RE: [PATCH v4 0/2] PCI/P2PDMA: Allow P2PDMA on Intel Xeon server CPUs

[Popov, Pavel E]

> Once the header files are properly exposed, clean up these other areas
> as well.

Agreed, that would be a separate change, for the x86 maintainers to
decide.

Thanks

> -----Original Message-----
> From: Leon Romanovsky <leon@xxxxxxxxxx>
> Sent: Wednesday, October 21, 2026 12:35 AM
> To: Popov, Pavel E <pavel.e.popov@xxxxxxxxx>
> Cc: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>; Logan Gunthorpe
> <logang@xxxxxxxxxxxx>; Lukas Wunner <lukas@xxxxxxxxx>; Jim Chow
> <jim.chow@xxxxxxxxxxxx>; Radu Rugina <radu.rugina@xxxxxxxxxxxx>;
> Alexey Makhalov <alexey.makhalov@xxxxxxxxxxxx>; Wei Liu
> <wei.liu@xxxxxxxxxx>; Michael Kelley <mhklinux@xxxxxxxxxxx>; Nathan
> Ciobanu <nathan.d.ciobanu@xxxxxxxxxxxxxxx>; bcm-kernel-feedback-
> list@xxxxxxxxxxxx; virtualization@xxxxxxxxxxxxxxx; linux-
> hyperv@xxxxxxxxxxxxxxx; linux-pci@xxxxxxxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v4 0/2] PCI/P2PDMA: Allow P2PDMA on Intel Xeon
> server CPUs
> 
> On Tue, Oct 20, 2026 at 07:37:32PM +0000, Popov, Pavel E wrote:
> > Leon,
> >
> > The #ifdef CONFIG_X86 in cpu_supports_p2pdma() came with the AMD
> check.
> 
> This is due to poking at `struct cpuinfo_x86`, which is not ideal but
> understandable.
> 
> >
> > Would it work for you if I move the includes into that single #ifdef
> > block, next to the CPU check? Some multi-arch code does the same:
> >
> >   https://git.kernel.org/torvalds/c/c7be64355fcc (drivers/acpi/mipi-disco-
> img.c)
> >   https://git.kernel.org/torvalds/c/7f3cfb7943d2
> > (drivers/cpufreq/cpufreq_ondemand.h)
> 
> Once the header files are properly exposed, clean up these other areas as well.
> 
> Thanks
> 
> >
> > Thanks,
> > Pavel
> >
> > > -----Original Message-----
> > > From: Leon Romanovsky <leon@xxxxxxxxxx>
> > > Sent: Tuesday, October 20, 2026 11:40 AM
> > > To: Popov, Pavel E <pavel.e.popov@xxxxxxxxx>
> > > Cc: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>; Logan Gunthorpe
> > > <logang@xxxxxxxxxxxx>; Lukas Wunner <lukas@xxxxxxxxx>; Jim Chow
> > > <jim.chow@xxxxxxxxxxxx>; Radu Rugina
> <radu.rugina@xxxxxxxxxxxx>;
> > > Alexey Makhalov <alexey.makhalov@xxxxxxxxxxxx>; Wei Liu
> > > <wei.liu@xxxxxxxxxx>; Michael Kelley <mhklinux@xxxxxxxxxxx>; Nathan
> > > Ciobanu <nathan.d.ciobanu@xxxxxxxxxxxxxxx>; bcm-kernel-feedback-
> > > list@xxxxxxxxxxxx; virtualization@xxxxxxxxxxxxxxx; linux-
> > > hyperv@xxxxxxxxxxxxxxx; linux-pci@xxxxxxxxxxxxxxx; linux-
> > > kernel@xxxxxxxxxxxxxxx
> > > Subject: Re: [PATCH v4 0/2] PCI/P2PDMA: Allow P2PDMA on Intel Xeon
> > > server CPUs
> > >
> > > On Tue, Oct 20, 2026 at 10:37:34AM -0700, Pavel Popov wrote:
> > > > VMware and Hyper-V expose passthrough devices to guest VMs without
> > > > an explicit host bridge device. Consequently,
> > > > host_bridge_whitelist() cannot automatically identify the
> > > > underlying host bridge structure, causing Peer-to-Peer DMA
> > > > (P2PDMA) requests to be rejected unless the device IDs are
> > > > explicitly listed in pci_p2pdma_whitelist[]. To enable P2PDMA on
> > > > Intel hosts, this series extends cpu_supports_p2pdma(), which so
> > > > far covers AMD Zen and newer, with an explicit list of Intel Xeon server
> generations.
> > > >
> > > > Patch 1 moves the CONFIG_X86 #ifdef out of cpu_supports_p2pdma().
> > > > Patch 2 is the v3 patch rebased on it, with Logan's Reviewed-by
> > > > carried over; the Xeon list and the check are unchanged.
> > > >
> > > > Changes in v4:
> > > > - Add 1/2, moving the CONFIG_X86 #ifdef out of
> cpu_supports_p2pdma()
> > > >   (Logan Gunthorpe, Leon Romanovsky)
> > > > - Keep the #ifdef around the asm/ includes, as these headers exist
> > > >   only on x86 (Bjorn Helgaas)
> > >
> > > I'm confident that this is wrong usage of asm headers.
> > >
> > > Thanks