XDP hardware hints discussion mail archive
 help / color / mirror / Atom feed
From: Larysa Zaremba <larysa.zaremba@intel.com>
To: Alexander Lobakin <aleksander.lobakin@intel.com>
Cc: bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net,
	andrii@kernel.org, martin.lau@linux.dev, song@kernel.org,
	yhs@fb.com, john.fastabend@gmail.com, kpsingh@kernel.org,
	sdf@google.com, haoluo@google.com, jolsa@kernel.org,
	David Ahern <dsahern@gmail.com>, Jakub Kicinski <kuba@kernel.org>,
	Willem de Bruijn <willemb@google.com>,
	Jesper Dangaard Brouer <brouer@redhat.com>,
	Anatoly Burakov <anatoly.burakov@intel.com>,
	Alexander Lobakin <alexandr.lobakin@intel.com>,
	Magnus Karlsson <magnus.karlsson@gmail.com>,
	Maryam Tahhan <mtahhan@redhat.com>,
	xdp-hints@xdp-project.net, netdev@vger.kernel.org,
	Willem de Bruijn <willemdebruijn.kernel@gmail.com>,
	Alexei Starovoitov <alexei.starovoitov@gmail.com>,
	Simon Horman <simon.horman@corigine.com>,
	Tariq Toukan <tariqt@mellanox.com>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: [xdp-hints] Re: [RFC bpf-next 07/23] ice: Support RX hash XDP hint
Date: Thu, 14 Sep 2023 18:59:27 +0200	[thread overview]
Message-ID: <ZQM77955STTiYebm@lincoln> (raw)
In-Reply-To: <6bee72b4-dbfd-b62b-932f-8ca0da705994@intel.com>

On Thu, Sep 14, 2023 at 06:54:21PM +0200, Alexander Lobakin wrote:
> From: Larysa Zaremba <larysa.zaremba@intel.com>
> Date: Thu, 24 Aug 2023 21:26:46 +0200
> 
> > RX hash XDP hint requests both hash value and type.
> > Type is XDP-specific, so we need a separate way to map
> > these values to the hardware ptypes, so create a lookup table.
> > 
> > Instead of creating a new long list, reuse contents
> > of ice_decode_rx_desc_ptype[] through preprocessor.
> > 
> > Current hash type enum does not contain ICMP packet type,
> > but ice devices support it, so also add a new type into core code.
> > 
> > Then use previously refactored code and create a function
> > that allows XDP code to read RX hash.
> > 
> > Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com>
> 
> [...]
> 
> >  	/* unused entries */
> > -	[154 ... 1023] = { 0, 0, 0, 0, 0, 0, 0, 0, 0 }
> > +	[ICE_NUM_DEFINED_PTYPES ... 1023] = { 0, 0, 0, 0, 0, 0, 0, 0, 0 }
> >  };
> >  
> >  static inline struct ice_rx_ptype_decoded ice_decode_rx_desc_ptype(u16 ptype)
> > diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
> > index 463d9e5cbe05..b11cfaedb81c 100644
> > --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
> > +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
> > @@ -567,6 +567,79 @@ static int ice_xdp_rx_hw_ts(const struct xdp_md *ctx, u64 *ts_ns)
> >  	return 0;
> >  }
> >  
> > +/* Define a ptype index -> XDP hash type lookup table.
> > + * It uses the same ptype definitions as ice_decode_rx_desc_ptype[],
> > + * avoiding possible copy-paste errors.
> > + */
> > +#undef ICE_PTT
> > +#undef ICE_PTT_UNUSED_ENTRY
> > +
> > +#define ICE_PTT(PTYPE, OUTER_IP, OUTER_IP_VER, OUTER_FRAG, T, TE, TEF, I, PL)\
> > +	[PTYPE] = XDP_RSS_L3_##OUTER_IP_VER | XDP_RSS_L4_##I | XDP_RSS_TYPE_##PL
> > +
> > +#define ICE_PTT_UNUSED_ENTRY(PTYPE) [PTYPE] = 0
> > +
> > +/* A few supplementary definitions for when XDP hash types do not coincide
> > + * with what can be generated from ptype definitions
> > + * by means of preprocessor concatenation.
> > + */
> > +#define XDP_RSS_L3_NONE		XDP_RSS_TYPE_NONE
> > +#define XDP_RSS_L4_NONE		XDP_RSS_TYPE_NONE
> > +#define XDP_RSS_TYPE_PAY2	XDP_RSS_TYPE_L2
> > +#define XDP_RSS_TYPE_PAY3	XDP_RSS_TYPE_NONE
> > +#define XDP_RSS_TYPE_PAY4	XDP_RSS_L4
> > +
> > +static const enum xdp_rss_hash_type
> > +ice_ptype_to_xdp_hash[ICE_NUM_DEFINED_PTYPES] = {
> > +	ICE_PTYPES
> > +};
> 
> Is there a big win in performance with this 600-byte static table
> comparing to having several instructions which would do
> to_parsed_ptype() and then build a return enum according to its fields?
> I believe that would cost only several instructions. Not that it's a
> disaster to consume 600 more bytes of rodata, but still.
>

