Re: [PATCH v3 1/1] PCI: dwc: Fix index 0 incorrectly being interpreted as a free ATU slot

[Niklas Cassel <cassel@kernel.org>]

On Wed, Apr 10, 2024 at 02:16:52PM -0400, Frank Li wrote:
> On Wed, Apr 10, 2024 at 11:33:41PM +0530, Manivannan Sadhasivam wrote:
> > On Mon, Apr 08, 2024 at 12:23:20PM +0200, Niklas Cassel wrote:
> > > On Tue, Mar 26, 2024 at 03:35:40PM -0400, Frank Li wrote:
> > > > When PERST# assert and deassert happens on the PERST# supported platforms,
> > > > the both iATU0 and iATU6 will map inbound window to BAR0. DMA will access
> > > > to the area that was previously allocated (iATU0) for BAR0, instead of the
> > > > new area (iATU6) for BAR0.
> > > 
> > > Nit: If we want additional clarity, we could also add:
> > > ""
> > > Right now, we dodge the bullet because both iATU0 and iATU6 should currently
> > > translate inbound accesses to BAR0 to the same allocated memory area. However,
> > > having two separate inbound mappings for the same BAR is a disaster waiting to
> > > happen.
> > > ""
> > 
> > Since Bjorn asked for the above info, it should get added.
> > 
> > With that,
> > 
> > Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> 
> Mani:
> 
> Do you need me rework patch? Or you can handle it by yourself when apply?

It appears that Krzysztof is the one picking up patches for:
drivers/pci/controller/dwc/

Since he has the bottommost Signed-off-by on patches in:
https://git.kernel.org/pub/scm/linux/kernel/git/pci/pci.git/log/?h=controller%2Fdwc

So I don't think that Mani can amend it, since he is most likely not the
one picking up this patch.

The fastest way is probably just to send out a v4.
(But I feel you... this patch has already taken too long :P)


Kind regards,
Niklas