Re: [PATCH v3 3/3] mmc: xenon: Fix 2G limitation on AC5 SoC

From: Vadym Kochan
Date: Tue Dec 13 2022 - 04:22:44 EST


Hi Linus, Adrian,

On Tue, 13 Dec 2022 10:16:57 +0100, Linus Walleij <linus.walleij@xxxxxxxxxx> wrote:
> On Mon, Dec 12, 2022 at 9:43 AM Adrian Hunter <adrian.hunter@xxxxxxxxx> wrote:
>
> > >> Right but I guess I meant what about 0x3_00000000..0x3_ffffffff ?
> > >> Isn't that also in DMA_BIT_MASK(34)
> > >
> > > Yes, you are right.
> >
> > So it would fail in that case? Is it possible to use devicetree
> > reserved memory or some such, to set aside 64k for the bounce
> > buffer DMA mapping?
>
> Yups spot on, reserved-memory can be used along with the kernel
> CONFIG_DMA_CMA to achieve exactly this:
> Documentation/devicetree/bindings/reserved-memory/reserved-memory.yaml
> kernel/dma/Kconfig
>
> Yours,
> Linus Walleij

Just in case, here is an old series with reserved-memory solution:
https://lore.kernel.org/lkml/20220806085818.9873-4-vadym.kochan@xxxxxxxxxxx/T/

But, what about to start with PIO solution (which is conceptually easier solution) with
checking on ram size and then develop better one ?

Thanks,