[Intel-wired-lan] [PATCH bpf RFC-V2 1/5] xdp: rss hash types representation

Jesper Dangaard Brouer jbrouer at redhat.com
Wed Mar 29 17:59:59 UTC 2023



On 29/03/2023 18.29, Jesper Dangaard Brouer wrote:
> The RSS hash type specifies what portion of packet data NIC hardware used
> when calculating RSS hash value. The RSS types are focused on Internet
> traffic protocols at OSI layers L3 and L4. L2 (e.g. ARP) often get hash
> value zero and no RSS type. For L3 focused on IPv4 vs. IPv6, and L4
> primarily TCP vs UDP, but some hardware supports SCTP.
> 
> Hardware RSS types are differently encoded for each hardware NIC. Most
> hardware represent RSS hash type as a number. Determining L3 vs L4 often
> requires a mapping table as there often isn't a pattern or sorting
> according to ISO layer.
> 
> The patch introduce a XDP RSS hash type (xdp_rss_hash_type) that can both
> be seen as a number that is ordered according by ISO layer, and can be bit
> masked to separate IPv4 and IPv6 types for L4 protocols. Room is available
> for extending later while keeping these properties. This maps and unifies
> difference to hardware specific hashes.
> 
> This proposal change the kfunc API bpf_xdp_metadata_rx_hash() to return
> this RSS hash type on success.
> 
> Signed-off-by: Jesper Dangaard Brouer <brouer at redhat.com>
> ---
>   include/net/xdp.h |   76 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>   net/core/xdp.c    |    4 ++-
>   2 files changed, 79 insertions(+), 1 deletion(-)
> 
> diff --git a/include/net/xdp.h b/include/net/xdp.h
> index 5393b3ebe56e..1b2b17625c26 100644
> --- a/include/net/xdp.h
> +++ b/include/net/xdp.h
> @@ -8,6 +8,7 @@
>   
>   #include <linux/skbuff.h> /* skb_shared_info */
>   #include <uapi/linux/netdev.h>
> +#include <linux/bitfield.h>
>   
>   /**
>    * DOC: XDP RX-queue information
> @@ -396,6 +397,81 @@ XDP_METADATA_KFUNC_xxx
>   MAX_XDP_METADATA_KFUNC,
>   };
>   
> +/* For partitioning of xdp_rss_hash_type */
> +#define RSS_L3		GENMASK(2,0) /* 3-bits = values between 1-7 */
> +#define L4_BIT		BIT(3)       /* 1-bit - L4 indication */
> +#define RSS_L4_IPV4	GENMASK(6,4) /* 3-bits */
> +#define RSS_L4_IPV6	GENMASK(9,7) /* 3-bits */
> +#define RSS_L4		GENMASK(9,3) /* = 7-bits - covering L4 IPV4+IPV6 */
> +#define L4_IPV6_EX_BIT	BIT(9)       /* 1-bit - L4 IPv6 with Extension hdr */
> +				     /* 11-bits in total */

Please ignore above lines in review ... they should have been deleted,
the new partitioning uses the enum/defines below.