It is not disasterous either way, I have added this table after a discussion 
with team members and would like not to throw this away now.

> Alternatively, you can look at how parsed ptype is compressed to 16 bit
> in libie and use those saved bits to encode complete XDP RSS hash enum
> directly there, so that ice_ptype_lkup[] would have both parsed ptype
> and XDP hash return value :D
> 
> > +
> > +#undef XDP_RSS_L3_NONE
> > +#undef XDP_RSS_L4_NONE
> > +#undef XDP_RSS_TYPE_PAY2
> > +#undef XDP_RSS_TYPE_PAY3
> > +#undef XDP_RSS_TYPE_PAY4
> 
> [...]
> 
> Thanks,
> Olek

  reply	other threads:[~2023-09-14 17:05 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-24 19:26 [xdp-hints] [RFC bpf-next 00/23] XDP metadata via kfuncs for ice + mlx5 Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 01/23] ice: make RX hash reading code more reusable Larysa Zaremba
2023-09-04 14:37   ` [xdp-hints] " Maciej Fijalkowski
2023-09-06 12:23     ` Alexander Lobakin
2023-09-14 16:12   ` Alexander Lobakin
2023-09-14 16:15     ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 02/23] ice: make RX HW timestamp " Larysa Zaremba
2023-09-04 14:56   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 16:29     ` Larysa Zaremba
2023-09-05 15:22       ` Maciej Fijalkowski
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 03/23] ice: make RX checksum checking " Larysa Zaremba
2023-09-04 15:02   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 18:01     ` Larysa Zaremba
2023-09-05 15:37       ` Maciej Fijalkowski
2023-09-05 16:53         ` Larysa Zaremba
2023-09-05 17:44           ` Maciej Fijalkowski
2023-09-06  9:28             ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 04/23] ice: Make ptype internal to descriptor info processing Larysa Zaremba
2023-09-04 15:04   ` [xdp-hints] " Maciej Fijalkowski
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 05/23] ice: Introduce ice_xdp_buff Larysa Zaremba
2023-09-04 15:32   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 18:11     ` Larysa Zaremba
2023-09-05 17:53       ` Maciej Fijalkowski
2023-09-07 14:21         ` Larysa Zaremba
2023-09-07 16:33           ` Stanislav Fomichev
2023-09-07 16:42             ` Maciej Fijalkowski
2023-09-07 16:43               ` Maciej Fijalkowski
2023-09-13 15:40                 ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 06/23] ice: Support HW timestamp hint Larysa Zaremba
2023-09-04 15:38   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 18:12     ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 07/23] ice: Support RX hash XDP hint Larysa Zaremba
2023-09-05 15:42   ` [xdp-hints] " Maciej Fijalkowski
2023-09-05 17:09     ` Larysa Zaremba
2023-09-06 12:03     ` Alexander Lobakin
2023-09-14 16:54   ` Alexander Lobakin
2023-09-14 16:59     ` Larysa Zaremba [this message]
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 08/23] ice: Support XDP hints in AF_XDP ZC mode Larysa Zaremba
2023-09-04 15:42   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 18:14     ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 09/23] xdp: Add VLAN tag hint Larysa Zaremba
2023-09-14 16:18   ` [xdp-hints] " Alexander Lobakin
