[xdp-hints] Re: [PATCH bpf-next v2 12/20] xdp: Add checksum level hint

[Larysa Zaremba <larysa.zaremba@intel.com>]

On Thu, Jul 06, 2023 at 11:04:49AM +0200, Jesper Dangaard Brouer wrote:
> 
> 
> On 06/07/2023 07.50, John Fastabend wrote:
> > Larysa Zaremba wrote:
> > > On Tue, Jul 04, 2023 at 12:39:06PM +0200, Jesper Dangaard Brouer wrote:
> > > > Cc. DaveM+Alex Duyck, as I value your insights on checksums.
> > > > 
> > > > On 04/07/2023 11.24, Larysa Zaremba wrote:
> > > > > On Mon, Jul 03, 2023 at 01:38:27PM -0700, John Fastabend wrote:
> > > > > > Larysa Zaremba wrote:
> > > > > > > Implement functionality that enables drivers to expose to XDP code,
> > > > > > > whether checksums was checked and on what level.
> > > > > > > 
> > > > > > > Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com>
> > > > > > > ---
> > > > > > >    Documentation/networking/xdp-rx-metadata.rst |  3 +++
> > > > > > >    include/linux/netdevice.h                    |  1 +
> > > > > > >    include/net/xdp.h                            |  2 ++
> > > > > > >    kernel/bpf/offload.c                         |  2 ++
> > > > > > >    net/core/xdp.c                               | 21 ++++++++++++++++++++
> > > > > > >    5 files changed, 29 insertions(+)
> > > > > > > 
> > > > > > > diff --git a/Documentation/networking/xdp-rx-metadata.rst b/Documentation/networking/xdp-rx-metadata.rst
> > > > > > > index ea6dd79a21d3..4ec6ddfd2a52 100644
> > > > > > > --- a/Documentation/networking/xdp-rx-metadata.rst
> > > > > > > +++ b/Documentation/networking/xdp-rx-metadata.rst
> > > > > > > @@ -26,6 +26,9 @@ metadata is supported, this set will grow:
> > > > > > >    .. kernel-doc:: net/core/xdp.c
> > > > > > >       :identifiers: bpf_xdp_metadata_rx_vlan_tag
> > > > > > > +.. kernel-doc:: net/core/xdp.c
> > > > > > > +   :identifiers: bpf_xdp_metadata_rx_csum_lvl
> > > > > > > +
> > > > > > >    An XDP program can use these kfuncs to read the metadata into stack
> > > > > > >    variables for its own consumption. Or, to pass the metadata on to other
> > > > > > >    consumers, an XDP program can store it into the metadata area carried
> > > > > > > diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> > > > > > > index 4fa4380e6d89..569563687172 100644
> > > > > > > --- a/include/linux/netdevice.h
> > > > > > > +++ b/include/linux/netdevice.h
> > > > > > > @@ -1660,6 +1660,7 @@ struct xdp_metadata_ops {
> > > > > > >    			       enum xdp_rss_hash_type *rss_type);
> > > > > > >    	int	(*xmo_rx_vlan_tag)(const struct xdp_md *ctx, u16 *vlan_tag,
> > > > > > >    				   __be16 *vlan_proto);
> > > > > > > +	int	(*xmo_rx_csum_lvl)(const struct xdp_md *ctx, u8 *csum_level);
> > > > > > >    };
> > > > > > >    /**
> > > > > > > diff --git a/include/net/xdp.h b/include/net/xdp.h
> > > > > > > index 89c58f56ffc6..61ed38fa79d1 100644
> > > > > > > --- a/include/net/xdp.h
> > > > > > > +++ b/include/net/xdp.h
> > > > > > > @@ -391,6 +391,8 @@ void xdp_attachment_setup(struct xdp_attachment_info *info,
> > > > > > >    			   bpf_xdp_metadata_rx_hash) \
> > > > > > >    	XDP_METADATA_KFUNC(XDP_METADATA_KFUNC_RX_VLAN_TAG, \
> > > > > > >    			   bpf_xdp_metadata_rx_vlan_tag) \
> > > > > > > +	XDP_METADATA_KFUNC(XDP_METADATA_KFUNC_RX_CSUM_LVL, \
