.
>
> If there's one thing I've never run out of, then it is bad luck with
> technology.
Careful! Someone might try to recruit you as tester. ;-)
Thanx, Paul
