Re: [PATCH v11 00/19] PCI/P2PDMA: Route peer-to-peer DMA by TLP class

[Leon Romanovsky]

On Mon, Oct 19, 2026 at 02:18:45PM -0500, Bjorn Helgaas wrote:
> On Mon, Oct 19, 2026 at 09:35:37PM +0300, Leon Romanovsky wrote:
> > On Wed, Oct 14, 2026 at 05:49:53PM +0300, Leon Romanovsky wrote:
> > > On Wed, Oct 14, 2026, at 16:23, Bjorn Helgaas wrote:
> > > > On Wed, Oct 14, 2026 at 09:53:47AM +0300, Leon Romanovsky wrote:
> > > >> On Tue, Oct 13, 2026 at 05:25:08PM +0300, Leon Romanovsky wrote:
> > > >> > PCI P2PDMA applies Request and Completion Redirect throughout both paths.
> > > >> > This misclassifies asymmetric and nested switches, and reports one answer
> > > >> > for every kind of TLP.
> > > >> > 
> > > >> > Three ACS controls act on TLP attributes the Requester chooses rather
> > > >> > than on the topology: Translation Blocking and Direct Translated P2P act on
> > > >> > a Request's Address Type, and Completion Redirect skips Completions carrying
> > > >> > Relaxed Ordering.
> > > >> > 
> > > >> > Evaluate each direction at the path divergence, decide every class from the
> > > >> > one walk, and treat a client with ATS enabled as translating unless its
> > > >> > driver declares per-mapping ATS.
> > > >> > 
> > > >> > This completes the P2PDMA side; dma-buf and mlx5 follow separately.
> > > >> > 
> > > >> > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxx>
> > > >> > ---
> > > >> > Changes in v11:
> > > >> 
> > > >> 
> > > >> Bjorn,
> > > >> 
> > > >> Could you please save me from the AI-bot attack? Every rebase seems to
> > > >> uncover another bizarre pre-existing issue. It never ends.
> > > >
> > > > Can you elaborate?  I'm not sure what you'd like me to do here.  I
> > > > don't control the Sashiko reviews, and I thought Sashiko was recently
> > > > changed to put pre-existing issues in a database but not in the email
> > > > responses.
> > > 
> > > AI reviews are not deterministic and every resubmission gives us slightly different result, which is in this phase of submission is  full of hallucinated flows not applicable to the upstream.
> > > 
> > > In addition to upstream Sashiko, i run everything through local one with slightly better models which hallucinating even more.
> > > 
> > > I would ask you to merge the series so it gets real exposure in addition to my kunit and specific hw setup.
> > 
> > Bjorn,
> > 
> > We're in the merge window now, so I don't expect such a large series to be
> > merged. What about the next cycle? Should I send v12 after -rc1, or wait?
> 
> After -rc1 would be good.

Thanks