> +
> +/* Lower 4-bits value of xdp_rss_hash_type */
> +enum xdp_rss_L4 {
> +	XDP_RSS_L4_MASK		= GENMASK(3,0), /* 4-bits = values between 0-15 */
> +	XDP_RSS_L4_NONE		= 0, /* Not L4 based hash */
> +	XDP_RSS_L4_ANY		= 1, /* L4 based hash but protocol unknown */
> +	XDP_RSS_L4_TCP		= 2,
> +	XDP_RSS_L4_UDP		= 3,
> +	XDP_RSS_L4_SCTP		= 4,
> +	XDP_RSS_L4_IPSEC	= 5, /* L4 based hash include IPSEC SPI */
> +/*
> + RFC: We don't care about vasting space, then we could just store the
> + protocol number (8-bits) directly. See /etc/protocols
> +	XDP_RSS_L4_TCP		= 6,
> +	XDP_RSS_L4_UDP		= 17,
> +	XDP_RSS_L4_SCTP		= 132,
> +	XDP_RSS_L4_IPSEC_ESP	= 50, // Issue: mlx5 didn't say ESP or AH
> +	XDP_RSS_L4_IPSEC_AH	= 51, // both ESP+AH just include SPI in hash
> + */
> +};
> +
> +/* Values shifted for use in xdp_rss_hash_type */
> +enum xdp_rss_L3 {
> +	XDP_RSS_L3_MASK		= GENMASK(5,4), /* 2-bits = values between 1-3 */
> +	XDP_RSS_L3_IPV4		= FIELD_PREP_CONST(XDP_RSS_L3_MASK, 1),
> +	XDP_RSS_L3_IPV6		= FIELD_PREP_CONST(XDP_RSS_L3_MASK, 2),
> +};
> +
> +/* Bits shifted for use in xdp_rss_hash_type */
> +enum xdp_rss_bit {
> +	XDP_RSS_BIT_MASK	= GENMASK(7,6), /* 2-bits */
> +	/* IPv6 Extension Hdr */
> +	XDP_RSS_BIT_EX = FIELD_PREP_CONST(XDP_RSS_BIT_MASK, BIT(0)),
> +	/* XDP_RSS_BIT_VLAN ??? = FIELD_PREP_CONST(XDP_RSS_BIT_MASK, BIT(1)), */
> +};
> +
> +/* RSS hash type combinations used for driver HW mapping */
> +enum xdp_rss_hash_type {
> +	XDP_RSS_TYPE_NONE            = 0,
> +	XDP_RSS_TYPE_L2              = XDP_RSS_TYPE_NONE,
> +
> +	XDP_RSS_TYPE_L3_MASK         = XDP_RSS_L3_MASK,
> +	XDP_RSS_TYPE_L3_IPV4         = XDP_RSS_L3_IPV4,
> +	XDP_RSS_TYPE_L3_IPV6         = XDP_RSS_L3_IPV6,
> +	XDP_RSS_TYPE_L3_IPV6_EX      = XDP_RSS_L3_IPV6 | XDP_RSS_BIT_EX,
> +
> +	XDP_RSS_TYPE_L4_MASK         = XDP_RSS_L4_MASK,
> +	XDP_RSS_TYPE_L4_ANY          = XDP_RSS_L4_ANY,
> +	XDP_RSS_TYPE_L4_IPV4_TCP     = XDP_RSS_L3_IPV4 | XDP_RSS_L4_TCP,
> +	XDP_RSS_TYPE_L4_IPV4_UDP     = XDP_RSS_L3_IPV4 | XDP_RSS_L4_UDP,
> +	XDP_RSS_TYPE_L4_IPV4_SCTP    = XDP_RSS_L3_IPV4 | XDP_RSS_L4_SCTP,
> +
> +	XDP_RSS_TYPE_L4_IPV6_TCP     = XDP_RSS_L3_IPV6 | XDP_RSS_L4_TCP,
> +	XDP_RSS_TYPE_L4_IPV6_UDP     = XDP_RSS_L3_IPV6 | XDP_RSS_L4_UDP,
> +	XDP_RSS_TYPE_L4_IPV6_SCTP    = XDP_RSS_L3_IPV6 | XDP_RSS_L4_UDP,
> +
> +	XDP_RSS_TYPE_L4_IPV6_TCP_EX  = XDP_RSS_TYPE_L4_IPV6_TCP |XDP_RSS_BIT_EX,
> +	XDP_RSS_TYPE_L4_IPV6_UDP_EX  = XDP_RSS_TYPE_L4_IPV6_UDP |XDP_RSS_BIT_EX,
> +	XDP_RSS_TYPE_L4_IPV6_SCTP_EX = XDP_RSS_TYPE_L4_IPV6_SCTP|XDP_RSS_BIT_EX,
> +};
> +#undef RSS_L3
> +#undef L4_BIT
> +#undef RSS_L4_IPV4
> +#undef RSS_L4_IPV6
> +#undef RSS_L4
> +#undef L4_IPV6_EX_BIT

All the undef's are also unncecessary now.

> +
>   #ifdef CONFIG_NET
>   u32 bpf_xdp_metadata_kfunc_id(int id);
>   bool bpf_dev_bound_kfunc_id(u32 btf_id);
> diff --git a/net/core/xdp.c b/net/core/xdp.c
> index 7133017bcd74..81d41df30695 100644
> --- a/net/core/xdp.c
> +++ b/net/core/xdp.c
> @@ -721,12 +721,14 @@ __bpf_kfunc int bpf_xdp_metadata_rx_timestamp(const struct xdp_md *ctx, u64 *tim
>    * @hash: Return value pointer.
>    *
>    * Return:
> - * * Returns 0 on success or ``-errno`` on error.
> + * * Returns (positive) RSS hash **type** on success or ``-errno`` on error.
> + * * ``enum xdp_rss_hash_type`` : RSS hash type
>    * * ``-EOPNOTSUPP`` : means device driver doesn't implement kfunc
>    * * ``-ENODATA``    : means no RX-hash available for this frame
>    */
>   __bpf_kfunc int bpf_xdp_metadata_rx_hash(const struct xdp_md *ctx, u32 *hash)
>   {
> +	BTF_TYPE_EMIT(enum xdp_rss_hash_type);
>   	return -EOPNOTSUPP;
>   }
>   
> 
> 



More information about the Intel-wired-lan mailing list