> > > > > > > +			   bpf_xdp_metadata_rx_csum_lvl) \
> > > > > > >    enum {
> > > > > > >    #define XDP_METADATA_KFUNC(name, _) name,
> > > > > > > diff --git a/kernel/bpf/offload.c b/kernel/bpf/offload.c
> > > > > > > index 986e7becfd42..a133fb775f49 100644
> > > > > > > --- a/kernel/bpf/offload.c
> > > > > > > +++ b/kernel/bpf/offload.c
> > > > > > > @@ -850,6 +850,8 @@ void *bpf_dev_bound_resolve_kfunc(struct bpf_prog *prog, u32 func_id)
> > > > > > >    		p = ops->xmo_rx_hash;
> > > > > > >    	else if (func_id == bpf_xdp_metadata_kfunc_id(XDP_METADATA_KFUNC_RX_VLAN_TAG))
> > > > > > >    		p = ops->xmo_rx_vlan_tag;
> > > > > > > +	else if (func_id == bpf_xdp_metadata_kfunc_id(XDP_METADATA_KFUNC_RX_CSUM_LVL))
> > > > > > > +		p = ops->xmo_rx_csum_lvl;
> > > > > > >    out:
> > > > > > >    	up_read(&bpf_devs_lock);
> > > > > > > diff --git a/net/core/xdp.c b/net/core/xdp.c
> > > > > > > index f6262c90e45f..c666d3e0a26c 100644
> > > > > > > --- a/net/core/xdp.c
> > > > > > > +++ b/net/core/xdp.c
> > > > > > > @@ -758,6 +758,27 @@ __bpf_kfunc int bpf_xdp_metadata_rx_vlan_tag(const struct xdp_md *ctx, u16 *vlan
> > > > > > >    	return -EOPNOTSUPP;
> > > > > > >    }
> > > > > > > +/**
> > > > > > > + * bpf_xdp_metadata_rx_csum_lvl - Get depth at which HW has checked the checksum.
> > > > > > > + * @ctx: XDP context pointer.
> > > > > > > + * @csum_level: Return value pointer.
> > > > > > > + *
> > > > > > > + * In case of success, csum_level contains depth of the last verified checksum.
> > > > > > > + * If only the outermost checksum was verified, csum_level is 0, if both
> > > > > > > + * encapsulation and inner transport checksums were verified, csum_level is 1,
> > > > > > > + * and so on.
> > > > > > > + * For more details, refer to csum_level field in sk_buff.
> > > > > > > + *
> > > > > > > + * Return:
> > > > > > > + * * Returns 0 on success or ``-errno`` on error.
> > > > > > > + * * ``-EOPNOTSUPP`` : device driver doesn't implement kfunc
> > > > > > > + * * ``-ENODATA``    : Checksum was not validated
> > > > > > > + */
> > > > > > > +__bpf_kfunc int bpf_xdp_metadata_rx_csum_lvl(const struct xdp_md *ctx, u8 *csum_level)
> > > > > > 
> > > > > > Istead of ENODATA should we return what would be put in the ip_summed field
> > > > > > CHECKSUM_{NONE, UNNECESSARY, COMPLETE, PARTIAL}? Then sig would be,
> > > > 
> > > > I was thinking the same, what about checksum "type".
> > > > 
> > > > > > 
> > > > > >    bpf_xdp_metadata_rx_csum_lvl(const struct xdp_md *ctx, u8 *type, u8 *lvl);
> > > > > > 
> > > > > > or something like that? Or is the thought that its not really necessary?
> > > > > > I don't have a strong preference but figured it was worth asking.
> > > > > > 
> > > > > 
> > > > > I see no value in returning CHECKSUM_COMPLETE without the actual checksum value.
> > > > > Same with CHECKSUM_PARTIAL and csum_start. Returning those values too would
> > > > > overcomplicate the function signature.
> > > > 
> > > > So, this kfunc bpf_xdp_metadata_rx_csum_lvl() success is it equivilent to
> > > > CHECKSUM_UNNECESSARY?
> > > 
> > > This is 100% true for physical NICs, it's more complicated for veth, bacause it
> > > often receives CHECKSUM_PARTIAL, which shouldn't normally apprear on RX, but is
