Thx Arnd,
On Tue, Jul 30, 2019 at 9:29 PM Arnd Bergmann <arnd@xxxxxxxx> wrote:
>
> On Tue, Jul 30, 2019 at 2:15 PM <guoren@xxxxxxxxxx> wrote:
> >
> > From: Guo Ren <ren_guo@xxxxxxxxx>
> >
> > If arch didn't define dma_r/wmb(), linux will use w/rmb instead. Csky
> > use bar.xxx to implement mb() and that will cause problem when sync data
> > with dma device, becasue bar.xxx couldn't guarantee bus transactions
> > finished at outside bus level. We must use sync.s instead of bar.xxx for
> > dma data synchronization and it will guarantee retirement after getting
> > the bus bresponse.
> >
> > Signed-off-by: Guo Ren <ren_guo@xxxxxxxxx>
>
> This change looks good to me, but I think your regular barriers
> (mb, rmb, wmb) are still wrong: These are meant to be the superset
> of dma_{r,w}mb and smp_{,r,w}mb, and synchronize
> against both SMP and DMA interactions.
Wow, thanks for correction, yes! mb, rmb, wmb is the superset of dma
and smp, and I should use sync.s for mb.
I check the arm64, it use dsb for mb and dmb for smp_mb and dma_mb.
I also check the drivers, you are right , a lot of them use mb() to
sync mem data with device.
>
> I suspect you can drop the '.s' for non-SMP builds. What I don't
> see is whether you might need to add '.i' for dma_wmb() or
> dma_rmb().
It's no need for sky, when there is non-SMP, the cpu is in reset mode,
and sync.s is equal to sync in efficient.
--
Best Regards
Guo Ren
ML: https://lore.kernel.org/linux-csky/