2023-09-14 16:21     ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 10/23] ice: Implement " Larysa Zaremba
2023-09-04 16:00   ` [xdp-hints] " Maciej Fijalkowski
2023-09-04 18:18     ` Larysa Zaremba
2023-09-14 16:25   ` Alexander Lobakin
2023-09-14 16:28     ` Larysa Zaremba
2023-09-14 16:38       ` Alexander Lobakin
2023-09-14 17:02         ` Larysa Zaremba
2023-09-18 14:07         ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 11/23] ice: use VLAN proto from ring packet context in skb path Larysa Zaremba
2023-09-14 16:30   ` [xdp-hints] " Alexander Lobakin
2023-09-14 16:30     ` Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 12/23] xdp: Add checksum hint Larysa Zaremba
2023-09-14 16:34   ` [xdp-hints] " Alexander Lobakin
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 13/23] ice: Implement " Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 14/23] selftests/bpf: Allow VLAN packets in xdp_hw_metadata Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 15/23] net, xdp: allow metadata > 32 Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 16/23] selftests/bpf: Add flags and new hints to xdp_hw_metadata Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 17/23] veth: Implement VLAN tag and checksum XDP hint Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 18/23] net: make vlan_get_tag() return -ENODATA instead of -EINVAL Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 19/23] selftests/bpf: Use AF_INET for TX in xdp_metadata Larysa Zaremba
2023-08-24 19:26 ` [xdp-hints] [RFC bpf-next 20/23] selftests/bpf: Check VLAN tag and proto " Larysa Zaremba
2023-08-24 19:27 ` [xdp-hints] [RFC bpf-next 21/23] selftests/bpf: check checksum state " Larysa Zaremba
2023-08-24 19:27 ` [xdp-hints] [RFC bpf-next 22/23] mlx5: implement VLAN tag XDP hint Larysa Zaremba
2023-08-24 19:27 ` [xdp-hints] [RFC bpf-next 23/23] mlx5: implement RX checksum " Larysa Zaremba
2023-08-31 14:50 ` [xdp-hints] Re: [RFC bpf-next 00/23] XDP metadata via kfuncs for ice + mlx5 Larysa Zaremba
2023-09-04 16:06 ` Maciej Fijalkowski
2023-09-06 14:09   ` Larysa Zaremba

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.xdp-project.net/postorius/lists/xdp-hints.xdp-project.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ZQM77955STTiYebm@lincoln \
    --to=larysa.zaremba@intel.com \
    --cc=aleksander.lobakin@intel.com \
    --cc=alexandr.lobakin@intel.com \
    --cc=alexei.starovoitov@gmail.com \
    --cc=anatoly.burakov@intel.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=brouer@redhat.com \
    --cc=daniel@iogearbox.net \
    --cc=dsahern@gmail.com \
    --cc=haoluo@google.com \
    --cc=john.fastabend@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=kpsingh@kernel.org \
    --cc=kuba@kernel.org \
    --cc=magnus.karlsson@gmail.com \
    --cc=martin.lau@linux.dev \
    --cc=mtahhan@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=saeedm@mellanox.com \
    --cc=sdf@google.com \
    --cc=simon.horman@corigine.com \
    --cc=song@kernel.org \
    --cc=tariqt@mellanox.com \
    --cc=willemb@google.com \
    --cc=willemdebruijn.kernel@gmail.com \
    --cc=xdp-hints@xdp-project.net \
    --cc=yhs@fb.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox