Re: [PATCH v2 02/17] psp: base PSP device support

[Willem de Bruijn <willemdebruijn.kernel@gmail.com>]

Jakub Kicinski wrote:
> On Thu, 26 Jun 2025 10:25:11 -0400 Willem de Bruijn wrote:
> > Preferable over the following?
> > 
> > 	struct psphdr {
> > 		u8      nexthdr;
> > 		u8      hdrlen;
> > 		u8      crypt_offset;
> > 
> > 		u8      sample:1;
> > 		u8      drop:1;
> > 		u8	version:4;
> > 		u8	vc_present:1;
> > 		u8	reserved:1;
> > 
> > 		__be32  spi;
> > 		__be64  iv;
> > 		__be64  vc[]; /* optional */
> > 	};
> > 
> > I suppose that has an endianness issue requiring
> > variants with __LITTLE_ENDIAN_BITFIELD and
> > __BIG_ENDIAN_BITFIELD.
> 
> Right, this part. Always gives me pause :(
> 
> > > > This makes sense with a single physical device plus optional virtual
> > > > (vlan, bonding, ..) devices.
> > > > 
> > > > It may also be possible for a single physical device (with single
> > > > device key) to present multiple PFs and/or VFs. In that case, will
> > > > there be multiple struct psp_dev, or will one PF be the "main".  
> > > 
> > > AFAIU we have no ability to represent multi-PCIe function devices 
> > > in the kernel model today. So realistically I think psp_dev per
> > > function and then propagate the rotation events.  
> > 
> > IDPF does support multiple "vports" (num_alloc_vports), and with that
> > struct net_device, from a single BDF.
> 
> Upstream? If yes then I'm very bad at reviewing code :D

I then don't think I want to focus your attention on this, but..

See use num_alloc_ports in idpf_init_task. Which keeps requeueing
itself until num_default_vports is reached. Which is a variable
received from the device in VIRTCHNL2_OP_GET_CAPS.