> > > treated by the network stack as a validated checksum, because there is no way
> > > internally generated packet could be messed up. I would be grateful if you could
> > > look at the veth patch and share your opinion about this.
> > > 
> > > > 
> > > > Looking at documentation[1] (generated from skbuff.h):
> > > >   [1] https://kernel.org/doc/html/latest/networking/skbuff.html#checksumming-of-received-packets-by-device
> > > > 
> > > > Is the idea that we can add another kfunc (new signature) than can deal
> > > > with the other types of checksums (in a later kernel release)?
> > > > 
> > > 
> > > Yes, that is the idea.
> > 
> > If we think there is a chance we might need another kfunc we should add it
> > in the same kfunc. It would be unfortunate to have to do two kfuncs when
> > one would work. It shouldn't cost much/anything(?) to hardcode the type for
> > most cases? I think if we need it later I would advocate for updating this
> > kfunc to support it. Of course then userspace will have to swivel on the
> > kfunc signature.
> > 
> 
> I think it might make sense to have 3 kfuncs for checksumming.
> As this would allow BPF-prog to focus on CHECKSUM_UNNECESSARY, and then
> only call additional kfunc for extracting e.g csum_start  + csum_offset
> when type is CHECKSUM_PARTIAL.
> 
> We could extend bpf_xdp_metadata_rx_csum_lvl() to give the csum_type
> CHECKSUM_{NONE, UNNECESSARY, COMPLETE, PARTIAL}.
> 
>  int bpf_xdp_metadata_rx_csum_lvl(*ctx, u8 *csum_level, u8 *csum_type)
> 
> And then add two kfunc e.g.
>  (1) bpf_xdp_metadata_rx_csum_partial(ctx, start, offset)
>  (2) bpf_xdp_metadata_rx_csum_complete(ctx, csum)
> 
> Pseudo BPF-prog code:
> 
>  err = bpf_xdp_metadata_rx_csum_lvl(ctx, level, type);
>  if (!err && type != CHECKSUM_UNNECESSARY) {
>      if (type == CHECKSUM_PARTIAL)
>          err = bpf_xdp_metadata_rx_csum_partial(ctx, start, offset);
>      if (type == CHECKSUM_COMPLETE)
>          err = bpf_xdp_metadata_rx_csum_complete(ctx, csum);
>  }
> 
> Looking at code, I feel we could rename [...]_csum_lvl to csum_type.
> E.g. bpf_xdp_metadata_rx_csum_type.
>

What about:

union csum_info {
	struct {
		u16 csum_start;
		u16 csum_offset;
	};
	u32 checksum;
	u8 checksum_level;
};

bpf_xdp_metadata_rx_csum(*ctx, u8 *csum_status, union csum_info *info);

One thing that is worth considering in my opinion is whether some hardware can 
provide both CHECKSUM_UNNECESSARY and CHECKSUM_COMPLETE. Judging by [0], this 
does occur. I such cases using an enum to represent the checksum status would 
artificially limit the capabilities. Now, imagine the situation:

- You want to use your XDP program with 2 different NICs

[...]

err = bpf_xdp_metadata_rx_csum(*ctx, &status, &info);
if (!err && status == CHECKSUM_UNNECESSARY)
	/* Do stuff */

[...]
- One NIC can both calculate CHECKSUM_COMPLETE and parse headers, another one 
  is only able to parse headers. Those can be very similar NICs from different 
  generation.
- You test your program on the simpler NIC, program works fine.
- You tests your program on the more advanced one and suddenly you need an 
  'else if' case with some additional calculations.

Please write, whether this makes sense :D and if so, we can work out a solution.

> Feel free to disagree,
> --Jesper
> 
>